Focus drift during warm-up

[Roy]

My recently restored 1951 Vidor CN4208 has a focus problem.

When the picture first appears, it is usually slightly out of focus, but settles down within a few seconds. Once in a while, however, it is particularly bad, to the point of being unwatchable, and then it takes much longer to settle. It does always settle eventually, with no need for adjustment, but the time it takes is unpredictable.

I have discovered that, when the focus is bad, it can be made to settle much quicker if I turn the brightness to maximum for just a second or two, then return it to normal. While the brightness is up, the picture can be seen to shift its position on the screen slightly as the focus steadily improves. Once the focus has settled, there is no more drift until the next switch-on.

The CRT is a Mullard MW31-74, the focus unit is a permanent magnet type made by ELAC, and EHT is quoted as 6.4kV in the service notes.

Do the symptoms I have described sound familiar? Any guidance would be much appreciated.

Thanks

Roy

[Ray Cooper]

I'm speculating a little here, but it does sound a bit as if the CRT is slightly soft, i.e. 'gassy'.

Gas in a tube has a well-known focusing effect - in fact, the earliest CRTs (pre-war) were 'gas-focussed' - a very small amount of gas was deliberately left inside the tube, and this assisted the focussing of the electron beam. Such tubes were not very stable, or long-lived, which is why the 'hard' tube was developed.

It follows that presence of small amounts of gas in a 'hard' tube can undo the best efforts of the electrostatic/magnetic focussing system. Why does the focus improve as the tube is run? It seems that the gas molecules are swept out of the way by the electron beam after a while, so cease to defocus it. Switch the tube off, and they slowly drift back, ready for next time. Giving a quick burst of high beam current, by turning up the brightness, accelerates the sweeping action.

Well, that's my take on what is happening - I may be quite wrong. But the bad news is, you have to ask yourself where the gas is coming from, and why the 'getter' in the CRT isn't dealing with it. There is a possibility that you've got a 'slow puncture' in the tube - possibly in an area where the lead-through wires are sealed into the glass. If so, eventually the focussing will get so bad that the tube becomes unviewable.

Has anybody ever tried to stave off the inevitable, I wonder, by attempting to re-seal the lead-through wires - perhaps with an epoxy resin?

[thyratron]

I have the same trouble with a Bush TV24. At first the raster does not fill the screen because of what looks like static charge on the screen . It reminded of the effect I have seen on a VCR97 when you touch the tube face.
The TV24 takes a while to come up and then operating at maximum beam current tends to push the rater out to fill the mask. The focus then can be adjusted cold but always needs another tweak after 10 mins. The tube is tired but after about half an hour all settles down and is quite watchable in a dim room. I call these 'winter' tubes!
Where can I buy a new MW22-16 ? and If they have all gone where does it leave our sets?
Perhaps they really will all end up in the bin one day? (ouch)

[Roy]

Thanks to all for your responses.

In some respects, the symptoms do suggest gas in the tube, but on balance I feel this is unlikely. I'm far from being an expert on this type of CRT, so please do correct me if I have misunderstood this, but I was under the impression that a slightly gassy tube would show symptoms similar to those of a low-emission tube, due to cathode poisoning. Such tubes improve with regular use, but the symptoms return if the tube is left idle for days or weeks. This MW31-74 does not behave like that - it can be left unused for any length of time and still produce a clear, bright picture at once, whereas it only needs to be switched off briefly for the focus problem to return.

I wasn't aware that bulb charge could cause temporary defocussing, that's very interesting. I had noticed that the raster is a very strange shape if the brightness is set to a low level during warm-up. There seems to be foldover at the edges or corners, this corrects itself as the brightness increases. I assume this effect is due to static buildup. Is there anything that can be done to limit the buildup of static on the tube face, such as an antistatic coating?

I have made some attempt at measuring the EHT using a calibrated resistor chain of 200M in total, it comes out at around 7.3kV at zero beam current, dropping to 7.0kV displaying the testcard at normal levels of brightness and contrast. The EY51 is new, and the raster only just fills the mask with the height and width controls at about 90%, so I am reluctant to try raising the EHT any further, in case there is insufficient scan power available.

Roy

[Studio263]

Check the tube heater voltage. It is not uncommon for Mullard tubes of this period to get a partial short circuit across part of the heater which in a set where the tube is part of a series chain has the effect of drastically lowering the emission (= poor focus in some cases). Normally the voltage should be 6.3V, it can fall to half of that when a short occurs.

