EHT reservoir cap - anything wrong with small ceramics?

[Helder Crespo]

Originally Posted by Techman

I knew you wouldn't. You've just added additional smoothing.

There's a difference between 'regulation' and 'smoothing' and adding capacitance across the output isn't going to give any extra regulation. Lack of smoothing causes a sort of cogging, well, more correctly bent verticals on high beam current. As you saw earlier I used 700pF to replace a 1,000pF Visconol type and it worked just fine, but that's not regulation.

You didn't add very much at only 400pF. You can get larger values of that type you were using cheap enough. Below is a picture of a 1,000pF type that I bought a while ago especially to fit in that TV, but to be honest the random red one seems to work well enough, so I probably won't bother. I can't remember what that one cost, as I bought it with some other capacitors, but it was something like a couple of quid, I think.

The hope was that the average voltage would go up a bit due to (hopefully) smaller ripple and perhaps not fluctuate so much during operation, which could correspond to better regulation. I was clearly misguided, as the effect is negligible. The LOPT + rectifier seem to be the limiting elements here. I don't agree that a 40% increase in capacitance is small given that my aquadag is already 1nF. That's why I stopped there instead of risking to overload the LOPT.
I am aware of the various voltage ratings, capacities and prices of the blue ceramic caps, and merely got 200nF ones (20 of them for 5 quid on eBay) so that I could increase the capacity more gradually. You can get them up to 1nF, like yours, and 30kV.
- Helder

[Heatercathodeshort]

The capacity between the inner and outer 'plates' of a typical tube is around 500pf, more than enough to smooth the 10kc/s line ripple.
I have some .01uf 15kv capacitors and just as an experiment, a while back connected one across the EHT supply. No improvement in EHT regulation, it just held a charge for a lot longer.
If you are experiencing poor regulation at normal viewing levels of brightness and you have replaced the EY51, PL38 and PZ30 valves it is likely that the LOPT is showing it's age. Deteriorated laminations, low Q windings aggravated by years of damp are likely causes. It is much harder to obtain good regulation with the TV12 series. The TV22 are usually very good. They are not computer monitors and a slight increase in picture size with maximum brightness [never used] is quite normal for a TV from 1950. Regards, John.

[Helder Crespo]

Originally Posted by Heatercathodeshort

If you are experiencing poor regulation at normal viewing levels of brightness and you have replaced the EY51, PL38 and PZ30 valves it is likely that the LOPT is showing it's age.

I have changed the rectifier, line output valve and efficiency diode and I still have poor regulation at normal contrast and brightness levels. I agree that the problem is probably the LOPT, even though I may have a bit too much beam current (slightly soft tube?), which shouldn't help.

Originally Posted by Heatercathodeshort

Deteriorated laminations, low Q windings aggravated by years of damp are likely causes.

Yes. I will be checking the first point soon. Winding issues are less likely in my case because I have exactly the same behaviour with two different LOPTs and one of them has been rewound to a high standard (primary and secondary only). I understand you are probably referring to the wave-wound overwind, but again these are different in my two LOPTs (one overwind has been repitched and the other has been thoroughly cleaned and dehydrated with white spirit and then varnished) and I still get the same behaviour. I am in the process of cleaning and insulating the laminations in one of the LOPTs and will then check whether the Q has improved and if this has resulted in better regulation. If this is inconclusive I may try a tripler to bypass the overwind and check whether the problem could indeed be there.

Originally Posted by Heatercathodeshort

It is much harder to obtain good regulation with the TV12 series. The TV22 are usually very good. They are not computer monitors and a slight increase in picture size with maximum brightness [never used] is quite normal for a TV from 1950. Regards, John.

Yes, I am aware of that, but I still think that the reduction in EHT that I am getting is too large, even though it may well be the best that my LOPTs can do right now. The Bush fault location sheet mentions that, for a normal tube, the EHT should be at least 7kV at normal brilliance and fall to 3kV at maximum brilliance, but I'm getting this for normal brilliance depending solely on the image content, with bright scenes dragging the EHT down to 2-3kV.
- Helder

[Techman]

Originally Posted by Helder Crespo

I don't agree that a 40% increase in capacitance is small given that my aquadag is already 1nF. That's why I stopped there instead of risking to overload the LOPT.

