Sign In Sign-Up

Welcome!

Close

Would you like to make this site your homepage? It's fast and easy...

Yes, Please make this my home page!

No Thanks

  Don't show this to me again.

Close

The following are repair briefs for various 
equipment.The infomation is directed
to technically competant repair engineers.Generic terms have 
been used to make this info less model specific,eg terms like 
replace transistor Q123 have not been used.
The equipment is Test Gear, Industrial and Boating gear, Domestic and miscellany , computer. 
I would be interested in finding 
any other repair listings on the internet structured as i have 
done ie intended to be less model specific. For convenience using search-engines, 
use keyword divdevrep to target these files.

There is no point in contacting me about any of the following, the 
repair job may have been done 15 years ago .
I cannot clarify or enlarge on any of the following. Mains is 240V, UK.


Should the location of this file change please use the keyword "divdev7" in 
a search engine to find it again

Test Gear

8802AJ Wireless Inspection Camera 
no name maker, but like this one
http://kgbcameras.co.uk/cameras/0Brand_1_19_1226538224.jpg
I cannot get the 2.4GHz link working, works wired in , clipped together.
The camera is outputting syncs only , rather than CVBS is the main problem,
anyone know if there is a secret way in ? I suspect as supposed to be
waterproof , then a matter of hacking in.
It would be much more useful with a clipped on close-up lens, for reading
script, surprisingly not supplied
Convinced myself hacking-in only.
I hacked in along the mould line and accessed the inside without further
damage. 22 way bridge between CCD plus 2x LED and rest of pcb, maybe
fudgeable for miniture 7x7x10 lens and ccd . Whole ,with current pcb, could
be easily cut down to 12x 7mm access port instead of 16mm original.  Anyone
knows of a simple "elephant trunk" mechanical manipulator mechanism? very
stiff and unweildy spiral steel trunking umbilical as is. Nicking the mechanism 
from a remotely variable car rear view mirror mechanism and miniaturising 
it would probably work, extending the bowden cables.
uses wbcam type lens and CCD and Vimicro is 48 pin , 15 thou/mil spacing
VC0703NLEB
cant get any closer than
http://www.vimicro.com/english/product/d_vc0703.htm#
also 8 pin u724
6M xtal
op should be CVBS with 3.3V supply
From this odd named file
http://www.sh2a.com/Microsoft%20Word%20-%202A-cmos.pdf
and pins 1,2,3,4, 7 tied to Vss and p8 Vdd , then the 8 pinner marked
U274
16B-1
probably an EEPROM but why ? NTSC/PAL selection would surely be just taking
some pin high or low on the 48 pinner. Contrast and brightness carried i2c
to the front end? will check later today
there is only 4 wires between this camera pcb , down the 1m long spiral
trunking to the monitor etc, 0,3.3V, illumination LED supply (variable), and
CVBS signal out.
Repaired another of thse endoscope things, just mechanical damage to the
metal barrel surrounded "S video" connectors used as small polarised 4 way
connections. The barrels are screwed onto the trunking and this one undone but 
wires had not broken (not worked out how to get into the 
barrels , probably hacking in as likely glued). "Unwind" the barrel , tightening up the wires 
until you judge to be too tight and then engage the spiral 
and wind on, finishing with protected jaw mole grips. 
Sticking a 20D/ 50 mm focal length lens over the lens cover
means you can focus down to a conventional 16 pin DIL package to read
script. Image is then linear x2 so fills the height of the LCD screen.
If you were to cut away the waterproof  lens cover and picked at the weak
glue around the LEDs and lens carrier you could unscrew it , web cam
fashion, to focus closer.
5.7R between Vdd1 and Vdd2 
CCD pcb board labelled GB8802C08
"melt" glue join to pcb to remove the lens mount.
Passing thin wire under the CCD probably 4x5 BGA
Anyway given up on this one , come to conclusion half the vimicro is kaput.
Annoyingly about 40 MHz clock and 9 output lines from the CCD work with
plenty of signal variation to content of view variation. But CVBS output is
corrupt , absolutely no image content. There are 3 sync pulses and 2 colour
bursts per line interval. Tried removing EEPROM as no clock burst , just
apparently matching L to H with 2 or 3 negative pulses only, then staying
high, on (presumably) SDA and SCK pins. Removed EEPROM and same signal on
vimicro pins , but then no o/p signal at all. Replaced with an as found
24C04 and same nothing out.
Probably too much of a workup to descramble those 9 lines to give even just
a monochrome image.
4 way 3.5mm , first and third ring supply, video on tip ?
Monitor on/off sw not hard wired
2.36V ADD1 on 5150
1.8V Vdd, 3.3V for Vi-o, AvDD p4 and p32
DVdd p20
D-gnd 19, A-gnd p3,p7
topmark SS14 schottke diode?
A1sHB = FET? (s is s with overscore)
Apparently not in the manual.
For 2.4GHz , remote viewing. All has to be connected up and operational 
before removing the monitor from the "hot shoe", maybe first time of use 
only - initialisation ?

ABI DDS 40XP, CMOS/TTL etc tester / Coutant RL 150/100 SMPS, 1988
Went to use it this week , first time for ages and not a dickybird, no magic
smoke/ ominous noises. Just  single cycle slow repeat ping/clicks from the
inverter transformer. Was working fine last time of use.
Anyone happen to have schematic of the ABI DDS-40XP Digital Diagnostic
System or more pressingly the SMPS, I only have the ABI user manual
I may be able to get away with 2 SMPS replacement fitted in there, the
Coutant (maybe type number B91510 ) from 1988 uses a lot of early days SM ,
seems very high power, I suppose for the situation if all 118 relays are
active at the same time plus CRT plus floppy drive.
-12V,1A
+12V, 3A
+5V,5A
and a separate +5V,12A
I'll have to confirm this ps does not require a load to start up, unlikely,
but not done so yet , no obvious fuses on main board .
So far cold checked the smps and nothing amiss on Rs,Cs or large active
Uses UC3842 supervisor, plenty of data, so should be able to confirm the
oscillator, with NE555, part of things easy enough.
R14 looks like wirewound R, but 1M I though it read 1R4 or 1K4 and I 
was measuring my skin conductivity. 
R21 ==0.3R measured, R3 180R
The IRF840 HS remove easily for checking both devices
opto isolator H11AV2A checks out with double DVM diode 
test about 0.05V variation. 
I put sone w/w Rs on the outputs and monitored voltages, a few tenths of
volts + on + and - on -, except one of the 5V, not even a mV, so perhaps a
duff C on board of SM that is obscured by heatsinks etc. Just with the 10 or
so pings a second of drive, that drops back to 1 a second with half mains,
the voltage supply over the 3842 stays constant at 13.6V so supplying itself
, not just startup supply. 
0.6V over C17 1000uF,25V but after that is an off pass 
tr SM2176A
Extended out the main transformer, soldering on sleeved 
wires on LV and HV sides and bending up on themselves 
so could solder on the underside of the board for access.
SMPS that as it stands
is in tick-tick mode, 1 ultrasonic cycle per second.
Putting a dropped LED on each output shows life on all rails.
I expected isolating the LED of the optoisolator would 
defeat the tick-tick but no difference.
So far confirmed the UC3842 functions properly on a bench 13.5V supply with
original 2nF and also27nF // to it. 
Requiring supply to 555 connected at the hot side of the optoisolator.
But UC gets overriden by a protect sub-cct
on the hot oscillator side. Also confirmed that the 2 switching powerfets are
conducting balanced( 2 sets of 500V diode+10nF 400Vcap+neon+100R wired between 
S-D and D-S of the powerfets ) Problem seems to be a goodness of oscillation monitor
on the hot side of the transformer that rectifies some of the drive and if
not balanced or large enough starts the tick-tick mode. Rectified it must
exceed 15V set by a ZDP15 or this mode cuts in , it would seem. Only
reaching 4.9V. But in the process of checking this area a small electo now
has an ESR of 200R wheras checking before any power on it must have been sub
2R or so sitting around idle. 1988 ps with a load of early SM and this cap
of course in an awkward section to replace rather than test.
One tool I don't have in my toolkit - a sniffometer. Nothing smelled while
in use or close handling of this power supply. But on removing the electros
, that distinctive fishy smell of leaked elecrolyte, with soldering iron
temperature on the pins.
Tiny space available to fit some capacitance so they used 2x squat format
22uF,35V in parallel. One leg pulled out of one cap on desoldering and the
other a telltale drop of brown gloop immediately under the cap but not
enough to be visible over the pcb.
Both bright green nichicon 105 deg C. Found space on the hot side for
replacement and powered up long enough at about 3KHz instead of 40KHz to
give full V on all o/p ( unloaded) with very audible 3KHz coming off the
transformer. As , presumably, the pair of switching power fets were
operating at least partially linearly at an undesigned 3 KHz they heated up
quite quickly. 15 V over that sensing zener . Removed the 27nF and now
operating correctly with substantial test loading on all rails.
One mystery remains , the original tick-tick was about 1 a second, on adding
the 27nF to the 555 timing capacitor the tick-tick rate increased to about
10 a second. I've not traced out the path from the "goodness of oscillation"
circuit to the 555 but what sort of intermediary cct would lead to an
increase in that tick-tick rate ?
There is a lot more SM around the UC and 555 than any of the datasheets I've
seen. There are at least 1 diode for changing the duty cycle of the 555
waveform. But increasing the single timing cap from 2 to 29nF upped the
error mode frequency. Perhaps the duty cycle changed so much that it
inverted and played havoc with timings or limits in the UC. I will use x5
rather than x10 next time . Incidently slowing things down so less likely to
be destructive to drivers if the problem is shorted turns in the
transformer.
Managed to replace the transformer and 2 daughter boards , omitting 
to solder the offset pair of pins on the opto hot side, 
be sensitive to touch and mains current draw to 1 amp 
and rail voltages go haywire. to 
5.1V, -14,14,14V (not 5V)ich 
blew , replaced with TIC126 but 2.5V Vgt may be wrong.
Some ps details one h/s had D15,TIP32,D5,D5,2176
Topcode SM a1 pnp, 68 pnp, ?6 diode, AG diode, /17 ? diode, large 
footprint SM marked AB
No load readings on the final output daughter board header pins
0,0,4.66,10.3,5.1, 0,0,-13.4,10,0,0,13.8,11.1,0,0,0,
5,5,5,5,0,10.9,0,10
inside th eABI
Why is NS +12V regulator marked LM2940CT -12 ?
not printed as LM2940CT /12 say.
Any other awkward ones like that, hopefully to lodge in memory, or just a NS
thing.
Not being familiar with that range the 9 in the type number triggers a
mental connection with 79 series negative regs
I later found a supplier listing of V regs by , indexed by maker name. Only
NS out of 6 makers adopted, still adopts, the use of the minus sign in some
positive device type numbers. The one that confused me is the wrap around
situation , but I still don't see why they could not have used underscore_
or slash/
5V, 12A supply is for the ABI, all those relays if on all together 
and 5A rail for banana sockets supply to any external boards under test.
Set the ps +/-12V rails to +/-13.5V for dropout to these extra regs
240V mains current draw , sitting idle with CRT at readable level 
160mA. 

Converting Avo Model 7 to accept Avo Model 8 probe leads.
ie convert hook connectors to 4mm pins
Unscrew the screw terminals. Find length of 2BA 
brass stud or length of brass screw and the central 
brass piece of a 4mm chassis socket. Grind part of the 
2BA thread so 4mm barrel section will fit on, Then 
high heat (hot air gun perhaps) solder the 2 
bits together. Screw in to Avo and push a couple of 
grommets onto the exposed 4mm socket parts.

Avo B183, RLC meter, 1984
2 stand sc 
4 corner sc, remove fuse
smaller board pulls out 
2 headers 
555 timer C6 R21 1 percent
C5 original 485pF 1%
Display board remove 4sc , mark, replace LCD cover 
and its 2 sc.
To convert to about 10KHz ESR meter , using R scale.
Changed C5 to 27pF
remember to replace the fuse for checking function
R27 ,D17 to IC22 p21 I/P for feedback path
p2 of 10W header to IC21 p1+3 ,4066 o/p via 1M to 40 pinner
With 485pF original, 1.8K across Hi-Lo inputs measure 
1.807K on display and 56.6mV on p21 of the ICL 40 pinner
with 1.8K on probes, 60uS 0.04V on C5 485pf,
scope ground to batt negative.
With 27pF instead of 485pF and 1.8K then reads 
-0.327K on R scale, over C5 8uS and 10KHz and 0.1V pk-pk on the probes with 1.8K there. 1.5V 10KHz with no loading, so for 
general ESR use add 2x Ge diodes, crossed, over the input
For "ESR" version 18R on probes, display shows -3.63R
known bad 1uF electrolytic measured -927R and 
18.8K on original 1KHz version
With crossed diodes
-0.318K for 1.8K
-3.36 for 18R


Avo CT160 valve tester
Needle sticking. It is surprisingly easy to remove the 
meter on these once removed the 4 base bolts and 
opened up and propped up in the lid.
Slackening the 2 bolts and studs and nuts that 
act as fixings and terminals. If you overtighten 
these fixings it is enough to distort the meter housing 
and disturb the delicate movement.
To reset the relay manually just flip it 
back once inside, clean the contacts 
at the same time. The mains primary is about 
46 ohms through these contacts for UK 240V. 
The 210V/200V bulb should illuminate 
if this relay is activated but someone had 
disconnected this one.
Maybe of use to someone this meter 
movement showed 50% with 14.76 volt 
dropped through a 920K resistor ( for 
checking meter movement without mains 
voltage around the place).
If the meter cannot be set 
exactly on the tilde setting/ mains voltage 
heater continuity especially if changing the 
EB91/D77/CV140/6AL5 rectifiers, reset the 
large preset on the rectifier board 
between 2.34K resistor and rectifier.
http://www.vintage-radio.net/forum/attachment.php?attachmentid=37962
and have left a note to myself, the next time I'm inside this Avo valve
tester, to modify the meter protection.
EL34 test that i did not notice 2 bright ie burnt patches on the metalwork. 
Bright white flash from the valve and blown left fuse and internal relay delatched. 
Next time , if burns are seen, must try with low heater setting and low anode volts firstly.
Started undoing the "feet" from the wrong side and nuts drop away. To get to the 
socket/thumbwheel sw board, remove 8 edge sc and 2 outside label fixing sc . 
With a screw driver angled in through the mains plug stowage "socket" earth hole 
lever up the panel from the perished / stuck gasket. 
270R measured between bulb wiring and mains inlet. Tried a 240V , 10w oven bulb 
and worked well enough, a 240V 5W would probably be brighter. 
Remove one fixing bolt for the bulb holder to allow the holder to swing out 
to get to the clamping screw, With a bit more dsolder added to the screw brass solder blob, 
fixed into the clamp , a wire soldered to the pip of the bulb and soldered to 
the pins on the exposed side of the paxolin. No lamp on in normal use, comes on and pilot light 
goes out  if 
you manually try opening the points via insulation, and fuses are good. 
Added , on the midway solder tags, a 1N007 over the meter , with the meter voltage , and 4 AAY30 Ge and 1N4007 all in parallel 
and in reverse over the meter to protect when it tries to go negative.

Avo M2007 DVM
Plastic must stiffen or bind or something and not too obvious 
coming back to it years later , 
how to get to the battery compartment. Lift out the prop-up leg 
and push a rud in the hole next to the serial number, while pushing 
at the prop hinge towards the top edge

Avo Valve Tester , the first one about 1936
Just the main meter section , no switchbank and valve base 
section. Someone wanted for use rather than original condition. 
Made up an outlet plug from some QM socket pins used 
in reverse as the right diameter, using the siolder bucket end 
as connector and soldering to the other end.
But unknown condition checked out cold as best 
as possible. Only one active device 
in this unit a Copper oxide rectifier
http://home.graffiti.net/diverse:graffiti.net/CuO_rec.jpg
steel bolt mount is horizontal.
Not for anode supply but for signal to the meter 
(only 6 discs so limited to 6x 6V reverse voltage.
Slow variac powering up and not much happening 
other than coming on (even that iffy due to the buried 
switch contact uner the first transformer, for continuity 
probes when connected through the top pannel).
If the mA/V control was placed in 1/3 of the scale 
then needle went heavily negative. 
Broken join in the 2 decade log , wire-wound, pot 
track. Removed from circuit measured 14.4R on heavy 
gauge wire and 90R overall. Break at the doubly weak 
point where there is disjuncture between the 2 gauges 
and also the log law change of depth taper is a jump 
there. Easy to undo, remove , rejoin 1 turn less , 
down tight , lacquer in and later grind down to remove 
any jump in height.
Back in working order but C/H leakage was a factor 
of 3 over s, simulating by adding Rs to known good valve.
May not be correct , and no obvious index mark for the 
outlet socket but going clockwise, viewed externally, from the centre pin of the 
narrow spaced pins.
Heater (low?) , heater, D1, D2, A2, A1, G, Screen, K
At 95 percent mains (with broken mA/V pot) and 80, Normal, 
10, 75 Screen, 6 heater, mid set zero ac voltages on these outlet pins
5.9,15.2,15,15.7,96,6.9,67, 5.9 relative to the "index" pin
Some R vales , black rod with purple band 74.7R
barrel Rs 115K, 1.5K, 400K to neon
Meterread 31R on dvm with 16.2mV on tips and 
5.8 on the meter scale
With repaired unit unpowered and settings of anode 150V, S 100V
heater 6V, normal, set zero at 1/2, set mA/V at 5
set mA/V pot across ends 31.6R
nearby W/W single 6.8R and nearby 
double 8.6R , 1.2R
CuO measure 294R one way round and 699 the other
46R w/w next to the rectifier
set zero pots 11.8R, 238R
24.2R across meter
Instuction of use pages in the front of  1957 Avo valve data manual.
Start with Set Zero turned fully clockwise and mA/V 
to full anticlocwise. Set mA/V scale with pointer at the figure 
in the AVT data column , and use the 50 mark "mA/V" when testing diodes.
Some useful info from a G8MNY
http://theskywaves.demon.co.uk/technical/Testgear/Early_AVO_Valve_tester.pdf

Avometer Avo model 8 Mark 5
Low or absent reading on all scales plus on AC ranges high variability with frequency.
New owner so unknown abuses. The select/cut on calibration etched 18.4 ohm resitor in ohm x 1 
range burnt out so replaced with mixture of wirewounds and low wattage. The 270 ohm 
resistor in the 3V AC line was charred to over 1K. One of the Germanium pair 
of rectifier diodes was o/c leading to the f susceptability.
Someone had hacked into the meter movement area (broken covering) 
probably to sort out false tripping cut out. Something was wrong with the movement 
100mV over the meter would give about 90 per cent fsd.
Whether loss of magnetism,deformed return spring don't know reason - the coil 
was not obviosly overheated (shorted turns). Put a 18 turn preset in parallel 
with the small internal dropper in series with the meter (mounted on reversing switch) to 
bring up the movement to 100 percent.
Notes - the basic idea is the chain of "current" resistors adds up to 8000 ohm accurate to 
the last figure,this in parallel with the 2677 ohm of the meter gives potential divider of 1/30 
in the 3V range with the 58K resistor .
With the meter in damped setting (shorted meter) foreward and "reverse" over each 
Ge rectifier was .95V and .29V on a DVM.
DC resistance of the AC divider inductances (transformer-like device) were
<0.2 ohm,0.7,40.5,873,1592 and ,<0.1 ohm.

Bruel & Kjaer 8306 accelerometer
Beyond repair, probably dropped and shattered most of the 
piezo elements at the core.
Probably originally 8 off 5/16 by 3/64 inch anulus discs 
in series along the stud.
Probably assembled by tightening up the internal cylinder 
with one ground/polished face 
that stud goes in one end and Allen bolt in the other. 
To undo, undo black covering and remove the pcb. Then stop the block from rotating and 
undo the allen bolt relative to that block and base with connectors
Tightened with loctite on stud thread to some predetermined 
signal output, then heavy block put in place and allen bolt tightened and varnished in place.
Just the number 4289 ,distinctive shape, in plan view, like a ship with a
blunt end and a sharp end, 4.5x2x2mm, black body 3 in line gold plated legs,
3 of these trannies used probably 2N4289 plus an E102 FET .

Crotech 3036 scope
Maximum intensity only
9x200V zeners gone high leading
to EHT of -2000V instead of -1800V
leading to punch through of 2kV capacitor and 
destruction of active components.
Replace all and uprate electrolytic across 100V  zener

Crotech 3132 scope
R422, 1kohm in 180V ps was broken.

Dawe 1405D sound pressure level meter, probably dropped
Sifam meter 1.45V, 660R 147843
Coil is fine. This is almost hair-fine
phosphor-bronze ribbon suspension , not hairspring and jewels. A ribbon, top
and bottom, holds the coil frame and passes current. Each ribbon is soldered
to the frame then passes through a hole in the offset-rotatable anchor ,
passes over the rounded edge of  an open end of a U-shape of manageable size
phosphor bronze spring , and soldered along the arm of that spring. The
solder point to the frame has failed from a jar / knock. So both U springs
are now opened to a V in comparison. I have tried solder to the end of the
fine ribbon and the stub on the frame, both take solder fine but not joined
up yet. There is enough length of fine ribbon on the broken side to resolder
to the coil frame , pass thru the hole and anchor off , on its u-spring
anchor farther along than originally.
But how to bring the open-U springs back to proper U before soldering the
anchor point. A matter of making a jig to compress or some sort of external
spring over them ? And of course I've not seen a functioning movement. I'm
assuming setting up as U and then opening out a bit with the jigs removed,
or would the jig need to compress to more than U and then open out to U with
jigs removed, normal use.
Pushing down on the spring of the good end and letting the coil frame drop,
with gravity,  it seems to be centrally aligned over the pole piece, when
the spring is a proper U viewed  on side.
Where would one find , even robbing, short lengths of such rbbon if it was
required?
Certainly a job fully under a x4 or so illuminated inspection lens. This
Sifam one you can rotate suspension anchors at either end , but there is no
user accessible adjustment of zero once the meter scale is set in the
housing.
I think the sensible route would be . A loop of copper wire around each U
spring and twisted ends until parallel spring arms. Offer up the ribbon and
solder up and then while a finger over each spring, in turn, cut the copper
wires and gently release finger pressure. I'd rather have too little tension
in the U springs and so the fine ribbon, than too tight. 
One wedge from a mini clothes peg pushed in the gap between cylinder core 
and coil frame on the broken side. Just the right allowance to solder 
replacement ribbon in place and then when wedge released the frame was central 
on balanced springs. 
Grub screw of  att. knob had threadlock, required 25 seconds of a a soldering 
iron tip , due to thick brass ring in the knob, heatsinking. 
Remove top cover to remove the original battery holder to replace with 2x pp3.
Replacement phosphor bronze robbed from the hair spring of another meter. 
Could not sensibly measure the thickness of  the original but width of the robbed about .4mm 
wide and about 0.03mm thick and original about .2mm wide. Retain as whole spring to make it easier holding ,placing 
and soldering and then cut off the surplus. Coil 673 ohm
Just for my own purposes so relative rather than absolute but if possible to
calibrate then all the better. No access to a calibrated one to
cross-calibrate, falling ball calibrator etc.
An old Dawe 1405D that sems to have little info out there. Mic and
electronics works and I've now repaired the suspension on the meter but  fsd
is now .586V but is marked as 1.45V .
As the replacement phosphor bronze is probably more stiff than the original
, if any thing , I would expect greater fsd . It may be .45V marked and a
type number 1 or something ahead of the .45, there is quite a gap.
Anyone experinece of the ipod app, toy? or worth borrowing someone's ipod
and obtaining the app if any absolute use for cross-calibration
Other than the B battery test mark on the scale for running off the 1960s
B122
type 22.5V battery . But of course no standard for what the B equates to for
light load and less than 22.5V , seems to work adequately on 15V , I intend
using 2xPP3 , 18V
Looks as though Dawe bought in these Sifam ribbon suspension meters ,  the
original label saying 1.45V and mention of an external pcb. I suspect Dawe
removed that pcb and the movement itself is 100mV. Battery test is simple
chain of resistors , protect diode and rest switched out. So if 100mV fsd
then for a new B122 battery 100 percent fsd would correxpond to 24.5V and
the B marker correspond to 15.5V which seems reasonable. As the mic is
capacitive , about 300 to 3KHz  3dB bandwidth , not electret, not worth
going to any effort calibrating this SPL.
B mark is 64 degrees of 90 degree arc of meter swing, 71 percent fsd
It is very basic one probably intended for schools use, no A or C wightings
just 40 to 120 dB att sw and slow/fast response sw (470uF across the meter)
and battery sw, but there is a calibration pot externally accessible.
I will use the calibrator preset via ps variation and the B mark to reset
the meter to deemed 15.5V and 24.5V B and 100 percent of the meter
I cannot make sense of the B indicator , the
resistor chain suggests 100mV fsd but reassembled and largely functioning
meter fsd is more of order 1V.
2 db ranges 60 and 70dB  are stuck together was due to my signal injection prod 
at the microphone input pin bridging 2 tags . But for good posistions
switching between ranges for differing 1KHz  sine signal in, consistent "0"
dB and +6.7 dB switching betweeen adjascent ranges. Don't know if that 6.7
instead of 10 is due to sine rather than noise source, will have to try
again with an attenuateable noise signal
Most speaker cabinets are rated for so many dBSPL with one watt input, at one
meter. Set up outdoors to approximate anarchoic and 
use a pink noise source (constant energy per octave) .wav files
seem to be on the web.
I can see that being as accurate as a calibrator that hasn't been calibrated
for 20 years say or manufacturing variabilities of ipod microphones.
Assuming the chosen speaker magnet does not decrease in power too much over
time.
30V ps 120K and 2mA for "0" dB about 40 percent fsd
24.1V 27K 10dB
17.3V 27K battery at 7 dB
6.1V , 27K, 0 dB
Beware poor contact with Al spacers on the meter terminals
with 18.1V ps after changing Q2 as B-E intermittant failure
Electronics Q1 marked DAW1 perhaps, 4 pin , something like an AF1 / 5.8, 0, .54V?
Q2 BCY 71 /17,16.4, 10.7 V dc
Q3 BC109 /8.8,9.1,16.8
Q4 BCY71 / 17.5, 16.8 , 8.8
Q5 BC108/ 8.2,8.8,17.5
Q6 BC478 /8.2,7.6,0
with af sig gen over mic , 1KHz
90 dB att, 24mV ac, 11mV over meter
100dB 97mV ac, 137mV dc
110dB 251mV ac, 112mV dc
on 80dB , 10mV in and varying f
100 Hz 38mV
300 100mV
500 120
1K 128mV
3K 99
5K 66mV
10K 21mV
Attenuator ohms measured at red/blue wires and then yellow to gnd
40dB 3R, 1.2M
50, 6.83K, 1.2M
60 , 9.04, 1.2M
70 , 9.04, 380K
80, 9.04, 120K
90, 9.04, 39.3K
100, ", 12/56K
110, ", 3/94
120 ,", 1.21K
To get consistent 0/+10 readings on changing att , through the ranges with sine generator. 
For the 680R on the meter rectifier board added 100R and 2K preset in parallel and adjusted 
for best fit.
For battery test added 1K over the 100K and adjusted the preset to give appropriate reading, arbitrary
Finally measured .64V fsd ,ie +10 dB




Delta GT7810A ultrasonic bath, 2010
burrowed inside out of curiosity
3 screws under plus 4 under feet, one is Torx
E141940 FR-1 ZD-90F 94V-0 130 deg C board
no HV electros
2W 100, 27K, 4x 22, 1R, 220K
1W 1R, 47R, 1M

