GENT OF LEICESTER XC408 CHRONOPHER 3 MASTER CLOCK REPAIR
My professional watchmaker friend Charles brought me another one of his valuable but old defect devices. An about 50 years old Gent’s XC408 Master clock he had bought a few years back.
The problem with these quite old electronic circuits is almost always that it is very hard to find any schematics. And even just finding a user manual is close to impossible. And the ones that still exist are often immediately sold at auctions on eBay or other places. And my friend probably already had tried all existing channels, online, or other ways in vain. The manual normally also contains 4 circuits.
He told me that the Master clock had worked fine for about 6 months until it started to become very slow and a minute no longer was 60 seconds. And the backup battery was already bad from the start after he got this device. And after he tried to reconnect the bad battery to see if the Gent’s XC408 Master would work again like it had, it no longer did anything right.
I started checking the PUSH TO SYNC button and the 3 position toggle switch with the ADVANCE, RETARD and the NORMAL selection. And it was clear that the SLAVE clock in previous photo on the left didn’t work like it was supposed to. And I am careful not to lose any valuable information of the present condition the Master clock was in, so I started to make photos of all boards, components, connectors. And only when that was done I continued to dismount the boards to also make shots of the board solder sides. And of course I also made notes of the voltages I was able to measure.
Above board is the timing controller starting with the old but rather complex CD4045AE. Also an already seldom used and hard to find component these days. How complex it is can be viewed on page 1 of its datasheet. And notice that the CD4045 with AE extension types are not mentioned here or in most of the CD4045 datasheets online.
Most of my digital CMOS and TTL IC testers can’t even test this special type of IC. And in most datasheet books it probably isn’t even mentioned anymore because of its specialty field.
So the only thing I could do to see if the CD4045AE was still okay was connecting it to my Tektronix oscilloscope. The 2.09 MHz crystal was fine and still exact. Also IC outputs pins 7 and 8 showed an about one second signal. But it is hard to watch as a signal on a +350 MHz oscilloscope because it is just a very slow signal. Pin 8 showed the delayed signal from the signal of pin 7. So it looked that it was working but if it was precise enough was uncertain. The easiest would be to just replace the unsocketed 4045 chip, but I had none other to test the Master clock device with. So I continued to check on component level. After that I made the circuits after I had drawn all components in the solder layer pcb’s I already had taken photos of. But first another photo of the Relay board of this Gent’s XC408 Master clock.
And I also checked both the used 24V Potter & Brumfield K2501 Relays of which I couldn’t find a datasheet sadly. The middle one was marked RL A and the other one RL B. The smaller relay RL C.
But looking at the wires to the 2 changeover switches, and to its coil it was clear how they were connected to the 8 pin round socket. In the solder side both switches are switched in parallel use.
The smaller relay marked RL C had 3 x 4 contacts plus 2 for the coil.
The middle 4 pins column were P1 to P4 from the changeover switches, and in the left column are the release contact for P2, the make contact for P2, the release contact for P4, and the make contact for P4. The right column is R(elease) 1 for P1, M(ake) 1 for P1, R3 for P3, and last M3 for P3. Of course the top 2 pins are the relay coil connections.
The transformer had a simple AC input selector that still is perfectly working today. The secondary output voltage was ~28 VAC. Just the wires on the primary side are badly insulated because just a small square plastic plate beneath the 2 wire input prevents it from shorting to the metal of the case below it. Potentiometers VR1 and VR2 on the relay board are respectively 5k6 lin. And 10k lin.
Above the result after all components were drawn into the pcb solder side of the Relay board. And followed by the solder side view of the Timing board of this Gent’s Master clock XC408.
One defect I found was a bad Q5 (PNP silicium BFX29, HFE 238 probably same as PNP Q2). And after I had replaced it by a compatible 60V 0.6A 2905A transistor, the ADVANCE toggle setting finally worked! Showing a SLAVE clock advancing about every 5 seconds with an 1 second jump.
And next circuit of the Relay board was made after checking the PCB solder side with all components inserted to the A4 photo dump by hand.
It still may contain a small error somewhere but it already gives a good view on how to analyze the Master clock. It will be followed by the circuit of the Timing PCB with the CD4045AE on it.
Both 24V relays RL A and B switch in one after the other in turn by switching the Master clock output connector from +24V to -24V about every second. With a short 0V period inbetween both 24V signals. So both 24V relays seem to work fine! And in 2 of my digital IC testers all IC’s tested fine.
Above schematic from the board with the 4045 crystal clocked timer IC on it may be not entirely correct if it is still missing some left out connection somewhere. That is explainable by the rather not real double sided pcb without any metalized through via’s. And visible by the soldered IC legs to the tracks of the component side. Because I couldn’t find any defect I continued examining the wiring to the toggle switch and the push button. And came up with next rough but correct result:
And that showed that the yellow wire itself wasn’t directly connected to the timing board with the CD4045AE. It only was used if it was indirectly connected through with any of the other wires of the 9 wires that were connecting from the Toggle switch and Push button switch panel to the connector of the 4045AE board.
