A Tale of Two Dynatrons

[Mikey405]

Hi everyone.

I thought I'd put a short write-up of the adventures of a pair of Dynatrons with Pye Hybrid chassis (if people aren't too fed up with too much colour stuff).

Tas had come round for a visit and we decided to tackle a Black-and-White Dynatron which was supposed to be a colour Dynatron. The set in question, a CTV5 using the Pye hybrid "691" chassis, had a colour fault (or rather colour faults), a convergence problem and a blanking issue. The colour problem was that there was no colour at all - a very common problem with Pye Hybrid sets and almost universally caused by dodgy BC148 and BF149 "Lockfit" transistors. Being the lazy swine that I am, I decided I couldn't be bothered to find out how to override the colour-killer (A 10K resistor between two test points I think, or using a 9V battery or some such other contrivance) and we dived in with the scope. The first path chosen was to check the bias on the chroma-amp transistor VT20 - It was negative, which indicated that the colour killer was operating. Measuring the stage before revealed no 7.8KHz PAL identity signal. (The 7.8KHz being derived from the ripple generated by the swinging burst and used to synchronise the PAL switch and also to operate the colour killer.) Checking the DC amp voltage in the 4.43MHz subcarrier regenerator APC loop revealed that the burst was being correctly gated and that the APC loop was locking okay, so why was the 7.8KHz identity amplifier not working? Very often this is down to a faulty ident coil (L27) (usually the wires need resoldering to the legs) but this time it turned out to the transistor itself (VT18, a BC148). This was the first of our dodgy “Lockfit” transistors. We removed the panel, replaced the transistor, fitted everything back in and switched on. No colour. Hmm. Checking the 7.8KHz on the collector of VT18 revealed that we now had a nice big switching waveform and the colour-killer / chroma amp transistor VT20 was now biased on but there was still no colour. Our attention then turned to the path of the raw chroma through the channel towards the delay line driver etc. We didn't have to look far - Although the chroma amp VT20 was being biased on, indicating that the colour-killer was now off and that everything was rosy in the subcarrier regenerator, there was no chroma on the other side. So we removed the panel again, replaced the chroma amp transistor (VT20, BF194) with a BF255 purchased from Grandata, put everything back together and switched on. No colour. Damn. Checking the bias on VT20 revealed that the colour-killer was still off and checking the base of the delay line driver transistor (VT21) showed a nice juicy chroma signal, but there was nothing on the collector. Out came the board again, out came VT21 (a BF195) and in went another Grandata-special BF255. Back went the board and the set was switched on to reveal... colour. Hooray. Well, mostly colour, but there was no G-Y and Carol looked a bit orangey and sunburned. This turned out to be VT30, part of the G-Y matrix circuit on the CDA panel. This being replaced we had spectacularly good colour on the brand-new A63-120X tube. (The tube came from SEME many many years ago when they were giving them away free for the cost of the postage and this was the first time it had been used in anger.)

Next we turned our attention to the convergence fault, this was a problem causing the blue horizontals to droop at the outer edges of the picture and no adjustment was possible. There couldn't really have been much causing this problem as the same waveform used for the blue line-derived convergence is also used in the red and green, which were both fine. Anyway, after lots of misleading waveform checks and cold component checks in the convergence board it was decided that the only thing that the problem could be was the Plessey convergence yoke itself. So, after butchering another Dynatron (a CTV16) we hung the CTV16's Mullard yoke on the CTV5's tube and switched on. The blue convergence could now be adjusted without any problems and the droopy blues were banished. Well, banished until we had to put the original Plessey yoke back in and put the Mullard yoke back in its rightful place in the CTV16. Tas used his magic fingers and adjusted the convergence in the CTV16 to better-than-new standard (as Tas always does) and then we wandered back to the CTV5 to look at the blanking fault. (As an aside, the poor old Dynatron still needs a bog-standard Plessey convergence yoke if anyone has one spare.)

Anyway, the next fault was a little bit of a puzzler to start with. The symptoms were that there were frame flyback lines visible in the picture, especially at low contrast. After frame sync pulses come in to the frame board from the IF panel, they are clamped and sent straight back out to the CDA panel for blanking purposes. Being the cheats that we are, before we attacked the set with the scope, Tas rescued the frame panel from the CTV16 and we swapped them over. This didn't cure the problem. Okay, we thought, the problem has to be on the CDA panel. After restoring the frame boards to their original sets, out came the CDA panel. On checking the blanking transistor we discovered that it was leaky between collector and emitter and was also giving some odd reading between base and emitter too. A nice new BC147 went in and we replaced the board and switched on expecting great things. What we got were not great things, we got exactly the same problem as before. Oh. How odd. Okay, we thought again, there is another CDA panel upstairs - we'll swap that over to see whether the fault lies in this area. We were very confident that it would cure the problem, but it didn't. How could this be - the sync itself was fine (or so we thought) otherwise there would be problems with rolling picture or poor lock on the frame, so how could it not be the frame board or the CDA board? What we should have done in the first place was to scope the frame sync pulse going in to the CDA panel (well, actually we did in a bumbling kind of way and then ignored the readings) because it turned out that although the pulse looked reasonable in an AC kind of way, we were only measuring it with the scope set to AC and it was actually floating up and down. On following it back to the IF panel, we discovered that the earth wire had come off the plug, and the reason that it didn't upset the vertical sync is that it is AC coupled through C252 on the frame board before it is used to synchronise the frame stage.