ESM1 luxmeter from 1985, badged "British Gas" no info found out there. Sort of things carried around by environmental health people for non critical measurement, combined Lux and contact thermometer.
States 0 to 3000 Lux on the back of the instrument (for inside rooms) , uses ICL7106 and 4.5 digit LCD with last 
0 permanently on. So can read up to 20,000 lux, but as only stated to 3000 , is it only linear in that range. 
Has 4 presets , presumably 2 for lux range. VN10KM off/on FET, 2x 4066 , 4030, LM358
Tried it this morning about 8am when there was 10 minutes of sun,
holding so sun axial to the PBW21 sensor, well over range.
With a 4 stop plus 3 stop neutral density filter , ie /(16x8) ,
7660 lux, with 4 stop only  ie /16 measured 15380 and with 3 stop only over range again.
So next to useless above 3000, or is it to be expected and could be used with preliminary cross-calibration source and 2/3/4 stop filters with a good stable bright light source? 
the BPW21 datasheet shows log-log plot of short circuit current v lux over 6 decades as linear, so something about short circuit compared with actual use? 
 short circiut current is very linear.  You'd have to share the circuit.  But if they just use a simple resistor across the diode to convert I to V, then it becomes non-linear as the diode becomes forward biased and some of the current goes back though the diode and not the resistor.
No known schematic but I'll dig out the data for the main chip and work out what applies to it and lux measurement, which should show which pair of presets and which is zero and which is range. 
The photodiode is in series with 18K and one end of those 2 is connected to an opamp input,p2  
 and the other end to another opamp input p6. So cannot be fully linear. 
p32 and 35 of ICL connected
p31 to 1M R6
p32 18K & RV3
p36 to R11then RV2 and IC3
Decided RV1 is temp scale, RV4 opto scale
I.m reasonably confident which preset relates to which function.
Overal zero and overall scaling, then one for scaling the temperature function and one for the lux. The extra CMOS for OR driving the always on 0 digit and the switching function and maybe auto-off function. As it stands with zero lux it reads -80 lux so something obviously adrift. Adjusting light levels up from zero , then from -20 to probably +20 it auto "zeros" to 0. The temperature scale is easy to check calibration, which leaves the lux scale.
Looks as though the nearest i will get to a reference light source is taking the average of a number of 60watt mains tungsten lamps, but then  www sources say that is betwwen 700 lumens to 900 lumens, mostly about 840 to 870, and monitor 1 metre away. Accept that a reading of 20,000 is nearer 100,000 lux and produce a corelation chart via some unknown but constant value of bright source and 1,2,3, and 4 stop (1/2 to 1/16) neutral density filters over the sensor . 
Changed the socketed LM358, changed the photodiode, perhaps something wrong with the 4066 for switching function feeds as all that is left active to change. The main chip was within 1 deg C reading , 20 to 100deg C on the temp function, but lux a good 1/10 of where it should be.
Ended up changing 3 presets completely from what they were to get the readings to agree with a 60W lamp and c850 lux, 1m from it, including the overal ranging preset which obviously messed up the temp function.
The room temp reads 80 deg C with minimum of its obvious preset, oh well , it was the lux I was after.
Could a 4066 go awry over 30 years, consistent on-resistance changing value on one switch and not another ? 
Ignored thermal function error and set the 3 RV with a new CFL lamp marked 850 
Lumen, 1 metre for average reading of 850 Lux
With Insolation at a weather station 10 miles away on clear clouless sky day, 
and at the time measuring 860W/m2 (on the horizontal)
Set the luxmeter horizontal and with 3 neutral density filters of 2 stops, 3 stops and 4 stops 
took measurements to make a correlation chart back to the assumed linear 0 to 3000 lux region
all 3 filters, 2,100 lux
4 stops 8,000 lux
3 stops 11,100 lux
2 stops 19,000 lux
Black silicone rubber 0.85 mm thick , 140 lux horizontal(300 lux pointed at the sun) maybe not neutral density, 
yellow caste

Farnell LFM2 Audio Generator
Short lived function after switch on
and change of multiplier.
Replace the x100k switch

Feedback FG600 
The remnants of a Fedback FG600 England made function generator of 1975.
Someone clipped the legs of one of those canned devices of the time leaving
10 longish pins in the board. Overlay is a circle with a tab. He wrote
RC4195 and RC4185 on pcb. Now a Raytheon RC4195 looks right there, a dual
tracking +/-15V regulator but what was an RC4185 I cannot find any reference
to it , but it may have been the original device and he intended replacing
with a 4195.
The drilling pattern was 8 like standard DIL and one between 1 and 8 and one
between 4 and 5 but of course may have been bent to that form , from a
circular TO96 type package.
7815 + 7915 will go in there firstly for proving purposes in all likelihood
I've looked in my contemporaneous sources and nothing
He actually wrote , very clear script,
RC4185NB
RC4195NB
which are valid suffixes for Raytheon
Although footprint is 10 pins , only 9 have traces to them, but wire size
legs rather than the TO66 package pins and overlay would suggest TO66. I
wonder if the 4185 was not-fixed +/-V.
My old Raytheon book seems to suggest they used TK suffix for TO99 and 
TO66
Another IC he'd marked on the board , but still in place is a Plessey multi
transistor pack
It is that chalky print of the time and wrote
SL304??    5C?
With good slant lighting , perhaps decades on I could confirm SL3045C
But RC4195NB is not a 10 pin pack if the 5th character was indistinct

Fluke 77
No function.
About 1/3 of pins on major ic had disconnected
due to a heavy object falling on the front face 
of the DVM.
Fluke 77 DVM
Otherwise trusty meter, whenever I try on a Weston Cell it is always the same voltage registered. But now on the ohms scale will sometimes read 10 or 12 meg with no probes in place. Good battery.
Something to do with the screen shielding? Without any metal covering, ie the top cover removed and no replacement for the now missing sticky backed foil over the shielded tinplate box, the meter will not pass POST, all segments lit on display permanent and no beep on changing rotary sw setting, normal ?.
So start with cover in place , turn to ohms, and remove cover , it goes to the floating 10s of megs. Twizzle sticking does not show suspect switch , preset or anything, and gradually drifts to over-scale, >30M.
Something going leaky with age? 
Since the last time I looked Fluke has placed the SM and schematic out there
http://transmille.net/ProCal/Procedure%20Library/Fluke%20Digital%20Multimeter%2023%20MKII%20%5B1.10%5D/Technical%20Data/Fluke%2070%20Series%20MKII%20(PN%20896204%20July%201991%20Rev%201,%2009-92)%20Service%20Manual.pdf
The uk has been permanently damp for about 2 months now. Perhaps something has gone hydroscopic. Brushed meths over both sides of the lower part of the pcb and allowed to dry off. Much the same 10s of megs.
When powered up and open I played a distant hot air gun over the dvm, held in hand at the time so only hair drier temp and the o/c ohms suddenly dropped to 10s of K . Let it cool down and gradually went to >33M and repeated switching on/off the last few times has stayed that way.
Reminded of a batch of Marshall amp pcbs that had contamination inside them combined with damp perhaps and marked increase in conductivity with relatively slight warming of the pcb, but HV there in that case 
When the dvm was in errant mode , saying say 15Meg, stressing rotary sw wafer or sockets etc with a plastic pen barrel made no marked change in the reading, wavering much the same +/- 10K or so, long term wavering 10M to 20Meg, whether twizzling or not. Will see tomorrow if the ohms scale opens at 10s of megs rather than >33 megs 
I've had that failure before on another Fluke and the symptom then is that readings are about 1 percent of what they should be, no wavering.
Another rain-soaked night in the UK, but trying out this 77, shows >33meg for open probes and Weston Cell reading 1.017V (on DCV scale) as it always has been over 20 years and using the same cell 
well over a month on of daily use and no return to wavering digits 

Fluke 77
Battery icon showing for ages , then bad readings with 9V battery , o/c reads 6V so very forgiving of low battery

Fluke 77 dvm
Functioning but displaying about 1  percent of true values.
Large bodied 1K ohm fuseable resistor near I/P o/c.

Gould OS255 scope
Random false triggering, coincident with mains borne spikes.
Someone had removed the power cord earthing
to the case.Also backlash on mode selector
switch due to worn nylon pot shaft extender.-
partially fill the worn "socket" with hot melt
glue and push in a metal "D" section pot shaft
to remould extender.

Gould OS4000 DSO, 1981
NATO No. Z4/6625-99-647-3625
This scope needs the connections made to the 4001 output 
section for the scope to work in normal mode.
To remove that section remove bottom cover and then the 4 
slot-head long stud/bolts that anchor into 'C' 'clamps'.

Gould / Advance scope
Model not known , from memory probably Advance OS15 or OS 25
False triggering due to faulty ps electolytic leading to 
mains hum on the power rail feeding the trigger section

GW GOS 543 40MHz scope
no display,low EHT
EHT diode pump circuit fashioned on GRP circuit board with the board edges 
exposed to atmosphere bypassing the encapsulation and ABS housing,allowing 
ingress of water vapour and condensation along the glass fibres to the 
component leads in a high impedance circuit.Rebuilt on paxolin board

Hameg HM103 scope
Lack of flyback suppression
Replace the optoisolator

Hameg HM412-5 scope
No T/B and trace off screen to the right
CMOS 4011 failure

Hameg HM412-5 scope
Loss of X span, both beams , contracted down to a few mm and X shift 
only moving a few mm. x5 expands the trace to 20mm or so. 
Focus and intensity ctls ok
Remove 2 cap nuts at rear and whole cover slides 
off rearwards.
Checking Z board terminal strip, the +260V rail was 23V
DC over C511 was -140V, +23V
over C512 -38V, +143V
Forgot to measure these V after repair.
All active and C and big R tested cold ok.
Turned attention to the only 2 earth points in that part of the 
ps. The 115K to ground was o/c. 
So 1/4W carrying 1/6W and 140V was too much for 
it. Absolutely no sign of over-heating. Scraping off the coating 
the break was about 25K from one end, not centre, presumably long term 
electrochemical corrossion until narrowed too much to sustain.
Replaced with seriesed 2x 56K (70 V over each 1/4W) 
and adjusted the 140V adjust to 
bring 140V and 260 V to near enough.

Heathkit IO 4541 scope 1977
No vertical trace.
Worn out pot on the v/div switch assembly.
Recondition as per tips file

HP 9872A A3 plotter, 1977, 14 Kg
Dead ps
2 screws at rear, remove rear plate to show 2 large 
head retained screws. Undo to allow hinging 
of top but beware not full hinge anchoring
Replace the rear plate to protect the rear runner as 
too easy to lift the top with this metalwork.
Prop-up square bar provided internally.
Switching on fan comes on and 6 LED momentarily 
flash on the main board and brief kick from the motors
Uses HP coded NS 
1820-0269
1820-0471
1826-0138
1820-0429
RCA
TO220 thyristor number, 64 240
and TO3 power transistor, 1854 97 67
LM309H
unscrew top , then 5 pillars but not the 
one wiht the pair of TRW caps
undo VR board , fan and power rails plug
wrap the mains switch after discoinnecting
lift ps board off the central interconnect
VR bioard large R is 825R
ps W/W 3R3,180R,1K
large Rs 12K,390,43K?, 390,100,27K,27K
3R3,3R3
1R8,1R8
150,20,20
1826-0122
1826-0146 (Cx 7733 date ?)
2N6308
2x 2N6306
RCA 1854 07 67 ( CX 7724 date ?)
RCA 64 240, TO220 thyristor ? measured 8R near 2N3725 cathode-anode
the problem was on large stud diode 1N5812 
Why would a SMPS high speed Unitrode rectifier stud diode 1N5812 , 20 amp,
50V, 35nS fail sitting around in a normal human occupation heated environment for 10
years unused?
Currently reads a stable 6.2 ohms. 
A Motorola 2N2904A outside the ps  had gone B-E ohmic of 77 ohms in the same
timespan of un-use. Would these faults have occured if regularly used , ie
the m/c would have needed repairing twice in 10 years ?
Dremmelled around the top of the can exposing
plug of ceramic or glass which easily came away. As did the central copper
pin , presumably the joint to the die failed with the Dremmel vibration.
The casing is presumably steel , but no corrossion on the outsde. But on the
inside of the can is about 90 percent covered in brown deposit, presumably
rust. No grease or anything like that inside. Before dremmelling, the
reading had changed to 10 ohm and it still reads 10 ohm between stud and
weld blob on the die. I wil dig out my microscope but I doubt I'd see any
sign of crept silver, a few atoms thick, especially as it is presumably on
the edge of the die.   I'd have to grind off the rest of the cap to view the
side of the die.
Certainly no obvious burn marks on the shiny sliver and no smoke/sputter
trails. As this failed in storage, or theoretically at the point of last
switch off , then no such damage expected.
Is ferric/ferrous oxide conductive? if some off that could magnetically
shift and migrate to the die.
I've now removed the remaining cap cylinder and unravelled it and the patchy
rust film, or whatever, stops abruptly at where the glass seal was.
Much more interestingly viewing the die bond area to the base with a 30x
microscope. This appears to be a gold metalisation that has been eaten by
snails , if you've ever seen paper that has been in a leaking shed and so
grazed. Roughly linear tracks and then blobs of missing gold that seem to
have shamfered or grazed edges rather than the plating lifting and then
leaving sharp edges. One side of the die seems to have some of this gold as
a sort of powder in a few lines like iron filings over a magnet, rather than
solid metallic tin whisker needles, collecting there. Something seems to
have etched away the gold into a dust and then electrostatically or
magnetically drawn it to the die edge. It still measures 10 ohm so not a
will'o'whisp, frail accumulation . Perhaps I will take some microphotos
before probing the accretion with a needle. These grazing marks are roughly
midway between die bond edge and the edge of the support pillar.
So not an atomic level process of metal migration , so what is the process
that went on here presumably mainly when unpowered and very slow .
Some pics of all this, 1mm graph paper background
The sides of the cap unfurled, A-A is the distinct cutoff line of the rusted
part, the roughness at the ends is my butchery, also clean originally
http://home.graffiti.net/diverse:graffiti.net/1N5812_a.jpg
The green glass filled seal around the anode pin
http://home.graffiti.net/diverse:graffiti.net/1N5812_b.jpg
Its unaffected partner, as far as not being ohmic anyway
http://home.graffiti.net/diverse:graffiti.net/1N5812_c.jpg
View of the gold metalisation over the base metal stub, Os are below some of
the largest areas of erosion. I don't know where the missing gold went to ,
no flakes or obvious dust lying around in the diode when first broken into.
The weld for the anode pin is off-centre as can be seen in the pic, is that
relevant ?
http://home.graffiti.net/diverse:graffiti.net/1N5812_d.jpg
Edgewise view of the die , showing gold metalisation up to the die but much
is eroded away , eg base metal showing between C-C and the most obvious
accretion bridging the edge of the die marked with the Vs
http://home.graffiti.net/diverse:graffiti.net/1N5812_e.jpg
looking down on the large gold metalised areas with Bs showing the most
obvious bits of base metal, the bright line is the die edge
http://home.graffiti.net/diverse:graffiti.net/1N5812_f.jpg
the wavy gold edge ,near it, is probably metalisation over some underlying
pad so presumably at manufacture rather than a later formation
These diodes were stored anode upwards in kit , untouched for 13 years, so
something made the dust aggregate and migrate vertically , against gravity
it is now reading 13 ohms
the image of the whole die on
the hex base is placed on an Al ring for the purpose of photoing
Another possible ingress route is via the anode pin. When I cut through it,
it was in 2 parts and loose, a pin surrounded by a tube, both of copper and
somehow combined somewhere near the solder eyelet end. Maybe cap is
assembled after central pin welded to the die, then the cylinder placed
over, then glass seal and both copper sections fixed across at the eyelet.
Replaced both with similar stud diode
Main ps DC rails +/- 128V 
uses 2N3725
main ps rails out
5,0,0,0,-12,6.8,12,5,5,5
The VR board pins
0,5,17.7,0,0,8
0,0,5,0,24.7,14.7
and othe rone
.1,6.2,6.2,0,8,5
.1,1.2,2.8,0,14.7,5
all 6 LEDs on main board
some other interconnect measured 27,0,0,-27,0
Pen up /down not working
solenoid measured 40R, disconnecting R/R leads 
at driver board and applying 12V worked
Line 34 from top to mid board high on pen up 
low on pen down
Solenoid drive stays at -12.7,-12.7V
Replaced 2N2904A with 2N2905A and drive output 
then -4, -12.7V
At end of confirmation test and before resetting to 0 , 
3 plus 1 LEDs lit

ISO-TECH IDM91 DVM
Dithering values on display
Clean the selector switch tracks and wiper 
with solvent and spray with switch cleaner.

Keithly 128 DMM.
Display showing erroneous segments and multiple decimal 
points.
Clean off corrosion on the pcb pads that connect to the display 
via the rubber / wire composite Zebra bridging strip.

Labgear CM  6052/CB, 1975
Used as video colour bar signal source 
rear chassis coax socket is frame syncs only
disconnected this wire and jumpered the socket to the pink wire marked 
VID on pcb for baseband video and syncs out, 500mV pk-pk, ascending staircase
While in there 74 TTTL, R reversed orientation
73,04,93,93,73,93,00
93,20,73R,02R,73R,72.72.72
TBA520
4433.619 Xtl
Selenium bridge rectifier bolted to chassis and also 2N3055 covered with tape
C6 and C12 zeners
p29 is syncs out 
p11 is unbuffered VID out , 500mV with scope but distorted 10mV 
trying to feed into 75 ohm.

Maplin 920R pocket DMM
Nothing despite good batteries.
Compressed foam on the battery cover, added a 
patch of felt to pack out

Maplin MS8221 DVM
loss of one segment and then loss of the K range in auto and manual of ohms so 
if not M range apparently showed ohms as anunciators are M K omega
so 33K ohm reads as 33 ohm
remove the zebra 
File back the plastic surround to the zebra to allow mor ecompression, and blow out 
the filings.
Another time lack of zero-ohms and drifting value. Opened up and cleaned switch tracks and 
contact strips with meths. Did not change the long-term minor problem on battery check, 9V anyway. 
On dc V would measure correctly but turn to battery test, ie with a load applied, the voltage would be half
 a volt or so higher apparently.
later problem
.458V on diode test with no diode and 1.591 on R scale with no external resistor
Changed batteries and cleaned the rotary sw but much the sa,me.
220R resistor measured 153R
read ok on volts DC , just problem on scales requiring superimposed DC ,so also temp scale.
With no batteries about 2300R measured with a Fluke across the probe terminals in R and diode 
mode. 
Q1 S9014 = STS9014? was very sensitive to touch , varying this 2300R down to 1.8K and about 20R 
difference measuring each way.
3 of those S9014 transistors. Carefully snipped away the old and the probe reading went to >3M 
capacitive variability. B-E pads are connected . Replaced with BC183L with B-E legs tied. 
IC2 TO92 marked 8503 1547 , perhaps LR8503  1.5V ? voltage ref. 
Measured and marked inside the 4 pots but decided not to try recalibrating . 
2.67K R measured on calibrated  Fluke and 2.66K on this. In diode mode .537V of a diode on the Fluke 
and .527 on this DVM, probably due to difference sensing currents

Maplin N19FR Infrared thermometer
Will run on 2 nicads wired in , instead of the 3V cell , nice and compact of course but failed 
when urgentrly required and no replacement cell around. 4V and starts malfunctioning as "contrast" 
increases to blacken nul segments. The internal battery contacts take 
solder well. Ground out a slot in the periphery of the battery well to take the wires and replaced the cover.

Kikusui COS5041TM, 40 MHz scope, 1986
Absolutely nothing on the display, even in darkened room at switch off.
Something wrong with wiring/connectors to the CRT heater, the orange glow is
visible under the shielding now attended to and the scope is working.
PDA/EHT of 7.2 KV 
EHT transformer cover easily removes.
TL081, 1R, D880 mica .14mm no grease, 3.3K 1/3W over B-E
EHT transformer secondary 20.5R
Heater continuity 5.7R and transformer when disconnected 1.2R. 
I never know how much a 300 meg/313Kohm,30KV, 1000:1 divider meant for 
20K/volt meters & 10V range , but following measured with DVM, would pull down EHT.
yellow (focus?) -1045 to -1182 on varying focus control
orange -1777 to -1765 on changing intensity control
red -1736 invariant
brown -1736 heater, invariant
ribbon feed 60 to 80V on changing Y shift
blue 40V,  green (astig?) 70V
red 50V, brown 40 to 120V on changing X-shift
Jitter in Y sense, both traces equally, that gets worse on warming 
until they disappear off the top.
Q313 , 2SC2704 had a Vbe in circuit of .85V 
and by desoldering 1.1V . Replaced with 2SD669A, also 
for anyone else, similar are 2SD668, 2SC2911 and 2SC2682. 
Replaced sillypads on both  with mica and grease. 
As DC coupled it made the 2SA958 supply transistor have 
a collector voltage of -58V and due to heat there , looks like 
a problem there instead.
3884 supply measures 17.9, 11.9, 18.5
D880 -19.6,-11.9,-20.2
A958 -1.5,-30,-1 (1.3V from TO92 tranny in Y section)
C-E resistance in circuit 1.3K , W/W resistor 2.7K
Paralleling 120K to the 120K that feeds into 4558 input 
alters the output, so OK (diode here is ordinary 
signal one ). Should be about -5V o/p
4 Y drive transistors should be about 6.6V and 122V
Crackled paint of resistor next to U6 100R
Scale illumination lamps blown, unloaded supply, at maximum of pot control
is 19.5V to paired up bulbs. Anyone ever tried droppered blue or white LEDs
for such purposes.  Would require a reflector on the LED axis to throw light
off radially. Otherwise it will probably have to be 24V bulbs going in
there.
Bridge rectifier DC 139V, -36 and -19V,20V
Illumination controls C1846
Mains secondaries 32V ac, 147V ac
EHT transformer driver D880 , no grease on the insulator.
Y push/pull sw , furthest switch
rotating one next in line
main wafer 2-3/2-4/2-5 repeated for 6 highest ranges
front wafer 4,3,3 connections
Horizontal fuzz of same amount whatever time / div , adjust 
astigmatism pot along with focus.
Chop/alternat oscillator breakthrough disappears 
on replacing casing
Its possible to recase this scope with panels on upside down.

Marconi TF2103 / Malden TSA625/2, 1972 to 1975
Needed 1 to 10 Hz range so added 2 nominal 22uF electrolytics 
selected from a batch, one high and the other low to give 
a back to back pseudo unpolarised cap of 13.5uF 
Using ac cap meter. To switch in parallel to existing 1.5uF (10Hz) 
cap. decade range cap series went 150nF, 15nF, 1/5nF
Sometimes no output due to failed output attenuator 
1K wire wound Colvern pot.
The internal pair of battery connectors and battery eliminator ps connectors 
are not reliable, turn on the rivets

Marconi TSA625 loss of 10 to 100Hz range after my 1/10mod, due to bad contact on the range change sw, staying c/c when should have been o/c

Marconi TF2370 110MHz Spectrum analyser
Display and storage OK but no counter.
2 problems due to bad thru' board connections of one of these 1975 double-sided PCBs.
One through board link connected the 5V from the analogue section to the digital 
section of board AG1 that had knocked out the first 74s112. Replaced but counter 
only working in range 160 to 260 approx. Another thru' board break to the 7411 on the 
same AG1 board. Resoldered all such thru' board points.

Megger BM6 insulation tester
no function, no inverter whine
One of the BC214 had gone C-E s/c should be 
about 650 ohm C-E. On microswitch is 2 outer pins 
close and 2 inner pins open in use. Solder 
3 wires across solder points to simulate 
rotary switch contacts in M position, 4th not needed.

Micronta 22-185 DVM
Ignore the circlips on the switches. Prize off knobs with white pieces 
under the skirt. To undo bottom end of  case put screwdriver blade under pcb 
and lever upwards to release 2 clips and lift to replace.

Nasty power trannie 'insulator'
I've never in decades of repairing seen the like of this one.
Insted of mica or blue/grey silicone sheet insulator under a 
2N3055 type trannie.,at first sight looks like zinc or 
mu-metal a dull greyish metal.
Just one of the 16 such trannies was B-C short and it was 
only this one that had been replaced 5 years or so previously by 
someone who had used this 'insulator' so presumably to blame.
On inspection/scraping it is probably aluminium with a thin 
hard anodised oxide top and bottom for electrical insulation.
The edges including the holes are bare metal so 
just 0.5mm or 20thou/mil out of position and the B or E pin will touch 
the metal edge of the through-hole . Also just the tiniest microscopic piece of swarf will 
puncture the oxide to make contact that way.
It is professionally 
cut and drilled/punched so not some misguided bodger.
I measured the "insulator" small holes.
diameter 2.45mm and the trannie pins diameter 1.05mm,sanding back
the oxide to metal the oxide layer each side and measuring
before and after ,Al thickness was 0.5mm ,the insulation layer
was only about 0.015mm or 0.6 thou / mil each side. Asking for trouble  I
would have said . On more of a safety note apparently there is a variant 
of the above anodised coating to aluminium. These are pink and a 
beryllium oxide coating so no scraping or sanding of that type.

Norma Isolationsmesser 667 001, 1965, not needing repair
2009, I decided to break into my 1000V DC, 10 Gigohm insulation tester. No
particular reason , just curious, as still in perfect working order.
Had to break a seal to get inside, never needed repair by me or previous
owners, just my adapting to 9V nicad use rather than pair of unobtainable
4.5V batteries.
Made in Austria  
Date inside is 15 December 1965, i thought it would be 1970s
I could not find a www pic of it, so here is one
I translated the centre terminal legend of
Kriechstrom-schutz to mean earth leakage protection, it is connected to
battery negative.
http://home.graffiti.net/diverse:graffiti.net/norma.jpg
foot/inch ruler included
shiny black finish reflecting the camera
Normal 90 degree centre pivot analogue meter so the only reason for the odd
shaped bezel must be for styling
Just disintegrated foam inside. Whenever I check calibration with a 1G glass cased resistor
it is the same cal setting, no switch problems or anything else.
0.1uF/1250V and 160uF caps on the underside of the board
centre probes terminal is battery negative
Uses lazy s H logo , Siemens AD130 / AC153 / OC75 and zeners? marked Z1.5 and Z5
2 black HV diodes ? marked I in a circle, 
2.4mm diam , 14.7mm long, 1.6V forward voltage on a Fluke 77
Dismantling leave central recessed nut in place and strap anchors in place.
With 10M in series with 100M and Fluke 77 over 
10M measures 50.9V dc on 1000V / scale 3.

OK SA 20-30-230 24V soldering iron.
Weak mains switch , replace with SPST in a part blanked off 
hole and a separate neon.
These basic single temp controlled irons can 
be changed to variable by bringing out one 
of the presets on the control board as a 
pot on the front (not noted which one). 

Philips PM3311 scope
Not able to change the 
(time/div) time-base setting. The knob has no effect, the 
contacts are clean inside.
Multiplexer ( 74257 ) in the time/div circuit 
was defective (reported repair).

Philips PM5770 pulse generator
Output on 10V setting but not lower settings.
Inside the screened attenuator box a secondary switch 
mechanism comes into action in the 10Volt position 
but was failing to return on spring action on lower settings.
Same symptoms for someone's HP 8116A 50MHz Programmable 
pulse/function generator but relay driver failure for 
that one.

RS 480-100 Breadboard
Loss of 5V power rail - problem with 
button bridge rectifier losing 2 rectifiers or internal 
connection.

Sabtronic 8610B frequency meter , 1982 LED
Uses 7216 , SA 10116 replaced with MC10116, 74S196 
and SAB1009 and SP8680B on pre-scaler
Beware of plasticer leeching where the wires go through 
pcb hole , likely to cause shorting.
Gate LED only (driven from inverted in operation pin) lit, 
no 5V to 7216 etc, maybe due to Gate LED pushed and breaking 
trace at that point.
FND357 7 segment LED have a diagonal internal interconnect
Added 4.9220 MHz (loads to hand) TTL oscillator module 
with dpst switch for 5V and o/p via 1K to 10M i/p for 
spot frequency calibration check.
For 100 second gate timing (not tried yet) maybe feed digit 1 
pin, through 74LS90 to range select pin and ignore DP 

Scopex 14D10 scope 
Low intensity and a rough sawtooth waveform at mains 
frequency instead of (no signal) bright line trace.
There are two large electrolytics on the mains side of the SMPS,
only held to the pcb by their wire leads.One had been jarred loose,
so resolder and beef-up the mechanical fixing of cap to the pcb with hot-melt 
glue.Also this pcb (with EHT supply) is only held by 3 stand-offs and 
with environmental effects the pcb had curved so add extra stand-offs 
between the board and the chassis.

