Horizontal Deflection and EHT question

[Philips210]

Hi.

I remember reading a few times and being told by colleagues that if the width of the display is correct then the EHT should be about right.
Now we know what happens if say the HT to the line output stage is correct and the line flyback tuning capacitor falls in value, there's an increase in the EHT and reduced width.
If there's no fault in the set and reducing the HT voltage and hence EHT also results in reduced width. OK so far.
What if there's a problem with the tuning capacitor resulting in increased EHT then could this lack of width have been adjusted out to mask the problem?
I was thinking the set would appear to be displaying a quite normal sized picture but the EHT could be excessive. A giveaway may be there would be an increase in the display's intensity due to higher EHT. Would that scenario have been possible?

Regards
Symon

[Argus25]

Yes, the EHT could be running high with loss of tuning capacitor value and the width been adjusted to compensate.

In magnetic deflection the width of the scan is inversely proportional to the square root of the EHT voltage. So if the EHT say increased by a factor of 1.3 the picture width (for the same amount of deflection current and field) would decrease by 1/(root 1.3) or be about 88% of what it was before all other things equal.

Also, the EHT generated for a particular width and scanning magnetic field energy is inversely proportional to the square root of the tuning capacitor value, so lets say the capacitor value dropped to say 80% of its correct value, the EHT would increase by 1/(root 0.8) or about 1.12 times higher.

Since auxiliary circuits run from the line output transformer's secondary and utilize rectification of the flyback peak, the voltages there are directly proportional to the EHT voltage and can often be easily checked with the manual for correct value with the meter. Versus say the collector voltage of the H output transistor that could be peaking over 1kv and the EHT itself, where both can be more risky to check without the right gear.

Depending on the design of the CRT's cathode and grid circuits, an increase in EHT and other horizontal output transformer secondary voltages, might increase or decrease the CRT's beam brightness, so that is not a reliable indicator if the CRT's EHT is up. Also the focus voltage tracks the EHT to some extent so it also is not a reliable indicator of elevated EHT if that is happening.

Also, sometimes, energy is taken from the line output transformer on the opposite half of the cycle to the flyback peak, for example to run some auxiliary circuits or the CRT heater, these voltages are of little help in determining if the EHT is up if you cannot measure it directly.

Try to find a winding on the H output transformer that has the flyback peaked rectified and compare the value there (hopefully a value your meter & probes can handle) with what is documented in the manual. Otherwise you will need high voltage capable probes.

[barrymagrec]

Originally Posted by Argus25

Try to find a winding on the H output transformer that has the flyback peaked rectified and compare the value there (hopefully a value your meter & probes can handle) with what is documented in the manual. Otherwise you will need high voltage capable probes.

Very valid point, Philips used to get lots of PM2423 DVMs back with blown input stages due to people not realising the peak voltages available.

[Philips210]

Thank you Hugo for a very detailed explanation. Nowhere in the various television engineering textbooks have I been able to find the answer to my questions.
I have a few EHT meters and HV probes and when I can find the time will do a few tests and experiments. Obviously, I don't want to overstress the line output transformer by reducing the tuning capacitor by too much but will try to create the scenario described in my previous post if only briefly.

Regards
Symon.

[Philips210]

Hi

Slightly OT but related to the EHT regulation, I'd like to ask a question about the tuning of the line output transformer.
In sets that have an EHT overwind that feeds a half wave rectifier I understand that 3rd harmonic tuning is employed. On sets with a voltage multiplier such as a tripler then 5th harmonic tuning is used.

I've seen and read cases where the old LOPTx EHT overwind has failed and has been removed from the LOPTx and a tripler installed with the pulse input taken from the top cap (anode) of the line output valve. Additional HV ceramic capacitors being connected from the anode of the line output valve to chassis to correct the horizontal width.
My question is, is the tuning of the LOPTx therefore not optimum anymore and despite a correct width display, the EHT could be some way off being correct?
From what I recall, the EHT regulation is also impaired ie picture size will alter depending on the beam current.

Regards
Symon

[Nuvistor]

With valve Line output stages the third harmonic tuning in early CTV’s used a shunt stabiliser valve regulate the EHT. With the tripler the tuning was changed to 5th harmonic to give the EHT pulse more of a flat top to improve stability. It was never as successful in the role has the shunt stabiliser but improved safety and reliability.

With transistor line output stages the HT supply to the stage was stabilised and overcame that problem, the transformer would have been tuned to 5th harmonic.

So yes if the LOPTX is tuned to 3rd harmonic in a valve line output stage and a tripler fitted the stability would be worse.

[Philips210]

Thanks for explaining that and confirming my thoughts on the matter Frank.

Regards
Symon.

[Focus Diode]

The solid state Tandberg 2/2 chassis was troublesome in that department. The tuning capacitor was a special component only available from Tandberg.

It was several capacitors in one resulting in progressive width loss and consequent EHT rise as each section of the capacitor failed.

I remember a fault report in I think was the February 1981 issue of, "Television". Lack of width was the initial problem. Nothing amiss was found so as increasing the width "solved" the problem the control was left at its new setting.

Inevitably the set bounced back with the original fault condition. This time setting the width control didn't fill the screen. They belatedly checked the EHT which went off the scale, over 30KV! Replacing the special tuning capacitor cured the problem.