So there we have it, a Dynatron with a Pye hybrid chassis and a picture to beat any set, and with the most amazing sound I've ever heard from a TV. (No, really, a Pye with one of those little Mullard sound modules but a huge elliptical speaker.) Even the EHT regulation isn't too bad on this set; not up to the wonderful regulation in one of Tim's Bang and Olufsens perhaps but apart from that the picture and sound are every bit as good as (I might even go so far as to say better than) one of Mr Jarman's finest.

Thanks Tas for your invaluable efforts and amazing convergence skills.

I'm afraid I'm a terrible photographer and the actual TV picture is an awful lot better than the attached image.

Thanks everyone.

Kind regards.

From Mike.

[af024]

Hello Mike,

Another triumph I see. Well done!

It's interesting to learn that you've have a few dodgy BC147, BC148, BF149 and BF194 transistors. I found exactly the same in the 'Classic' branded Bradford that I have - all very low gain. What that the main problem, low gain?

I wonder how common convergence yoke faults are?

Regards,

Andy

[G8VAT Wolfman]

That was a great description of a very complicated set of repairs.. you should be in teaching, oh the nostalgia of it all.

I used to spend hours doing just that sort of thing on that particular chassis, probably couldn't tell my harness from my EL80 now.

[neil29]

Hi Mike. would the convergence yoke from a 26" tube be any good to you. as ive got one you can have if its any use. cheers neil.

[Mikey405]

Thanks Andy, G8VAT and Neil.

all very low gain. What that the main problem, low gain?

Well, the BC148s and BF194s all just seem to go open circuit between random electrodes. I don't know whether it's anything special to that Lockfit design or whether the metal-can variety of BC108s etc. would be just as unreliable if they were used instead. The BC108 in the CDA panel did go leaky, but the can looked a bit rusty.

That was a great description of a very complicated set of repairs.. you should be in teaching

Hmm. I don't know about that but thank you very much anyway. It's just remembering what signals should be where really - nothing much in the way of skill involved.

probably couldn't tell my harness from my EL80 now

I'm sure it would all come back in no time G8VAT.

would the convergence yoke from a 26" tube be any good to you

Thanks Neil - Well if it's a Plessey kind (black and three pieces joined together with strange-looking circular springs) then I'm sure it would be perfect. If it's a Mullard then probably not I'm afraid old mate. I'll see if I can post a picture when I get the back off again.

Kind regards.

From Mike.

[neil29]

Mike here is the item. it is one of the coils that are open circuit im assuming?. as the part that secures them to the tube has one piece missing. cheers neil.

[Mikey405]

Hi Neil.

They look just the ticket and just right to get the old Dynatron working well again.

That's fantastic. Thanks Neil.

Kind regards.

From Mike.
PS. I'll send you a PM about those.

[BGmidsUK]

What a fantastic looking set! I would love one like that. Maybe one day ...

Is it a dual-standard set?


Brian

[Tazman1966]

Certainly did cause a bit of head scratching eh Mike?! Thanks for the very kind words too. It must be said I was only there in the background offering moral support more than anything although I really do like to get to grips with convergence. I will add though that the picture (blue convergence error aside) is superb and with the replacement yoke from Neil fitted, it will be perfect.

[Hybrid tellies]

This has bought back many memories from my early working life in the TV trade. It is amazing to see one of these working so well and your write up was just brilliant.

[Mikey405]

Thanks for the kind comments Brian, Tas and Simon. That is most kind.

Is it a dual-standard set?

The set is a single-standard model Brian. Only the CTV1 and CTV2 are dual standard and I'd murder for one of those.

Thanks Brian, Tas and Simon.

From Mike.

[Mikey405]

Originally Posted by neil29

Hi Mike. would the convergence yoke from a 26" tube be any good to you. as ive got one you can have if its any use. cheers neil.

Spot on thanks Neil - The coil assembly has now been fitted and the convergence and purity are all tweaked up (again, not up to Tas standards, but not bad if I do say so myself ) and the set is displaying a fantastic picture. Actually, the frame was jittering all over the place so I gave the 4 pots on the frame board a bit of a squirt with WD40 which sorted that out completely.

Thanks Neil again for the coil - Top man.

Kind regards.

From Mike.
PS. I almost forgot - After I'd fitted the convergence yolk coil I had yet another "no colour" fault (again again again). To cut a long story short, it was another duff BF194 in the burst gate - open circuit base to emitter. Replacement with another BF255 brought Carole and Bubbles back to life again.

[Tazman1966]

Those pesky "loc-fit" transistors strike again! Glad to hear it's giving a good account of itself.

I now need to sort out the burst gate problems on my own Pye hybrid!

[neil29]

Hi Mike. good to know that its finally sorted . cheers neil.