You won't overload the LOPTx - the EY51 will easily limit things, it's a dreadful valve at the best of times and doesn't have a long life anyway. It could even be the EY51 that causing the problem, but from what you've now said in the previous post I would much more suspect the LOPTx. You say that you've sealed it in pitch and/or varnish, but the trouble is how certain can you be that every last microscopic drop of moisture was completely removed from those overwind windings. The trouble with sealing the windings with modern products is that there's a risk of sealing in forever any slight moisture that may still remain. It's best to give the set a good running for a few weeks before sealing to make sure that the transformer is completely dry and to make sure you don't have a problem like the one you now report. I know you say that you've dried them out, but I suspect that you haven't quite achieved it.

[Helder Crespo]

Originally Posted by Techman

You won't overload the LOPTx - the EY51 will easily limit things, it's a dreadful valve at the best of times and doesn't have a long life anyway. It could even be the EY51 that causing the problem, but from what you've now said in the previous post I would much more suspect the LOPTx.

Yes, you are right. In this case it's really the LOPT, as I've replaced the EY51 twice to no avail; I've also placed a HV diode across - same thing.

Originally Posted by Techman

You say that you've sealed it in pitch and/or varnish, but the trouble is how certain can you be that every last microscopic drop of moisture was completely removed from those overwind windings. The trouble with sealing the windings with modern products is that there's a risk of sealing in forever any slight moisture that may still remain. It's best to give the set a good running for a few weeks before sealing to make sure that the transformer is completely dry and to make sure you don't have a problem like the one you now report. I know you say that you've dried them out, but I suspect that you haven't quite achieved it.

I can't be sure. But the varnished LOPT was left in white spirit for 3 weeks (this displaces water) and then varnished over a couple of days, followed by drying over another couple of weeks. When placed in the set, it took a couple more days for its behavior to stabilise and for the EHT to rise to its present 8-9kV at low/medium brightness setting, as the remaining solvent evaporated. I should point out that my set does not show the EHT falling over time (this is sometimes called "LOPT saturation"), as do sets with damp overwinds.
- Helder

[Jac]

Helder,

Are you sure it's not the CRT?
Relatively easy to check by using the EHT from a known good set.

Check the EHT of this set to power a CRT of the known good set to make sure it is problematic.

Jac

[Helder Crespo]

Originally Posted by Jac

Helder,

Are you sure it's not the CRT?
Relatively easy to check by using the EHT from a known good set.

Check the EHT of this set to power a CRT of the known good set to make sure it is problematic.

Jac

Hi Zac,

I'm not sure if the CRT is to blame or not as I don't know how much beam current is considered too much beam current for the MW22-16. I suspect this may be the case, even though the overall image quality is really very good and bright. From my readings, this is not uncommon with slightly soft tubes though.

I don't have a known good set. My whole "collection" amounts to 3 TV22s of different vintages (Mk.2, Mk1.5 and a recently acquired late Mk.1) and only the Mk.2 has been restored and is working. Testing the other two MW22-16s with the working Mk.2 did cross my mind. Their heater resistances measure ok, which is promising; they will surely need to be left working for some time to "wake up" but this should not take more than a few hours.
I must confess that I feel more tempted to do this than to finish treating the 48 laminations of my presently disassembled LOPT. There is still a chance that all three tubes are soft and the results are inconclusive, but as long as its fun to do and nothing gets necked, well, that's one of the main reasons why I'm into this hobby.

- Helder

[Heatercathodeshort]

I didn't find the EY51 that bad! Many old receivers still have their original EY51 valves still working fine. It was used in the Philips G6 colour chassis till 1971.Now if your talking about the Mazda U25, that was a bad one!

Giving the massive difference in EHT voltage between normal and high brightness it may be the tube but if it gives a very good focused picture when viewed normally I would suspect the LOPT first.

[Helder Crespo]

Originally Posted by Heatercathodeshort

I didn't find the EY51 that bad! Many old receivers still have their original EY51 valves still working fine. It was used in the Philips G6 colour chassis till 1971.Now if your talking about the Mazda U25, that was a bad one!

Giving the massive difference in EHT voltage between normal and high brightness it may be the tube but if it gives a very good focused picture when viewed normally I would suspect the LOPT first.

My experience with the EY51 was not great either. I'm an my second NOS after only a few months of use. Last night I measured the EHT again. I get 8-9kV under normal viewing conditions with an average brightness image (Test Card C), which drops to 5kV when I turn up the brightness to maximum. This is still within the margins given in the Bush fault location sheet (>7kV normal and down to 3kV for max brightness). I do get a very good and homogeneous focused picture when viewed normally. The focus is lost in very bright scenes due to the change in EHT though.
- Helder

[Heatercathodeshort]

This is one of my TV22s. It has it's original cracked and baked LOPT and the EY51 looks original. First picture with blank screen 9kv. Second with average brightness level pic which is excellent but the camera does not show it. 9kv and the last is complete white out with brightness on full, just under 7kv. Tube is original DAG coated Mullard MW22-16 with ion trap.