Scopex 4D10 scope
No Y amp
Broken split inductor L101
Check drain-gate2 voltage on 2x fets if low replace one or both

Scopex 14D10 scope
Nasty noise from the ps and nothing else.
Using an isolation transformer and observing on a scope 
there was low level oscillation of the smps and one burst of 
oscillator per cycle of the mains.This smps is the type with 
the oscillator on the mains side of the pulse transformer.
All active components seemed ok and no leaky caps.
Putting a dc supply across the oscillator and varying 
the voltage the drive to the main trannie changed state 
at about 20V.Disconnecting the pulse transformer and 
testing with a megger (high v insulation tester) there 
was no interwinding breakdown and the inductance of 
the coils looked right (no shorted turns).Eventually found 
the O/P was being loaded by a faulty opto-isolator that 
gates the beam current.It was ohmic between I/P 
and O/P so removed and cracked open with mole grips 
(vice grip locking pliers).Looking at the transparent 
bridge under a 30x microscope there were tiny circles of 
mold or fungus that had grown and coalesced forming 
a resistive bridge between I/P and O/P.

Scopex 4D10A scope
T/B jitter
Due to oscillation in T/B
Replace fet BF244 (TR503)

Scopex 4D10A scope, 1977
loss of channel A which meant of course loss 
of trigger to channel B unless cross-linked inputs.
Trace locate returned both beams.
The dual FET E421 at the input was duff.
The S voltage of both FETs was about -6.7V 
instead of about 2V on the good channel.
Replaced with a 2N5189 which brought back 
the second trace but very low gain., the 
FET (S) dc was still -0.6V, changing the 
bias by changing -20V supply droppers 
2x 6.8K to 2x 36K brought the V up to 0.5V 
and useable operation, up to 1MHz anyway.

Scopex 4S6 oscilloscope
Trace is only a diffuse glow
The 3K9 resistor connecting 200V rail
to the CRT astigmatism preset was O/C
from long term overheating so uprated to 1W.

Seaward PAC500 PAT Tester
False negative indication
Replace the CMOS 4000 series SMT device

Shimpo DF 5.0R digital force gauge, 2Kg
Bad battery corrossion on contacts
To get inside to clean up
Remove battery cover and 2 screws inside
Remove 4 sc on back cover behind display, then 2 long black 
sc at the corners and 2 long cadmium plated sc at the end of battery box

SOAR 3210 DVM
Feint 1 second beeps when switched off. Probably indicator of  low battery.

Tektronix 221 palm top scope
Batteries 55x19x83mm 2x5 nicads
64 mm between locating pips in the pcb
Works if you charge up the nicad packs out of the machine, then introduce
them, but the charge circuit is not working. The full service manual is out
there , but wondered if anyone had worked on one.
At the moment I'm working on the power supply board in isolation from the
other boards, anyone know if this is valid. The schematic does not seem to
show any control lines back to the mains-hot oscillator. Seems it would be
operational (if it was working) all the time the scope is connected to the
mains, regardless of the on/off switch , which seems to be for the sake of
batteries only. Otherwise a matter of making up extender connections for the
interboard connectors, to work on the whole scope un-wrapped.
Everything, so far tested, in that oscillator in the way of transistors and
coils etc seem to be ok testing cold and plenty of high voltage DC from the
mains on the main supply cap, but no oscillation
With about half mains voltage , for a bit of personal safety. About 20V over
C-E of one driver Q and 120V over C-E of the other, whenever powered up.
Plan of campaign will be to try biasing on the off one and see what happens.
Could a leaky driver transistor be the most likely cause of non-oscillation?
one sliver-mica cap could have metal migration , others are polyester and
ceramic. Otherwise loads of diodes and coils.
I've decided to
desolder the pair of  TO220 2SC2333 500V/2A and try a pair of TO220 BUV46
850V/8A switching Qs
The batteries are modern replacements and good order, the scope works well
on battery power but cuts out , as it should, when the batteries run down.
Although this range of scopes started in the mid70s , the one I have here
seems to be made in 1989 with the later protection against self-imolatation.
Try-out replacement Qs in there , but will be tomorrow before I power up
again
Same static set up with changed Qs. Same ones high and low. Taking the base
of the off one high will switch it on, then returns on release, doing that a
few thousand times would put some charge in the batteries. Changing the
silver mica? cap  made no difference. There are a few probably multilayer
ceramic caps that are the next things to change for the same reason,
especially C622 near the base of that off Q. I wish I knew what all those
+5V taps are doing on the hot side of the charger unit, the +5V only comes
up with oscillation of the main power supply off the batteries
Some sort of problem with the Motorola "3-layer trigger" CR622 , in the
manual as 32V SPT32R and marked around the glass barrel , 1N4148 sized with
a blue central band, and neat printing around the barrel as letters ST2PR-
so how to read ? could be PR-ST2. Anyway replaced with  a 32V diac robbed
from a Philips compact flourescent lamp,  diac marked BLDB3 ,  DB3 on the
overlay. 
tested on bench ps 
180uA , 34V either way
60u at 33V either way
<0.1uA at 32V
Voltage over the battery now racing away upwards while I let it, will have
to tidy everything up to get it back into its shield and TO220s heatsinked
and take some readings as not much useful technical info out there on these
compact scopes. I'll have to try the original on a >32V bench ps as well
3 pin battery conns, centre negative ,outer the same +ve , so either way round.
100V mains initially. Fuzzy blob on CRT, moved SEC/Div sw then solid unmoving bright unfocused spot
varying with power pot slightly.
Due to unconnected interboard pins at R416 end of amp board
+200R to ) and 5V on amp board for 4 seconds and meter wangs to charge posistion.
Sw off and repeats and decay glow in CRT, realised the batteries were not charging.
Recharge batteries from bebch ps
Too much DC offset perhaps as in the High Gain of V/Div cannot get display in centreat max of up/down pot.
All diodes on mains section DVM ok
4x TO92 and RS marked TO5 ok
20/9.1/20/270 around Q634
thick loop fuse F630,3A ok
Test point BC is neg of battery of  board interconnect side of the pair
other battery -ve on cathode of large diode
+ve of other battery at C613 near interboard pins , pin furthest from mains fuse
2 Rs with wire wound over measure 19R in circuit.
Caps to mains L & N C615,C614
Confusion around pot core wiring is due to 3 layer constructiuon of the pcb
175V dc over C617 for 130V of mains
HV retained on this cap if no oscillation
14R across the 1K resistor slung under
9.4R across back to back zener diodes 1N9638
Shield/screen sc is under the Danger label
18K and 1K are "shorted" by the T623 and T633
changed C266 from 100nF to 10nF
R622 is 1M not 1M8 of schema
Q(A)(C) with isolated test scope 
70V pk-pk curved sawtooth , 1.2KHz for approx 50% mains, increasing to about 2KHz for 80% mains
at sw off, f & V slowly decrease
sawtooth on .0033cap CR623 end but not the T633 end
CR622 32V 
With 120V ac ,1.5mS oscillationwith .2mS pulses 120V pk-pk
Someone had epoxied a broken knob to a shaft so could not get to that board, broke away enough surround of the knob to get 
a soldering iron to the shaft to heat up and pull off .
On pwering again a HV crack and no display.
70V 1.2KHz sawtooth but no Tx pin5/6 signal and no charging current, (added .1R in line)
Hot +5V is at the end of the battery chain
With power sw on and current through the 9.1R then Q634 can turn on
Use the shield mounting pillar for isolated monitoring , sawtooth on one side of 3.3nF 
Disconnected CR621
at the 10nF, 18Vand rises at switch off

Tektronix 2467B (info from B Clarke)
 late model with the smd Control/Readout board. The 
RAM chip is used as storage for the on-screen readout characters, 
though not directly. It stores a set of addresses which are used to 
address a ROM, which in turn gives the necessary XY deflection data 
for the DACs to write the characters on screen. The bottom 3 bits of 
the RAM had gone high so the ROM was producing garbage, which showed 
as 4 lines of garbage on the screen together with cursor lines all 
over the place.
Replacing the TC5565FL-15  RAM has restored sweetness and light!

Tektronix 5110 scope
Jitter in vertical sense of both traces.
Jitter present with Y plug in present and pressing 
the beam finder and absent without the y plug-in 
pressing the beam finder.
Poor connection in socket J604 of the small 
auxilliary board behind the pcb with the sockets 
for the plug-in modules.

Tektronix 7603 scope.
No display.
On the low voltage regulator board there are a number of 
ribbon leads including 3 8way with no identifying sockets,2 8 way 
and 1 10way.Someone had positioned the 8 way  onto pins 3 to 10 of 
the 10 way and put 130V on the 50V rails.Numerous individual 
failures associated with the 50V and 50V sense line including the 
transistor array satisfactorily replaced for proving purposes with 
a 3046 array.Unfortunately one of the transistors in the U450 
Tektronix hybrid (3 push-pull amp stages in a can) was kaput so
no Y trace movement and defocused spot.

Tektronix 7834 Storage scope
No power supply
The HV inverter transformer had shorted turns,a matter of rewinding another 
as quote for replacement was 760UKP.
610 turns of 40AWG / 45 SWG wiring in 4 "layers" anyone experience of such
winding and thoughts on resulatant (lack of) structural integrity,
slip/sliding turns with temp cycling/magnostriction etc leading to failure.
I've never seen a transformer design brief, so perhaps interesting to others
on that aspect alone
Tektronix 7834 scop HV transformer,(flyback misnomer)
http://www.ko4bb.com/Manuals/Tektronix/Tektronix_-_7834/Tek7834_Flyback_Tran
sformer_Winding%20_Spec.pdf
some general info including bank winding, I've only ever seen on high DC
chokes so structurally sound
http://focus.ti.com/lit/ml/slup125/slup125.pdf
I was trying to visualise the mechanics of the winding machine, backlash
error for the reversing cam , once per bank,150 times for this coil. Runout
error trying to keep the overall traverse in step with the banking over the
width of run. And all the time just surface tension , assumed wound wet, to
stop the turns avalanching down the bank slopes.
40AWG wire is probably 'served' with polyester yarn which gives it a high surface
friction so the turns tend to lock together. 
Alternative maybe to use 'Pi' or 'universal' wound coils for reduced capacitance.

Telequipment D32 'scope
No fuction due to one of the 2 BDX36 transistors in the ps oscillator ,replaced with a 
BDX35. Also 2 of the 6 Nicads were shorted (dendrited) needed replacing. 
No intensity control due to ohmic 130V Zener. No Y deflection of either 
beam below centre line of the display; no negative going traces. One of the collector 
resistors to one of the Y-plate drive transistors was O/C ;replaced with double the wattage 
and similar replacement for the matching dropper.

Telequipment S51B valve 'scope, 1974
Central spot only , no movement.
Main DC line at 480V instead of 380V.
I don't know if as original but the large 500 ohm 
and 1K dropper resistors in the main DC lines were 
mounted underneath and the 1K one had at some point 
scraped the outer cover of the m/c both need mounting 
on the other side of the board.
Replacing the DC fuse and removing the 1K resistor but 
including the 500ohm then spot became a larger square 
and X-shift movement but no T/B or Y-shift/deflection 
other than the fixed square splurge. Replaced 1K and 
slowly powered up via Variac in case of problem with large C4 electrolytic.
No problems and full function so maybe the 500 ohm shorting to case and 
blowing fuse and not knocking out the mains transformer 
winding - someone had looked before me so may have replaced 
a blown mains fuse.

Telequipment 61a scope
No trigger
Check integrity of connection between
PC226 daughter board and motherboard,
and set R74

Telequipment D63 scope.
3 faults
1	High voltage discharge crackling noise. Even in the dark it was not possible 
to see where the discharge was so probably in the step-up transformer or the 
quintupler. The display was too bright and 
post deflection accelerator voltage was too high so much reduced trace size 
in X and Y sense.The CRT 2cnd electrode should have been 
-1400V but was -2200V so EHT probably 19KV rather than 12KV. The EHT oscillator 
transistor had been replaced by a BD131. Changing the biasing pot to the preliminary 
drive transistor made no effect. The primary of the oscillator o/p transformer 
was 56V pk-pk not the correct 38V. Although the DC voltage varied from 60V to 
100V altering this pot. One of the DC divider chain of 3x 5.1M was o/c so no -ve 
feedback component to the oscillator.
2	Traces in X sense squashed up to left of screen. One of the SPS5286 final 
X plate drivers was o/c replaced with BF338.
3	The third trace (Right hand PI) was static bright line with no Y displacement.
A trace to  330 ohm,1.5W dropper from 105V to one of the Y2 driver transistors was loose 
and touching another trace. With this 330 removed and no 105V then no 3rd trace, only 
appearing on beam finding activation.

Telequipment D66 scope
No timebase
Y deflection and X deflection function but no timebase.The high 
ohmic resistors biasing the timebase oscillator FET had gone high.
The drain of the FET needs to be about 2.5V before oscillation starts.

Telequipment D75 and Telequipment D83 scopes as far as p.s. and CRT drive circuits and faults 
are almost identical.

Telequipment D83 scope
Variable and fading image similar to changing
grid cut-off pot R315
Failure of D304 6KV,10mA diode

Telequipment D83
Inability to focus
R323 gone high ohms

Telequipment D83
After 1 hour loss of image
Associated with switch (trigger-auto sw.)
on T/B plug in

Telequipment D1011 scope 1981
No ch1 trace and no trigger.
Initially checked the 8 pin DIL dual FET packages.
In manual as SU2603M but on board were Teledyne 1125-00 (8006 datecode?) 
pinning 1-8 S,D,NC,G,S,D,NC,G so 180 degree replacable.
In circuit DC .1,5.2,.1,0,.1,5.2,0,0
I could not find data for SU2603M or 1125-00 so for anyone 
trying to find replacement I tried 2 2N3819 bent legs to match 
pinning - no trace and the S voltages rose to 1.7V and 2.1V .
Both problems due to physically broken front panel lever switch.
Neat design so choosing 1 of 2 positions of the dedent spring 
can select 2 ,3 or 4 way. Choice of phosphor bronze contacts can be 
1pole or 2 ,3 or 4 pole . The phosphor bronze bridging contacts 
are only held to the core by 2 plastic nibs that had broken. 
Each nib only 2mm diameter so failed as aging plastic.
Reformed 
bent contacts ,used broken nibs for alignment and well glued in place 
with hot-melt string. The detent spring is positioned around the small 
nib for 4 way operation and up against the nib for 3 way (ch2) operation.
3 of the four front switches were broken so rebuilt and reinforced 
remaining contacts including the fourth OK switch. One of the contacts 
was broken and irretrievably bent so fashioned a replacement from 
a phosphor bronze leaf of a lever switch but part from a relay 
would also suit cut ,formed and glued in.

Thurlby DSA524 , 1987
White "fungus" around a ni-cad
What exactly is it and what , if any, damage does it do to anything
adjascent/ same box.
Thurlby DSA524, digital scope adaptor From 1987 . Nickel dendrites like Tin
whiskers ?
Went to use it for the first time in 5 years and no response to any input,
was ok previous use. A vaguely staircase output on an analogue scope , but
thats about all. Composite and trigger out work as do some anunciator
readouts. So presumably ROM, RAM, PS, Clock are ok. So is it realistic to
suspect the A-D a Ferranti ZN449. ? One of those "better than nothing"
service manuals is out there but thats about all
This happens to be right next to one of those pcb soldered oval shape blue
cased nicads of that era, before EEPROMs came in.
Relevant looking signal (alternate time slices varies with varying DC at ch
1 and ch2)gets to the analogue data-in of the ZN449 , enable, clock etc ok
but invariant data out on the 8 data lines.
I've some SMD ZN449 , any tips for converting to 18 pin conventional DIL?
Will have to deal with the conductive contamination on the pcb around
the nicad, down to 300K over the polyester of the pcb, soaked in
capilliary-wise?. Removing 4x 74HC as well as the ADC
Maybe I'll be lucky. Removed 5 ICs and the nicad from that area. Washed
locally with meths and while still wet dremmel and rotary wire brushed over
that area . Now I cannot find any conduction to/between any isolated pads or
the plastic/glass less than the 30Meg of my DVM. Made good the IC holes
ready to receive 5 turned pin sockets
Well I suppose that is progress. Back to exactly the same malfunction as
before , but with the 5 original ICs in sockets now. Varying activity on the
analogue side of the ADC and no variation at any of the 8 output digital
lines . Will try a SM ZN449 , which strangely is wider package than standard
DIL. , see tips files
Using the original nicad at the moment, slung underneath. I like the idea of
a silicone bund. I think , if I get this working, I will replace with a
new one , stout-wired upside down, right side of the board  , over a
silicone sweet or cupcake mould
2 main pcb with plotter board removed and insulation interleaved
ZN449 p8 2.5V ok
p4 high
3051 to LF351 to p6 of ZN449
slightly up and right of centre of  DSOA processor board on murky schematics
Eprom HD6303, 
84256, HM624, VL65C22
ignore the ZN449 business above
, I did not realise
the large UVC3120 was the main ADC not the ZNA449 which is probably just for
overlaying ch 1/ch2 levels of Y control.
http://diverse.4mg.com/dsa524_inline.jpg
shows the test-bench layout with 4mm threaded stud brace to the right 
and the 2x screened signal leads are long enough for this layout
remove the plotter board
join the earth tags together
remove the BNC sockets from the rear so the midle pcb can lay next the bottom board
With no power to the unit, the input relays are grounded at the relay
For monitoring the input amp/attenuator sections use the 100R resistors at the top 
of each section
Yellow lead is the trigger BNC one
pics , before I grabbed a tripod
http://diverse.4mg.com/dsa524_front.jpg
showing LEDs as set for the test point scope traces
http://diverse.4mg.com/dsa524_leds.jpg
The variation at ZN449 p5 / 430R with changing ch1 and ch2 offset is not coming via 
the outputs of ch1 or ch2 preamp sections
Error ? in the schematic D8 1N4148 is in series with the ADC analogue input 
not to analogue ground
.1V rms 400hz at input ch1, .13V pk-pk at ADC screened wire
the datasheet for the ADC+DAC uvc3130 which is out there, is for the pins
used in this unit, the same as the uvc3120 same pinning
some UVC3120 dc
all ADC digital outputs 3.4V dc
p20 n/c
p22 1.98V
p25 1.95
reset attenuator to 100mV and p21 UVC .4V pk-pk
change attenuator to 10mV and p27 has signal
M5165 varies with input
74AC374 o.ps are ok to 6264 
p22 (06) of HM6264 from 74HC14 that feeds Y & G of 74HC244
Cut the trace to bridged p1 & p19 of the 244 and took low
p2 of the UVC had a semblence of sine wave input 
244 gating signal is mainly H with trace remade
http://diverse.4mg.com/dsa524_36.jpg
Trigger out 5V pk-pk
7 negative going pulses over 7mS then 5mS H
scope 2V / cm , 20uS /cm using the DSA ext trigger for triggering scope 
in RUN and HOLD mode
In working order and normal view on ext triggered scope
http://diverse.4mg.com/dsa524_21.jpg
signal going to scope with internal trigger
http://diverse.4mg.com/dsa524_33.jpg
or maybe the DAC o/p of the UVC , (mangled notes)
Started getting cyclic relay clicking and nothing else
74HC4075 , one of the 5 I had socketed as around the nicad, replaced with 
CD4075BE as maybe dodgey connection there, functioned ok
Removed the VL65C22 , lightly abraided pins and reinserted 
and relay clicking stopped
4KHz , .1V rms input ch1 , 50mV attenuator, AC 100uS TB
scope 50uS and 200mV / cm
p15 of Con1 between boards in HOLD and RUN  mode
http://diverse.4mg.com/dsa524_32.jpg
con1 dc in RUN or HOLD mode
1,3 ....
3.9V, 5,5,5,0,0,3.1,variable
2,4, ....
5,5,0,0,0,-5,7V
legend "both traces off" measures 170uS

Variac V5H
Could not remove brush housing.
Presumably should be simple push against spring, turn and 
pull but aluminium corrosion didn't allow any such movement.
Matter of hacksawing/ dremmel thin disk cutting a slot in the 
aluminium arm down to the brush housing to give enough slack.
If clipped sine waveform in ch1 or ch 2 , press Gnd + AC/DC for AUTOCAL , 
all LEDs go out. 

Wavetek 187, 1984 function generator
No functions
Shorted supply rail C22 tantalum 22uF, 20V uprated

Weller magnastat soldering irons
Open circuit iron after being dropped or knocked.
It is usually the magnastat switch that is faulty not an o/c element.
These switches are very frail and it does not take much to knock them out of alignment 
or even break apart. Take apart ,realign the contacts and reassemble.
Magnastat soldering iron bits 
I cannot find any major suppliers of these bits anymore in the UK, despite the likes of RS selling magnastat irons.
I have a few hundred NOS , plated copper long cone bits, not Weller or Cooper AFAIK, that take about 10 minutes to convert to magnastat , assuming you have the curie temp magnastat end caps from old bits hanging around, easily removed when you know how.
If any interest I'll post more details, I'll not be around long enough to have used up a few hundred 
It was about a year ago I checked on wwwRS and they carried the bits then and also 24V heaters which I was actually checking for as the heater on mine is perhaps 20 years old. Who wants to use 240V magnastat irons in post_CMOS days, especially without stocking replacement bits, they do stock those though. 
 http://uk.farnell.com/jsp/search/productdetail.jsp?sku=416435
 http://uk.farnell.com/jsp/search/productdetail.jsp?sku=416691
I've never seen a "Frenchman" style before but the tip diameter and general conicity? looks much the same as the popular long conical type 
I got the nice man at Hedge End, England RS counter to check and no 24V bits or heaters. Just as well I checked my spares status, bits no problem as I can make my own brand ones easily, but what I thought was a spare element was a spare new switch.
So I'll have to look into determining what W/W resitor/s to unwind or resistance wire, for 24V 50W, plus fire cement and strips of mica to make one. I cannot see me moving over to Metcal 
I dismantled a 240V soldering iron heater. All these years I thought it would disintegrate into a pile of mica but now I see the secret, should be no problem making one. How do you form mica into a tube, simple. You part score mm apart lines into the material, then it will roll neatly.
This one accepts a magnastat tip so may even be reusable in itself with the right resistance wire for 24V 50W instead of this one of 4,000 ohm .
Mica tube 6.7 and 5.8mm diameters as a former and 2 layers of wire with an interlayer of 10x35mm scored sheet of mica rolled up, terminals just clipped over the central core. 
http://uk.rs-online.com/web/p/soldering-iron-accessories/0661986/
 http://uk.rs-online.com/web/p/soldering-iron-tips/0661447/
To convert Solon? bits to Weller/Cooper magnastat bits. 
Long conical sharp tip form with a hollow stem and a rim. Grind back the stem until of right length to match the main part of a magnastat bit. 
Copper nail , as used in clinker built dinghy production/repair. 
Hammer nail into the hollow. Cut off leaving a stump to take a salvaged magnastat. With the double action cut off pliers used after clenching operation in boat making , or a dremmel+disc cut slots into the nail stump and open out so the magnastat just clears it. With washer over 
the cone , up to the bulge, squash in a vic so the 4 way split of the nail wedges in the magnastat. Never throw away an old magnastat bit. 
With Dremmel+disc cut into the 3 or 4 crimp/swage patches to reduce the amount of metal, then pull/lever off the magbnastat cap.
For 240V Weller PU-3D, shorted secondary overheating primary before failure.
Tx dismantling. Grind diagonal line across E edges in case order is critical.
Razor open free E end . Brush over and soak in Methyl Chloride paint stripper in a dish overnight. 
Brush over with washing up detergent and then water to neautralise before handling. Remove the first I plate.
To assist removing the razor opened E, grind 2 0.5mm part slots in the E ,half 
way between E edge and bobbin face. Set on bench against an end stop that abutts main lump of E plates.
Punch and hammer , angled into the grind-slots , outwards enough to grab with mole-grips . 
Rest of E & I follow easily.
Unwind between end hooks of known separation, to measure the length. Next time count the turns off (I expected a 
spark welded lump in the middle of the windings but just touching contact)
13x12.62m + 3.24m =167.3m of 0.32mm diameter wire, but only .28mm replacement
( For Weller PU-1D primary wire ,22 to ,24mm wire diameter 
With TCP iron 24V, 45W , with leads 12.1R load )
Bobbin 27.9x27.9x
Rewinding compensated for smaller diameter by less wire by ratio of squares of diameters plus 
10% to play with = 141m. But on testing with all but one E only needed more like original length 
as hand winding in effect, although using a coil winder , not exact layup
1,120 turns 49.9R 
trying air cored bobbin on audio gen
0.754Vac of 5.5KHz , output of 23mV
It turned out ,for same voltage out and for my ,in effect, manual winding, the same length of wire required 167m or so , 
going down from .32mm to .28mm. Using just the Es minus 1, and no Is , magnetising not complete and 
could only run the variac up to 70 per cent or so, but gave the voltage ratio at 50Hz. Next time I'll probably 
add the Is as well, as its easier inserting the Is than the Es. Tip for final E insertion, force a razor blade in 
between the interim last E and the bobbin, then insert the last E between the razor, bent a bit, and the in place E. 
Loaded with 2x25R in parallel to test
magnetising buzz , near to 240V , needs all the Is added to return to normal mains use.
4 bolts temporsry ,Laqcuer over , especially the last Es and Is

Workzone 6149 DVM
It had previously survived +1.5KV on the 1000 volt range but not -1.9KV, killed 
it. D1 shorted and 30R over the battery rails. Removed the 358 (LM358 ?) 
just in case it was that, but main epoxied lump hid the problem. 
Salvaged LCD and soft case

Industrial and Boating

Advenda AD1 photographic enlarger timer.
Failure of relay to always pull off after timed interval.
A stock fault apparently-company advice to bang front panel with 
a metal object over area housing the relay after yet another spoilt print. 
Although 24Volt relay due to 
unregulated supply the relay on voltage down to 15V.My replacement 
24V relay failed to cut in unless whole unit was upside down.Replaced with a
 12Volt and small dropper R in series.

Avenda AD1 photographic projector timer
Intermittant erroneous timing say set for 20 secs ,start down count and sometimes 
would reset itself to above 20 seconds still counting down giving say 30 seconds count.
The circuitry is driven from a dropper straight from the mains. Simulated dirty mains 
was the only way to cause this count problem. Found space for a small 12V 
transformer and bridge rectifier to take the logic isolated from the mains except for 
the timing input 1M dropper. Still don't like the way this feeds straight into 
the clock input of the 50Hz divider IC.

Advena AD1 photographic enlarger timer
No display or function other than fan is active.
DC is derived direct from the mains and was only about 2 volts.
Powering up from a 15V bench power supply there was too 
much current drawn for CMOS circuitry with the relay 
coil disconnected. The 4025 was loading down  the DC rail.

Boxford BUD lathe.
Failure of contactor latching on.
2 sets of microswitches , any of which will lock out on failure/ misposition..
Single one on fron door. Set of seriesed 3 , 
side door for gearbox, headstock end of bed for saddle endstop and 
one in the headstock for A/B gear ratio select lock down (activated 
by retaining hex bolt). Rear 3 junction box 
is under the rear cover of the headstock. 

BT DF100 Fax machine
Dead machine.
The external 24V ps was ok.
Disconnected main board and powered up outside chassis.
No 5V on the 74HCT chippery.
The Vcc pin on the UC3845N SMPS supervisor was only 8V ,
should be over 12V,cutting trace to this pin and trace rose to 19V.
Replaced the 3845,normally Vcc of about 14V.

