1975 PYE CT-219 “Chelsea” 717 Chassis

Just a thought Chris before diving in too deep, check the supplies with a scope and be suspicious of any small electrolytics decoupling any of the supply rails. Should be possible to run your cap tester over any decouplers for a quick check.

Hi Rich, I've been through the electrolytics even breaking my rule, and just replacing them. Three remain but are in the frame cct which is rock solid. I've also removed, tested and replaced, around the 920, caps that were marginally out of tolerance due to them being video coupling, and checked resistors. I've even replaced the TBA920Q to eliminate that from suspicion. No luck.

I've also installed a spare IF panel to also eliminate that and the associated TCA270, the fault remained. The problem has to be somewhere in the timebase, but for the life of me I'm blowed if I can track it down. John even popped by the other evening, and he too is flumoxed by it, so at least my misery has company. And we both just ended up muttering "barn finds pah!" 😉 

Something very strange is going on. Upon switch on I have a locked TCF albeit with bent verts. Over a few mins this deteriorates until lock goes. The waveform at switch on, at pin 5, looks text book, though 1.8V not 1.6V. The waveform on pin6 however is very odd indeed. This waveform also shows the effect when the lock is lost (final pic). The sequence of photos follows, power on, scope of 920 pins 5*6, followed by lock going and waveform 6 showing the effect of that.

Well, this is proving to be yet another hump the wumpus, really don't know where to go with this now.

As I said in my previous post on page 3, finding a cct that matches my version is difficult. I do now have the final version of the cct, but as I say, mine is an early final version. The main difference I see is C616 (10n) is not present on my board, not missing, it never existed. My version is this last revision as the previous versions have subtle component value variations that my board does not tally with but does the final version.

Question 1: why did PYE/Philips add that decoupler to the line phasing, to the final, final version, maybe I should add it and see. Well I did and it not a blind bit of difference.

Question 2: Is there any possibility this is a fault elsewhere, am I being fooled into thinking it's timebase? Although I have to say, I'm so sceptical of that thought as It really does look like timebase. I've run out of any sensible ideas now, so fiction is what's left to me, although I've ruled out the IF by swapping in another board. That only leaves the decoder, so it stretches credibility.

I had a similar picture on a little ITT black and white set, there was a chip that failed in the IF, which I had replaced, but the IF coils had been twiddled, eventually I got it back, mostly by trial and error! I know you have tried another IF board, but could they both have a similar problem? Could there also be a problem with the sync separator?

 Regards 

Lloyd 

Posted by: @crustytv

Upon switch on I have a locked TCF albeit with bent verts. Over a few mins this deteriorates until lock goes.

It could be temperature related and a careful localised use of freezer spray might help narrow it down. Whether it be a passive or active component. Could be a high value feedback resistor somewhere in the line stage, is there such an arrangement obtaining pulses from the LOPT windings?.

Still no further on with this, in desperation I removed C628 which is a 10n high precision cap, thinking perhaps this had drifted, nope. I've not gone down the freezer spray route as I was advised elsewhere, although useful as an aid for intermittent problems, not so for faults like the one I'm presented with.

I've continuity tested the traces out from the IC and all appear to be OK, this is one baffling fault and as John pointed out the other day, "just defies logic".

As an aside; I don't know about everyone else, but I find scoping pins on ic's raises my anxiety levels tremendously, one slip etc. I do have tiny little clips you can hook on, but that's time-consuming and sometimes access is not easy. There are scope probe shrouds which also aid locating you onto a ping, again not always easy.

As a result, I've bought a little tool to help aid me in taking readings from 16-pin ic's. It makes the whole process so simple and aids quick connection to what ever pin you desire. It simply clips over the chip, you slide down the lock and job done. For what it is, it's not cheap, but to me worth every penny.

Not that it's going to provide a solution to this stubborn PYE, but it will certainly ensure easier access to taking chip readings on this and other future sets. I'll have to look out for other clips for other ic's now.

Here's a short video of the problem, the colour coming in is just me noticing I had it off, so adjusted it in.

Hi Chris,

Oddly enough, that effect is not too dissimilar to the fault of poor smoothing I had in the line oscillator stage on the Murphy V410. Totally differing sets of course, but that was an O/C capacitor giving reduced smoothing.

I wonder what the effect could be if the frame pulses are not filtered, or somehow increasingly getting into the line sync on the Pye.

I don't have a circuit diagram but looking at the symptom and seeing the diagram of the components around IC601 something rings a bell in some distant corner of my brain that C623 caused weird sync problems.  Anyway just thought it was worth mentioning. 

Hi Jim, good memory 👍 

C623 going o/c or print associated with it will cause a ragged line sync. C624 going s/c can cause a similar fault or be responsible for the bent verticals should it go o/c.  Both, I have replaced.

Posted by: @pye625

Hi Chris,

Oddly enough, that effect is not too dissimilar to the fault of poor smoothing I had in the line oscillator stage on the Murphy V410. Totally differing sets of course, but that was an O/C capacitor giving reduced smoothing.

Hi Andrew, did that not produce a ripple on the rail? The main smoother on the PYE was removed and reformed. I have checked the 155V rail and no detectable ripple is present.

Out of sheer desperation and basically running out of ideas, I'm breaking all my rules. I've started blanket replacing around the TCA920 with 1% 1W resistors, caps and even the TCA920. Only two resistors (R621/R622) and two caps (C625/C627) left to replace but thought I see, before doing those.

Results this far, pretty much the same. I shorted out PC7 A/B and adjusted the line hold R634 (10K and it's been replaced) until I had a float. It's a perfectly static float, and the verticals all look pretty good as well.

Removed the short and tried adjusting the line phase R613 (2K2 and it's been replaced) but to no avail.

Edit: Now replaced everything around the TCA920, this is just too weird for words, where on earth is the fault located.

Are any of the PCBs earthed via the solder to the chassis, or indeed are all the 0V returns good between the boards? - because this is beginning to look like an earth current where it shouldn't be.

You will not believe in a month of Sundays what the problem was

Posted by: @crustytv

 

 

Are you going to put that little lot all back? 😋 

Was it an earthing problem? Was it a cabling problem with an earth returned to the wrong place?

No, for some unknown reason, the PYE TV objected to the video signal being looped through a broadcast monitor. All my other TV's are fine with it.

@crustytv Had the monitor 75 ohm video input terminator switch been knocked to the on position thereby double terminating the signal by chance? 

Incredible given that the frame sync was solid as a rock, very surprised. I thought the SAWF would have coped with anything you threw at it signal wise.

But what I find stranger still is why would the "fault" become worse over a short period from switch-on?

You know, I very nearly asked earlier if you had tried another signal source, but decided it was a silly question, so didn't ask! Good to see you found it though!

Regards,

Lloyd