TV22/24 metrosil

[Helder Crespo]

Originally Posted by Panrock

I take your point about the high Curie temperature indicating that losses in the windings themselves could be the dominant factor in raising the temperature here. However, these losses would not be primarily resistive but due to the 'partial shorted turns effect' caused by leakage in the degraded waxy encapsulation (as previously mentioned). Otherwise, a new LOPT would suffer just as badly.

Please correct me if I have misunderstood you.

Steve

One thing I did notice as I repaired my second LOPT (the one which has not been rewound but has been fully stripped of pitch and then varnished) is that the resistance of the overwind was strongly dependent on the temperature, as expected due to its large length / number of turns. Just by touching / handling the practically bare overwind, the resistance would increase by tens of ohm. Given the temperatures that can be reached inside the line output stage screening can, I think the change in overwind resistance will be quite significant. This will of course be due to everything that contributes to heating up the can, including the PZ30 and the PL38, so you don't really need to have a heated core for this. One thing that I have always thought about is that perhaps the EHT dependence on beam current I'm getting is still within the normal range for the TV22. I really can't say for sure but I have seen a couple of other sets that don't seem to have this behaviour.
Helder

[Ed_Dinning]

Hi Helder, tempco of copper is about 0.4% per 100 degC, so not asa much as you have seen, that could be doen to poor insulation on the very fine wire.
Most of the TV 22 LOPTs that I've done had badly carbonised primary windings.
These gave the fault of quickly reducing EHT over about 1/2 hour until the picture vanished.

Ed

[Helder Crespo]

Originally Posted by Ed_Dinning

Hi Helder, tempco of copper is about 0.4% per 100 degC, so not asa much as you have seen, that could be doen to poor insulation on the very fine wire.
Most of the TV 22 LOPTs that I've done had badly carbonised primary windings.
These gave the fault of quickly reducing EHT over about 1/2 hour until the picture vanished.

Ed

Hi Ed. Thanks, but I think you got that wrong by a factor of 100. It's 0.4% per 1 degree Celsius, so a 1 degree change will give a difference of 3.2 ohm in a 800 ohm winding. And you can imagine how much the winding resistance will increase inside the screening can! 30 Celsius more (not unlikely) would give you an extra ~100ohm, which is more than 10%.
My first LOPT had a shorted primary (basically the carbonisation issue you mention) but was rewound. My second LOPT is original but was fully stripped and varnished; it has been running for 8 hours or so every other day since more than 1 year ago without any problems. I'm actually watching BBC news on my TV22 right now. I do get a reduction in EHT over a period of 1/2-1 hour with both LOPTs, but this stabilises at a relatively high value of around 7kV (vs. 8-9kV after a 5 minute warm up from cold). This is fine, I guess, and my issue is the much larger decrease in EHT with beam current - I have measured EHT down to 3kV or so when the brightness is set to maximum.
Helder

[Ed_Dinning]

Hi Helder, my Kaye & Laby and a couple of other books give it as 0.004 per deg C at 20C, so yes 4% per deg C.

I've just rewound a TV22 Lopt and with first light, raster only I have 7KV DC eht at full brightness and 7.5KV with zero beam current, measured on an electrostatic meter.
The acceprable figure is, I believe anything above 5.5kv

Ed

[Panrock]

I can't see how the losses seen, either due to load or heating of a decrepit winding, can be resistive in the classic sense. Even at 1mA of beam current - a huge amount - the loss across 800 ohms would still only be 0.8 volt. Yes, I know it's not smooth DC in the winding but am I otherwise missing something?

Steve

[Helder Crespo]

Originally Posted by Panrock

I can't see how the losses seen, either due to load or heating of a decrepit winding, can be resistive in the classic sense. Even at 1mA of beam current - a huge amount - the loss across 800 ohms would still only be 0.8 volt. Yes, I know it's not smooth DC in the winding but am I otherwise missing something?

Steve

Hi Steve. You are right. But there are two different things here. One is the decrease in EHT as the transformer heats up - we know that the core and all windings can get quite hot - which reduces the EHT. In my two transformers, neither changes in wax nor water ingress can be behind this effect since one has been dried & repitched a couple of years ago and the other was dehydrated & varnished. On the other hand there is the beam current dependence, which in my case occurs both for the cold and the hot LOPT. This huge reduction in efficiency at larger beam current is definitely not resistive. Since a breakdown in the primary/secondary and overwind insulations is very unlikely in my two recently treated LOPTs, the losses may be in the core. i hope the ongoing treatment of the laminations of my rewound LOPT will shed some light on this.
Helder

[Helder Crespo]

Originally Posted by Ed_Dinning

Hi Helder, my Kaye & Laby and a couple of other books give it as 0.004 per deg C at 20C, so yes 4% per deg C.

I've just rewound a TV22 Lopt and with first light, raster only I have 7KV DC eht at full brightness and 7.5KV with zero beam current, measured on an electrostatic meter.
The acceprable figure is, I believe anything above 5.5kv

Ed

Hi Ed. The EHT value that you get at full brightness is much better than anything I'm getting with either of my LOPTs. I imagine you are talking about a fully white (beam fully on) raster, correct?
Does your CRT have good emission? I'm asking this because if it's low emission, the load is not the same as for a higher/normal emission tube.
Helder

[FRANK.C]

Hi Helder
I will be interested to see if your treatment of the laminations will help. But this effect is also seen in transformers that have cores that are not laminated. I had a similar problem with the EHT transformer in a Ekco TSC48. The core in it was made from a material that looked like ferrite.

