Ekco T344 plus another one

The detector out polarity look correct.

Not much more progress to report right now, except that I tried the detector the other way around out of interest - the sync was virtually non-existent and the picture inverted. Also, I received the OA91s, but fitting one has made no improvment - so it looks as though, for reference, a modern silicone diode may work as a detector in these sets! Interestingly, I notice that on the circuit for the Ekco T368, the detector is labelled "MR1" as though it's a metal rectifier there rather than a germanium diode.

One thing to note for anyone else tackling a T344 is that there is an error on the service sheet's diagram - R26 is labelled as 470k whereas it is in fact 470 ohms. I was about to take mine out and replace it with the "correct" value until I checked on the very similar T368 diagram and saw it labelled there as 470R.

I did also take some voltages from the tuner stage, and the valve voltages there are all pretty normal except that 125v on the anode of V1.

Something I tried this evening, which I think rules out anything beyond the IF stage as the cause of low contrast, was to inject an amplified signal into the grid of V8, the video amp. All I had available to do this was an audio preamp. Feeding in the video signal from the Hedghog worked with the simple vertical grey bands test card, which seemed to be just within the limit of the amp's frequency response. The sync was iffy, but what this proved was that with a higher amplitude signal being passed into the video amp, high contrast blacks and whites are possible.

I haven't spent much time on the TV lately, I've been a bit stalled with it to be honest - and I have had a side project, a 16mm movie projector about which I've just posted elsewhere on the forum.

I measured voltages in the IF parts of the circuit, and they all look more or less spot-on; I've also tried swapping valves V1, V2, V3, and V5, nothing made any improvement, so I'm thinking that it's worth removing the tuner box and trying to sort out that low voltage at V1's anode (125v versus the 165v specified).

A very quick update on this - I'm trying to get back into it again, after a summer of other projects & duties. As mentioned previously, I'm trying to get to the bottom of the low voltage anomaly in the tuner section (because despite evidence to the contrary at this being to blame, I'm a bit stuck at what could be causing the poor contrast). I've just replaced R9 which was a little high, at about 126k rather than the rated 100k, but unfortunately the voltage has now dropped further to about 107v at the anode of v1 🙁 .

What happens if you apply an external AGC voltage?

That's a good idea - and something I haven't tried. I've only tried increasing the AGC to 0v with respect to chassis, and that didn't make enough of a difference to affect the contrast greatly (from what I remember, this was months ago now!). I'll give that a go, it may be a few days till I get the chance however.

0v on the AGC line and not enough contrast sounds like low gain somewhere, which would also explain the low HT on the tuner since the RF amp is running with low or no bias.

I'm in Oz so I don't know that set. What's in the IF? 3 x EF80s?

@irob2345
Service sheet is in the Data library.

Hmm. Special Mazda valves in the 2 stage vision IF! They will be clapped out by now.

Looks like the closest to the 6F19 in the first IF is the EF85 or 6BY7 in Oz speak. 2nd IF is EF80 / 6BX6.

Since it's a 2 stage IF you could consider replacing them with the newer, higher gain, pin-for-pin frame-grid types, 6EH7 (EF183) and 6EJ7 (EF184) in that order.

Thanks for the suggestion, I'll have look to see whether I already have any EF183 / EF184s at the weekend. Although I have already been through all (I think) the valves in the vision circuits and tried substitutes (several in some cases!).

It's definitely a gain problem, I had the scope on the video amp back in summer, and the peak voltage before it started clipping was way lower than it ought to be. Logically the problem ought to lie in the IF stage, however the lack of any voltage anomalies there is what got me stuck.

Ah I see. So it clips early? Since you already tried the valves and the voltages check out, that leaves the detector diode. Drop a 1N4148 in there and see how you go!

Those newer frame-grid valves were developed to provide good performance from a 2 stage IF. Earlier generation valves were pretty marginal when there was only 2 of them.

Could you try injecting a video signal into the stages and work back from the video output valve to see where it drops.

You mean a video signal modulating a carrier at the IF frequency?

You can use an ordinary VHF sig gen with AM modulation of about 500 Hz which the generator may already have. That will give you horizontal grey - black bars, about 10 on the screen. No sync of course....

The fact it clips at a low contrast level does sound very much like the diode.

Thanks, both of you, and sorry I haven't replied before now - I've only just got to sit down at the computer this weekend, and it's a bit late to start on the telly now unfortunately. I just cannot find the time to get back into this properly at the moment, especially with making more of an effort to get out of the house every weekend due to being shut in for months...

I need to read back through my own replies on this thread to really get back into it. I have already substituted the detector diode, but I can't remember now what type I ended up with. I do remember though that the set seemed to work better with a silicon type rather than a germanium one, perhaps that's a clue as to what's going on.

I see from my inventory that I have one single EF183 in my stock so I will give that a go, just in case.