Unfortunately there is not much you can do about this when it happens, a careful tap on the neck can clear it for a bit though. Don't try a heater transformer whatever you do, this will just open-circuit the bit of heater that you have left.

[Roy]

Yes, I remember reading about the partial heater s/c problem in another thread.

This particular set has no heater chain, however, all heaters are parallel-fed from the mains transformer. I did check the heater voltage during restoration, it was 6.5V. If the CRT did develop a partial short, I wouldn't give much for its chances of survival.

[Steve_P]

Are the power supplies OK. If they're down a bit, it could cause picture size variation. You then correct it with the EHT and the problem is 'masked' so to speak.

Also, try new Line O/P, Efficiency and Frame O/P Valves too. Check for High Value Resistances and any caps you may have missed.

Cheers,

Steve P.

[Roy]

The power supplies do seem OK. All voltages measure slightly higher than quoted in the service notes, but I put this down to a combination of higher mains voltage (around 254V here) and the use of a DVM. The line and frame o/p stages run from the main HT line (320V) and everything else is supplied at 270V. There is no efficiency diode. The set has been the subject of a thorough restoration, with all capacitors (except mica and one or two electrolytics) changed for modern equivalents. A handful of resistors also needed changing.

Apart from the temporary focus problem, the set works beautifully, giving a bright, sharp picture which is difficult to fault.

I have attached the circuit diagram (just a couple of pages from R&TVS).

Roy

[Steve_P]

Why don't you start it 40s/50s style?

1: Switch on and leave brightness down, with sound up.
2: Leave for 5 minutes.
3: Give it Test Card C, and slowly advance the Brightness.
4: See if it works that way.

Cheers,

Steve P.

[Roy]

Originally Posted by Steve_P

Why don't you start it 40s/50s style?

1: Switch on and leave brightness down, with sound up.
2: Leave for 5 minutes.
3: Give it Test Card C, and slowly advance the Brightness.
4: See if it works that way.

Thanks for the suggestion Steve, but if I do that, the picture comes up rather oversized, with severe foldover at the edges, presumably due to static. It still has to be left at normal brightness for a while for the focus to reach optimum.

I think I need to sort out this static problem before going any further. I suppose I need to remove the CRT from the mask and spray the screen with something - any recommendations?

I have just remembered something else which might explain the static buildup. The original CRT (MW31-16) had an aquadag coating which reached right into the rubber mask, close to the edge of the screen. The coating on the MW31-74 stops well short of the mask, which means that the inside of the mask is insulated and presumably could hold a static charge. Should I perhaps consider extending the aquadag coating closer to the screen? If anyone has experience of dealing with static problems on this type of CRT, I would be grateful of any advice.

Thanks

Roy

[Roy]

I had another go at this over the weekend. As suggested by Danny, I tried a solid state rectifier in place of the EY51. I "borrowed" one of the sticks from a BRC 1500 tripler for this.

Success! - up to a point. The increased EHT does indeed make the focus settle down much quicker. Unfortunately, the raster no longer fills the mask, even with the width control at maximum, so the EY51 will have to go back in. The problem remains, but at least the cause is known.

I have reluctantly concluded that the CRT, whilst being perfectly serviceable, is not ideal for this particular set, and would be happier in a receiver with a higher EHT. Unless a NOS one turns up, which I feel is unlikely, I shall have to continue to tweak the brightness control to get the picture into focus. It's not such a big deal - provided it doesn't get any worse.

Roy

[Steve_P]

There are 2 sorts of tripler in the BRC1400/1500 series. 5 stick (small and square) and 3 stick. Or, you could put the EHT Winding back in and try a one diode arrangement, using a BRC1580 diode?

Cheers,

Steve P.

[Niechcial,Steve]

Does this set have a sepeaate EHT smoothing cap? I once had one of the oil filled visconol types produce exactly this fault, the focus getting worse and worse over a period of about 15 minutes. Presumably there was increased leakage across the dielectric as it warmed up. Replacing it completely cured the problem.
Steve

[Roy]

The tripler was a 3-stick type, so a doubler really. I didn't use it complete, I just opened the casing and removed one of the sticks, and fitted it to the existing LOPT overwind in place of the EY51.

The problem is that if the EHT is increased to improve the focus, more power is needed to scan the tube. The very basic line o/p stage of the Vidor cannot provide this.

No separate EHT capacitor, thankfully. The CRT is aquadag coated.

Roy