Frank

[Helder Crespo]

Originally Posted by FRANK.C

Hi Helder
I will be interested to see if your treatment of the laminations will help. But this effect is also seen in transformers that have cores that are not laminated. I had a similar problem with the EHT transformer in a Ekco TSC48. The core in it was made from a material that looked like ferrite.

Frank

Hi Frank
In your case eddy currents are not an issue but there may be other losses, e.g., a failing insulation and/or water ingress in the EHT winding, which gives symptoms similar to a partially shorted turn, as mentioned by Heatercathodeshort and Panrock. The reason I suspect the laminated iron core is because such insulation related losses are highly unlikely in my LOPTs due to the treatments they have been subjected to (one of them has also been rewound).
I will post my findings in the forum once I finish the laminations and test the LOPT.
Helder

[Panrock]

Yes, I agree that in your case the partial shorted turns effect in the overwind is not a factor (and as we have seen the effect of DC resistance on regulation will always be insignificant). Therefore, impaired hysteresis in the core could already be present when cold - as shown by the poor regulation v. beam current. Then, when it then heats up, this gets still worse - causing reduced EHT among other things.

I am not absolutely clear about the present condition of your internal (non overwind) windings? 'Carbonisation' here (as pointed out by Ed Dinning) could bring back the partial shorted turns effect with a vengeance, with consequent heating.

You're doing sterling work here, Helder. I look forward to the results.

Steve

[Ed_Dinning]

Hi Helder, tube on this set is not brilliant, but should give a watchable picture when finished, emission showed as quite good on a crude tester.
The figures given were for max brill and tube cutoff. Measured on a 25KV EHT meter, so could be over reading a bit at 7KV.
New valves fitted and this set has a PL81 LOP valve.
Transformer had pri winding rewound; EHT as original. Some rust on the Lams but not cleaned off.

Ed

[Helder Crespo]

Originally Posted by Panrock

Yes, I agree that in your case the partial shorted turns effect in the overwind is not a factor (and as we have seen the effect of DC resistance on regulation will always be insignificant). Therefore, impaired hysteresis in the core could already be present when cold - as shown by the poor regulation v. beam current. Then, when it then heats up, this gets still worse - causing reduced EHT among other things.

I am not absolutely clear about the present condition of your internal (non overwind) windings? 'Carbonisation' here (as pointed out by Ed Dinning) could bring back the partial shorted turns effect with a vengeance, with consequent heating.

You're doing sterling work here, Helder. I look forward to the results.

Steve

Thanks Steve. Shorted turns in the primary/secondary section of any of my two LOPTs are very unlikely, for the following reasons:

1) my first LOPT (the original one on my set) was not treated in any away and developed a short between primary and secondary a few days after first light (not uncommon with untreated TV22 LOPTs); it was then rewound with new materials by Mike Barker and has been working since then.

2) my second LOPT came from a scrap chassis that I purchased for parts, and even though it was in a relatively poor state, all of its windings measured ok resistance and inductance wise, as did the windings in all remaining wound components in the same chassis (including the usually O/C frame blocking oscillator transformer). This second LOPT was dehydrated in white spirit and varnished prior to being tested in my set and has also been working since then. While it would be possible for the primary/secondary windings in this second LOPT to have a weak/damaged paper insulation, this usually only happens if you power up an aged LOPT without first treating it, or at least drying it thoroughly, which was not the case here.

3) I did a ring test to both LOPTs, including simulating a shorted turn with a wire loop around the core. The observed damped oscillations without the shorted turn are virtually the same for both transformers, and similar to other ring tests performed on TV22s and reported elsewhere. Below you have two scope screenshots of the ring test on my second LOPT (the electrically original one that has been varnished), without and with the "shorted turn". The third image is a photo of my first LOPT (with pitch removed) undergoing a similar test, where I shorted one of the secondary windings - this results in a shorter period of the fast oscillations under the damped envelope. These tests were performed with a 5V square wave as the excitation voltage, and so one cannot rule out the possibility of leakage for nominal operating voltages. In any case, such leakage would have to occur in both the varnished and the rewound LOPTs, since they both show exactly the same behaviour (poor EHT regulation with increased beam current), which is very unlikely since the two transformers have very different primary sections (new rewound vs. original+varnished).

Once I manage to get some results with the treated core I may post them in a dedicated thread where I will start by giving a clear graphical description of the two LOPTs and their characteristics, to facilitate checking/understanding what they have in common as well as their differences.

Even though I believe that core saturation issues may be behind the poor EHT regulation observed with both LOPTs, I think that a proper study should involve measuring/quantifying the different waveforms at various points in the line output stage, in the line of the approach and results given in Hugo Holden's detailed TV22 report.

Helder

[Helder Crespo]

Originally Posted by Ed_Dinning

Hi Helder, tube on this set is not brilliant, but should give a watchable picture when finished, emission showed as quite good on a crude tester.
The figures given were for max brill and tube cutoff. Measured on a 25KV EHT meter, so could be over reading a bit at 7KV.
New valves fitted and this set has a PL81 LOP valve.
Transformer had pri winding rewound; EHT as original. Some rust on the Lams but not cleaned off.

Ed

Thanks Ed. Even if the 7kV may be a bit optimistic, those figures are still very impressive compared to what I'm getting with my LOPTs and set. A key point here is the contrast (gain) setting. If I reduce the contrast on my set substantially, then I get much better EHT stability, but the contrast in this situation is way below the value for adequate viewing even in subdued lighting conditions. Your stage could be stiffer due to the PL81 (which has a screen grid network not found in the PL38 version). It would be great if you could provide a couple of images (or a short video) of your set in operation, e.g., showing a test card for different brightness values, so that I could have a better idea of the amount of change in the image (if any) for different brightness settings in your set.

Helder