It's taken me longer than I would have liked to get back into this, but I've just spent a few evenings on the TV and last night was very productive - the problem with the poor contrast is now resolved! I am very happy!

I've been trying a lot of the suggestions from everyone, including replacing the IF valves with EF183 / EF184s. The 184s worked OK but not noticeably any better; the EF183 just completely messed up the sync, the TV didn't like it at all (just as I believe @katybush said a while back). Looking at the AGC it became apparent that it seemed to be working as I'd expect, changing in line with contrast adjustments and dropping to about -0.5v at the maximum setting.

Reading back my comments, I got it wrong above when I said that the video signal was 'clipping' - it wasn't exactly doing that, it was more that it was compressing the upper part of the waveform at higher contrast settings, so effectively distorting and amplifying only the sync pulse part of the wave. I was getting only about 1.5-2 volts peak-to-peak at the detector.

As the signal was compressing and it was before the detector, I became convinced the problem was around the second IF valve and with compression, I wondered whether it was a power supply issue. The DC voltages all measured pretty close to spec, but when I put a scope on the anode of V5, at higher contrast the voltage was dropping just a little along with the signal (a rough waveform being visible on the scope), maybe 5 volts or so, but I thought it probably shouldn't be doing that. I tried taking C43 out of circuit and it didn't make much difference - this made me suspicious. Given that it said 'Hunts' on it, I wasn't going to put it back. I put in a brand new cap, and hey presto - the issue was resolved!

The waveform at the detector now - it's like night and day - I get well over 8 volts peak-to-peak at maximum contrast. In fact, the timebase is now overloaded and loses sync if I turn the contrast up to max!

I'll post some pictures another day, I've stayed up too late again.

I think I have a little work to do on the sound now, the level isn't great and I think it's distorting at high volumes. It could just be the old 6F23s that I relegated to the audio sections - or maybe the several other identical Hunts capacitors that I didn't replace...

Very nearly there though, then just a few minor cosmetics to sort out.

Thanks for all the help so far, everyone!

Congratulations you have that fault sorted out. As for Hunts paper capacitors, or any paper caps of that vintage they will likely all be faulty in one way or another. Philips black tar and Wima toffee types were paper and suffer the same fate. The arrival of the polyester type like the Mullard ones in the 1960’s did wonders for reliability.

Well done for persevering, thought you replaced all the Hunts already !

Thanks both!

I'm afraid I only replaced the larger of the paper capacitors previously. This was a fairly tiny one of which there are still a few.

There's actually one less as of this evening as I've just sorted the audio out too. The problem was R46, a 2.7M resistor which had gone high. I had to move C57 out of the way to get to it, another small paper cap, so I replaced that too. The sound is now much better.

The TV is now working pretty well, good enough to put back in the cabinet I'd say. I've been watching it for over an hour as I type this. As mentioned though, there are about 4 of those small paper caps remaining; I can't decide whether to replace them - I know you'll say I should, I just kind of like the idea of preserving as many original components as possible. But, I know what will happen - they'll fail at some point soon necessitating the removal of the TV from the cabinet again. I suppose I'll probably replace them...

It's ironic that when these passive components became reliable in the 60s, they went and introduced early transistors into everything!

Here's how the picture looks now...

Now of course you could have simply replaced all Hunts capacitors on sight and have a good working set almost immediately. But wouldn't that have been boring?

At least this way, you have made it interesting for us all and no doubt learnt a good bit from the fault-finding excercise.

Well done sir ! 👍 

Posted by: @pye625

Now of course you could have simply replaced all Hunts capacitors on sight and have a good working set almost immediately. But wouldn't that have been boring?......... [ ] no doubt learnt a good bit from the fault-finding excercise.

Hi Andrew, that's the drum I'm always banging.

I was often criticised, occasionally still am, for my "fix on fail" approach rather than following "They'll all need replacing anyway so do it". It's OK for seasoned pros and ex trade folk to do that, they've earned their "irons" and built a large amount of experience, but the likes of us amateurs need the lessons. 

So indeed Chris, well done for sticking with it. 👍 Dogged determination is a key ally in the face of mounting frustration on a stubborn fault. It's clear to see how you've progressed from the start of this thread to the end and what's more this thread will now serve to help all those that follow in your footsteps. So, another thumbs up for adding to the Vrat pool of knowledge 👍 

Posted by: @helloekco

I can't decide whether to replace them - I know you'll say I should,

Not me, personally I would leave them. If and when they fail, they will present symptoms on screen, thus teaching you even further about their function in the circuit. Then you can replace them why throw away an opportunity to learn. Perhaps the exception to this approach would be to replace the line output coupling and frame coupling. The only downside is some folk looking on develop a twitch, I've lost count of the times folk have had kittens when they see "Callins caps" in my Thorn 2000's but the sets still work perfectly fine and when they don't they get fixed.