Canon FC 220
Vertical blank stripe in image.
Toner running low and so allowing the developer (iron filings) to become 
insufficiently diluted with toner leaving a gap around the magnet cylinder.
Broke open the toner/OPC/developer cartridge to redistribute the filings 
as beyond just turning the cartridge over a few times. Used a mixture of 
toner intended for Mita 512Z and Toshiba BD5120 but left a feint grey 
background to the image. To break open the cartridge. The toner hopper 
is under the embossed instructions so lay this downmost ,OPC upmost. Release 
the 4 plastic catches along the width and 2 each side in different positions .
The OPC section will then comme away from the rest. Hold back the OPC shutter 
flap and scoop the toner topup into the exposed slot. AE means Auto Exposure.

Car alternator (Ducellier but probably general) repairs
Firstly to test alternator out of vehicle and 
on the workbench.Feed a low current less than 1 amp
into the exciter coil (the current path that
goes to the instrument panel lamp display, 
this lamp is functional as well as for display ).On spinning 
the pulley by hand with this current on there should be
substantial resistance to rotation and some
output volts.Most common failure on this test
is worn brushes
For s/c failure of one of the 6 3phase rectifier diodes
leading to no o/p. You cannot prove this sort of failure but occured coincidentally 
when the car was in for welding . I suspect using arc welder or TIG /MIG on 
a car can inject spurious currents in the body and blow even these high current 
diodes. Makes you wonder about expensive CD players etc could be knocked out 
in similar fashion.
I could not find replacement 50 amp barrel
diodes small enough to replace but
a TO-220 type packaged epitaxial double
rectifier from a switch mode power supply paralled up 
fitted the bill
Another failure mode was failure of the regulator
giving some but insufficient o/p.Fairly
straightforward discrete circuitry inside the 
soft sealed package.Actual culprit was an o/c .3W
6.2 volt zener so replaced with a 1W and resealed.

Claude Lyons TS2 mains voltage stabilizer.
Output following input rather than maintaining at 240V ac.
Problem with the 20mm fuse-holder with the 500mA fuse 
supplying the servo-controller.

Courtenay Solaflash 2 photo studio flash unit, 1982
Xenon lamp tube 
8cm major and 1 cm minor diameter , only markings 3X on positive end and 2J on
the other end.
Fails to flash.
Worked out how to discharge the caps and then remove the innards.
Remove 2 screws near the IEC inlet and 3 other periphery screws 
 to release end 
cap. Discharge the caps via insulated probes and at least 100 ohm/10W 
resistor before taking further apart.
Remove 2 screws on front reflector around 
the edison screw socket and disconnect the wires to the tube. 
Remove one screw near the handle and 4 around the slide out section 
and slightly turn to slide out the rear. 
1982 with components left over from the 1960s, those, don't even know the
generic term, flange stand-off flesh-tone background rod resistors.
At this stage probably problem with the 400V , 2A 2P4M SCR.
Nothing much else active, MPSA56, 2x 1N5407, 2x 1N4006 , BZX61C68 1.3W
zener.
The caps hold probably charge for hours after disconnecting from mains.
Putting a neon bulb and dropper over the trigger transformer gave plenty of
strike even in room lighting.
I've an old reserve piezo gas lighter that the metal grounding barrel on the
end falls off, ideal for testing this tube, but no flash over.
With 2 compact camera xenon tubes in series and a 1K, 10W dropper and the 2
camera trigger surfaces connected together to the existing trigger circuit,
it will reliably trigger the tubes to rather unstable continuous light ,
until I switch off , as 10W is not enough for continuous dropping.
I once triggered with the piezo gaslighter, proper trigger disconnected, but
didn't work the next few times.
The owner reported it would intermittently not work before giving up totally
, so I assumed it could not be a tube problem , well not gasifying anyway.
I've now tried 2 xenon beacon tubes in series and it works every time on
full or reduced settings. For continuous high rate of fioring 
it would need 4 such beacon lamp tubes
Also works every time with the piezo gaslighter
when its ground plate is connected to the flash unit ground
Now I've separated the tube from the reflector enough to clean the tube in
the area where it descends through the reflector
http://www.diversed.fsnet.co.uk/xenon_tube.jpg
I can now see all is not well.
The cloudiness in the cathode end of the tube (only) is on the inside and
there is a patch of distinct blackening marked with a yellow line.
General black background is conductive foam for picture contrast , the
yellow wire is the trigger wire connected to the loops around the tube, no
silvering. The gaps 2 turns up of the spiral wrap at both anode and cathode,
would they be breaks or like that at manufacture?
It is after all 25 years of commercial photo studio use.
The slave action via LDR also checks out
Powering up, switching off, and triggering with piezo and measuring before
and after V then 42 joules per discharge per beacon tube. It may be
advantageous to wire in a pair more of these beacon tubes to make 4 , to
spread the load, at 2.50 GBP each compared to 70GBP for a not exact size and
fit single tube replacement.
Neon extinguishes or flashes when cap V is low.
With the 5407 board removed it is possible to work 
on the trigger cct etc without having the 660V on the caps.
330V plus 330V. 2P4M replaced with TIC126N
1R5 big dropper in series with 660V and discharge current 
flips the reed relay which in turn trips the 48V 
relay (5.6K in cct)for recharge lockout.
Full/reduced power shorts the 100R and 1.5R
Globe lamp 150W, 240V
Also 44R dropper.
LDR feeds the base of the MPSA for secondary slave 
action from primary flash lamp.
Mains transformer in cct 980R 117H, 500R 52H secondary 
163V ac in use 

Demolux D6070 overhead projector
Intermittent light o/p.
Replace the thermal switch on the lamp reflector
assembly.

Diamond force by Diamond Guard Corp. ,diamond checker
Complete non- operation .
The rear panel has to be broken free from the front and middle 
section to gain access.
All 3 NiCads dendrited (short circuit) also poor physical 
integrity of the spot-welded tags on these cells

Digisonic 502 ultrasonic metal thickness meter
Display only shows 1 in all modes.
1978/1979 meter using double ADC ICs ICL7103A 
and ICL8052 4 1/2  digit converters.Luckily 
these were OK,the 74C00 on the display board had 
one NAND gate o/p stuck high,replaced with a 74HCT00 ok
The conformal coating may have been breaking down and causing an 
oil film on the board which could have seeped into the IC
Another conformal coating breakdown situation
Racal RLC1/2 PIR unit to rob the RPY93, 6.5 to 14 micron long-IR pyro to play around
with, for remote temp monitoring
and the pcb was covered in this waxy stuff  now oily and green on some pins. 
The TO5 pyro had one pin corroded away
Supposed to be there to keep damp/condensation out but over 20 years or so
does the opposite. The waxy/jelly type coating (not the epoxy potting stuff)
breaks down to an oily film that is probably hygroscopic - plenty of green
copper carbonate and rotted IC pins. Gets inside chips capilliary-wise and
rots them internally as well. Anyone happen to know what this oily breakdown
product is chemically?
Translucent appearance, so can make out form but not detail of components
under it, same when new. There is also iron rust brown colour as well as the
green.
If it was a mixture of paraffin wax or petroleum jelly and silica gel I
could see the silica taking in water and the other breaking down to an oil.


Electric fence "Electric Shepherd" ESB 250, 1986
no output
Input side measures 5Mohm, HV output side 12.7R, but looks like a standard
mains transformer.
As I cannot see how the high current side of a mains transformer could fail
I drilled out the rivets.
If originally a mains transformer, looks like one , it had primary 0.4mm
wire and secondary 0.9mm wire. 0.9mm wire has failed .
Fence output voltage o/c 8KV on the label. 9 seriesed 0.22 nF P/D Z3200(no  V
stated) across the output and also 47K, 1W dropper to 2 seriously blackened
defunct neons in series.
Well I would not have believed it unless personally witnessed.
As the o/c must be on the outer layer of the transformer, I stripped off the
cloth tape. There is indication that damp air has got in this unit but not
flooding or internal drips. But in the middle of this outer layer, with no
sign of corrosion or damp under the tape , generally, there is one spot of
blue green corrossion and a small break in the 0.9mm wire.
Maybe a spot of acidic something at that point on assembly and then years of
damp air. Bridging the break gives resistance of the winding of about 0.2
ohm.
Powering on a variac, could not take higher than 150V (saturating)  so
assuming it is 110V , 50W transformer ( or problem on the original pimary ) output
read 4.7V ac , no load, for 110V ac input.
Back working with proper bridged repair, new neons etc , at least it fires a neon via 100M ,
(0.1G ) dropper. The casing screws mayvbe
UNC which may indicate USA but label has UK address. Electric Shepherd
printed on the PCBs.
1000:1 EHT divider to scope showed about 1.5V pulses which presumably is
consistent with no load 8KV. Left the bank of Cs and neons in circuit ,
giving about 1.5Kv peak pulse. Nice strong blue spark with a spark plug 
(not known gap as lost feeler gauges)
Of course the 110V primary is now used as the secondary here, for the 
voltage step-up. Internal construction looked the same as mains
transformer.
0.1R, 3R3, SGS 50 679 9 501  (TO220)
47K 1W, 20uF 400V, 4584 S2512NH 

Elu MOF 96/10 router.
Excessive arcing on the armature.
First problem the owner had inspected the brushes and had broken 
the pcb holding the interference suppression components. Bad design 
with small neck in part of board that seats in the casing . Needed 
hardwire soldering and reinforcing the board.
	The main problem was a short on one of the stators. Now how 
to disassemble the housing. Drift out the roll pin through the large 
diameter slideway pillar then the base will rotate around the small 
pillar to access the 4 screws under the motor. Mark and unplug the 
wires going to the stator,mark and remove the brushes. Remove housing 
and then circlip retaining small slideway pillar to remove the base.
2 long screws retain the stator housing.
On one of the 2 stators the exit wire crosses the first turn of the entry wire 
where there is maximum voltage across the enamelling and with vibration 
etc had broken through shorting out all that stator (bad design again). 
Forced apart and introduced some thin insulation material and varnished 
in place. Did same on the other stator pre-emptive.Reseated brushes 
with armature seating stone gradually raising drive voltage via 
a variac supply.
Measured Inductance,resistance for this 220V,3.6 A motor.
Each stator 6.1mH,3.5 ohm. Armature 28 section .75 ohm per 
section and 5.8 ohm across diameter that is across 14//14 sections.

ETA Tool Co, Leicester, coil winding machine approx 1920s, later than 1912 anyway.
( An internet reference gave info about wording form on brass 
name plate on ETA lathes relating to the age, on this m/c the plate is 
just to left of the 7 digit turns counter in top pic below )
This machine had been in a garden shed for years and was 
seized up with surface rust, but brushing and lubricating 
brought the leadscrews, traverse reversal clutch etc into operation.
The contiuously variable gearing between turns and traverse is 
2 plates, one driving and the other driven, with a lead screw 
positioned jockey wheel connecting the two. The tyre on this 
jockey was totally perished. Standard 'O' rings are too small 
minor diameter. But filling the periphery with 2 small O rings and 
a rectangular section VCR pulley tyre was the nearest I could get. 
A rubber suspension band for car silencer suspension would stretch 
and fit to this pulley but would require cutting back the cast-iron 
frame to give space. Decided to use VCR tyre and pack out the gap by 
gluing a platter from a 5 1/4 inch hard drive to the driven plate. 
Ground down the edge a bit as a bit over size, also the thou per turn pointer 
holder needed cutting back. Paked out the 4 prong spring plate 
with 4 thick washers. Machine lacked the supply spool 
holder back tension arm. Made a pefectly satisfactory 
spool holder by mounting on a couple of ball races and spool 
mounted vertically. Then removing the felt slip clutch section 
from a VCR, mounted touching the spool and other part 
tied back to the support arm and different weights on top of 
slip clutch. Calibrate with different weights and monitor tension with a lever arm 
load guage. As low as 1 oz back tension is possible with this method.
To even out the wire flow for heavier gauge wire, pass over a pulley 
pivoted on a swing arm, pulled back by a light spring, position 
or change spring strength to match back tension.
The traverse scales on this old machine are 2 to 15 thou per turn 
and with the other 3:1 gearing and spindle 6 to 45 thou per turn 
not SWG as I first thought. Opposite sense to SWG , high SWG number 
is low thou per turn. Ruler in pictures is cm and 5mm markings.

If the wire supply kicks back( better than forward and breaking 
fine wire). 
Check the anchor bearings at the end of the 
traverse leadscrews. Take up any slack that allows the screw 
and so follower to move axially on the reversing cam kick. 
Too much will give too much back torque to the variable 
gear drive system. For very fine wire I use a small PTFE 
thru chassis terminal with wire going thru where 
the pin was, ptfe squashed and 
then mounted on variable/settable metalwork 
for the correct angle, closeness to former. 
With as straight as run as possible between 
mandrel , PTFE and supply spool. To give some slack in the wire, 
to allow for snatching/uneven slip clutch by allowing the supply wire to stretch 
I mount the supply spool vertically about 16 feet away with 
slip clutch and adjustable anchored weight on top and then a felt clutch 
about 2 foot from the spool, far enough for up and down 
motion along the spool. Felt clutch made from a strip of 
felt wrapped a few turns around the wire and then held 
in the clothes line notch of a sprung clothes peg , fixed 
to a support.
From Avo/Douglas coilwinder data, going on the lax side of the spread ,  
for the excess amount of traverse over the width of wire 
or the winding will upset and overwind without it. 
Because of dirt, non-linearity, inaccuracy of drawing die, 
enamel thickness, irregulatity in back tension etc.
50 swg 20 percent
46 swg 8%
38 swg 6.5 %
30 swg 5%
25 swg , 4%
UTS of 42 AWG / 47 SWG wire, 0.05mm diameter is about 150 gm.
UTS for 40 swg copper wire about 300gm, I measured UTS of some measured 0.09mm 
(classed 45 swg or 0.071mm so assume measured gauges in the above table)
diameter wire about 120gm (by placing a weight, wrapped in some soft felt, on some 
kitchen scales and then lifting up with a loop 
around the weight and soft anchored around a finger and half 
decrease in scale reading . So set back tension at about 1/5 of UTS 
for sub 0.1mm gauge. Thicker gauge data from 
a powered coil winder machine maker for 0.2mm copper wire back tension of 300gm, 
0.4mm 900gm, 0.6mm 1.8Kg, 0.8mm 3Kg, 0.91mm 4Kg
After a trial run or two it may be possible to decrease the traverse 
rate a bit, less excess traverse.
Incidently I think the ETA reversing mechanism may be better than the
Douglas one, for fine wire anyway. I seem to remember that if everything was
not just perfect, then sometimes with the Douglas, there would be too much of
a kick on the traverse arm and so break the wire, at reversal. I discovered
that the kick I was getting on my ETA one (in the right sense of slackening
rather than stretching, in both directions of traverse) was just due to a
slack end bearing of the leadscrew. The ETA uses a rubber rimmed cam that
engages with a flywheel to flip a dogclutch across to engage cleanly 
with the reverse set of cogs reversing the direction of the chosen 
leadscrew. The Douglas reverser throws half nuts between leadscrews 
so not as clean engagement.
A tip for coil winding fine wire. Grind/melt into the 
core of the spool , if a large one, so you make 
contact with the first layer and solder a small piece of wire to 
it. Then you can check continuity , no wire breaks in 
the coil , at any stage. I foregot to make a lead in 
of intermediary thickness wire between the primary and HT wire 
of a magneto and the wire broke at the solder point , due to 
tension building up on the winding. For fine wire robust  joins , scrape off enamel over a length so that 
you can loop 4 times  and twist together with intermediary wire fed in , before soldering, rather than 
simple overlap of wires and solder
For this winder it is convenient to fix a piece of steel tube into the drive "chuck" , centred. 
Then for different sizes of formers/bobbins , round or square , use different size ball bearing ball 
up to 1 inch diameter, between tube and axis of bobbin. Then wire some point on the 
bobbin to an anchor point on the drive chuck to rotate the bobbin. And the free spinning 
cone on the other side.

Genrad 1659 RLC tester (quadtech ) , various models 1990s
The advantage of these older bridges in a production 
environment is the auto-ranging is much quicker than 
the more modern ones.
I have the service manual with schema and repaired 
a lot of these so a lot of info on paper for 
power-up failure display codes of 
555 5 .00xx
222 4 
just a 0 , sometimes flashing 9 instead
dealt with but would have to be on a consultation 
basis as a lot of paper records. The internal/
external voltage bias is confusing at first sight 
giving wavering displays in certain settings, 
but with no appled dc V, would 
seem to be as designed. Resetting the SRAM 
by lifting the IC , disconnecting its supply 
is permissible if erroneous data gets in 
there while still operating but a voltage spike 
on the D.U.T has killed some of the electronics, 
before blowing the "protection" fuse of course.
Otherwise problems associated with dirty keypad, 
key bounce and poor 6 line indirect connection 
to the keypad. Ground down the subboard standoffs a bit to give 
deeper socket contact. Very feeble toggle action 
switch that looks like a slider switch. Can be replaced 
with an ultra-miniture DPDT toggle switch with a 
bit of hacking/grinding back and repositioning of 
the pcb terminal holes. The fans have poor 
bearings but fan can be replaced with small 12V 
pc smps fans driven off the "12V" supply cap rather than 110V.
All ICs are plated through soldering so tough 
going removing duff ones but plenty of room for 
turned pin sockets before replacing.
Added a "bund" of a large wide rubber belt 
around the test slot, glued to the pcb, 
to catch any stray wires as used in a production 
environment. One of the nearest chips is a 40GBP 
LH002C IC.

GSP Graphix 4B vinyl letter cutter
Failure to complete self test on power-up
This m/c is like a beefed up Y-t plotter with extra stepper and double belt and splined 
shaft to turn the cutter knife to match the x-y angle of cut. One revolution of 
the cutter matches one rev of the this stepper motor and the notch on its spindle 
plate gives the initial 
index position.Failed opto-coupler so failure to find index position.

IGG (going by my memory, years later) moving LED sign
I tried repairing one where damp air had got inside and corroded pins ,
mainly on the data line expander/multiplexer chip , inside the sign. So parts of letters
were missing. Removing the big chip  , abraiding pins, and soldering to an
intermediary turned pn socket worked for a year or so and then corrosion got
down inside the encapsulation - could not find a direct pin for pin
replacement, but with some research a functional replacement could probably
have been found.
By expander I think it was something like expanding 8 data lines from the
serial receiver up to 96 lines for 80 columns and 16 rows or whatever the matrix
was, so sometimes column/s would be missing and sometimes row/s missing

Ilford Ilfospeed 2001 photographic processor
Smoke and crackle coming from terminal of fixer tank immersion heater.
Severe pitted electrolytic ? corrosion of stainless steel tank around the mount of 
the immersion heater leading to small leak of solution including seeping  
between live terminal and ground leading to chemical smoke.
Demounted heater and cleaned mounts and stainless steel 
tank. Cut rubber washers and pad from lorry tyre inner tube.
One rubber washer between flange and inside of tank on 
unaffected side. Other side cut 5/8 inch hole in a steel plate 
large enough to extend beyond  porosity of the st/st. Remounted with rubber 
washer next to heater flange then st/st plate then large rubber pad then 
inside of tank.

Inflatable boat tip
If inflating near a mains supply , connect the outlet 
side of a standard domestic vacuum cleaner to pump 
up the main volume. Let run for 10 seconds first to 
clear dust .

Interphase 20/20 depth sounder and speed log
Missing transducer connector
The owner's dog had chewed the end off the lead and the owner 
had thrown the remnants of the plug and not the dog overboard.
Worked out which of the 7 leads went where.
Piezo black (RX) and blue (TX)
Hall effect Red(5V) green (O/P)
Brown and white 10K NTC thermistor for monitoring sea temperature.

Jobo CPA2 or D5270 Professional photographic processors.
7 of these m/c with similar faults due mainly  to condensation /corrosive 
chemical vapours mainly.
1	Erratic motor speed due to errant CMOS 4001
2	Catastrophic flooding in base of m/c.Replaced about 6 traces 
on the base PCB.Replaced combined motor switch and pot with separate 
standard pot and waterproof covered toggle switch as original 
switch disintegrated.
3	Slow and erratic motor drive.Replaced relay driver BC337 with higher rating 
Tr after checking back emf diode.Broken pot track so rplaced as in above example.
4	No motor drive due to defunct op-amp.Drive trannie and motor can be checked 
by taking the base of the TOP66 trannie via 1K to rail voltage.
5	No motor drive-Replace the motor switch mechanism yet again with 
standard toggle switch with waterproof cover through a convenient part 
of the top cover.
6	Intermittand motor drive with RFI on nearby radio.The 20V transformer 
primary had fault buried in the epoxy.Replaced with a 18V,25VA standard transformer 
mounted diagonally.
7	No motor drive. Motor failure due to build up of brush graphite dust from 
commutator wear and 
excess lubricant getting into one of the brush housings and gumming up the 
brush so it could not slide. To remove brushes bend back the 3 nibs on 
the back plate. Before remounting wedge back the brushes and with fine 
wire anchor back both the brushes then release when replaced in housing.
General note: to work on these machines disconnect one side of the heater.When 
finally checking without a water bath,with capillary sensor touching the element 
power up for short durations with the thermostat set to minimum and the thermostat 
should cut out.

Kestrel MIG welder
Erratic speed of feed motor
Broken pot control , replaced 2.5K pot with 
10K wire-wound in parallel with 3.3K .
 CEM Electronics board 
uses 7808,BU806,CNY17-3, LM311, LM324
MCP3021Z, GD4093

Kodak Ektapro 3000. 1992
Took apart as water got inside when unpowered
Lamp 82V , 300W, 1.1R
Removing lamp housing , press top catch and slide outwards, before removing 
top chassis cover. 
Disassembly of top much as Ektapro 4020,5020,7020,9020 service manual, 
but leave opto code disc in place. Stepper motor comes off  with top cover. 
When tx and ps board is loose remove loose loop of wires with joggling 
thru slots in chassis to give room to remove.
2 loose sw covers in top cover
remove and refit J3 and J10
200R on main pcb
L494DV5 / ST on ps
LM317T on sw panel
deliberately crushed the ferrite ring to allow reseating of  pair of ribbon connectors 
on bottom pcb
Tx pins wrt thermal device
.1,1,1.2,1.7,3,3.1,3.1 sec .4,.3,.4
4 screws holding top down 2 long, 2 short
black arm , zero lever, radial slot in small gear adjascent angle at end of flat arm 
when index crescent of  ? ? is next to the thick end
spring short arm under arm , hook long between top pcb pillar and end face of shelf supporting 
top pcb. Its only for setting up so no great force on the spring as only acting as a follower.
White arm is focus indicator, shows thru clear panel in top cover front L of lens
Spigots thru deck and hook slides over brass manual focus spindle.
Slot 2 sw covers and lens surround ring , in place, at point 
of closing chasis halves.
If you remove the carousel when not at "0"
Remove any slide in the gate , push advance sw to bring back main ring 
square protrusion back to 0

Kodak ESP 3250 copier,2010
Only used as a stand alone copier, put away for 6 months now does not get past POST. Displays error "Printhead carriage jam...."
Some sort of mechanical carriage latch is not de-latching,firmly holding carrier in place, how to delatch or whatever , without taking the machine apart. Carriage is jammed at park position. 
If only someone had taken a pic of that area, under or behind the carriage, I could perhaps fashion a joggled probe and fiddle. I can lift the top cover enough to see the leftmost edge of the carrier and nothing hooklike there.
I was wondering if it was some latch that was under Software control , unlatched in POST but they don't use batteries these days , that could run down in 6 months 
Looks like gummy ink problem.
I unclipped the ink cartridge carrier that contains the printhead, contacts to the right in pic, purpley colour is ink
http://www.diverse.4mg.com/kodak_esp3250.jpg
Then unclipped a 40x35mm simple black cover to the left of the dump sump pad. The dump pad is not recessing and so jamming the parts marked o either side of the pad, on the carriage.
Also shows 2 wiper pads marked W, with spring, that must slide over to wipe the printhead, so a complicated and compact mechanism , now how to free it up 
I've now got to the problem area. The final cog to the rack that moves the cleaning carriage is binding in too tight in a small housing around it, plastic against plastic, either plastic on plastic " bearing" or plastic sides of cog against plastic housing, presumably chemical reaction with the ink or time. It will just about turn with parallel jaw pliers but that force would jump teeth farther back along the gear chain, let alone driving by motor force. So what squirted in there will dissolve the compaction or whatever. Moving by hand via pliers , has not eased up on a couple of traverses. 
kill or cure time. I squirted some PTFE lubricant spray in where the cog/rack tooth engages. Now rack is moving with the lightest of finger turning of the cogs, will it still be free to move tomorrow? a few months down the line?
The relevant spot is on the underside , so I doubt you could squirt in-situ , without dimantling 
 Third attempt , at reposistioning the cleaning carriage in the different parts of its cycle , to get it copying again. 
Joggle left hinge prop/stop until it comes out, to allow 
removal of 3 rear sc
unclip cover over ribbons
left ribbon pulls out, right black section pivots up and ZIF
guide ribbon thru ferrite filter ringnot motor on end but peristaltic pump
1sc under and 4sc at top including near pcb to remove whole deck
2pk + ms to release peristaltic
1sc at axle and remove the rack drive.
With parallel jaw pliers on axle the cog will turn just.
Next time without dismantling . With screwdriver lever cleaning carriage rearwards by jumping the rack teeth, exposes last of rack and a tooth of the drive cog, squirt ptfe in that hole.
Will go through initial mechanical movements , without the whole top 
scanner section or its cables connected.
Reassemble with cleaner pad ,in the carrier , with registration of its 
cycle , with pad in its up position.
I managed to reinsert the rear cleaning roller section the wrong way 
round and that produces "paper jam" error also.
If cleaning carriage is jammed due to having put back sections in wrong part of cycle, just prior to the pad lifting, black lever at the top to 
manually shift.
Paper needed help into the entry slot, rollers need cleaning?
Magenta only copy output. Washed the carts with meths and then went 
through powered cleaning cycle a couple of times. Just one jet blocked, can live with that

Kreonite CPE31 Photographic processor
Blown fuses F1 and F2.
The QM connector at the rear of the control pannel had corroded contacts that had 
then overheated causing charring and adjascent pins arcing across.
A general note for photographic processing equipment:- Beware of the bleach solution leaking 
on to electrical areas as it leaves a high conductivity film when evaporated

Kreonite CPE31 Photographic processor
Pumps and heater not working.
The line and neutral fuses to the pumps,drives and 240V to 12Volt transformer 
were O/C. Nothing obviously wrong other than catastrophically 
blown fuses. Replaced fuses and all ok while in attendance but another day 
a repeat performance of blown fuse.
Suspected intermittant problem with a pump so put 1Amp 
fuse and in-line holder in line with each pump. Again occured 
but only one of these 1 Amp fuses blew. It was an intermittant 
seizing bearing problem with one pump.
Another pump problem (Little Giant Pump) on a different Kreonite 
machine was no fluid transfer because the multipole magnet had 
become unglued from the impellar and was rotating on its own.
Cut 3 pieces of silicone sleeving arranged 120 degrees apart and pushed 
the magnet onto the rest of the impellor. Trickled in runny epoxy to bind 
in. No reported problem with chemicals in use breaking down the epoxy.
Another problem with the geared drive motor-no function. There is a Klixon 
relay as a 3 tag contactor external to the Bodine Electric fractional motor 
and gearbox. A small plastic plug nest to 
one of the blade tags had dropped out allowing the blade to move and break 
the wire to the coil.

Ledu type 271 magnifying inspection lamp, Sweden
Shallow form of rocker sw. Replacement main push-push sw mounted into space or proud rocke rsw. 
Beware that the underside cover holds the lens in place
Perhaps generally for such inspection lamps it is possible to fit 22W  60mm circumference 
lamps, eg FCL/T5/22W  instead of the 98 mm one. Unscrew the clamp between the glass tube ends 
and twist so the 4 pins are in the right orientation making sure the wires to the pins do not 
cross , 

Lokata CC/ADF Radio direction finder
Rotating ferrite rod housing - opening.
Don't unscrew the top metal piece. Undo 2 screws 
to remove the farings and then the multiple nuts, bolts and plates.

