Omega Checker 1 aka...

[767Geoff]

Also known as a WITSCHI Q TEST 4100.

Got this off Ebay for $400 with the displays working but in untested condition. No guarantees on accuracy or if fully functional. Well I love a puzzle and knowing the displays were okay I thought my chances of getting this fully functional would be 50/50.

So it arrived well packaged and shipped from Europe.

I noticed they writing in places on the main bezel for the supply voltage test section was corroded and missing. Case was very dirty along with the wood case. The wiring was for 220 volts but is switchable to 120 volts and with a European to American plug converter.

Plugged it in and noticed the on off switch was hooked up backwards, that is the off was on and the on was off! This told me that someone had been inside the tester other then the makers.

When opened I noticed some wires were broken from their soldered positions. The timing display function for 18000 and 26600 mechanical, 32kHz and Accutron tuning fork was displaying correct daily and monthly errors, more on that later. The voltage testing and internal voltage supply for testing were non functional (the side with broken wires).

The mirror was loose and foam for the mirror was decomposed and also the white collar was completely missing on the test side of the checker. The white collar reflects light illuminating the under side (display side) of the module under test.

Both LED red filters were loose and came off when the case was disassembled.

So problem when I received it:

1) wires not connected internally.
2) loose ICs (they are all plug into sockets, amazing).
3) mirror on testing side loose
4) foam underlay for mirror which pushes up and into the large PCA decomposed.
5) collar which surrounds the internal mirrored testing side missing.
6) on off switch hooked up backwards.
7) whether correct or not the electrolytic are all 33 years of age so they may be suspect.
8) NO MANUAL or INSTRUCTIONS (any help here would be appreciated).
9) wood cover stained and finish deteriorated.
10) metal bezel had corrosion on surface as though acid had been leaked onto it.
11) supply voltage writing missing due to the corrosion.
12) red led filters both loose or separated from the upper face plate.
13) missing probes for the test side of the checker.

What I did:

Electronics

1) disassembled components.
2) replaced all capacitors with like capacitance and 5% tolerance.
3) resoldered broken wires to correct probe locations.
4) upper PCA which contains the LED modules.
- Removed LED blocks and cleaned all push button contacts as well as battery knob pot and PCA.
5) main PCA cleaned and all ICs reseated after being treated with detox-it electronic preserver.
6) cleaned the sensor pads being REALLY careful about the wiring under the detector PCA.

Case:

1) lightly sanded wood and lacquered with satin finish.
2) cleaned corrosion off the aluminum bezel and cleaned up the engraved supply voltage writing.
- Then filled in the writing.
- lacquered the metal bezel upper and lower in satin lacquer.
3) polished and reattached the LED red lenses.
4) cleaned mirror and mounted on new foam.
5) found a spray paint cap of the perfect size and colour to replace the missing collar around the mirror.
- Cut to size and mounted it.
6) added rubber feet to base.
7) will remake the probes for the test side and already secured test probes.

So the BEFORE pictures:

As it arrived EXTERNAL:



Upper right closeup:



Upper center closeup:



Upper left closeup:



Over view: note the white collar on the right side surrounding the mirror on the testing side is missing.



Damage to the aluminum bezel at the supply voltage side of the checker (yellow box):



Wood with damaged finish:



As it arrived INTERNAL:

Capacitor on main PCA (replaced).



Other parts of the main PCA with capacitors and dirt on PCA.



Broken wires to testing side of the checker:



Main PCA with shielding for smaller sensor pad PCA;



Sensor PCA pads obverse side:



Sensor PCA reverse side:



Upper display PCA with push buttons all cleaned:



Reverse of the upper display:




AND NOW THE REVEAL:

The overview with lacquered wood and bezel. Supply voltage writing redone. All probe receptacles removed, sanded and polished then reinserted.



Three views of the pocket watch (18000bph) with the seconds per day displayed. You will notice the balance wheel button (far left
is selected) as is the 60 second sampling button. Also note the green signal led (which beats in relation to the watch beat) as well as the red led indicating seconds per day:







The LCD push button is selected with green signal and red seconds per month now indicated. Sampling time is 5 seconds:



Voltage of a battery under test, the EXT V push button is selected:



Internal supply voltage being tested for accuracy. Output is being tested by my multimeters. Note the selector knob is abeam the 1.55 volt indicator, not bad accuracy! The SUPPLY V push button is selected:



Here the voltage knob is selected to the highest value, almost 3 volts. These two voltages and all between are typical LCD and LED voltage requirements.



Here is the accuracy of the voltage measurements:



And the capacitors removed:



The missing probes, which I will be making later, would connect to the module + and - testing points. The module would be upside down and the display would be visible in the mirror.

There is an alarm button as well as a display button, again I need a manual to check these for functionality.

So I think I have it working. The tested seconds per day and seconds per month in the above examples check well with the watches actual performance. I also tested a Bulova Accutron, Omega 300 hertz and my 2.4 analog Omega Marine Master all check. The Omega Marine Master clocked in at 1 second per month on the tester and that coincides with its actual rate!

The quartz watches are divided into 32kHz, LCD and analog. The analog was used for the Marine Master so I guess it automatically detects the rates. Again I need the manual to determine what is what.

Hope you enjoyed the posting, I will do a YouTube video with various aspects of its capabilities demonstrated.

Best regards,


Geoff

[gjlelec]

I have no words Amazing job Geoff

[767Geoff]

Thanks Gilelec!

Thought there would be more interest in this extremely capable checker.

