Marconi 2955R Shrunken Text/Linearity Woes - LOPT or H drive issues?

[TheBay]

I recently purchased a Marconi 2995R Test Set, this is the same as the 2955A but with the Off air receiver.

I was aware of a number of faults when I purchased it and am slowly working through everything.

There is a strange issue with the CRT, where a portion of it has shrunken text, it looks like a linearity issue but making adjustments on the linearity magnet makes no difference.

If I adjust the V position this shrunken part of the picture remains in exactly the same place on the CRT and moving the image up and down shows that.

First port of call was to take the AC1 board out and check the capacitors, I replaced
2x 470uF/16v
1x 1000uF/35v
1x 2200uF/16v
1x 100uF/50v
1x 3.3uF/50v

There are a few other electrolytics there I didn't replace as I don't have that value or they are axial and it would be quite a messy stretch to fit a radial in it's place due to the already long lead on the axials.

All those capacitors that I had removed tested fine with my EVB ESR Meter and my DE-5000 LCR meter.

I checked all the other capacitors and they all seemed fine, even C20 and C16 which are on the one end of the deflection coil, but I will replace these when I order some parts as I don't have them here.

Looking at the waveform from the output of the BU806 and before and after L1, between the pulses there is another small pulse and it's also not flat, it does look like where this extra small pulse is would be in that part of the CRT i'm seeing this issue. But I do not know what a pulse looks like on a perfectly working 2955A/R

I flipped the wires around on the deflection coil and the image flipped upside down as expected, now I was suspecting the deflection coil on the yoke as being the issue, BUT this issue now flips with the image to the bottom of the CRT. So I think I can rule out the CRT/Yoke?

I've attached a picture of the pulse, but this has been shown with the current yoke connected as I do not have any other way of loading this circuit up.
Please ignore the voltage reading on my scope, it just a digital overlay on an analogue scope. It is not indicative of the actual voltage.

As most people who are familiar with this model of test equipment know, the Vertical on these is actually what would be the Horizontal drive on a traditional CRT, Marconi rotated the CRT 90 Degrees.

I'm a bit baffled with this and don't know where to go next, I'm not sure if something is creating a dodgy pulse like the 74LS211, transistors or a passive like a capacitor is acting up at a certain frequency.

Hopefully someone would have come across similar, it's not an issue I've ever encountered on a CRT.

Here is a Youtube link to a video I made of the problem while adjusting the V position to show it is not in the video signal but part of the scan system.
https://youtu.be/EF3Ke5V2aJg

Photos of scope measurements:
https://photos.app.goo.gl/9Zg3SJDJb4zrtgxb6

A version of the AC1 schematic attached, this isn't up to date and has some slightly different capacitor values but the circuit is the same, I didn't want to attach the whole service manual as it's huge.

[TheBay]

Images on imgur for anyone who can't see them on my Gdrive.

[TheBay]

Another image here, checking the Base and Collector of TR3 (BU806) that drives the LOTP and Horizontal.

I may be wrong but I thought the waveform at the base would be more of a Sawtooth and the waveform from the Collector would be taller and thinner, higher voltage.

[TheBay]

I think some of these measurements may be leading me up the garden path.
Looking at the cold side of the H Scan coil, the current waveform is perfectly linear with no distortion.
So I'm starting to suspect there being an inductive issue somewhere within the H scan coil winding.
Some kind person from the Marconi group is posting me out a scan coil to test which will be with me in a few days.
And I will report back, if that doesn't cure it I'm going to suspect the either the BU806 or the 74LS221 next.

[CJay]

It would be useful to post a picture of the actual screen.

[TheBay]

Hi CJay,

Good point, attached some pictures to this post.

[TheBay]

Swapped over the Yoke/Scan coils today and the fault remains, so I can definitely rule that out.
Something is just not working correctly in the H deflection circuitry and I can't seem to find it.
As I mentioned the CRT is rotated 90 degrees in this so what you see as the Vertical is actually Horizontal drive (In a traditional CRT Display)

[fzabkar]

The collector voltage on TR4 should never exceed 12V, yet you are showing 249V. Are you sure your annotations are correct?

The flyback pulse should be smooth and parabolic. There should be no step, so that's where the non-linearity is occurring. I wonder if D11 is leaky?

[TheBay]

Quick update today!
I was about to do an order on Farnell today and swap all the poly and ceramic capacitors, including the S correction capacitors.

Before I did that I thought I'd swap out some with values that I already have.
I thought I'd replace the decoupling capacitors on IC2, C12 and C13, which made no difference.
I added another capacitor parallel to C27 (5V decoupling on IC2, only on some AC1 Boards) and noticed no change.

Put the original capacitors back in C12/C13 and C27 as they tested perfectly.

Then got out the Freezer spray and a heat gun, still couldn't find anything.

Then I grabbed a plastic anti-static brush instead of my usual wooden one by mistake, just to check for dry joints.
As soon as I got near the LOPT the picture would start to collapse, in EXACTLY the same place as this linearity issue is.

So something is not right somewhere with the LOPT I think!

Video here on YouTube:
https://youtu.be/TWW2ebBtDIY

The collector voltage on TR4 should never exceed 12V, yet you are showing 249V. Are you sure your annotations are correct?

Please ignore any of the onscreen readings, it is a purely analogue oscilloscope with a digital overlay, they are meaningless and only a rough guide if I move the markers and select the correct multiplier depending on the probe I'm using.

[fzabkar]

As in my previous edit ...

The flyback pulse should be smooth and parabolic. There should be no step, so that's where the non-linearity is occurring. I wonder if D11 is leaky?