Makita DC9100 Drill ni-cad charger.
Dead
The click switch pressed to initiate needed replacing.

Metz 45CT5 professional flash gun.
Stuck in one mode setting.
The plastic radial linkage under the top setting dial had disengaged.
Reconnect and beef up with hot melt glue.Unfortiunately I had disassembled 
the compound switch and PCBs to get to it.This is not necessary as there 
is enough clearance to reposition the lever now I know how it works.Reregistering 
the position of these PCBs and switch elements is finnicky.

Minolta EP320 OPC drum 80.1mm diameter. Phosphor bronze grounding contact 
re-engage by feeding some hot melt string to it via the screw hole and then hold 
in place with a small  finger tip while screwing into place.

MTE UC 15? / U15 ?, 440 50 , 3 phase contactor for a 5 hp lathe, approx 40 years old
Open circuit coil.
Bobbin and cloth and wire 74.4g, Bobbin 12.9g
Core of bobbin 0.8 x 0.85 inches x 0.9 width
Outer 'diameters' coil with cloth wrap removed 
across 'flats' 1.2 and 1.35 inch and across 'corners' 1.54 inches
Counting off gave 6,250 turns of 40 SWG, summed total because the 
varnish presumably had part melted over time and coalesced so wire 
often breaking on unwind.
Outer, approx half, about 96 turns per gram and innner section 
about 113 turns per gram (see tips files for counting-off ).
Rewind of 6,250 turns gave 876 ohms, coil voltage 240V
30 gms of back tension was not really enough for  
the coil bulk to seat well inside the spool end pieces.
A second coil rewind ( for the reversing contactor coil which had 
failed previously ) using same wire and 
40 gms of back tension gave a more compact coil 
and 840 ohms.

Nasa Video Navtex (marine radio data terminal)
No display.
Black screen except at switch-off with a flare of green.5V 
regulator OK.Disconnecting the ribbon cable to the digital 
board and local mains interference from nearby fluourescent 
light starter would induce hash on the screen.The two casing 
fixing screws at the top of case were too long and had cut 
two of the pcb traces.

Nielsen Kellerman Cox Box 
Instead of cox shouting at crew.
Failed ultra miniature 4 pin connector to headset, due to salt water corrossion . 
Binder 719 type, superglued threads. 8mm diam plastic socket , 5.5mm 
plastic socket section
Squash wire cover to break glue join to undo. 
Squash centre section to break that glue 
For socket(plastic, plug pins)  ,from large slot G,R,Y,B wires, 
G pin broken , carrying the current
Used Hughes WP 22Y 22C 00 from 1978 to replace
pins on 2mm square centres and .7mm diam
5.5V o/c +ve centre pin
B BNC outer, G inner.
If the boat sinks under the cox then this wiore must disconnect but to improve waterproofing 
used a 1 inch long piece of rubber sleeving expanded with Hellerman pliers 
over the head , rolled back to double up and rolled over the other section.
Mic box is superglued closed and not explored , maybe possible 
to wire to the other socket inside.
Gortex seems to be all that keeps water out of the mic.
Unpowered diode test on speaker line ? (stroke-sensor, speakers and charger not seen) 
1.06V each way
Main connector , presumed pinning, A gnd, B3.1V, C3.1V (B-C speakers) , D 4.6V (stroke magnet/reed relay?), 
E 6.2V charger? (remains 6.2V at sw off)
Mic diode test .6V/1.4V ohm test 1100/1400R
up to 5V ac on dvm between B-C by whistling into mic , full vol

OJ ELEKTRONIK  Thermostat (Denmark)
Elektronisk, the screw terminals ,mains and signal 
rely on one pin soldering so torque on screwing 
can easily twist/ break solder joint

Olympus pearlcorder S909 dictaphone
Audio howl from the speaker on play,FF and REW.
Worn function switch ,cured in simillar manner to the m/c above. 
No need to remove the PCB as the pins of this switch are exposed.

OPC problem on Toshiba A3 photocopier BD5120
Scratch lines on the OPC but of course prohibitive cost to 
replace. A much older (by 15 years or so) sliding top Rex Rotary 7320 
m/c otherwise worn out had a good OPC drum. Dimensions 
of the drum were the same but the original central rod was 10mm 
and Toshiba is 7mm so packed out with rolled up brass 
shim to suit. One plain circular spiral and the other flatted to 
engage with the drive pulley,both heated up and solder admitted 
capillary fashion to make robust brass/solder packing pieces.
Rex rotary m/c is the same as Mita 152 as far as the 
mechanicals are concerned.
Toner compatibility for the Toshiba BD5120 ie T80P 
OK but neutral exposure setting may shift from centre
Mita 152 OK
LZR 1200 (Dataproducts USA toner) OK
Kodak Ektaprint K Toner ok but needs 50K ohm 
or so beween toner sensor signal line and 0V to stop 
the false toner replenishment motor on signal.
Canon ,unknown model,container 150x105x60mm with one corner cut away OK but grey rather than black
Agfa X35 no good ,grey background ,not white

Philips 390 Dictaphone.
No functions
2 possible faults as neither looked right and worked afterwards.
The internal /external switch on external dc input socket was iffy,the 
owner only ever used batteries so bypassed this switch and blocked socket 
semipermanently. The mode change switch under the speaker was loose.
To get open remove 4 screws and pull away the mode swich knob. To get to 
the mode switch desolder the 2 speaker terminals.

Philips Genie 8W compact flourescent lamp
Curious to see if anything useful inside, unreliable operation.
Anyone who wishes to break into one with minimum damage to the mineral/ 
plastic casing. The previous one I broke into , the first generation, 
much bigger, ten years before used a ceramic body and thick glass cover 
to the tube.
Drill a hole into the closure recess directly opposite the logo, 
the is a recess beneath that would take a screwdriver blade to 
lever off the make once wedge catches, cut the mains wires.
Reason for failure was probably 2 of the wires to one end of 
the tube were crossed over going through one hole on edge of pcb, 
other pair 1 to 1.
10R, 1W, 2 x Si 13001 TO92 high V bipolar Tr? 105 deg C electros 1.8uF 400V, 
2x 22uF lower V, 1cm cube choke , ferrite ring, 4.7mH choke, probably 
diac as well, as "diodes" not checked
A different type GE 11TBXT 3/827 10Y, 240V, 105mA CFL gave these "useful" bits
10R fusible R
2 x BU102  high V, TO92
2.8uF, 400V, 105 deg C with nice long leads
32 V diac BLDB3 / DB3
high temp textile sleeving
4mH inductor
ferrite ring
2x 0.5R
The white cone part is one-time fitted via pawls to the main part. Dremmell
grind a small hole into the cone at the break, where in case of both GE or
Philips logo is , seems to coincide with a pawl. Then lever off with
screwdriver. Cut the wires to the outside connector or desolder I suppose.
These days fibre reinforced plastic or mineralised plastic, used to be
ceramic in the original ones, Googling for BU102 + TO92 gets nowhere, I
assume as diac in there then triacs
Returning to that Philips application note
http://www.nxp.com/documents/application_note/AN00048.pdf
I can only find SOT533 version of BUJ101AU, eg
http://www.datasheetcatalog.org/datasheet/philips/BUJ101AU.pdf
a power transistor format rather  than TO92 format with 1.5A/3A : 400V/700V
rating , perhaps for this specific use and known peak/av duty cycle then
could be packaged in TO92
Broke into a 15W GE one
240V, 144 mA
FLE15TBXT3/827 10Y
PbF desolders easy enough on the bayonet buttons.
This time 3.3uF , 400V, 105 deg C and 2x BU103A so whatever current rating
of BU102 the BU103A is likely 33 percent higher, still TO92. Much same else
inc 32V diac
Myplanet 11W, 96mA , 2008 CFL
o8W43 SU111 2700K
Touching contacts for 4 tube wires
fusible R <>.1R, 2.2uF, 105 deg C , 400V
Black 221 300R NTC 
2x 1300 3H TO126, 2x 1r SWR

Philips LFH 0510 desktop dictation machine
No tape functions
The owner assured me that he had been using the m/c and 
the jammed tape.After taking the m/c apart it was obvious that 
the tape head assembly is fixed to the deck so not a matter of 
failing to retract.The owner had 
tried to use a Sanyo cassette in a Philips machine,they are 
not compatible.

Plastimo AT50 boat autohelm
Dead after rudder hitting rock.
The back stop of the lead-screw failed breaking free and 
breaking the wires to the compass etc. Compass wires red to 
the 7808 8V rail and black to negative. This volt. reg. 
needed replacing as no o/p and the backstop anchor refashioned by 
drilling 8 holes in the adjascent plastic and holding together with 
4 cable ties plus gluing up. Replaced housing screws with small 
O rings under the heads for water seal rather than plastic covers.
To dissassemble this unit remove the 4 screws holding the compass 
to the body and disconnect the 3 wire connector. Remove end 
clip and ring on lead screw cylinder so 
the tube can pass totally through the housing. Release the 4 
screws on the end cover where the power lead goes through and 
remove all internals through that end.

Precision Grinding (PG Ltd ,Mitcham) Optidress E 
Microscope and X,Y  Tangential and Radial positioning for precision diamond tool grinding.
Long term problem loss of radial transducer reading but more immediately total loss of display 
in all traverse senses.
Total loss due to break in the 8 wire 9 pin Amphenol M9P and M9s lead. The whole 
system is initiated by turning the tangential transducer fed by this 8 way lead. It also 
caries the signals to the radial transducer. These transducers are Moire fringe 
structures. To test off machine connect 5V to positive of Tx and 180 ohm to low side 
back to earth. This supplies the 2 IR Tx LEDs. The receive side seems to consist 
of 2 phototransistors with 2 optical gratings 1 for mm the other for imperial and 
for directional sensing. The grating is observable under a x30 inspection 
microscope. These o/ps each need 20K or so to earth to bring 
down the o/p voltage from 5V. Changes sinusoidally with position. Because the radial Td is 
closest to the active grinding area 
this 5 wire cable had been wrenched and a short circuit at the point of entry into the Td 
housing which was broken also leading to ingress of grinding dust. Apart from a good clean out with meths and 
replacing with beefed up cable and new end plate (cable end bent at right angles 
and seperately anchored to the end plate, it worked again. Beefed up all theese 
Amphenol plugs and sockets which were all broken in some way, with 
cable-ties embedded in hot-melt glue. A couple of the locking skirts were 
broken but same thread as some cut-down ,military connectors so used these.
Wire breakages due to bad cable anchoring in these plastic Amphenol connectors. 
Removed the inadequate rubbery grommet/compression rings in all these connectors. 
They do not really stop the cables pulling or rotating in the barrels. 
Replaced with 3 small cable ties tightly pulled around the screened cable. One inside 
the first cover,one between this cover and outer,hot melted in and another 
on the outside again melted in with the end of nylon spiral-wrap cable wrapping 
the length of the cable. Noted the ohmage to 0V of each of the 8 internal presets before 
adjusting as per instruction manual. The wiring of the 2 separate 5 pin ("domino" rather than ring)
connectors is not mutually compatible

Protimeter dampmeter masonry prope, with sliding hammer action.
Don't use in an awkward position so the pins are other than 
right angle to the surface or the high carbon steel pins will 
shatter.
Making-do lower grade pins , presumably pucker ones have tungsten tips.
The proper pins are 2 inches long and stepped diameters of 
0.08 and 0.098 inches. Start with 2 pop-rivets with highish 
carbon pins of 0.09 inch diiameter. Remove the aluminium 
caps and grind points on the other end. Push through 
the plastic housings and place 2x2 washers in the innermost 
recesses to localise the ends of the pins

Rexon SM2150 E, sliding saw, 2005
Intermittant failure and then permanent non-saw 
action.
There is a ventillation path around the electronics 
in the handle, totally clogging the internal 
space plus the 3 switches with sawdust. Contacts had arced across 
and burning the sawdust across the contacts.
Replaced the microswitch and reconditioned the 
other two.
Uses triac BTA26-600B and controller IC marked 
FEGO 153S 0439W BA49K1
2x 15K, 100, 25R, 5.6V zener plus 
another unknown zener.
Some unpowered DVM resistance measurements at mains plug
laser on, 1.2M and 6.4M in reverse sense
1) 5M, main power sw 3.7M
2) 5M, " 6 ohm
then reversiing polarity
1) 1.2M , 1.1M
2) 1.2M, 6 ohm

Rexon SM2150 E, sliding saw, 2005
A heavy lump fell onto the control box, saw ran for a while 
when next used 
then packed up with a puff of smoke.
One of the ferrite rings pushed down and partially 
broke solder contact which then overheated and 
burnt up. Unwound a turn on the ferrite and burred 
away burnt patch. 
BTA26600(8)? / (B)? triac
62R measured across triac. Ferrite filter coil 
between motor (L) and 5,000/3,000 switch common.
100R resistor to triac gate. 
"laser" 1.5V diode test on 2 lead supply to the light
Heated and unstuck the label over the control box 
and glued a sheet of aluminium on the outside covering 
the crack and then label back over. To check out, initially, 
disconnect the saw motor and replace with a 60 watt 
lamp - glows half bright and flickery on 3,000 setting.

Ricoh Articio MP C4502 photocopier.
In a public library so 10p per A4 and 20p per A3. It took much effort to work 
out how to copy A3 "landscape" original down to A4 paper with rotation in the optics and 
only the "portrait" orientation of A4 hopper. With hindsight it is obvious of course.
Make sure only 10p in the coin mechanism.
Select Reduce option menu and select 71 percent
Back to the first basic menu, select Auto
Then press print

Rock-ola 451 , 1970s juke box
The main fault that I was asked to sort out was the owner realised that the
deck motor suspension grommets were perished.
He first tried removing the whole deck drive sub chassis but failed ,
reassembled and just removed the motor and replaced the grommets.
Unfortunately he had twisted something along the way , the mechanism
seriously jammed and he gave up on it for some years.
He (using levers) or  the mechanism  had managed to bend the the deck
chassis sublate so the extended angle plate/latch
from the disc-centre lifter was twisted. This latch then could fall and
catch at the end of the plate with the long slot so jamming things good and
proper. Also twisting the angle plate at the double cogs drive that connects
via long rod to the stylus brush arm etc. 
To remove the deck sub-chassis plate remove the 2 bolts at the 
transfer gearing housing the one nut on the stud through the deck 
and the nut on the far end of the other stud, access from beneath.
Gain access to this area by disconnecting lamp plug and speaker 
plug and lifting off the whole front panel.
Straightened things out and reassembled and that all works fine.
The chromed plate with the raised curved sloping section is positioned 
so that the chamfered lower section of the bronzy speed change jockey 
assembly moves over this chromed ridge, bit confusing if removed trying 
to find a second stud nut holdig the deck sub-chassis..
Secondary problem
Failing to grip disc off the deck to replace in carousel.
The gripper was kicking up before the arm retracts inwards so was above the
disc rather than peripheral to disc before grabbing and lifting.
The m/c has not been used for some years and and the grease in the
paired cog area had gone sticky.
Holding the gripper arm down with a bit of finger pressure allows gripper to
work.
Also using service switch to step the drive the gripper arm repeatedly lifts
and then drifts back (when motor switched off) during the gripper contraction sequence  
until the point at which the 
full lift drive engages. Squirt of WD40 betweeen the cogs and mechanism cycling  
cleared that problem.
I was looking at the manual and there are 2 ICs just marked as reference
numbers 45762.
These are pinning compatible with MC1437 or MC1537 it would seem, Motorola
of that era.
May even get away with MC1458 or MC1558.
Although not a problem with the amplifier electronics, just though I'd put
that info on the net as it does not seem to be out there, whether the actual
devices have original Motorola markings or not I don't know 
as I didn't look in the amp housings.

Rovtech Systems diving camera viewer and comms , 1999
Hirschman 4way mains connector 
ident and then clockwise of line conn
Gnd ,N/C, N, L
Remove the DVD player by removing Stainless cover - to replace move the earthing point to 
the mains connector so can then ignore the non captive nuts. 
Robust casing exploercases.com
Had been dropped and DVD problem see Toshiba D-R160 and intermittant dead display , see 
Pyle PLVW 1342R cut down to fit the lid of the case.

Rutland Electric Shepherd ESB55,2006, 0.24 Joules , electric fence, PbF
No function
2V diode test on 12V supply lines
T5 2N6027 , PUT / TYN825 TY1, SCR
large Tx .1R//23.3R
small .1R//1R
LED momentary flash at se on and sw off, normal
1.6V only across the paired up reservoir caps
2x 250V , 2.2uF yellow epoxy sealed polyester? caps used as intermediary
high DC storage in an electric fence unit. Failed as one of these caps is
fairly consistently ohmic at 370 ohm.
Now repaired 195V dc over them ,peak. 
9.9V supply and low batt LED comes on, normal use 25mA draw.
Used in the wettest part of the UK ,
Cumbria, so assumed due to condensation getting inside the cap, no direct
rain ingress traces seen.
But removing the cap casing and pulling back the leads , no trace of
green/corrossion. Heating , with low heat of a hot air gun , ohms drop to
350 and then back up again on cooling.
Failure due to damp or HV punch through? .
This is PbF solder but is RoHS a consideration inside such caps , ie
dendrite formation? If punch through then does the oscillator need knocking
back a bit, 2 presets on the board. Replaced both of these caps and another
one used in the pulse shaping drive to the "pulse Tx" , the usual backwards
mains Tx. Thinking perhaps if unit is in direct sunlight perhaps the pump
circuit gives out 300V say. Schematic not seen but HV zeners in that area
BZTD3C270V + 85C16V plus another , voltage not seen but probably 16V.
I was inclined to drop the value of the 270V one but I assume these are
fudge values, "selected on test" for nA  leakage at less than combined knee
voltage of about 300V and has been working for 5 years , then presumably ok
to leave as is , as these zeners DVM diode test at least as normal.
I've come across a near lightning struck one before and it knocked out a
custom PIC/ASIC with no work around. I would gave expected  that more likely
than high V caps failing from spikes. This 2006 made unit oddly could have
be made in 1976 all discrete , assuming a Programmable Unijunction Transitor
for the SCR  was available in 1976 . I had asked the owner about lightning
in area and he reckoned not.
I wonder what other test to determine if punch through or metalisation
creep/dendrite ? only for professional interest reasons, I'm reasonably
content to return, as is , without any mods. As lightning is "act of God"
cannot sensibly mitigate against that. One advantage of old circuits -
nearly always repeairable
I've started unwinding this and it would seem to be wide scale
electrochemical breakdown. Each turn I unwind the ohmage goes up each time I
measure it, now 700 ohms or so. The unwound film looks like a piece of 35 mm
projection film of white image and no sprocket holes. Milky clear patches,
not clear polyester, between "silvering" that is wavy at the edges,
periodicity of the half the length per turn, and a repeat strip like 35 mm
film strip, twice per turn. The silvering of each large flat face seems to
have disappeared leaving it at the tight bends, twice per turn
Of course I don't know what it should look like and will have to break into
a new, different make one, to check and also one of the other 2 of the same
batch as the failed one , both measure near enough 2.2uF.
Here is a pic.
http://www.diverse.4mg.com/duff_cap.jpg
The leftmost is the bad cap, centre an older but unused Philips also 2.2uF
250V, and the third is from the same batch as the bad, not ohmic up to
30Mohm anyway
The 2 red >s are "frames" 10&11 that show the curving and the graph paper
behind where all the "silvering" is absent, so film colours consist of
silver,  grey colour , hazy white and clear .
Centre is good "silver" and crystal clear polyester , only colours, with
alternate gutters for the edge connection, metal lead masses retained so
ragged edges from my unwinding.
Right one shows the same effect on the outer 4 frames but beyond that the
metal foil is entire, but going home  as the other.
Yellow case is the bad TC make and blue is 3 sides of the Philips puter
case. The philips actual capacitor volume is much less than the TC so epoxy
filling , inside the block casing is complete, but both the others you can
see the active film in a number of places , after removing the block plastic
casing. This epoxy?  inner encapsulation was much easier to break away than
the Philips. I suspect the Philips used 100% epoxy and the TC had fillered
epoxy so maybe making porous. No green corrossion products seen but I assume
capillary migration of condensation through the poor epoxy from the open end
with the leads or along the leads which were bright shiney in all of them.
So this will be a generic fault with Electric Shepherd ESB55 , first
becoming apparent in the wettest county of Cumberland in 5 years of use and
then other counties later presumably
I'm assuming, although no water trails seen, that it starts with capilliary
action bringing dampness between the outer casing and the capacitor proper
as in Philips and TC the epoxy fill does not bond to the outer case , it
pulls away easily.
Then something to do with the 200V, perhaps the curving edges , maximum
width at the middle of the flat faces is something to do with charge
distribution. As powdered aluminium is grey I'm assuming the grey colour of
the depleted metal is finely divided Al and the white is Aluminium oxide.
So you reckon millions of point discharges have erroded away the metal ?
sounds logical but would not the plastic show signs of roughening, I will
have a look under a microscope. The discharges have to be through the
plastic, would dampness enhance that, considering the effect starts at the
outer layers , not the inner, it would seem
burn test , this sort of discrimination
http://www.boedeker.com/burntest.htm
Along the way I seem to have misinterpreted the meaning of "self-healing" as
the plastic is punctured and then melts to reseal. But seemingly it is the
aluminium that "spark erodes" away to something, what material  ? and then
cannot discharge to that point again and termed healing.
Via microscope lots of random tiny holes in the aluminium in the bad
sections, no holes seen in the plastic.
Next thing is to identify the insulator film as I assume some plastics are
porous or whatever the term is under the presence of water/water vapour to
reduce the electrostatic breakdown potential.
revised pic on that URL
the Philips 373 MKT was polyester as was the yellow TC probably, by that
flame test.
Yellow flame , sustained after flame removal, no drips, slow
progress,brittle remains, although HCl not smelled.
This film 15 microns, WVTR permeability of 25 micron films in g/m^2/day
polyester 4
Polythene 20
Polycarbonate 170, I didn;t realise this was so "porous"
From the boating industry epoxy has much better permeability figure than
polyester.
Now just to find the quantative  deleterious effect of water/vapour on
dielectric strength.
What happens to the "evaporated" aluminium from all those "35mm frames" ?
perhaps some of it migrates and settles at the edges as a slightly
conductive coating. It was interesting that as I unwound each layer of 2
"frames" the resistance incremented higher each time on a regular basis.
The real fault of these TC is the capacitor proper had to be squashed in the
outer shell and hence the moisture path in at the centre of those "frames"
as the potting was absent at those points , the silvering remains at the
tight curving sections of each turn . I'm amazed at how spectacularly
cleanly all the Al has erroded from those frames
, the remnant silvering is finely punctured though , looking via microscope.
Once an island of silvering emerges with  a ring of eroded around it , what
mechanism ensures the erosion of the islands?

Sanyo Sanfax 100 fax machine
Symptom copy functions ok but no telecoms
Replace reed relays

Sanyo TRC 8700A 'dictaphone', 1986
No play or record - broken belt.
Replaced both belts. Quite well made inside 
2 motors. The 2 header connectors DC voltages in play mode 
and 0 is approx 0.
0,32,25,13,0
0,0,0,5,0,0.7(average),0.4,0.4

Seadoo ? Dolphin underwater scooter
No function
First check whether reed relays are functional with the magnets (magnetised 
reeds may lock on or off in theory) by firmly pushing the magnets against the body
2x ST P60NFD?? , ST LM324N ?
1N5408 S/C from owner connecting motor to battery wrong way round, original problem 
probably magnet positioning being out
4mm wide trace acted as a fuse, replaced with a solderable piece of  15 amp fuse wire
motor on 12V 12A start down to 5A

Sealine MC5500 Marine VHF tranceiver
No receive audio
LFB12 ceramic resonator gone resistive due to 
Silver migration round edge of one (thinnest) of the
4 internal resonators. For silver migration , research 
the term "silver mica disease"

Seavoice RT100 marine tranceiver
No function Rx or Tx.
This 12V unit had been run on 24V for some time.
Replace 2 1N4002 diodes near the relay,2 trimmers in
the Tx section were duff,one s/c the other broken.
On Rx the area around the TBA641 audio power amp
was totally charred,cut away the mess and built
a new o/p around a TBA820M.

Seavoice Seafarer RT100 old marine band workhorse
No transmit.
No relay click over on press to talk.
Break in wire at Bulgin connector of the mike.
Remade and filled the contacts area with hotmelt glue to act 
as cable anchor.

Sharp SF 7750 photocopier
Intermittently stopping showing fault code H4
The general Sharp reset code Clear repeat 0 Pause 14 did not work.
Near the main hinge at the rear of the m/c is a 2x9 way signal 
connector. Disconnecting this ,powering up,off,reconnecting and 
powering up reset the H4 condition.Fault was due to dodgy 
contacts on the fuser section thermal switch.As not a standard thermal switch 
i opened up the available space and put in a standard lower temperature switch 
(as not physically rubbing on the fuser roller)
I've since found the reset code for this m/c is CP0P Clear,Pause,0,Pause.Other 
Sharp reset codes for different m/cs are CROP13,C00C also sometimes these codes 
preceeded by pressing the print button.
Incidently the 
page count on this m/c is by pressing 0 continuouslly for 
a sequence of 3 two figure numbers.

Sharp SF 7750 photocopier
Don't know correct way to remove developer chamber as "access" space about 
10mm too small. Flip back green lever to undo 2 screws for slideway. 
Remove 2 paper splitters by pressing both retainers inwards and undo screw for the motion sensing one to remove.
Remove front axle anchor and bend the curved and flanged part of  the developer 
chamber access to clear the OPC drum and cog drive, cover with paper 
before removing , flange can stay bent as does not interfere 
with anything.
Remove top glass , 2 screws retain black cover
Remove fron cover by slackening one hinge mount screw.
Tip machine over onto its pivot end , fairly stable , but tie 
to something substantial , gives better working access.
Exposure lamp 7.1R in cct

Simda 3215/2 slide projector 
No mechanical slide transfer mechanism also broken focus adjust.
The main mechanical drive ,not the carousel stepper or inter-slide 
black-out function,was mechanically broken.The main cam of black plastic 
was broken diametrically initiated at the centre D shaped steel spindle.
This plastic was not fibre reinforced ;there was enough space to glue back 
together with hot-melt string and reinforce the boss with a couple of small 
cable ties.Similarly the plastic closure on the focus knob spindle had sheared off.
Made a steel bracket anchored to the nearby chassis screw and packed 
out with a block of PTFE to bear against the focus spindle.Crude in both 
cases but stronger than the original poor design.

Sub Sea Services Diverphone, diver communicator, model D8808 ? 2002?
a pic here is the only info I could find 
 http://anuncios.coyoto.es/adpics/4dc2c793f27f42db3b2ff00d0.jpg
Swamped with sea water and crackling topside to diver link. 
TBA810S , TIP121, 4016, TL071
Sub Tech Systems compatible diver headset measured 2.6R
4R each ear speaker and mic (small speaker ?) 5.2R alll paralleled together. 
2 wires only to each diver in the umbilical
replaced diver vol 10K log pot and TL071 , green to 071 p6 o/p 
purple to 4016 p1 and p2 to p8 i/p of TBA810
TIP121 4 pole c/o relay driver ?
5mV 1KHz injected onto a diver line is hearable on speaker
Battery state monitor is via LED at sw off monitoring charge in 1000uF , 
probably the 2x TO92 as oscillator. So sw off with relay active then only brief flash. 
Owner had replaced press to talk switch with standard toggle, replaced with 
a proper biased off , for safety reasons (rope or clothing could toggle sw wrong way cutting off diver to topside link)
as legends wearing off
col 1/ diver 1, diver 2, ext speaker, headset
col2/ mode (d1,both,d2),crosstalk 1to 2, accessories socket (for tape recorder etc)
col3/ speaker off,crosstalk 2 to 1, BNC isolated audio in for feed from ships PA (via the 5W 1 to 1 transformer)
PTT
LED
Speaker vol and on/off, diver vo;. 12V supply
5mV 1KHz on diver line noticeable on speaker but not headset socket/
Battery voltage monitor flashes at switch off if good. Probably TO92 x 2 until 100uF cap is discharged.
Monitor PTT is o/c and relay is off normally

Tektronix 4663 X-Y plotter, c 1980
I own one of these monsters and thought I would 
just make an internet notice. I only have a user manual 
rather than repair manual. But it may be useful for someone 
out there with a non-working one to be in contact with 
someone with a working one. I do have a 4662 repair manual 
which superficially is a scaled down 4663 but don't know 
about the electronics 4662 to 4663.