They were other reports of catastrophic set failures caused by capacitor failiure including a fractured CRT neck plus the demise of several components to add to the misery.

It goes to show how essential it is to have EHT measuring facilities when handling colour sets, the saying being if the correct boost volts results in lack of width find out why!

[Philips210]

Hi Brian.

I recall reading that article when it came out and have just been having a read through it again. What a disastrous outcome due to that 'special' flyback tuning capacitor. I bet those sets were feared in the trade. It would have probably been safer to use two or more decent polypropylene caps in series to obtain the correct value if there was room to accomodate them. I suppose it was designated as a safety component so for insurance/litigation reasons, must be replaced with the exact type approved by Tandberg. I wonder if Tandberg launched a recall if that was standard practice back in the early 1980s.
It was interesting though to read through the article by Derek Snelling.

Regards
Symon.

[Argus25]

Symon,

One of the reasons why talk of harmonics in the tuning crops up is because of the design differences in transistor vs valve line deflection systems.

In both cases when the active device or damper diode are conducting the damping is very heavy and the current increases at a rate of V/L where V is the power supply voltage and L the total inductance of the load. When the active device is cut off at the end of scan, the system resonates due to the inductance, its self capacitance and any added tuning capacitance to give the half cycle of oscillation seen as the flyback peak. (the next half cycle is prevented by the damper diode) It is this half cycle width that determines the H flyback time. And the peak amplitude value determines the EHT.

In transistor sets, the operating voltage is lower than valve sets, so they have lower inductances of the yoke and line output transformer (less turns and less self winding capacitance). This also requires higher peak currents to attain a similar deflection field than a valve set as there are less yoke turns. Most of the tuning capacitance is therefore typically added external to the H output transformer and yoke and added across the transformer primary circuit, often seen across the H output transistor's collector-emitter connection.

On the other hand in a valve set, because of the larger inductances required (due to the higher operating voltages), larger self winding capacitances occur and the natural resonant frequency of the yoke & transformer combination can be low enough not to require additional tuning capacitance, or perhaps just a little extra such as a few hundred pF added. However this combination results in much higher peak flyback voltages on the output tube's plate as the Q is higher than in transistor sets, sometimes 6kV or more vs say a few hundred volts to 1200V in transistor sets, even colour ones with 27kV EHT.

When two tuned circuits with different resonant frequencies are placed on the same core you can observe two different resonances depending on how tightly they are coupled.

Due to the light loading on a EHT transformer's secondary (overwind) only damped by coupling on the waveform peaks, then during flyback often the secondary (overwind) resonance is not even visible, and the waveform you observe on the output device during flyback is mainly the primary circuit resonance, downshifted somewhat by the presence of the overwind. In fact you can run a lopty without the overwind for test purposes and the flyback peak often looks little changed.

If there is tight enough coupling and the EHT overwind has a significant self resonance you can see it superimposed sometimes as a higher frequency disturbance riding on the main resonance during flyback. There is a photo of this on page 13 of this article top right. The peaks are upside down because the transistors are used as high side switches:

http://worldphaco.com/uploads/The_19...o_monitor..pdf

[Maarten]

Philips had already discovered the "Tandberg problem" in their 1972 K9 set. Two mixed dielectric capacitors in parallel were used. When one went O/C, the picture got a bit smaller, very sharp and bright, until ****! Suitable polypropylene capacitors didn't come to market until some years later.

[Argus25]

Originally Posted by Maarten

Philips had already discovered the "Tandberg problem" in their 1972 K9 set.

A joke at the time was that the K9 was a dog of a set !

[Focus Diode]

Originally Posted by Philips210

I recall reading that article when it came out and have just been having a read through it again. What a disastrous outcome due to that 'special' flyback tuning capacitor. I bet those sets were feared in the trade. It would have probably been safer to use two or more decent polypropylene caps in series to obtain the correct value if there was room to accomodate them. I suppose it was designated as a safety component so for insurance/litigation reasons, must be replaced with the exact type approved by Tandberg. I wonder if Tandberg launched a recall if that was standard practice back in the early 1980s.

Not sure if Tandberg did a recall or not. In this modern era I wonder what would be classed as a suitable replacement(s)?

Keith Cummings did a follow up letter with his own experience on a set only 18months old. He ended up giving the owners six months interest free payments for the repairs required including a new CRT, all because of one capacitor!

I saw one of these sets in operation as late as the 1990s!

[Peter.N.]

The picture size reduction was very common in early '50's receivers that used a selenium HT rectifier, they went high resistance for a past time, replacing them with a BY100 diode very often caused the HT to rise to an excessive level but the picture size was still OK because the EHT increased as well, even when fitted with a series resistor the voltage was often still excessive especially before the valves warmed up, heard a few 'cracks' from the smoothers but never actually had one fail.

Peter

[Maarten]

Originally Posted by Focus Diode

Not sure if Tandberg did a recall or not. In this modern era I wonder what would be classed as a suitable replacement(s)?

Polypropylene impulse types. MKP or KP/MKP. A well known and very good impulse capacitor is the WIMA FKP1 series but Vishay/BC and Epcos have suitable types as well. New old stock axial polypropylene types from ERO are also a good match.