TV22/24 dimming picture

[Ian - G4JQT]

My rebuilt TV24 has developed a 'feature' that I suspect may be the new-old-stock tube I was lucky to get a couple of years ago.

Image starts OK then gets dimmer over 30 mins or so. The picture geometry is quite good and remains stable after the 15 minute warm up for the rest of the evening.

The focus also starts to drift. Increasing brightness/contrast just raises the black level to greys.

The set hasn't had a huge amount of use since I fitted the NOS tube, so I'm hoping the tube may improve, although this seems unlikely after probably 50 hrs or so use. Although the measured emission was pretty good (can't remember what it was) and the picture was quite good when first fitted, something is definitely deteriorating.

Am I barking up the wrong tree and it's something else in the vision circuits? If it is the tube I guess a boost wouldn't do any harm...?

I've got two Aurora converters and it's the same on both.

Thanks.

Ian

[Heatercathodeshort]

My thoughts are not directed towards the tube. A NOS tube is unlikely to be faulty unless of course it was a faulty new one that was not returned to Mullard under guarantee. These are often found in old radio shop basements and must be treated with caution. This is very common especially with 'NOS' valves.

First I would measure the heater volts across pins 1&12 either side of the key providing of course it is fitted with a Duodecal base, MW31-16? It should read close to 6.3V under the fault condition. If it reads much under this the tube may have a partially shorted heater but as it drifts slowly into the fault condition, I would not put this high on the fault list.

Next I would check the A1 voltage on pin 10 [Duodecal base] under working and fault conditions. Did you replace the A1 decoupling capacitor hidden behind the tag strip?

Last I would check the EHT voltage especially as you say the focus drifts possibly pointing to a failing lopt. Just my first thoughts. John.

[Ed_Dinning]

Hi Ian, I think there was a discussion on this type of fault a year or so ago. In that instance it was caused by faulty insulation between the LOPT primary windings that damped out the line osc action

Ed

[wd40addict]

It's almost certainly drooping EHT, mine does the same. One day I'll get brave enough to do the white spirit dipping of the LOPT that has been described here before:

https://www.vintage-radio.net/forum/...d.php?t=172523

[Graham G3ZVT]

'Picture gets dimmer'. By that do you mean there is little availible brightness, or is there plenty of screen illumination, but with a washed out low contrast image?

[Helder Crespo]

Originally Posted by Ian - G4JQT

Am I barking up the wrong tree and it's something else in the vision circuits?

Something worth checking is the state of the valve pins / holder contacts in the vision circuits. Over a couple of years they can suffer from slight oxidation and this can result in heat induced jumps in gain that resemble EHT changes. A few gentle wiggles in the RF and all vision IF valves should reveal any issues there.

[Heatercathodeshort]

The clue here is the focus drift pointing to a drop in EHT after a period of time. J.

[Andrew2]

Wouldn't a drop in EHT cause the picture size to increase?

[Heatercathodeshort]

Agree if it were just the EHT that was failing as would be displayed if the EHT rectifier was low emission. The low efficiency line output transformer caused by age, rust and dampness produces a massive reduction in 'Q' reducing scan power and EHT in similar proportions. J.

[Andrew2]

Ah, that's a good point, J. Although that would affect only the line scan amplitude, so couldn't we see an increase in vertical size? I know the frame osc often gets a boosted supply from the LOP stage but it seems odd that it would perfectly cancel out, leaving the picture size unchanged.

[Heatercathodeshort]

The frame timebase oscillator is normally fed from the boost rail on later receivers from all manufacturers. From what I can remember it's fed from the HT line in the Bush 20 series.

If you turn the brightness up under fault condition I think it will result in a height increase and a width reduction as the failing LOPT cannot supply enough EHT current. Think of it like a shorting turn around the core.
Many years ago I did a lot of work with a Ferguson 998T LOPT the results of which showed that the main cause of the problem was dampness discussed many times on this Forum.
These transformers are now 70 years old and have had a very hard life due to the crazy badly ventilated screened compartment around the LOPT in the TV22 and years of bad storage especially as it has two very hot valves as close neighbours.. The only totally guaranteed cure in most cases is a reliable rewind. J.

[Panrock]

Originally Posted by Heatercathodeshort

If you turn the brightness up under fault condition I think it will result in a height increase and a width reduction as the failing LOPT cannot supply enough EHT current. Think of it like a shorting turn around the core.

Right. So the failing EHT increases the height and changes the focus. The width meanwhile is reduced, because the picture 'ballooning' tendency is overridden by the reduced line scan drive.

Originally Posted by Heatercathodeshort

These transformers are now 70 years old and have had a very hard life due to the crazy badly ventilated screened compartment around the LOPT in the TV22 and years of bad storage especially as it has two very hot valves as close neighbours.. The only totally guaranteed cure in most cases is a reliable rewind. J.

Indeed so.

You can also reduce the transformer heating by inserting a small fan over a 1-inch hole punched in the LOPT lid. The 12v to power it can be derived from the heater chain.

Steve

[Ian - G4JQT]

Thanks everyone for the replies - some good ideas. This is one of my stripped down and rebuilt sets. No missed components although new parts can of course fail. LOPT has been rewound, but I can check HT again.

[Heatercathodeshort]

Please let us know the conclusion. John.

[Ian - G4JQT]

TV24 MKI update:

Fixed - I think.

EHT seems OK at about 6kV measured on my EHT probe on Fluke multimeter. I tweaked this with TC1. Seems Heatercathodeshort's suggestion (on the TV24 metrosil thread) was to check the PL38 sorted it. Swapped with another and picture stable for hours. But oddly replacing the 'faulty' PL38 didn't return the fault, not during the observation period anyway. Still watching and with a new Ferranti PL38 all is still good after 30 mins. Fingers crossed.

While out of the cabinet I did the Dr Holden mod for better interlace (though not sure I can see much difference) and reduced the tone corrector from 0.01 uF to 0.002 uF for a bit more treble as it's a bit boomy compared to the smaller TV22.

I never did locate the PL38 mod I scribbled down of putting two 160 k and 0.1uF all in parallel into the PL38 screen grid feed, ​or what it was meant to achieve. I think it was probably from this forum though.

The 2.5 MHz frequency is only just visible once all the variable mechanical settings are adjusted - which takes a while as they all interact, as do the electrical adjustments!

My TV22 is a MKII chassis and 2.5 MHz is clearly visible on that.

Anyway, hope it's going to be OK for Christmas.

Ian

[Heatercathodeshort]

The EL/PL38 is a remarkably reliable valve often surviving the entire life of the receiver.
They are still very good but of course time marches on! Like silver mica caps, the odd one that has thrown the towel in will always to turn up. Strange the original one has now decided to work ok but that is valves for you. John.

[Helder Crespo]

Originally Posted by Ian - G4JQT

TV24 MKI update:
While out of the cabinet I did the Dr Holden mod for better interlace (though not sure I can see much difference)

The mod, as described in Dr Holden's report, is aimed at correcting a timing error and hence reducing the line pairing. It assumes you have good mains filtering and is not aimed at correcting poor interlace due to stray pulses.
The effect of the mod is more noticeable when the vertical hold control is set close to the upper edge (clockwise) of the stability region.
- Helder

[Andy - G8MNM]

Hi,

I tried the mod on my TV22 and the frame oscillator objected and wouldn't lock!

It wasn't the change in capacitor but the additional resistor in parallel with the existing resistor that stopped the locking.

Any idea why? The frame lock was good without any known issues before the mod.

I have removed the resistor but left the capacitor in place.

Andy