3 phase converter
Boost Electrical Engineering 3 phase converter, 1993
Converting single phase to 3 phase to run a lathe from domestic supply.
Hopefully just replacing the 415V 20A contact ratings 4 pole relay but I'd
like to know a bit about how it works , in general terms, as both coil and
power useage contacts are u/s.
2 pairs of contacts paralleled to bring up to 40 amps rating. But I assume
over time with bias to one set closing before the other one pair failed and
then the second pair failed soon after but why the o/c coil , would there be
a thermal fuse buried in the housing somewhere, lots of heat was developed
from the bad contacts
I assume the converter is peincipally an autotransformer to create the other
phase in conjunction with some capacitance. There is a boost function that
is manual biased switch at start of motor or automatic if the lathe cuts
through a hard spot, this brings in more capacitance via the relay it seems.
There is also a reed relay with a few turns of the power amperage passing
around it , presumably the auto function ,but the mystery is a small TO5
size bridge rectifier and electrolytic for a 100V or so DC supply for an
offset or something
this textually sounds the sort of thing
http://www.patentstorm.us/patents/5545965/description.html
including boost switch and
"FIG. 1 is a 3-winding induction motor supplied from a single-phase power
supply with a phase control consisting of a capacitor connected to a tap on
an autotransformer."
and
"The current in line L2 goes through a coil around a reed-relay so that the
reed-relay contacts close when the current exceeds several times full load
current, and the contacts open when the current is less than twice full load
current. The coil is selected so that the contacts open when the motor shaft
speed is greater than 80% of rated speed. The reed-relay contacts control a
solid-state-switch which inserts the starting capacitor CS between terminals
TC and T4. "
not Solid State Switch in this one
Crabtree Ceicon 41400/.-
my numbering of terminals
1 and 2 to R
3 and Bn  to 6K8 and then to coil and Be wire 
4 Y
5 &6 R
7 Y and Bk to coil
8 Y
39R to the bridge rect and 15K over the 100uF 200V cap
Replacement 4p relay 120V so added 4x10K 3W in parallel
no motor load
97V on relay coil when activating boost sw
, on release relay stayed on with 85V ac over the coil
Level 1 setting Br and Bk neons on
Sort of working order but will not drop out of boost mode. The voltage over
the relay drops but not enough , it would seem , to make this replacement
relay and so different coil demand, drop out. Will have to do some on site
fudge adjustment to the original 6K8 dropper for this relay and check that
strange bridge rectifier and offset DC function as no ac or dc without a
load motor
Is there an online calculator anywhere ? or failing that what sort of
"packing factor" for 0.08mm (+/- .005mm ) winding on a relay coil.
I have a good idea of the weight , subtracting an estimate of the plastic
former and this would give the length from the density of copper but number
ot turns ?.
Impregnated coil so cannot count-off turns
Rectangular section to coil , on the inside anyway, 14x 16.4mm , 16.8mm
width, outer layer is curved at "corners" and bulging (from the winding not
abuse) so more scatterwound than precise regular lay-up .
Outer dimensions of 21.4mm bulge / 20.3mm , or so, at outer edges one way
and 23.4/22.2mm , the other, a bit of geometry would give a good idea of the
volume of this space but what ratio of that volume would be copper and what
air+varnish, then what sort of weight would be contributed by the varnish?
coil has a burnt/sputtered patch, where hot gases from arcing contacts/burning plastic had passed through
The copper filling factor would be the ratio of copper as weighed to the
weight if that volume was totally filled with copper. The volume is easy
enough to calculate, but how to get an estimate of the number of turns from
the filling factor. The length is calculatable from the coil weight,
ignoring weight of the impregnation
In a coil winding manual I have a table
of wire gauge v the minimum excess advance per turn to avoid upsetting of
wound layers. Then sum the layers in terms of the number of turns per width
and this excess , including in a notional paper layer per turn (p), out to
the outer layer set an average value . Including p being negative (layers
settling into the gaps of the previous layer) and determine what value of p
gives the right length of wire to fill that volume, then that gives the
number of turns. At the first approximation ignoring curvature at the
"corners".
I doubt the increase in "layer length" of wire per layer goes up lineally
with each layer but again first approximation could assume so and so simple
arithmetic series summation should be all that is required. I just measured
the outer layer "circumference" ie rounded rectangle and not as much
difference as I thought, 77mm around the centre bulge and 74mm at the edges.
Hopefully I can now do the maths and then if I mess-up grinding across a
section of the original coil ,
so unable to count wire endings, then will at least I will have something to
go with . I don't know how the impregnation will react to a cutting disk ie
smearing.
For
44SWG .08mm wire I assumed the manual coil winder excess turn advance of 8
percent. From the weight of wire then the length. Then using my formula ,
came to 30 layers and a p value of -.013 mm
That gave 5790 turns
I disc ground through the coil and because I can weigh to 0.01 gm counted
off a sample of  952 turns and then ratioed by weight to 6420 turns .
Perhaps for machine wound coil can drop the 8 percent to 7 percent for 44
swg to give a better result.
So now know the inductance as well as the DC impedance (plus a useful
formula for the future , for 44 swg anyway)
Bobbin measured 14.1x16mm
width 16.8mm 
outer dimensions with bulge, 23.4/22.2 x 21.4/20.3
23.86 gm with bobbin
bobbin weighed 4.60gm
n=h/(d+p)
d=diam, h-height of windings, n = no of turns, p= "p" factor
Ln=L1+ nx8xd
75.5=60.8+n*8*.08
so n = 30 layers
p=-0.0133
vol= 16.8[(20.8x22.6) - (16.4x 14)] = 240
55swg Cu .093lb/1000 yds and 2985 ohm/ 1000 yds
so 432yds = 395m
After discing thru the coil , opened off the bobbin , tying round with 
2 ploops of lacing cord. Fanned out a small section, counted, twisted together and bent back. 
so counts of 94, 131 etc totalling 958 turns 2.81gm
remainder = 15.99gm so 5452 turns by scaling
total N = 5452+958 + about 10 broken from the burn patch

Toshiba BD5120 photocopier
presets on the HV driver board marked AC cleaning voltage, marked TR transfer 
voltage, marked B Bias voltage, all clockwise increases V
From working m/c measured with 1000:1 divider
top corona +8.3KV dc
transfer +8.7KV dc, "TR"
cleaner 2kV ac ,"AC" (measured on DC/50Hz divider and DVM, not meant for 40KHz use)
iron filing+magnet centre rod 220V dc (bias voltage) "B"
But from makers labels on HV generators
corona unit Murata MPH 2536 (uses 2SD526) 5.55KV. 0.3mA
Matsushita EUK MBB 300H
input 31.5V 
TR 6.06KV, 380uA
AC 4.65KV, 550uA
B 180V 1uA
Reposition the bottom wires outside the m/c after 
removing the double HV wire tray , to monitor the voltages.
Corona _ remove the HV wire carrier and make an insulted 
rod to push through to the back to contact the socket.

Toshiba BD5120 photocopier, 1989
Paper in feed roller works up and down but main feed motor drive inactive.
Nasty clicking noise with the top section hinged up probably slip clutch 
for optics drive of the leadscrew.
Without the black to brown ribbon to the mask motor in place  at the rear , the 
start up cycle does not get to activating white cleaning lever.
3.15 A blown fuse on rear ps , replacement glows orange at switch on and blows at the point of normal 
commencement of main drive , then the dreaded red spanner.
Overlay 33A = 33V dc ?
C2555 B686 A130 4, 78DL05AP, LM305AP, 330R, 5W
56R , 68R over green O
J27 connected, but not J28 then no powerup at all.
Single orange wire , 33A, from J28 , 1R to ground. 
3 pole DC motor failure mode?
Or internalised self destruction of one pair of drivers and nothing to do
with motor playing up?
Motor supply is 33V and 3 amp fuse , main drive motor for a photocopier, 3
poles and magnets. I have a parts mule for this copier but do not want to
take all the frame apart to change the motor unnecessarily, the driver board
is easy to get to.
Other than lubricate the motor bearings and change the
driver+speed/direction processing board over , anything else to check?
Both motors measure 1.5 ohms all 3 ways between coils , also inside is 3 off
4 pin, presumably hall effect devices, can only see the pins. The
electronics is for speed control, direction and braking. One pair of NPN+PNP
TO220 drivers , 3 pairs in all, is all-pin shorted , well 1 to 3 ohm. Failed
at switch on , knocking out fuses , no overheating or odd noises/smoke etc,
Of course may have failed at final braking of the motor at last useage but
as presumably braking is once per sheet cycle as not used for batch copying,
seems unlikely it would happen to fail on the last of dozens of sheets
Maybe something to do with lack of lubrication, seems to run quiter now,
perhaps a bearing stuck enough for overload. Seems to turn, by finger
,easier than before.
Decided to solder 3 tell-tale small 28V bulbs to each coil and ground and
power up with changed board. Overdriven by 33V at power up and half
brightness but balanced in operation.
3 coil ,30 degree stepper motor rather than DC motor to be more exact. Will
power up a few more times and then remove the telltales
The hall-effect presumably for phase monitoring and an external 12 vane
interupted slotted opto for speed and a PTO solenoid clutch to transfer
power presumably at the right operational speed. Must vary that speed for
different size paper and magnification. Anyone know where the braking
fusnction comes in? perhaps the PTO system will not disengage reliably at
speed and operational torque.
If there is no binding of  bearings etc there should be no load to speak of.
It just turns the OPC drum and some rollers that feed paper through the
system. The main load is probably the fuser roller as it is pushed against a
backing roller to fuse the toner into the paper. Worst normal situation
would be scrunched up paper jamming in the fuser section
 turning by hand it
has a stepper motor feel against the magnets and 12 magnetic steps. Ground
is mechanical to the frame if any ground. 11 wires 3 are related to the
coils, 1.5R between each to the other leaving 8 ,
the 3 hall effect have one common I could detect by probing with a needle,
probably one other in common leaving 2 for each sensing part of the Halls.
Separately there is a slotted opto.
Without scoping or strobing or something I cannot tell whether PTO is
coincident with starting or stopping. When I had the tell-tales in there
(from common collector of each driver pair to ground)
bright at startup, half bright for motor run , then off or too low a voltage
to light the lamps for half a second then a second of half bright. Perhaps
the bulbs off is the braking stage.
In normal operation you see the axial fan which is directly coupled to this
motor , ie not via PTO, kicks back before stopping.
The motor control chippery is TC9192P, TA75358, NEC uPC494C, TA7712 and a
4013
With this copier there is paper size selection from A5 to A3 nut more
interestingly reduction or magnification from 64 percent to 154 percent by 1
percent steps, so need some pretty precise control of speed. I always
thought it was feather action microswitches that set other parts of the
operation as paper goes through the machine but they may be for timeout
error determination and the real sequence process timing done via the
hall-effects and slotted opto/vane system around the motor
The motor is this, from the stock number, but no technical info known about
the internals/specs
http://www.eatik.org/index.php?page=shop.product_details&category_id=0&flypa
ge=flypage-ask.tpl&product_id=218&option=com_virtuemart&Itemid=1
Toshiba BLM-I450B3-A02 A-320 24V, 3500 rpm, 11 pin wiring
3 wires to slotted opto for 12 vane disc.
Slide belted pulley away from motor to lubricate .
PWA-C-MOJ--320 pcb board under main board at rear
J16 p3, C2655 , 3 more pairs of B1022 , D1412, Tr11 and Tr12 ohmic to 3R
pur/or/pur wiring 1.5R/1.5R/1.5R either side wrt orange , to fan motor
Small black cloth covered cylinder is solenoid , 2 black wires, to push the 
driving dog along the motor shaft to the driven dog withthe toothed belt.
Red LED1 comes on with solenoid activation
3x28V telltale lamps soldered over the protection Ds
Bright 33V at switch on , down to 15V or so ,balanced in normal drive. 
If the hinged top housing of the copier will not close , check the whole copier 
is level and that the holes at the paper outlet face engage with the cattches under, with 
plastic cover removed.

Toshiba BD5120 photocopier
Gradual loss of image to no image in about 20 pages.
Failure of the driver to the 24V motor pushing the toner from 
the toner bottle,both leads at 24V at all times.
The monitoring of the toner sensor / switch would apparently 
be 2 stage first voltage switches on the 24V line once every page 
if sensed lack of toner and then a higher voltage sets the 24V line for a specific 
time and then if Voltage not dropped back to low voltage disables the m/c.
The first stage was being triggered too late for the time slot to 
pass enough toner into the m/c to replenish,so adjustment was needed.
Also probably due to owner fiddling there would appear to be no way of 
holding the complete developer section onto the rear cog train so 
was sliding to front panel side and disengaging drive so attached a 
haevy duty compression spring that pushed against the main front access door.

Toshiba BD5120 photocopier
Total loss of image particularly at switch on. 
By switching off m/c half way through a cycle the latent image is on 
the OPC but no transfer voltage. To gain access to the HT driver board at the base 
of the m/c remove rubber conveyor belt section and dual HT wire carrier. Remove 
the nylon standoffs from under the base of the m/c. There are 4 wires that go between 
this board to a 4 pole switch assembly that is an interlock for full closure 
of top and bottom of m/c near the fuser area. This switch assembly is actuated by 
a triangular plate jutting down from the top section of m/c. This is held with one 
tapped screw which was loose. Replaced with a through bolt and double nut.

Toshiba BD 5120 photocopier.
Toner going everywhere in the toner hopper area but not into 
the archimedean entry. Perished foam that allows the toner 
bottle holder to slide into the archimedian position, so 
allowing toner to get into the housing .

Toshiba BD 5120 photocopier.
Clattering noise at switch on then engineer spanner symbol and cease.
The paper entry drop down arms would clatter up and down a couple of times 
and the cycle would stop there before moving on to the main motor start-up.
Under the toner hopper is a microswitch actuated off the cam/cog that operates 
the paper entry arms. Because of spilt toner or wear on between cam position and 
uswitch button the system was failing to get a set/reset signal from this uswitch.
Another problem - maybe preceded by rucked paper as that fault 
rectified after cure. Main fault was paper jam at O/P LED 
lighting but paper stuck at I/P with leading edge near OPC.
Cleaned rubber and clear covered rollers at entry. Also 
after cleaning the paper when fed in by hand showed less 
resistance to ingress before being rolled in.

Toshiba BD5120 photocopier
Intermittant loud squeal after 10 minutes of operation.
This fault and solution is probably applicable to all 
photocopiers. The squeal was so loud and short duration and 
no consistent part of the machine cycle. All bearings were 
lubricated and no pile of filings to indicate metal rubbing 
against metal. Just by listening I could not even determine 
whether the squeal was coming from the top or bottom,front 
or back of the machine. I obtained a medic's stethoscope and 
unscrewed the diaphragm and fixed on a length of polythene 
tube. Waving this around the squeal was loudest in the region 
of the driven end of the fuser roller and silicon rubber 
backing roller. The fuser lamp was slightly off centre inside 
the fuser cylinder and presumably after heating for 10 minutes 
was sagging and rubbing the cylinder.Propping up the two rear 
feet of the machine helped as well.maybe a bent baseplate.

Toshiba BD5120
OPC 78.00 mm diameter.
Turn drum by hand firstly if replaced to check for gap clearance
to developer drum , adjust at the front end stop anchor.

Toshiba BD5120
Document glass leading edge motorised indicator panel 
dropped into the path of the travelling mirrror.
Relies on 2 plastic pivot points so just ordinary 
leaning on the wrong area can easily break a pivot.
Make up a support from this metal (to clear the mirror) 
to support the edge of the pannel.
Toshiba,Gestetner and Mita may have badged the same machines. 
Equivalent toner for Toshiba 5110,5120,7610,7720 , Types 64, 80 Universal 
is not a useable equivalent to toner for Sharp SF 7300,7320,7350 ,7370 
both by Katun , USA. No idea if the proportions are the same but both types contain 
Styrene Acrylic Resin, Carbon Black, Polyethylene, with same code numbers 
but differing code numbers for the organic pigment component. 
The Sharp toner in the Toshiba produces too feint an image for use and 
too high a throughput/wasted toner and excess build up of a toner film on the OPC.
LZR1200 (Dataproducts USA) toner required the exposure 
setting at the lightest end of scale for image.
Agfa X35 leads to black spots of dumps of toner on image.

Toshiba BD5120 photocopier
Fails to start the cleaning? cycle, OPC does not turn , spanner in display
Rod under toner supply/ paper ingress area rotates
optics leadscrew operates
outer white cogs at hinge/rear operate but not the inner ones
white lever to scraper? blade operates
fuser heater comes on 
erasure lamp comes on for a second then m/c stops
relay or solenoid noise , preumably usually masked by the main motor noise and fan

Vibrodorm VRT/3 bed vibrator, 1996
Intermittant no control over the controller although 
red LED always flashes on r/c, sometimes worked 
if r/c right nest to controller.
All IR LEDs checked working.
Used only 6 (12) commands of a SAA3004P.
Assumed firstly it was problem with switch in 
r/c battery compartment or recessed on controller box 
intermittently swapping code channels, 
there for 2 kit use. Problem was due most likely 
to corroded sprung DIL socket for the PIC IC , 
replaced with turned pin one.


Volvo ,Sweden 1986 car door speakers
Litz wire problem soldering  between tag terminal and cone

Wolsey Spitfire AF0323,electric fence
Internally HT discharging from HT overwind to transformer chassis.
The water ingress seal was compressed so packed out to regain
a seal.The HT overwind at point of arcing was touching the chassis so
forced a steel rule to make a gap between overwind and laminates
 and then introduced some thin paxolin (FRPB) sheet.

Computers and related

Acer 950CX power supply ADP 36GB by Delta Electronics.
No o/p.
Short on the o/p,isolated possible culprit / false leads ie low voltage 
side of the transformer and one of the electroluytics. After cutting a 
couple of traces isolated to one of the diodes in the double centre tapped 
high-f rectifier GI UG18DCT. Replaced with a GI SBL1040CT 10 amp device.
Also in passing for anyone with burnt out mosfets they are 2SK2056 and IRF9Z34.

Acer Aspire ZG5
Battery UM08A71 not charging, rated 2200mAh/23Wh. Got inside the battery pack without too much colateral damage. Put a bit of current limited charge into the cells and retried but the same.
battery over ends 7.07V and individual 2.29,2.27,2.52
I did not expect a computer inside a battery. 24 pin bq29312APW and a labelled over memory? of 38 pins etc.
What I thought would be the pass powerFET was a heated fuse , marked 12AH3, presumably
http://www.cyntec.com/product/protector/download/FC543212AH3C.pdf
I've never come across the like. Looks as though the heater has failed.
So why the heater? Fast fusing of metal in overload situation if its already close to fusing? It is designed specifically for L-ion cell use.  I assmue this PC in a battery monitors for heater failure and then locks out from charging.
There does not seem to be independent monitoring of the 3 cells. There is a thermistor , well something that is 7.8K sitting there. What would be the kapton covered piece of thin ribbon going to the ,still mounted, unobserved side of the middle cell.
Seems the powerFET is 2 off p-ch , marked A2715, 8 pin devices = NEC uPA2715 ? 
foil in kaptan is individual connections, thermistor , in cct, cold 7.8K
Got further into it and made good contact with the fuse thing and its heater at 7.8R is ok. The foil plus another is tap-offs from the 2 other cell joints so individual monitoring/charging?. Another SMD 4 or 6 pins,skew view, as longish and lumpy perhaps a crystal ,marked
Q5B
H8
463 
The heated fuse principle would seem to be used as a crowbar
 http://ip.com/IPCOM/000004821# 
I'd not current (50mA) and voltage limited external charging of the battery to high enough minimum voltage. Needed at least 10.5V for this 11.1V battery. 
No ps and pc 2 orange battery blinks on power on push.
Connect ps and power on, power meter says no battery , for a few minutes and then power down and remove ps.
No ps , but battery present , 4 orange blinks on batt LED.
With pc off leave charging , for 5 min, disconnect ps.
PC starts up on battery only, so reconnect ps
11.28V on p1&5
Mechanical internal batt present usw somwhere under the spigot?
With hindsight, next time, just melt a hole at each end to engage 
some interim current and voltage limited charging probes.
normal operation orange goes to green with battery recharged
Fn+F6 to kill the display light

Now taking charge from the notebook and all working normally.
Next time a similar fault , I'll try just melting holes in the battery casing to make contact with the cell ends and not separate the casing into 2. 


A H Electronics Backpacker 1 Metal Detector.
No function,no audio output at any setting.
Both ICs had the markings ground off ,one leaving a marking of C and the other 
a final figure 2. Replaced the DC level amp marked C with a 741C. Note if the outer 
casing is too near the large 2.5nH RF choke then no function although OK before 
mounting back in the casing.This became an intermittant fault with the machine 
only functioning when cold and not when warm. Desoldered the choke and remounted 
on the other side of the pcb to cure this odd fault.

Automaston Hopping Bunny toy 
Sorry about this one.
After a couple of bars of "music" the bunny should start hopping via 
the motor driven rack and half pinion - spring return mechanism but 
just cuts out at the point the motor should start.
Probably some current monitor in the circuitry and motor start current 
too high and monitor sets the circuit to stand-by mode. Motor worked 
fine on external ps,no shorts etc. Added a 2.7 ohm in series with the motor
resistor in series with the motor.

Bios problem with old computer.
I decided to use my old computer as a file back-up. 
So gave it a present of a new 40GB hard disk. But 
whatever I did I could not Fdisk / Format to greater than 6.9G. 
My two tame pc experts couldn't help. I decided it was a 
problem with the Bios, no website for the motherboard 
manufacturer and couldn't find a site for updated bios. 
If I went into the bios / Standard bios settings I could 
change the HD parameters in user option. Too 
large a setting and available space would be even less 
than 6.9G , one time only 45M.
Gradually increasing (by suck it and see )
cylinders and sector numbers 2100 cylinders and 189 sectors 
would give 31GB and 21,500 or 192 sectors would not work. 
So settled on 31GB. Then found a "hard disk exerciser" discex.exe 
as discex.zip to check if non-standard settings would throw a wobbly 
and it didn't.
 
Another BIOS problem
BIOS protected with , now forgotten, password 
and unable to enter BIOS setup page. Changing the 
jumper settings made no difference.
PC unpowered , disconnect the CMOS battery, power up 
and reset the BIOS , switch off to replace the 
battery . While at it, measure the voltage 
and mark the date and V inside the pc

CMOS problem
At switch on flashed up checksum error in CMOS, will 
install defaults. Didn't notice any operational difference 
so thought no more of it. Then went to use the USB port and 
it was dead, disabled in the BIOS settings pg up/ pg dn to select enabled 
rather than disabled

CMOS battery heavily soldered to motherboard 
so could cause collateral damage if desoldered.
 2 wires soldered to the
existing , in place, soldered-in battery and a 3V battery wrapped up and
stuck somewhere else. 10K safety resistor put in line.
When I did that on a pc I could not be sure how the original battery would affect
things, would it quickly drain the new battery? But I did that on this pc
some years ago and the clock time still advances marginally, so presumed all
ok. 
Second occassion on another machine. To avoid loosing CMOS contents when occassionally 
CMOS would go to default and then clock running slow. With machine still powered up 
remved the case. Stuck a needle between the 3V (measured now 2.05V) and the 
plastic battery skirting socket. Prepared (filing the metal) 2 soldered and wired together 1.5V AA batteries 
and wrapping up with insulation tape, no 10K this time. 
Jumpered between case and needle for an hour to charge up the original battery. 
Safe to power down the pc and solder the plus wire to the existing battery (spot filed) , 
left in place. Tied the battery pair to the framework
.
COMMODORE A 500 ps
No output.
470Kohm dropper feeding control cct was open circuit.
NB the 3 pin regulator on the o/p side is used as a voltage
monitor.

Commodore C128
Malfunctioning key
The usual problem with conductive rubber keys.Repaired as in repair tips for 
repairing conductive pads but because of the extended resistance between 2 
contact pads on these keys used aluminium foil instead of compressed conductive foam.