If yours is reducing to as little as 3kv it's my guess that the LOPT has lost it's grunt.
Regards, John.

[Helder Crespo]

Originally Posted by Heatercathodeshort

This is one of my TV22s. It has it's original cracked and baked LOPT and the EY51 looks original. First picture with blank screen 9kv. Second with average brightness level pic which is excellent but the camera does not show it. 9kv and the last is complete white out with brightness on full, just under 7kv. Tube is original DAG coated Mullard MW22-16 with ion trap.

If yours is reducing to as little as 3kv it's my guess that the LOPT has lost it's grunt.
Regards, John.

This is invaluable information John. Many thanks!!!
The last picture is saturated of course, due to the high brightness of the image, but how much has the image changed geometrically? Could you please take a picture with a smaller aperture/exposure to avoid saturation, if I'm not asking too much?
- Helder

[Heatercathodeshort]

If the EY51 is failing after just a short amount of use it may be overrun. A small resistor was shunted across the heater to absorb excess current. I must admit in over 50 years repairing old TV's I never had any problems with the EY51. It used to go O/C but no more so than the B9A EY86.
Some manufacturers that had originally wound their line output transformers for the 2V Mazda U25 often modified replacements to take the 6.3V EY51. This usually put a stop to EHT rectifier failure. The U25, fatter than the EY51 failed in various ways. O/C heater, internal flashing and gassy creating a bright pink/purple glow. John.
[PS It remains the same size. you can just about make out the edge of the test card on the right.]

[Helder Crespo]

Originally Posted by Heatercathodeshort

If the EY51 is failing after just a short amount of use it may be overrun. A small resistor was shunted across the heater to absorb excess current. I must admit in over 50 years repairing old TV's I never had any problems with the EY51. It used to go O/C but no more so than the B9A EY86.
Some manufacturers that had originally wound their line output transformers for the 2V Mazda U25 often modified replacements to take the 6.3V EY51. This usually put a stop to EHT rectifier failure. The U25, fatter than the EY51 failed in various ways. O/C heater, internal flashing and gassy creating a bright pink/purple glow. John.
[PS It remains the same size. you can just about make out the edge of the test card on the right.]

The heater on my EY51 seems to glow a normal colour, very similar to all other valves. Not a very precise and objective comparison, I know, but I don't think it's overrun.
Your same-size ultrabright image is very impressive! Having struggled with the low EHT and image/focus distortions ever since I started this hobby 3 years ago, I cannot but be impressed. And this shows that it is indeed possible to have very good EHT regulation on the TV22, in spite of the lack of active regulation in this basic TV. When I have some time, possibly next month, I'll try my other CRTs with the present TV22 and LOPT, and in case this fails I will finish treating the laminations of my presently disassembled extra LOPT. In case this fails too, I'll have to live with it as the problem may well be in the overwind, which is hard/expensive to rebuild. I don't think anyone here in the forum can do it, but it could in principle be done elsewhere, e.g., by this company, although I have no idea how much a custom job like this would cost (btw, the photo in their website below is very interesting and definitely reminiscent of the overwind in the TV22 LOPT):
https://agw.co.uk/news/article/new-m...winding-at-agw
- Helder

[Helder Crespo]

Yesterday I tried the tube on one of my unrestored sets (the Mk1.5) and the results were similar to the original tube, although the decrease in EHT with brightness is a bit less. This of course depends on image content, but for Test Card C, the EHT drops from 8.5kV at normal brightness to a bit below 7kV for maximum brightness. So, as John (Heatercathodeshort) suggested, the bottleneck here seems to be the LOPT. The good news (for me) was that the tube from the Mk1.5 seems to be in excellent condition, and gives a very bright and focused image. One thing that surprised me quite a lot though: I was expecting to have to adjust the focusing magnet x-y position, as well as the H and V linearity and height controls for the new tube. However, I didn't need to touch a thing as the image was linear and well centred in the mask. Either I should play the lottery today or the manufacturing tolerances of the MW22-16 are really tight. AFAIK, these tube require a lot of glass turning and spinning, coating is partially done by hand, etc., so I was believed the differences in e-gun positioning and angle would require readjusting the external field to optimise the image.
Ayway, the next step will be to try to improve the insulation in the laminations. It also seems that energy is being lost in the primary section and not the overwind, as I can perceive a reduction in the brightness of the EY51 filament...
- Helder

[Helder Crespo]

In my previous post, I meant "Yesterday I tried the tube *from* one of my unrestored sets..."