In short I noticed that of the 3 changeover switches in the Toggle switch following was happening when we select the RETARD (middle) position:
*White + yellow + green are connected (grey is unconnected)<= so white+green because yellow itself isn’t connected to any board, and also red + red/green are connected to (3th toggle switch of which the blue wire is unconnected).
In the ADVANCE position we get following:
*Grey + yellow+ white are connected, so grey + white got to the timing board., and on the 3th toggle switch red + red/green are connected leaving blue unconnected.
And finally in the NORMAL toggle position we get:
*White +grey, and also yellow + green of which yellow is not connected to the timing board so again not significant. And also red is connected in the 3rd toggle switch to blue.
____________________________________________________________________________
And the 9 wires that connect to the 4045AE board connector have following meaning on that board:
RED <= to hex inv buffer 4049 pin 3 (1A input).
RED/GREEN <= Q outputs 4024 and 4013.
GREEN <= + VDD
WHITE <= RESET pin of 1st 4013 after 4045.
GREY <= Output 4045 pin 7 and CLK 1st 4013.
BROWN <= Reset 2th 4013.
BLUE <= H signal to RL C relay (smallest relay).
_______________________________________________________________________
Next some photos that shows the still rather complex Toggle and push button wiring of the switch panel with the ADVANCE, RETARD and NORMAL selection.
In ADVANCE mode the Master slave clock advances 1 second step in about every 5 seconds.
In NORMAL mode the Master slave clock advances 1 minute every 60 seconds.
The RETARD mode is not really doing anything. (Keep in mind that we had no instructions found).
And in above advance and normal selection mode both the 24V relays RL A and RL B in turn are active outputting +24V , a 0V rest period, followed by a -24V signal every second or every 5 seconds depending on the selection etc. (the way these old Master slave station clocks work).
Conclusion: This repair article became a larger project but with a good end! And with a valuable and rather unexpected conclusion! DON’T trust your digital CMOS/TTL IC testers!
After 3 days of examining all parts and all connections and finding no reason for the NORMAL mode not working, I suddenly was thinking WHAT IF my digital testers say that the 4024 counter is okay but in fact they probably never tested all 2^7 output positions of this 7 bit binary counter IC?
And because the relay board tested okay with the toggle switch in ADVANCE mode I thought it was only logic if only the counter IC was somehow failing.
So I replaced that chip with a new 4024 chip I had in stock and guess what? The GENT’s XC408 Master Clock was completely working like it did already about 50 years ago!
After this conclusion I am thinking to recheck my DCF Exachron repair I made a few years ago also for my friend that too was very large, took a lot of work and even got Youtube videos added but failed in the end because it wouldn’t let its 4 potentiometers adjust correctly. And that device sadly had no schematics or instruction manual either and didn’t have any bad parts like in this Master clock repair.
I phoned my friend with the good news and he will come and take his perfectly working device tomorrow home. And he finally will be able to run a very large slave clock on this Gent’s XC408 Master clock.
Anyway this article will be a great help to any other Watchmaker in need of information to fix a bad GENT’s XC408 master clock too. Since most of these devices after so many years are sold without any matching documents.
Final note about the mentioned backup battery that consisted out of two times 10 x type 450 DK cells packed in series. Making a rechargable battery of 32.2V. Following text was printed on 2 battery packs with each 10 of those type 450 DK cells:
I carefully tried to recharge the packs but of course all old battery cells were gone and no longer useful because their voltages already had dropped to about 0V !
Albert van Bemmelen, Weert, The Netherlands.
Please give a support by clicking on the social buttons below. Your feedback on the post is welcome. Please leave it in the comments.
P.S- If you enjoyed reading this, click here to subscribe to my blog (free subscription). That way, you’ll never miss a post. You can also forward this website link to your friends and colleagues-thanks!
Note: You can read his previous repair article in the below link:
https://jestineyong.com/repair-of-a-toshiba-windows-10-laptop/
(63)Dislikes
(2)
16 Comments
Leave a Reply
Parasuraman S
April 23, 2020 at 3:17 pm
OMG! What an analytical trouble shooting you have done and to what extent you have gone, is beyond comparison! This is a total R&D on an electronic servicing and hat's off to you, as you are the only one in the world to do this sort of study and rectification!
Albert van Bemmelen
April 23, 2020 at 5:26 pm
I just was lucky having the fortunate opportunity to examine this masterpiece of a Master clock device from the past dear Parasuraman. And having a professional watchmaker friend who has faith in my repair knowledge. And glad having found the culprit being the bad 7 bit binary counter, a 4024 cmos chip. Which wrongly tested good in 2 of my IC testers.