Compaq Evo laptop LCD screen, exploring the backlight area for the experience
F P16 FDS 8958A 
Is there a specific name for those sheets
of diffuser/refractor plastic. Look like a conflation of "straightened"
Fresnel lens and stationary image of TV picture noise but semirandom under
x30 viewing. I've never understood why no apparent unevenness of
illumination over the height of the screen and still don't see why that is .
Something to do with , in this case, 3 sheets of different grades and
thickness of diffussion sheet?. Also anyone know how much shorter a light
tube replacement you can get away with before the edges of the screen become
a/ unviewable , b/ noticeably less lit as a band at screen edges
layers 
1. next to the glass , lightly misty , no ridges to fingernails and a semi-mirror silvery appearance 
2. Ridges towards LCD
3. Denser mirror effect , ridges to LCD side
4. Frosted finnish, heavy misty, rough face to LCD
5. Thick clear sheet with ridged side away from LCD, next to backing white
6. White backing sheet
Dismantling, remove periphery screws
beware of ribbons under and cut or lift the coppery sticky foil, white tape with shielding 
and black plastic sticky tape. 
Bend back tangs , mark inverter orientation, wire ribbons and removable metal piece 
before removing. 
Pink end of lamp remove rubber spacer/locator , don't pull the tube as the 
wire at the hidden end is looped around the end of the white plastic "reflector" . 
Disconnect the wide and narrow ribbons, mini circlip pliers to release 6  way and simply 
lever off the black filler strip for the wide ribbon. LCD 50 way and 6 way.
Hitachi TX36D79VC1CAB, Evo LCM 14.1 CTFT, 2121C, 71039, Compaq N610C, 
SPS:291262-001
Uses OZ965G, FDS8958A dual driver , 6 way conn and 2 pots , conn pinning not as in OZ datasheet
CCFL 2mm diameter lamp, uses Mashusita 4KV 10J25 High V rect
Inverter 7868825K Tx ,5R/446R
Fl tube 280mm long, 2mm diameter
OZ 5V  p16, gnd p8, enable p12, dim p5
22R dropper to oz
With 5V bench ps (upstream of  2.5A fuse) start up current .95A , up to .98A and when warm back to .95A
With middle of tube 40mm from light meter sensor gave 1200 Lux

Compaq portable 3, PC,failed dedicated ps
Someone's cherished early LCD screen computer that i would have called a luggable 
rather than portable. Previously owned by a famous scriptwriter 
complete with the drafts of some of his plays. 
For diagnosing fault powered from variac at 100v ,half normal mains for safety.
It was cycling every 5 seconds or so to kick in but not maintain supply.
One of the 2 optoisolators for the half second when ps functioning was reverse driven off. Bypassing 
the output C-E transistor 
by 5K allowed the supply to fuction but as drawing 0.5 amp from the mains at half supply voltage so 
only left on a second intermittently. Enough to show problem with -12V rail. Not the 7912 1 
amp regulator but the 3amp high f rectifier supplying it shorted.
Incidently there are high V supply rails for this sort of machine as well as 5,12 and -12V rails 
measured without load as 200v,220V and 5V going to the screen

COMPAQ presario CQ60-307EA
Laptop power supply  output plug
Somewhat chewed up, seems to be 7.3mm outside diameter, 9.5mm deep, 5mm
inside diameter , about 0.5mm pin but tip recessed 5mm into body, would it
be recessed by as much as 5mm ?
Number on ps body
PP012L-E
http://www.laptopjacks.com/
I'll try some calculations on the foreshortened image on there but not far
off 5mm recessed just going visually.
I wonder if they do a splicing kit, to avoid trying to break into the box.
Some sort of 5 amp compact connector and a 2 part black rubber formulation
and 2-part mould that fits around the cable.
The most chewed up area is the cover. 3 connections at the active end, pin
and inner and outer rings not connected at the plug. Two leads coming from
the supply but are the 2 rings connected to one supply line or just one and
which one ?
from the laptopjacks pic then recessed very approx 0.38 the length of the
exposed barrel in the skew-on view, which makes it 5.1mm recessed going by
the straight on view, so I'll take that as pin tip recessed 5mm
The minor, blue covered central lead , i think, was turned back within the
fill of the connector shroud, not connected to the pin, safer leaving
disconnected without any specific knowledge.
Just leaving the polarity of the 2 concentric cylinders taking the 4.74 amp
rating
Just looked at another known Compaq (Compaq Presario 2700EA) and DVM testing
the simple 2 way DC power connection , nothing on diode range.
no reading on R range one way
but other -2.8M on 30M range , DVM+ to known neagative and chassis ground
-260K on 3M range
and with  another DVM
-6M on 20M range
-600K on 2M
-25K on 200K
how universal is that ? if not able to power up with current limited bench
power supply, monitoring voltage.
The remnants of the connector had a black covered wire and a blue covered
wire going to the cylinder parts of the connector. I don't suppose there is
a standard for these sorts of things , but it looks as though the blue one
was not connected through to the power supply so just one lead but connected
to the inside or the outside cylinder of the connection. This is a pin
centre conductor , not a female connector.
My guess is the inner surface as then both conductor surfaces are recessed
from casual touching anything when disconnected.
not found a www ref specifically to that Compaq DC polarity.
Did just DVM the blue lead back to the mains connector and 1.3M regular
resistance back to mains ground pin, nothing showing on any other routes.
I assume the small diameter core wire and central connector pin are for
static discharge to ground, when used, not in this particular case
That 3 wire problem , not a problem. Have the compaq with me now , took dvm
readings, chassis ground to outer face of that 3 way  line connector, red
DVM
plus probe to chassis on diode test 0.63V forward V, nothing on neg to
ground . Current limited power up on bench supply, needed to switch on pc
before it would charge (maybe my inadequate bodged up dummy 7.3/5mm
connector). No requirement to have third pin connected for this use. No
popup my side of Vista password entry anyway, perhaps a Dell problem and
third wire problems.
it didn't bounce back so presumably no Vista induced hickup, so centre wire
not required for this Compaq and HP 90W ps.
Outer sheath 0V, +20V with no load , 19.1V on load
central core about 17.5V , no load, down to 2.7V with 47K load
ps unpowered and DVM diode test like diade 
and cap in paralllel. Unplug and ps reatains 20V on DVM with 
no load for about 30 sec then drops to 1V in 10 sec.
Ground to outer contact at power connector of the laptop.
Red ov DVM to ground reads .63V on diode test, no forward V on 
Fluke 77 the other way round. Needed to sw on laptop 
before charging light would come on

Compaq Presario 2700EA, 2001
service manual , as EVO N180 series
http://bizsupport1.austin.hp.com/bc/docs/support/SupportManual/c01137202/c01137202.pdf
Checking the battery as only 4 of the 5 test LEDs would light.
Remove the stick-backed plastic label and scratch back the plastic 
coatings to make test contact points. Should measure about 4.2V 
per cell, open circuit on the 4 exposed on that face 
and a pin contact in the covered contacts should show 
8.4 , 12.6V and 16.8V or so. Lacquer over the cuts.
3.5mm brass expansion insert.
Broken hinges and torn mobo to display phenolic ribbon lead.
On reassembly , loss of keyboard, 
beware of interconnect under the "Chips" board
 structural repair to the broken and
missing Aluminium extension from the friction pintle mount to where it joins
the plastic of the LCD surround . Fractured at the screw point so only half
the screw hole remains , a highly stressed point, going by the amount of
force required to turn this pintle rod.
using some expanded
aluminium, anchored into the aluminium, with a fresh drilled small hole or
two and small nut/bolts. To give a scaffold for epoxy to anchor onto. 
If I replaced the whole hinge, I'd have to find out how to take the laptop
apart to get to the other part of the hinge, C-clip or whatever is buried
inside. Anyone's guess what chance of colateral damage just doing that.
Obtain a part, without being ripped off and having the correct one supplied.
Whereas all I've to do is find a way of building up the lost few square mm
of aluminium of the hinge anchor plate and make good some of the broken away
and missing plastic of the display surround/lid. All nicely exposed and easy
to work on. Why go to all that bother if a bit of epoxy and some Al mesh/
minimal hardware/drilling is all that's required.
Original was 1x 2.5mm screw into a boss on the display outer, boss
and small part of external case broke away.
2 x 1.5 mm holes drilled through the large remnant of the friction hinge anchor
plate, 2 nuts and bolts and washers holding a returned loop of expanded
steel, ex ghetto-blaster speaker cover, of area about 15x35 mm epoxied through/around and over the scored
remnant Al and bridging onto drill-pock-marked original part of case and
into the reinforcing webs, not just flat plastic. When cured, then a skim of
hotmelt glue plus dystuff and thick ptfe tape for moulding a patch, to patch
the seen outside part of the display surround that broke away.
I also tapped the 2x 1.5mm holes with 8BA threads for the screws.
For the simpler hinge on the left, requires 2 stages as not possible to 
align and glue at the same time as on the inside. Temporary location 
"hinge" made from a solder tag, slid over the pintle and epoxied 
in place through the access from the outside with stout rubber 
bands around the whole case with clasp closed. When cured it is possible 
to add expanded Al mesh and epoxy to beef it up , knowing 
the positioning is nor correct.
There is a silly powered gizmo on the display casing that you'd have to pay
for also , I ignored that in the following. 
To remove LED gizmo, remove LCD screen and push outwards the white plastic 
logo boss pins.
Managed at different times to short the 5V supply to the LCD driver board 
and the 19V to the inverter, but other than pc dropping out , 
no damage.
Inverter uses oz960s 0.1R 592R under, 1.5A on 19V/14V rail
When working without connecting the tube, 100M and neon 
will give strong neon light at striking.
100M , fast diode 1N3068 and 220nF builds to 
3.08V dc on Fluke 77
Wiring at the inverter (pin 6 connected to 5)
gnd, (enable/brightness not checked order), 5V, 14 to 19V
19V on all the time, 5V at power up
To remove the long bezel above keyboard that includes the on/off button, 
remove 2 screws from rear pull upwards one end to unclip 
and then the other end. To reassemble 
push these ends in first. The 2x 50 way connector for the 
display board is under it, so no need to undo the whole m/c.
Someone had been in there before because the hank bushes 
to retain the main heatsink/cooler are on the wrong side 
of the board and pull out. He had drilled through to the 
rear of the casing and 4 long bolts and nuts to hold in place.
For LDC driver board downloaded SN75LVDS88B pdf 
that shows the 3 data stripline pairs and the clock pair 
and 2 control lines then 2 separate supplies and remainder are 
guard band ground lines.
Used fine coloured wire from screened 15 way cable to make 
a ribbon, more flexible than standard rainbow ribbon. 
Pulled the original phenolic off the mobo connector as 
neatly as possible. Alternately bend the lands to give 
more space to solder my ribbon to. Did inverter 5 lines 
first and then the 20 odd driver lines after checking out the 
inverter. Epoxied over the solder points and 
eventually hotmelt over the lot . 
Commoned all the ground lines at the display 
socket end for convernience. On checking out there was the 
semblence of a display, useless broken image with 4 way, H and V, 
repeat like wrong frame setting. Wrapped some Al foil around the bunch 
of lines , grounded it, and image much improved, slight yellow caste 
and bit of jitter on lock. 
Made into something like stripline by cutting a strip of fine  
brass mesh thinly coated with hotmelt and then more thin coat of 
hotmelt over the other side, so still flexible at the hinge. 
Retaining the order of "flatlink" pairs of lines and grounds. 
Worked very well , a perfect didplay image after that.

Compaq Presario 2700EA
loss of left click mouse action and loss of charging light and 
no battery charging. Mouse problem probably due to part pulled out , angled, 
ribbon at socket at the mouse end. Reinforce with cloth tape 
before reinserting. Charging problem due to  multiway interboard 
connector part loose. With keyboard and top removed and both CD rom units 
and HDD then sometimes would light charger LED or cut it by flexing 
the rear panel by holding front and back. Cental interboard connector of 
card that joins to HDD back to the 2x7 inch Al h/s cover area.. Double 
sided tape and a pad of silicone rubber sheet laid over it to give some 
retaining pressure.
Loss of certain keyboard buttons
PCI chip board carries keyboard signals to main board . The black dense foam 
holding board and hence header in place is not enough with flexing of case. Remove that pad 
and replace with a deeper disc of hotmelt glue 

Computer mouse problem
Where cleaning has no more effect on curing lack of positional 
control of the mouse.
Remove the ball and in the space above the ball mount a piece of foam covered 
in plumbers PTFE tape so when ball is replaced there is something more tangible 
than self weight to keep the ball in contact with the mat .
Or where it responds one way properly but refuses to do the reverse, a ratchetting effect, make sure 
the roller corresponding to that direction is not only clean on the roller 
but no gunge , hair etc at the " bearing " on the end of the roller. 
Failing that lightly axially abraid the active rollers with fine grit glass paper

Connector repair dodge/bodge for broken pin(s)
on those 15 way high density D plugs
on monitors etc. 
I gather this is
a fairly frequent problem - i don't
as a rule repair PC equpt. The moulded 
on plastic connectors that would be a pain 
to break into to trace the wiring and 
solder on a replacement.
Obtain 2 solder type 15 way high density D cnnectors
one plug, one socket to make a back to
back "adaptor" then plug and screw
down the socket
onto the damaged plug. 
23swg tinned copper wire is ideal to 
wire in between. 5 short and ten long 
pieces. To solder the central pins you 
need a filed down solder iron tip.
Before soldering the second connector 
(socket) reverse and introduce the free 
ends of copper wire into the socket holes 
to align the copper neatly. 
Swathe in hot melt glue to hold in 
register before turning round and soldering 
on the socket.
Firstly use a cut down
high density crimp type plug pin
swaged/crimped onto the stump of the broken pin
then soldered as well. Pack out a bit
in other parts of the mating surface 
area to allow for the swaged lump. The "repaired" 
section will not of course mate fully home 
so needs bolting down to the original plug 
with the retainers extended with small stud 
and hex stub anchors. Bathe back to back 
bit fully in hotmelt glue leaving space for more retaining stubs 
to be tightened to the pc

Dell DX 6000 printer ps , salvaged for reuse ,modified as about 42 V , 1.5A output
N1308=NCP1308, ON Semi
p5 to powerFET gate
chamfered edge p1 to p4

Dell Inspiron 7000 
To use a HD that is a snug fit in the Dell recess 
and cannot use a proprietary caddy and adaptor.
Adaptor for 2x22 way 2mm pitch HDD connector to 
the sort of 2mm pitch double sided edge connector fingers of the Dell
Remove the Dell battery.
Find 22 plus 1 DIN4162 phosphor bronze receptacle pins
Grind off the stick out retainer arms . Feed over some 
cord or something that will stiffen on application of super 
glue. Cut some 1mm lengths of 1mm silicone sleeving . 
With another pin flatten the splayed ends so you can feed 
a sleeving ring over, then open the blades with a pin to allow 
the , to be used pin , inside and let the 
sleeve back onto the pin, to get sleeving over 
the splayed out ends. Repeat 22 times. Align them all on 
a steel ruler , apply super glue to the cord and then epoxy 
round enough to hold all together. 
http://home.graffiti.net/syxygy:graffiti.net/2mm_pitch_adaptor.jpg
Next time I will change the order, a line of hotmelt string 
to keep in place, pull the cord tightish , then at superglue, 
the epoxy. So avoid the snaking cord , in that image. 
When set, replace the steel ruler with a bit of old credit card plastic 
for convenience
From turned pin DIL sockets cut away 44 pins and grind back 
the protruding rings so can slide onto the HD pin array 
without shorting. Use a scrap HD for alignment purposes , ( 
beware some pins may be interconnected).
 Grind away the crimp/solder sections 
of the DIN pins and now isolated conductors, bend a bit and solder 
to the DIL pins. Before soldering the 2 sections together 
add some Kapton tape to divide the pairs of lines of pins , 
to avoid solder bridges , inaccessible to remedy
When all aligned then add some plastic 
to fit in the outer shell of the Dell and can be glued 
to the adaptor. Inside a part mold of 
some sort bulk up the adaptor with hot melt glue or filled 
epoxy. Check for shorts after soldering and before final 
gluing and again before use.
Exploring the IDE handling Intel FW82371 

This is a 324 BGA , ball grid array, 324 solder balls
312 points come out to the periphery.
50 along each top edge and 28 on each underside edge.
tends to be outer 3 rows to the top and inner 2 rows to the bottom. Safest
to scratch away at the edge of this secondary pcb to expose the traces at
the sides, scratching-in from top or bottom may scratch through a trace.
Maybe just resin rather than glass reinforced board.
The underside line-up is more regular, tending to be 2 traces between each
grid ball. Upper edge is more erratic, the rings probably mean something ,
on some of the upper traces, but not decyphered.
I happened to have a scrap pc with this chip on and decided 
to desolder it to see how BGA soldering works.
2 loops of fine wire between pcb and chip, to slightly tug on 
so it pulls away from the board, no more than an ounce or so of force 
or the pcb tracks are likely to lift before solder balls melt.
Heat the rear with a hot air gun moved a bit circularly 
to evenly melt. 

Dell laptop ps PA 16, F970, 3A, 19V dc
Broken lead at the DC power connector.
Lucky for owner any short at point of 
breaking did not destroy the ps or pc.
Because of the ferrite ring not 
enough room to put on a new connector with 
cable protector protrusion.
Remade connection over the break, well 
insulated and then packed out with hot-melt 
covered with spiral wrap, in all enough 
to allow large piece of heat shrink 
over to cover the ferrite and the connector 
and lead to shrink down to one long 
conector in effect.
Diode test at o/p one way 0.25V or so 
and other way after charging a cap about 1.4V.

Dell PA 1900 -05D, 2002
20V, 4.5A ps
broken lead, crack open the case (ITF)
notes - LED uses light guide
remove one outer screw and lift away the Al 
to separate the heatsink parts
uses K2843, 2x 45NQ10 plus something else obscured

http://elektrotanya.com
problem downloading file. Don't click on enlarge view as interpreted as one of 
 your 2 downloads. If you have "veiw pictures" disabled in your browser then that is 
interpreted as cookies disallowed, so engage pictures before looking at the site.

Dialogue box problem in windows.
Especially freeware applications which assume 
a standard screen size/resolution, yours is smaller.
The box is too large on the screen so the bottom 
lie with OK,Close, Cancel buttons or whatever is 
down below the bottom edge of the pc display.
Pressing Alt-O,Alt-S or any such combination 
does not work. Try pressing the Tab key 
and step through to the last visible entry box 
and then one more , to the first unseen one , 
and press the enter key. Check correct dimensions in the 
My Computor/ Control Panel/ Display/ Appearance/ Screen 
area , go larger on the sliding scale , not smaller.
Returned with a site not made for Firefox browser and some dialoge boxes were 
out of posistion or something , so impossible to change or enter values and as some 
were starred for mandatory entry, impossible to progress. 
Tabbed through from the last box it was possible to enter normally. 

DR W monster display calculator
1 inch LCD display robbed for a repeater use
8 digit 7 seg+2
pins 1 ,26 and 27 backplanes
BP1 pins 2 comma/"E"/minus,3 seg A,4 B, etc repeated each digit triplet to pin 25
BP2 dp, C, D etc
BP3 E,F,G (maybe different order, not noted down properly)
3x8 +3 = 27 ways

email problem
This might as well go on the file as it might help others.
I run an obscure internet specialised site and part of the activity is 
circulating over 100 people in an email circulation list. In July 2002 
someone in the group but never identified had a rogue server/settings .
His machine/network was echoing each incoming message to up to 12 times 
a day to others in the group. 
All of them thinking the repeats were a system fault at my end so i 
revealed the hidden headers on these echoed messages coming back to me.
To reveal the full hidden header if you use Outlook Express goto File /
properties / details / highlight (mouse click and drag)  then Control+C to
copy and then paste to a text file.  
To get round the problem i used Blind Cc (BCc) instead of ordinary Cc 
and in the text of the email which contained everyone's email address within the 
group i replaced @ with ^. Apparently there is standard practise to use Blind Copy Circulate 
but no one had told me. With hindsight it now seems obvious the rogue recipient had lialinc.com 
as email address which correlated with lialserver. They had a mail box that was full 
and echoing input back to the internet and to internal servers which presumably gradually 
filled up more internal mailboxes for a runaway closed loop. The company closed 
down for holidays and the problem built up with no one to correct it.
BCc also hides each recipient from seeing the email addresses 
of the others in the circulation, not important in my 
situation but can be an important consideration.

Epson DX600 printer/scanner
Remove top via obvious screws plus 1 under 
the LH white pivot insert and side one of the RH pivot block.
Disconnect ribbons and remove main board.
No actual hidden screws for the main mechanics but not 
too obvious.
ps probably 42V, about 1.5 amp

Epson Stylus Photo 915
QE368 ink level resetting system worked on T008 cartridge and refilled with generic ink, 
yellow well and the non-photo grade wells of  the other 2 colours, heated needle to make 
injection hole and spot of hotmelt glue over the holes after.
Another time paper would jam up to the right hand side of the shute. Or 
sometimes a sheet would enter and then jam with overprinting of image as static.
Unknowingly a piece of cylindrical packaging foam had dropped in the paper entry 
slot and hooked over a finger of plastic over where the paper runs.
Foam packing/shipping piece dropped unnoticed into the paper entry slot. 
Caused serious paper jams until problem source rectified. Lost Yellow ink 
in th eimage. Ended up with Err 1Ec in the display, despite topping up the yellow 
in both Y chambers. As this cart had been reset, dug out the QE368 re-reset and the 
problem disappeared.
No colour cart , the message is Err 1C
In the process, playing around fo r a factory reset.
Press Green i button and lower LH button at power-on , produced clockwise "snake" 
symbol in "Select" graphic

Epson Stylus color 800
Something fell on the corner over the ps, 
would not index , leaving power and 2 lower LED lit 
and the top one flashing.
Paper grabber arm was not rotating.
Removed the 4 screws to remove top cover 
and unstuck and unplugged the ribbon.
A loose black cog in the space next to the ps 
and the top slotted opto mount near the ps was bent.
The black cog had burst off the end of the long black 
shaft that caries the paper inlet rubber cams.
The whole paper carrier tray was loose towards the ps 
end as the retaining boss had sheared, undoing the 
other special bolt allowed that to be removed.
Note the position where the cog fixes, indexed to the 
rod and mark both before reassembling.
Pushing the whole paper carier tray ps-wards allowed 
a peg or 2 to engage to locate and locked 
in place at the other end after replacing that screw and 
swathing in hot-melt glue.
So it would be possible in other situations to cut that 
boss at the ps end to avoid removing the ps to remove 
the paper holding mechanism.
The 3 wires go to an indexing opto on the end away 
from the ps.
The previous owner, now throwing it out, had put the printer in the loft
when the black started running out a few years back.
No trace of black on refilling and many blocked jets on the  R,G,B
Doing the utility cleaning dance made no difference.
Manually sliding the carriage over 6 sheets of cardboard soaked in ammonia
solution made no difference, same with ammonia in the docking/cleaning
hoppers.
It is quite easy to remove the large round slide-way bar on these and turn
the inkjet heads upside down.
Then putting enough ammonia solution to cover to meniscus level, the black
and RGB heads, and very lightly moving a piece of cardboard around in the
ammonia water cleared the blocked RGB jets but no change on the black.
I cut a couple of small squares of sewing/habberdashers/miliners felt,
intending to lay both on the B&W cleaning tank but unpowered moving the
carrier across, one transferred and landed neatly on the RYB (not RGB of
coarse) tank so left as is. Placed half a drinking straw of ammonia solution
on each piece of felt, moved the carrier fully to the right to engage the
tanks uplift mechanism and left overnight.
Bit of a puddle of ink in the base this morning but on running the ROM dump
print test all is working fine except two jets on the black delivery - I can
live with that.
Had been in a loft for over 3 years.

Epson Stylus Photo 890 , 2000
probably applicable to 1290 also
Hanged with "unknown printer error" LEDs ; doing the 2 button , 1 button
procedure resets something and printer goes back to normal. But for how long
? or is it a page count thing like deemed lamp life "error" on video
projectors, reset there  resets the hours counter. Or is it like the excess
toner sensor on a photocopier by monitoring for a sense current through the
rising toner heap . Is a build up of gunk on the sponge falsely saying the
dump tank is full or piping is clogged and producing the error.
Lay over lrge sheet in case of ink etc spillage.
Unclip top hinged cover
4 screws to remove top cover
unstick R ribbon if removing end cheek for better visual and spring manipulating access to dump tank area 
unclip the exit tray when end cheek is loose
mark the ribbon before removing from the sw + LED pcb and unscrew to reassemble
Removed the right cheek end and exposed the sponge area and a rocking sponge
carrier mechanism .
Rotate the large cog at L end to free the cartridge carrier
It looks as though you have to separate the whole underneath plastic section
from the upper metal framework to get to the tank
I managed to unclip the rocking sprung loaded sponge carrier and could then
see the path of the silicone drain tube and hardly a stain on the base of
the sump. Filled the sponge hopper with meths via funnel and it drained away into the
sump. Tied thread to the spring under the motor pinion so does not fly off or 
drop into the tank. There is a separate swing arm under the sponge carrier 
with 2 clips that you need the end cheek removed to be able to clip back and 
replace the spring. A wire through the side access to pull the spring and then hook 
over the end peg.
So just leaving what sort of reset does that button pushing sequence
produce ?
green (power / data) and red (ink / paper feed) lights flashing alternately
   1. Switch off printer with panel switch.
   2. Switch off printer at mains switch - leave off for 20+ seconds.
   3. Hold down roll feed and sheet feed buttons.
   4. Switch on at mains.
   5. Switch on at printer panel.
   6. Continue to hold down roll and sheet feed buttons until both green and
red lights flash together in synchronisation, not alternately.
   7. Switch printer off and then, after a few seconds, on again at mains
switch. When red and green lights are both flashing in synchronisation,
release the roll and sheet feed buttons.
8. Immediately after that, press and hold down the ROLL FEED button (the one
at the extreme left) and keep it down until the machine finishes whirring,
cleaning etc and the green light comes on steady - process takes more than
30 seconds
   9. Release the roll feed button and all should be well.

Epson Stylus Photo 915
To get to the underside of the carriage, remove both carts. 
Remove both top clamps , remove ribbon cable cover by flipping 
the unseen clip at the rear. 
2 sc at the botom and sub assembly with ribbon will lift out. Washed 
underside with meths.
Trying to stop problem with serious colour imbalance
Lack of Red and excess of Yellow
A perfectly adequate get-around is +30 percent of Red and 30 percent less
each of Green and Blue in the graphics package. It all started with a postal
packing foam "doughnut" dropping unnoticed into
the paper entry slot and numerous paper jams before finding the culprit. The
ink-jet cartridges are re-set , re-used/ refilled  multiple times and
refilled and reset since this foam business. Built-in "cleaning" and the
other maintainence routine has made no difference.  The colour imbalance is
stable and invariant over perhaps 30 sheets and 4 separated sessions.
Each time a colour or black is refilled, use the resetter , seems to have removed 
this imbalance problem.
Out of curiosity for when the battery fails  looked indside the resetter 14 pin micro EM78P153SNJ , 10x1K, small C and 10uF 
, LED and 3 x button cells
Unfortunately seems to require resetting the colour cart each time before using. Switch on 
, press maintenance button to shift carriage and leave with mains on while the cart is removed and reset.

www.eserviceinfo.com
Their search is organised as OR use instead Google implied-AND instead
term1 AND term2 site:eserviceinfo.com
in searchbox cuts down the excess spurious "results"

Favicon creation
No good just creating a 16 x 16 pixel GIF and changing .gif 
to .ico as the first 3 characters etc in the file remain set to "GIF".
Easiest solution I found was to download small viewer and file converter 
IrfanView , load as gif or JPG and save as ico.

Fresco Logic FLUSBVGA-1.1.216.0 
USB to VGA adaptor
software on CD, copy to thumbstick into some subdisrectory.
Takes 5 minutes to install, requiring access thru firewall, and confirmation clicks, 
then restart pc

FSP 250-60GTV atx pc ps dead not even a flick of the ps fan
Chicken or egg
There are 3 HF transformers and an opto-isolator bridging the HV and LV
sides.
The startup oscillator, on the HV side, is a KA1H0165R whose o/p crosses the
divide to give a Vcc of 12V on the main SMPS controller a KA3511DS, both
datasheets downloaded.
This controller o/p should then cross the divide to drive the HV side main
oscillator to drive the main transformer back to the LV side supply rails.
The SMPS "flag" output pin goes high meaning no-function and no o/p pulses
for either push-pull output of this KA3511.
As the KA3511 monitors over-voltage, and more to this point, under-voltage
of the supply rails , how should it ever get started if it never outputs any
main drive pulses, so no main supply rails.
I can see the one large and two small HF transformers tied up with the above
process.
The opto would seem to link, DC fashion, from the 3.3V rail back to the
feedback pin of the low power startup IC to shut that one down.
This IC, with the reservoir caps, will keep the main smps IC running for 10
seconds after removing the mains supply (in the present unfunctional as a ps
state)
With the KA3511DS connected purely from a 12V bench supply
5.6K from 12V to pin4
REM low via 50R
pin 7 and pin 9 then 3.5V
Q1,Q2 both still 2.2V
did not get around to fudging 3.3,5,12V on p15,16,17 as realised 
it was a problem in the wiring to the pc front sw not a ps fault.
20 pin ps connector P1, pin14 (green) was not going to 0 
was the trouble, there was the standby 5V on pin 9 (purple) 
and no ps fan on in standby.

FSP black compact plug-in dial up voltage power supply.
Beware the polarisation chamfers are not large enough on the 
DC plug socket connection and it is easy to place in the 
wrong way round

HP PPP012HA
18.5V 4.9A ps
Case cracks open if squashed in vice over mains inlet end 
Make sure the copper tray is bent back to contact internals thermally if  
the sheet is bowed in vice-squashing.

Irfanview
Beware of changing JPG compression factor and then not remembering 
what it was as this app. does not remember and 
sets the default as the previous useage. 
If likely to be changing this factor then keep 
copies of previous before re-working an image.
To copy/paste and merge. Select area of one image , copy, 
then open similar area box in second image, paste, adjust 
perimeter to match then repaste, repeat until fitting 
and then cancel the box to fix in.

Indexing large pdf files on CDROMs.
A number of CDroms at a radio rally being almost given away 
as there was no externally accessible indexing - you could not 
see if an IC was included without loading up each CD.
Widows explorer can "see" specific words in a pdf file.
FoxitReader pdf reader to read whats there, where the index/cross references
were.
pdf2text to convert the first 15 , 50 or whatever pages to text.
It loses spaces but that doesn't matter in this use for keyword indexing.
Unfortunately 2 pdf files  were 1,500 pages long with the index at the end
so had to go away for 1/4 hour and then delete 1,450 small text files as you
cannot preset pages on the free pdf2text version.
Then used AF5 batch file-renamer to put some structure to the file names.
(aside, A.F.5 is useful for creating a text file of a large batch of file names - 
use explorer to pick out files and copy to some directory, 
AF5 /Add/ mouse click [move cursor] shift+mouse click [to highlight blocks of files] and then  [select all/batch/save as ] to save as a text file)
So I won't be throwing them out now they are usable.
No wonder someone was disposing of them at a radio rally for next to nothing
as useless before externally indexing as plain old text that is easily searched 
before loading up the relevant CD.