Since the posting and more pics to follow I got the complete 1986 manual from the the manufacturer WISTCHI in PDF which I printed off in a 1986 style binding.

In the manual it states the importance of a three pronged grounded plug. Well upon inspection of the Euro plug (round prongs) I notice only the -ve and +ve prongs and no grounding prong. Someone had neatly filed it off flush with the original plug. This contrary to the setup instructions of the machine.

This error might have impacted the performance. I was using a Euro to US adapter and had switched the power input and fuse to match North American standards (built in to the checker with a switch selection 110 / 220).

So the checker has a new grounded plug. I also ordered 20 inch stainless steel rods of the correct thickness to make the two new probes that are used in conjunction with the far right mirrored section. This will make module, LCD and alarm testing a breeze!

So the project is almost complete.

AND LOVING the LED display typical of the large displays of the 80’s.

The newer LCD versions which accomplish the same functions but without the glow are upwards of $3000 USd!

So very happy with my $400 dollar purchase plus $10 in parts and solder

Geoff

[Adam]

Amazing work Geoff as always you do such good work.
The forum seems pretty dead now. My forum at my digital-watch.com has very few posts now either. Most people are on Facebook groups. This is such a shame as the information thats posted isnt kept in the same way. Finding old posts is much more difficult. Problem is that seems to be the future and Facebook is in control.

Again lovely project and thank you for sharing.

Adam

[jadney]

Really nice job. Thanks for posting all these. I find myself envious of your finding this; wish I had one.

(My wife would kill me if I found one and tried to add it to the clutter already in my life....)

Can you tell me what the pickup for the quartz crystal watches is? Is it just a microphone that can hear the 32 kHz hum or is it sensing an electrical signal?

[Ole Joe]

Hi Jim,
Back in the early '70s when I worked at Frontier Semiconductor producing LED watch modules, we used two different types of equipment to set the modules 32.768 K Hz oscillator frequency. Both pieces of equipment, manufactured by different companies, used the “radiated” signal from the oscillator to detect and display the actual frequency that the oscillator was tuned too. Needless to say, the radiated signal was in the low micro-volt region and required very high gain and noise suppression to acquire and process. The signal detectors front end started with a FET with a tuned 32.768 K Hz resonant circuit, followed by a number of active crystal filter gain stages, in one case, 10 in all. Following this, the signal was digitally manipulated for display on a HP frequency counter. One of the two pieces of equipment we used, the Valpey - Fischer A-600, was mostly used for engineering lab work with a counter which would display to three places after the decimal. The production line equipment was much more sophisticated and expensive and would read out to more digits after the decimal but drove the production workers crazy because of the constant changing of the digits after two or three decimal places. They ended up obliterating all digits displayed after three past the decimal point for sanities sake. In the 13th picture from the top that Geoff posted earlier, the one labeled “Sensor PCA reverse side”, it looks like the sensor on the right is connected to a transistor, most likely a FET, which is connected to a couple of resistors and what appears to be a tuned circuit composed of a capacitor and inductor. The output then goes to the main board via the four wires to the rest of the circuitry in the cabinet.
Ole Joe

[jadney]

By radiated signal I'm guessing that you mean an electrical signal. The tuning fork crystal also radiates a sonic signal, even though it's way too high for us to hear.

Tuned circuit amplification certainly makes sense, either way, electrical or sonic. The input to those stages could be either and I can't tell from the photo what kind of sensor is there. It seems likely that the signal will be extremely weak, either way, but it's hard to imagine that 10 stages of amplification could ever be useful.

In the place where you worked, was this sensing done with the modules in their cases? If in, was it done thru the crystal, so adjustment could be done without flipping the watch over?

And yes, more than 3 decimal places is pointless in this application. Even the third place jumps around a lot.

thanks for the response,

[Ole Joe]

Hello Jim,
Frontier Semiconductor only produced the LED watch modules, not complete watches ready for retail purchase. We
sold modules to many companies who were well known, Omega, Hamilton etc, and those who only got on the band
wagon for the money that was to be made and had never sold a watch in their history. Our modules were tested
and the oscillator tuned by simply placing the module in a designated sensor area on the test head with the
variable tuning capacitor of the module exposed so that it could be adjusted. Both types of equipment we used
detected the electrical signal (radiated), from the oscillator circuit within. The sensor area was no more
than a piece of copper clad board about 2 inches square which acted as the antenna for detecting the very low
level electrical signal emitted from the oscillator circuit. The "antenna" signal was then processed via a
chain of 32.768 KHz crystal filter gain stages to limit the bandwidth, suppress the noise and amplify the
signal to usable levels. Both pieces of equipment would also detect the oscillator signal when the module was
installed in any type of watch case, metal or plastic. Of course if you wished to tune the oscillator while
in the case, you had to remove the back cover to get to the trimmer capacitor. All metal cases did have a
slight effect on the frequency of the oscillator due to the stray capacitance introduced which would pull the
oscillator frequency a bit, but not enough that it made much difference. As policy at Frontier, we spec'd the
modules oscillator frequency to be within 3 PPM out the door. The production operators doing the tuning would
strive for a .032 Hz reading on the counter (approx. 1 PPM). Because of the popularity of, and the demand for
all electronic watches at the time, most watch manufactures could have cared less about the accuracy of the
oscillator frequency as long as it was close. If you Google patent #4,041,767 issued Aug. 16th, 1977, it will give
you and idea of some of the instruments used to tune the oscillators in watch modules at the time. This patent
shows a fairly complete schematic and explains the operation of such.

Ole Joe