About these failing IC testers: most digital IC testers only recognize the more complex IC's. Which can be helpful if we can't read any identification on them anymore. But they do not fully functional test all of them! They however correctly do test the less complex digital ones.
Albert van Bemmelen
April 25, 2020 at 2:39 pm
By-the-way: Only 2 of my portable battery operated universal digital IC tester failed totally in testing the bad CD4024 7bit counter correctly!
I also have at least 2 other PC controlled IC testers that passed the test in disapproving the bad chip. They are the Elektor March 1998 IC tester, the Xeltek SP/L Super Pro L universal programmer. (and probably also the Genius G540 and the TL866A, but I just have not tested them yet with the bad counter chip).
Bruce Saxby
April 23, 2020 at 5:07 pm
Albert your articles are amazing as always, I enjoy reading them you do some incredible stuff - it tests my memory and 40 years past knowledge (thin now) well done!
Albert van Bemmelen
April 23, 2020 at 8:24 pm
Thank you Bruce for that compliment. Being 60 myself now, I probably can match your knowledge of the past. No less still eager to learn new things. like learning about the great on Ada based programming language VHDL. And never forgetting the knowledge we already obtained in our younger years.
Yogesh Panchal
April 23, 2020 at 5:38 pm
Albert.......No words...............to appreciate.
Albert van Bemmelen
April 23, 2020 at 8:40 pm
Thanks Yogesh. Correct me if I'm wrong but I think you, or maybe it was Beh?, in one of your older repair articles had mentioned a similar problem about a digital semiconductor or IC Tester not correctly working. Like in my current repair where two of my universal IC Testers completely failed to disapprove a component (the counter IC CD4024) that in fact was defect. But I can't remember what the article was about, or what Digital Tester it was that also failed the
component Test?
Greenford William Mafuleka
April 23, 2020 at 10:07 pm
As indicated before you are great men. I tried and did halfway clock and watch repair Hological training with UK online training. It was tough because parts are very small and need to be very careful. Regards. Greenford William Mafuleka
Albert van Bemmelen
April 23, 2020 at 11:34 pm
Thank you Greenford William, but therefore I think you are, like my watchmaker friend, a much better educated man in clocks and watches then I am. I never followed any training in fixing those tiny special mechanical and/or electronic watchmaker parts. Fixing mechanics is just not my thing. My best regards to you.
Andrea Del Corso
April 24, 2020 at 8:52 pm
Compliment Albert!Another big repairing with complete explanation of the job.
Only your passion for electronics can do!
Albert van Bemmelen
April 27, 2020 at 5:07 pm
Yes, Thanks Andrea! And often we are only able to fix circuits and devices after a bit of reverse engineering was done first. Which I like too!
Humberto
April 26, 2020 at 8:14 am
Hi Albert, this article recalled me when I was just graduated from the University, I worked in a huge Factory and one day we had to fix the Master Clock similar to this one(Czech made)and later we synchronized about 60 Slave Clocks inside the whole Factory. Thanks for the article.
Albert van Bemmelen
April 26, 2020 at 2:40 pm
I am glad I could save this XC408 Master clock for my friend, Humberto. My watchmaker friend Charles told me that they chose this higher 24V pulse output signal because a strong pulse is needed to sync-control many Slave clocks in series over long wire distances. Given your example it would leave only about 24V/60V for every slave clock. Which I guess is no problem if the signal is amplified electronically in every clock. But I've not seen the internal circuit of a Slave clock yet. Maybe fixing a Slave clock will be the subject in a future article. We'll never know what my watchmaker friend has in store next.
Albert van Bemmelen
April 28, 2020 at 1:11 pm
Important note: I noticed that the TL866A programmer as IC tester is unable to test most counters like the CD4024, the 7490 and the 7492, because it can't check any of them. (not in the support device list!). But the Genius G540 universal programmer tester does in that respect a bit better because it would recognize a bad CD4024 counter chip. Sadly doesn't do the 7490 or 7492 either!
Alireza
September 11, 2021 at 5:45 am
Hi.
I don't writing and speak English,
Excuse me.
I repair battery of clock xc408
Consist of:
1-Power supply and ups mode mean well
2-Battery 12v2.3a
3-Module xl6009
4-8 slave clock original for xc408
Power supply is charger for battery and supply 13.8 v DC
Out put of power joint to xl6009 that convert 13.8 to 28 v DC
Out put of module joint to diode bridge on the board of clock.
For another information please ask me.
Thanks
Sherif
August 9, 2023 at 4:14 pm
please , can you tell me what is the output from the device ? , I mean the clock input , thank you in advance .