Javascript junk
Where they try to force you to use it so they can propel more junk in your direction.
Where you want to repeatedly access a site, and without Java they lock you out 
or page latches up/ no hotspot activation etc. First time of use , engage Javascript 
and then on returns without Java , just use the URL box , backspace deleting the URL and 
letting Windows recall the sub-file idents and you select the relevant ones

JBL Creature, pc/ipod 3.5mm stereo, line level input, amp/woofer and satellites
How to get inside the casing?
8 screws under removed and I've got a plectrum between the 2 main case parts and can see presumably fixing catches but soon I'll be splitting the case, rather than it coming apart. What looks like hidden fixings under the label are moulding injection/blow-hole marks.
Very low level sound on all 3 speakers is the problem once I get inside it. 
Initial separation point along a side , not under the logo , that I decided on. Internal corner posts seem to be the holding points, screwdriver blade at those points and prize apart.
Full metal, not matalised plastic knobs, are glued to the pot shafts, small blade digging into the metal at the lowest point in the casing hollow and twisting against the plastic case, forces them off.
2004 is the date , will work on the electronics tomorrow 
Must be mark 1 version, not recent 3.
0.1V 400Hz on combined pins 9&16 of the  TDA8512J produces 1.7V over 4R speaker so that presumably is ok.
Twizzle sticking elicited nothing.
+4V and 26.6V DC rails looks a bit odd but only one main cap, maybe the 4V should be 9V or so
Then to find what the Q12 N448AB transistor does, also on the heatsink.
Why some digital chips in there?
This funny bell shaped thing was working, put on an indoor shelf for a few years and then next to nothing when brought out of retirement. 
main caps ESR'd ok. The 4V dc is probably mains ac sampling for some compensation purpose. What is the 4011,4013 and a couple of other 14 pin digital devices for? 
Looks as though a case of forgetfulness and RTFM.
Put aside as one satellite had failed, forgotten about, notorious problem with these , glue fails and the speaker magnet is pulled to the front by the suspension and then cone movement is bottomed out, so just  amechanical problem, plus bad ESR cap inside the orange satellite at some point.
The digital is for the vol control , that is stored NV, and still set as low years later. Control is via 2 little buttons on the "orange" failed satellite. Bass and treble pots on the main unit 
Broke bush off treble pot , not realising the matal knobs were glued on. 
Small blade against knob and case will remove with no obvious damage.
Bent back top housing of hte pot plus 2 small cable ties around.
.1V 400Hz on p9/16 of TDA8512J
N448AB Q12
Gnd connection via Q12(C?) and CEO behind h/s
orange sp 3.1R
ST 339 
H449(M) hundreds of ohms ESR
for digital 1000uF 16V,105 degC, Lelon green cap.
Don't put the sw cap in place before reentering pcb, or will 
break the on/off switch.
Found a different type that would fit in there , wired and hotmelted 
into place and use same knob.
Resonant noise on one speaker, added 2 thicknesses of bike inner tube 
under the plastic wedge closer, also added to the othe rone

Liteon 8x external DVD/CD W-R
Failure of pc to recognise it. USB is underslung and difficult to see whether the lead 
is fully home and seems to need more force than usual for a mini USB. Leave the short 
lead with mini USB permanently connected.

Microsoft Office 2000 installing on Windows 7 problem. 
Failing to install and a Q: drive problem. Uninstall , change self-chosen drive to 
c:\ Program Files\office 2
and install using the custoim selection option

Outlook Express / usenet problem
Failing to update a newsgroup (check via dekrailer / google etc)
Unsubscribe the group, close OE, open OE , and then in list of groups choose 
GO TO the group rather than subscribe , then subscribe when requested later or in properties

Problem downloading emails from freeserve/fsnet, maybe orange/wanadoo.
If you get error message
"... Server Response: ' Disconnect because authentication to back server is too long', Port: 110 ..."
Then on repeated attempts you get 
"... Server Response: '-ERR [IN-USE] Unable to lock maildrop' ..."
telling you you have the wrong password, but you know 
full well its a problem at their end.
Especially if still a problem trying to browser access via likes of 
http://www.mail2web.com/
Firstly go to 
Tools/accounts/mail/properties/advanced/server timeouts
and increase the time.
Secondly
Leave a gap of about 3 minutes or so before attempting 
to repeat access, go away and do some browsing or something else in the meantime.
Makes no difference , what time of day or night, same deliberate? system delay, 
2, minutes is too short and 4 minutes too long. My guess is that so 
many servers are now linked into a chain with company taking over company 
etc. Towards the end of reject boxes, the time gap between sending and 
reject reply may lengthen a bit before finally being accepted.
Early Sunday morning is likely to be more successfull than 
during the week.

Praktica SlimPix 5200, USB symbol on the cover
Smaller than the smallest USB connector I've seen up to now, on compact
cameras etc.
The connector must pass through 5.8 x 3mm hole in casing. Not trapezoid in
cross-section, but rectangular with 2 chamfered corners on a long side
As recessed in its difficult to see as so small , perhaps 8 ruts with 5 or 6
conductors. But smaller than the USB mini AB type. Under 30x as end on can
read 2.2 for the electrical type of USB 
Maybe similar to Konica/Minolta DiMage E 500 camera . That tip
measures  about 2.1mm by 4.9 mm.
It has five gold fingers (probably two are power-ground) with
asymmetric placement (1 x x x 1 1 1 1 ).
I'll go with the name micro USB but until one appears a matter of 
removing the SD card and placing in another camera as a card reader.

Foreign unicode script on a file which corrupted the Google cached version 
of otherwise English page. 
I downloaded Hex Editor XVI32 from
http://www.chmaas.handshake.de/delphi/freeware/xvi32/xvi32.htm
That allowed me to remove the 2 characters ÿþ / hex FE,FF / ASCII 255,266 / y diaresis and p with
ascender that clogs up the front of the file, which you cannot see let alone 
edit out in Word or Notepad.
Apparently this is appended to denote the file contains unicode, 
the BOM Byte Order Mark and also Zero Width Non-Breaking
Space (ZWNBSP) . Google cached interprets this as inter-character spaces throughout 
the cached version and consequential loss of HTML action. The preview pane on Google 
is also corrupted because of the spaces mangling HTML. I'm surprised there is 
nothing on Google FAQs pages about this. Putting "ÿþ" and "h t m l" in Google 
produced 206,000 hits. Randomly selecting 5x10 of those showed 44 were mangled so 
perhaps about 180,00 such affected files.
 With Hex editor also "Replace All " inter-character 00 to (blank/empty) which also 
reduces the file size by half. 
Then a matter of converting the foreign code characters like hex
code [ 05D2 ] to decimal/decade code [ & # 1 4 9 0 ; (no spaces) ] which Google Cached seems to like and also
browsers. For smallish amounts of text for conversion: - in Word convert all 
end of line ^p to * , to concattenate and then break up to lines of about 100 characters. 
Submit each line in turn to Google ( much more than 100 is a Google illegal op) 
and it returns search string as &#....; form, highlight and copy back. 
In Word convert back * to ^p , saving as non-unicode text in a non-unicode HTML file 
and compare the result when viewed on a browser with a .png, .gif , 
or .jpg form of the script to check. Then add to English file.
For a load of foreign text use the block routine in XVI32 
and copy Hex to Word as a .txt file after removing FE,FF and converting all the 00 to 0D0A 
and any spaces/punctuation to 2020 or whatever as 4 characters. 
Giving a file of lines of 4 characters after converting 0D0A to ^p. 
Then make a macro for converting adjascent 
quad alphanumeric characters to decimal numeric. Finally changing ^p to ;&# and tidying up punctuation etc.
I used this Yale file as a model which reads correctly as English plus foreign script on a browser and is
cached by Google correctly
http://pclt.cis.yale.edu/pclt/encoding/
and a bare minimum of HTML eg not even LANG designation.
So with hindsight just save the foreign Hex text as unicode file and convert 
to decimal/decade form before adding to full English file and then can continue to save 
as ANSI and retain correct caching of HTML on Google.
For anyone else with this problem but where they have no foreign 
text on their file and accidently saved their file as Unicode.
Without a Hex Editor you will not see the ÿþ or double zeros that Google sees.
Suggestion: rename your file from XYZ.htm to XYZ_old.htm
View it in Internet Explorer and click View / Source, 
"Select All" the text and copy to notepad and name
the file XYZ.htm saving as ANSI and not Unicode.
If you want to check the file then download the XVI32 hex editor
 ( link above) - its only about 500KByte so
only takes a couple of minutes and compare the two versions of your file.

Google no longer indexing files in list form
After 10 years of indexing my files , only more text like files
continue to be indexed, same with MSN etc but not Gigablast, Ask etc.
 What is going on ?
 If you put a few words like
 stilsons pilewound
 into Google you should find my UK/USA tools translator site
 http://www.divdev.fsnet.co.uk/tool_terms.htm
 but Google is no longer indexing it.
If you know what site its on ie adding 
site:divdev.fsnet.co.uk
in the searchbox then the results appear as
These search terms have been highlighted:  
These terms only appear in links pointing to this page: 
or
These terms only appear in links pointing to this page:
despite the words being in the text of that page
I've identified some issues associated with the PRE function and will
address those first as it is, in my case, affecting files that are basically
listings. For my own convenience, looking in a text reader most of the time,
I wanted the files to be readable in a text reader and browser
simultaneously.
ie not a string of characters with no end of line characters. line breaks
and cluttered with markup characters.
Another point is my files have basically had the same structure for 10 years
with just the odd additions or deletions. Somrthing changed with
searchengine acceptance end 2007/start 2008.
Returned to Google indexing after a month 
and a number of crawl visits.
Turned out the BODY html function was missing, META statements 
should have been in the HEAD section and the problem with 
the PRE function was that there were HTML functions in 
between PRE and /PRE which is not allowed.
Eventually checked out with 
http://www.htmlvalidator.com/
after adding html 3.2 !doctype statement at top
but not via
http://validator.w3.org/
as too many URLs with ? etc in the URL

Hi-Grade LM 156$, 1995, pc CRT monitor
Wavering brightness , going up and down fairly slowly ie thermally , with basic 
image itself rock steady.
Remove the 2 obvious rear screws and then prize open at the top 
where there are 2 square notches. Retaining flat-U clip about 
1/2 inch either side of the notch and 1 inch in.
Cleaned the CRT pins and redid all the CRT socket 
soplder points although nothing obviously wrong.
So much for thermally derived fault, looks as though it was 
poor plug/socket connection on the back of the pc, some 
capacitive effect slowing things down ?

HP Deskjet 660
How to remove plastic casing.
Rear panel covering main circuit board - push a rod into each of the two 
holes in the baseplate to release that part.
For rest of cover first remove the small half inch strip at the left hand base by 
lifting and pushing to the front of the m/c,you can see the "hook " pieces 
that lock this fillet into place by removing the paper carrier and viewing 
into the left side interior. That unlocks the main cover which you release by 
poking a screwdriver in the slots in the base plate.

HP Deskjet 720C no paper roller motor action.
Someone had managed to wrench one of the 2 pivotting paper guides. 
These have "cam" end actuators and one had ended up above the 
actuating arm that runs parallel to the main paper rollers. This could 
be wrenched back again but the misplaced cam had jammed the secondary 
cog system and stalled the drive motor and knocked out the driver chip.
Motor OK but 44 pin plcc Texas SN104558FNHR running hot and no 
output drive to the motor. Could not find a replacement so no repair.

HP Laserjet 4P
Toner refill hole melted in with soldering ron tip on 
top flat surface with the temperature moulding to the right of it.
Direct mid centre of this flat area hit some aluminium 
foil ? to the body side of the hole, perhaps 
better position 5mm further from body of cartridge but still 
centred in that clear panel section, so about 1/3 from left edge

HP Laserjet 1020, 2007
Maybe full cartridge, certainly a lot of toner left. 
HP Q2612A, 12A weighed 611 gram cartridge and toner.
various www sources gave
Dimensions W x D x H	357 x 105 x 140mm
Empty weight	605g
Weight	690g
Dimensions W x D x H	357 x 105 x 140mm
Empty weight	685g
Weight	700g
Dimensions W x D x H	357 x 105 x 140mm
Empty weight	685g
Weight	690g
So not much help
Added a mechanical page counter, plenty of space inside end cover.
Cut the milometer section 48x22x30mm off the solenoid part of an 
impulse counter. Hole through side to take some tuner dial cord 
to a PTO solenoid or whatever engages once per sheet delivery. 
Cut a slot and some more shaping to a nylon rod to extend 
the action out by gluing to the pole piece and 
then the cord glued to that with a spring and adjuster in the 
cord. Other end of cord glued part way along 
the ratchet arm for about same actuating movement there 
and the solenoid.

Hyundai HMM413 VGA monitor.
Loss of image but would momentarily flare to proper brightness at switch off.
Mechanically broken preset marked sub-bright at rear of casing.
Replaced 500K satisfactorily with a 100K preset.

Iiyama Vision Master Pro 17 monitor
3 unlabelled buttons on front for preset settings, call A,B and C
Press B or C to get to category
Press A to get to black slider bar
Multiple pressing of B to lower or C to raise value

Inno3d  tornado geforce4 titanium video card
Low resolution text screen reads alright but after switching to Windows 
high resolution breaks into an overlaid multiple image of wrong line rate.
Bearing on the fan was knackered and would stall.

JPG problems of apparently truncated jpegs
They display the perhaps about 80 percent of the original on
Firefox or IE (same cut off point in each browser) but you can't save as jpeg. You go through the motions of
saving as jpeg , no error message appears, but nothing saved in the relevant
directory. Other than saving to file as a rather useless .prn, printing off
the screen and scanning in , or photoing off the screen and montaging,
is there a work-around to this.
One of those mysteries of the internet/applications.
The emailed files were to a browser based account of mine. Downloading to firefox or IE produced truncated but
viewable images that could not be saved to disc.
Forwarding those attachements from browser account to my main email account, 
with filters temporarily lifted to download, then saving those attachements
to disc as jpg,  then all 5 were readable ,complete un-truncated pages. I've
never had to do that before.

Kensington lock , well using the hole as an anchor point for loosable Dongles etc, rather than for security
3 3mm nylon cable ties. Grind back the 4 edges about 1mm , farthest from the main tie, on 2 of them. So they are narrow 
enough to pass through the hole. Introduce 1 and turn 90 degrees to lock into  position. Repeat for the second one , the other way about. 
Then cut out the pawl in the catch of the third tie (assuming you may wish to easily remove this 
anchor at some point), and cut off the straight piece. 
Slide this down one of the fitted ties . Then a Hama bead or similar tight fit slide 
down the doubled-up ties, Introduce a ring or some thread between the ties. 
Push another bead down and cut of the excess straights. Cut a pointed end off  one of the 
straights and push in between the open ends of the cut ties to stop the beads sliding off (again 
assuming you wish to easily remove the anchor at some time

Keyboard shortcuts
Ctrl and A to highlight all
Ctrl and C to copy
Ctrl and V to paste
If the right click function of a mouse fails try Shift and F10
Alt and F4 to close the current window

Modem - loss of modem speaker
As a result of a rogue dialler the modem speaker was disabled 
so could no longer hear the FSK dial out and ingrained familiar 
and so consoling distinctive beep/chime response.
Disinfected but speaker still switched off regardless of what the options 
about volume were saying.In 
Control Panel
Phone & Modem Options
Properties
Advanced
Extra Settings was 
S7=240
I changed it to 
S7=240,M1
then Change Default
and 
produced SD & RD LEDs flashing and "connection was closed".
M and S are Hayes modem codes.
Removing the comma limiter leaving just
S7=240 M1
brought the modem speaker back to life and normal service resumed

Modem - Failure of external computer modem.
Resulting from interupted power to the modem. After sorting out low voltage 
supply ,LEDs lit on the modem but no one was home. Computer unable to 
make connection to modem. Standing 50V on the phone line OK, on opening the modem 
there were 12V,-12V and 5V rails so probably OK. Put an RS 232 LED checker in line with the 
pc to modem lead. From the pc active negative on lines (pin) - TD (2), RTS (4) and DTR (20)
and on the modem on its own active positive on CTS (5), DSR (6) and active negative 
on RD (3), DCR (8) ,so all ok so far . No flickering LEDs when clicking to dial out. 
Windows 2000 has a failing (feature) of "disconnecting" modems 
if found not powered up when the pc is powered up and does not advice you of this when repowering the pc. 
Windows 98 does not have this "feature".

Mouse troubles
Fails to move cursor unidirectionally
That is moves easily ,say,l-r but not r-l or up-down 
but not down-up.Reduce the spring action on the third 
dummy balancing rotating wheel that pushes against 
the mouse ball.

Mozilla Thunderbird and Windows 7 and usenet eternal-september
Inability to use a pre-existing account as going around in circles and not being able to 
enter the password . Only e-s newsgroups ar eavailable. 
Delete the account that you've failed to hook up with.
Start again making a new account in TB and making sure that in
 Account Settings/Server settings, "Always request authentication when connecting to this server" 
is ticked , before attempting to connect to the e-s server

TBBmeter wifi/mobile up/down data transfer monitor and alarms
I could not get it to run a proper daily alarm . Had to settle for 2 days with twise the 
allowance I wished for plus a bit of headroom
Set ENABLE
Set start DAT and TIME
Duration 2 day
Repeat Daily
Retain (Reset on start)
Trigger twice daily + headroom
Visual + sound file
Runs in 2 day blocks 
Check "daily" total in the "Running Alarms" panel
If the alarm is triggered then go into settings and up the trigger level until the start of the 2 day cycle again, 
to avoid repeated alarm
For the monitor reduce the size of the window, position at the top of the screen, opt for 
inclusion on start-up menu,  select stay on top , and change opacity to 40 percent or so

Thunderbird , deleting usenet messages
 Tools > Options > Advanced > General, and click on Config Editor.
"Here be dragons " warning
 In the Config Editor,  news.allow_delete_with_no_undo
 Double-click on it, to set the value to true.
To select multiple threads to delete, open complete thread if  long , or just one message in the thread will be deleted in each block deletion
click to highlight top one in the index
Press keyboard keys + then move cursor and click to highlight bottom one
or Press keyboard keys + then  or  
Right click and select Delete Selected Messages

Thunderbird, copy folder, as basic Win7 is useless for this sort of thing, 
use SearchMyFiles or something instead of win7 non-explorer
C:\Users\\AppData\Roaming\ folder (Windows 7/Vista) 
    Go to one level above your profile's folder, i.e. to %APPDATA%\Mozilla\Thunderbird\Profiles\

AppData\Roaming\Thunderbird\Profiles\fg5jb???.default\Mail\Local Folders
??? = disguised

    Right-click on your profile folder (e.g. xxxxxxxx.default), and select Copy.
    Right-click the backup location (e.g. a USB-stick or a blank CD-RW disc), and select Paste. 

Nero 6 install of CD writer
CD writer not recognised, the relevant last page before 
is whited out in the drive pannel and "Burn" button 
greyed out.
Go More/Configure/ expert features
Tick "Enable all suoorted recorder formats 
for the image recoreder".
If the apply/ok button is missing below the bottom 
edge of the screen press ALT-K for OK , 
other panes it is ALT-O

Nintendo Game boy DMG 01
One line and click on switch on then a few lines and another click then nothing.
The DC top DC converter seemed to be working 6V to 5V and -26V.
Did all usual checks making and breaking connector,switches 
etc ,reassembled and worked.Possibly poor ps input socket as similar 
result is obtained with using a current limited 6V suply as original fault.

Nintendo Gameboy DMG01
Zilch on battery power
New set of ni-cads had slightly smaller positive 
end terminals and not engaging with the contacts.
Packed out behind the interconnect busses and removed 
part of the plastic surround to the 6V terminal to make 
consistent contact.

Nokia ACP-12X , ps , 2004, 5.7V 800mA
other numbers 06752964409049L392Y2 
2601 0052699
2x Torx then thick blade inserted to open
no load output 6V
pc123 opto , 400V supply cap , ST 1NK60Z  TO92 600V , .25A Tx mosfet driver

No name Mini Digital Camera, pcb overlay 2007
F2.8, f3.9 on lens surround
Disc driver with Arcsoft and PCCAM 
Black disc to front surrounding 2 character LCD and 2 rings around mode sw and lens
This disc needs prizing off to reveal a screw to before opening 2 parts of case
2194A2 on pcb , ST 4 on lens, cut into glue spot to try turning lens, then locally heat with 
soldering iron - worked, to free lens screw.
Uses HY57641620ETP-H and powerfet? marked CA33P/ 733D7
Bump on LCD edge is to the left, when freed from the casing recess, zebras will 
easily drop away from pcb lands.
No micro A-B USP lead available. Wired up to 5 way part of DIL turned pin socket, 
4 active 1 for polarising, fitted to outside of case. The 4/5 way turned pin pis section 
used as a plug wired to a salvaged sttandard USB

Onetel dial up
Maximum file upload/download time is limited to 2 hours maximum, 
connection cut after 2 hours of use

Pogo/ FW7207/05 optronix smps
squash ps on long dimension in a vice enough to get something like 6 inch
steel rule in opened up gap, then foot rule, bit deep for screwdriver blade
. Plastic weld bond line is
about 5mm from the main label face, not near the the join line, so you have
to press in deep with the rule before twisting/levering/pushing. 
The 10V cap would need replacing too for different o/p V
16CT2100 , 2SK2700, TH 2028.SA(8P)[9931] ,z60N (4p)
Added 27R over o/p for testing, odd seeing 4 band SM R
Adding 510R over 715R next to the rod D and 2.6V drops to 1.47V and o/p
drops to 4.11V
 From the TO92 KA431 shunt regulator data.
5.1V = (1+715/681) *2.5, how high towards 37V could you push it?
 and uprate the 105 deg C 1000uF ,10V
electro, probably split (space) to downstream also.

Potrans ITE power supply for Time Cybernote 953E laptop
Dead ps due to short in DC lead at point of exit from 
the ps box due yet again to over tidy owner (always seems to 
be female) wrapping the cord tightly around the ps case.
Replaced with high current dual lead rather than co-ax

Relisys TE770 monitor
Only 3 years old, LED blinking and relay clicking on and off.
The "on/off" switch short-circuits the mains 
SMPS side of the opto isolator
to switch "off" the monitor, bit unusual.
Stable 7V or so on the driver side of the opto-isolator
and a 7805 gives a stable 5V for the micro area.
Failed to repair, the NT6827 OSD chip was 
loading down the 5V line. Cutting the rail to that 
chip meant the monitor would power up as far as high V 
generation but nothing on screen. Forcing the 3 RGB 
lines from that chip made no difference.

Remote linking problem.
If a host platform does not allow remote linking to a picture , for 
which you have the URL but you don't know which page it appears on. 
Remove the directory + file name + pic name  from the URL 
leaving just the site name, enter that giving the index page and then 
retry the full URL for the picture.

Rogue Dialler and killed modem speaker
Rogue dialler but first indication was the 
dialling & beep/chime acknowledgement disappeared 
from modem. Bit of a  giveaway really.
Nothing I could do in modem options would bring 
it back. 
In rasphone.pbk Speaker=1 and Dialmode=1 seemed 
correct. 
In Control Panel 
Phone & Modem Options 
Properties / Advanced / Extra Settings was 
S7=240 
As this was a Hayes code , I changed it to 
S7=240,M1 
then Change Default 
and 
produced SD & RD modem LEDs alternate flashing and "connection was 
closed" dialogue box.
Removing the comma limiter leaving just 
S7=240 M1 
brought the modem speaker back to life and 
normal service resumed. 

rnaapp.exe and Outlook Express problem
leading to an assortment of error messages and 
dropped dialup connection when connectinfg to 
first or server or one, later, further down the connections.
error 678
The computer you are dialing in to is not answering.
error 720
error 645 
Dial-Up Networking could not complete the connection to the server.
Check your configuration and try the connection again.
"Another program is dialing the selected connection" etc
Some carry-over data gets corrupted from a 
previous session and the rnaapp thinks there already 
is a connection when you dial up, fist time of a new session.
Instead of uninstalling etc just try a change 
of operational sequence.
Dialup as the first application and then open OE.
If composing emails/replies prior to dialup , then 
close down OE and then dial up.

Samsung Notebook, crash to black
NP N102S, PbFitis? Win 7.  Presumably hardware fault, display died with pc just sitting idle. "grey" light present of backlight on, going on and off, on closing the screen down and hinge so switch operating. cTL-Alt-Del no function, start-up button no function. Had to remove the battery to switch off.
Powered up via "normal" selection of recovery page.
Next time will try F4 and also an external monitor.
Another time , next day,backlight went, screen contents viewable with torch . Again no response to C-A-D, but pressing hardware start button brought back the backlight and previous workings on the display.
No problems before concerning the hinge switch or backlight going, but twice over the months blue-screen windoze type crashes, recovered via normal option each time.
The fan seems to be on noisier/faster. Outlet air is 16 deg C above ambient, via thermometer, higher than remembered, with pc just sitting there, charged battery, 30% or so CPU useage. Don't know what is normal for this though. Is there a route in win7 to see some hardware temperature monitor readings? What is the likely failing part or cause for this? 
Decided, at this stage, could have been caused by loose clothing up against the fan outlet grill. So glued on 2 halves of a bumpon either side of the grill to avoid a simple cloth drape blocking the feeble fan flow. 
Broken  SVGA ext monitor output
I broke it using it as an active feed to determine the pinning to an unknown proprietary input to a video projector.
There seems to be low level syncs ie not full level, RGB outs are normal. What may have been the damage, maybe I connected it to the 5V undroppered supply on the projector connector ,or a +/-9V RS232 line on it or perhaps just hot-plugging. Broken TTL chip/fused fusible R/VTS diode failed ?
I don't intend getting inside this notebook just to fix the monitor out, I'll wait until something more disabling occures before I go venturing inside, just curious what sort of damage I may have done.
No great problem as I acquired a USB to VGA adaptor plus driver SW(useful for the situation when someone else's laptop may not have an SVGA , to feed into this projector). Reminder to myself, next time, use an old desktop pc with a sacrificable output card. 
When it hasppened screen went black , pressed Fn + F4 then lost mouse control
CTL+ALT+DEL but as no mouse, direction , tab and enter keys to shut down 
and restart in safe mode

Shotz Smart Digital Camera ,SS 2000, installing on Windows 98SE
A page relating to problems with 98SE came with the instructions.
RPC stub error relating to oleaut.dll dating, the solution was 
supposed to be, download files from their site rather than their CDROM.
This may well be possible but I could not make it work, first time 
used a USB port on this pc compounding matters.
Downloaded mcrepair.exe from microsoft and selected backdating 
option. Then, as I don't consider myself stupid a tip relating 
to Microsoft "Scanners and Cameras" installation wizard. 
I was exiting at point where you have to select port Serial 1 or 2 
or Parallel port - no USB showing up althogh apparently installed. 
Ignore, by selecting Parallel, and the camera drivers can be 
installed employing USB by default anyway. 
For people not often using the camera - notes
1/ When starting to download the software is not specific 
and you have to select from menu. There is no specific SS 2000 
to select and thinking SS=Single Shotz then leads to "no device found" 
error - SS=Smart Shotz.
2/ The other default selection is "download and delete camera memory", 
you have to actively select the other option to retain camera contents 
which seems bad arrangement.

Shotz Smart Digital Camera ,SS 2000
Converting the "shutter release" for macro photography.
Too much pressure required to manually press the release switch 
leading to camera shake in a magnified set-up.
Remove 5 obvious screws plus the one under the battery cover.
Pull the cover off and unattach the sounder, incidently the same process 
to put an external break in the line to the sounder for covert use of camera.
Remove screw through internal battery holder and the 2 near the viewfinder 
and pull away from the USB/AV socket hole and bend out the chrome section 
to free the plastic lugs 
so both PCB sections can be removed in one for minimal disturbance to 
the boards. Remove the now loose AV rubber cover, 
and viewfinder.
A lead to the earthy side of the switch is easy but the other side needs 
finely melting away some of the plastic of the switch to expose 
the contact to solder to.
Find a space to drill 2 holes through the casing to mount 2 connectors 
from a butchered turned pin DIL socket for external connection 
in parallel to the switch

e-mail

ncook246@gmail.co.....m