[TheBay]

As in my previous edit ...

The flyback pulse should be smooth and parabolic. There should be no step, so that's where the non-linearity is occurring. I wonder if D11 is leaky?

Thank you for your reply and the edit.

I completely agree, I've been chasing my tail trying to find where that step is.
It's easy enough to replace D11 and worth trying.

I assumed today that this whole issue is due to a fault with the LOPT as if I put something in close proximity to the LOPT the linearity gets worse in exactly the same place as the current issue, so something must be happening capacitively or through inductance. Even if I leave something next to the LOPT it remains affected without me holding it.

The LOPT has an aluminium shield that clips over it and when I put that back on the linearity gets worse, but when this cover is on I cannot influence it by holding things near it as it is interacting with it and also shielding it.

The only things I can think of are, wound incorrectly during manufacture, shorted turn, wire crossed over or poor ferrite material.

It's all very strange and something I have never come across on a CRT.

[TheBay]

I'm about to order some diodes, D11 MR845 and D3 a BYW95C.
But I'd really like to know if it is a LOPT fault or something wrong in the H/LOPT drive circuitry before I spend more money on discontinued parts.

This linearity issue is massively affected by placing something near the LOPT even if I am not holding it and it is far worse when I put the factory aluminium shield on it.

Is the LOPT faulty and if it is what could possibly wrong to cause this strange issue, or can something in close proximity to the LOPT affect another part of the circuitry. I've worked on CRT's for 30 years and this has really baffled me as I've never come across anything like this before.

[fzabkar]

MR854 is a 3A, 400V fast recovery rectifier. I think any diode meeting those specs should be OK.

https://www.onsemi.com/pdf/datasheet/mr850-d.pdf

I doubt that the tuning diode, BYW95C, would produce the observed symptom.

BTW, I've never seen anything like this, either. I have seen ringing in the horizontal trace, and I've seen non-linearities, but never a step.

[TheBay]

A little update, I haven't had chance to play with this the past week so decided to put this on the bench today with a fresh mind,
I changed the diodes and it made no difference, I guess I was clutching at straws with those.

I got the HV probe out to check the EHT and it seems quite low, the service manual states 15KV but I'm seeing 8.25KV so it really is starting to look like the LOPT is faulty.

Voltages are as follows:
15KV = 8.25KV
475V = 445V
150V = 150V
50V = 44.7V

[TheBay]

Couldn't resist having a little play tonight, I wanted to double check the H/Lopt drive frequency and it is absolutely spot on at 15.625KHz
Although I knew it was 15KHz and not any higher as I know what 15KHz sounds like but it doesn't hurt to check the basics as it could have been a frequency issue causing saturation of the LOPT or not enough drive. So I can rule that out.

What I'm wondering is whether a rectifying diode or capacitor built in to the LOPT is at fault or whether there is a resonance issue with this and was not manufactured correctly, possibly gap issues or poor core material.

[james_s]

That's a very strange fault, in years of repairing CRT displays I've never seen that before. Is it possible that you have a crack in the ferrite core of the flyback transformer (LOPT)? I'm struggling to think what could cause that behavior.

[TheBay]

That's a very strange fault, in years of repairing CRT displays I've never seen that before. Is it possible that you have a crack in the ferrite core of the flyback transformer (LOPT)? I'm struggling to think what could cause that behavior.

It's extremely strange, I've worked on all kinds of CRT's over the past 30 years and it's thrown me.
I can't see any cracks in the external part of the core, but something is not right with it, especially if I put something in close proximity to it.
Usually I take it with a pinch of salt when someone tells me this issue has been there since new, but I'm really starting to think that's true.

Unfortunately a replacement LOPT is unobtanium, it wasn't used in anything else apart from 1 other Marconi product and a HP scope.
I'm getting tempted to find a way to take it apart and find out what's wrong with it, but it looks almost impossible to open without breaking it.

[fzabkar]

Try injecting 1mA into the EHT winding. I don't know what the forward voltage of the EHT diode(s) is likely to be, so I would use a current limited supply. BTW, the diodes seem to be drawn with incorrect polarity. :-?

Edit:

I have seen plenty of cases where a tripler has failed with a low output voltage, but these are usually external units connected between the output of the FBT and the CRT anode. Moreover, a low EHT always causes blooming, IME.

[james_s]

It would be really unusual to have any sort of multiplier with a small monochrome CRT, the EHT is only 9-12kV in most cases.

[TheBay]

Try injecting 1mA into the EHT winding. I don't know what the forward voltage of the EHT diode(s) is likely to be, so I would use a current limited supply. BTW, the diodes seem to be drawn with incorrect polarity. :-?

Edit:

I have seen plenty of cases where a tripler has failed with a low output voltage, but these are usually external units connected between the output of the FBT and the CRT anode. Moreover, a low EHT always causes blooming, IME.

I really wish the rectifying diodes were external or a doubler/tripler used. I was considering making a doubler/tripler out of some diodes and capacitors and putting it inline with the EHT lead as a quick test to see if a higher EHT voltage would make a difference.

I do have a ring tester here, I might desolder the LOPT and see what each winding rings out of curiosity, but I think this is probably a core, diode or capacitor fault internally.

Talking of blooming, it's not as easy to spot on this as it's all text driven and all screens are pretty much the same. Though when relays are energised or any slight change on the 12V rail the screen will change size slightly, it also resizes as it warms up then settles. So it could well be blooming and as the EHT is that low it is getting effected by any subtle change on the 12V rail.

[TheBay]

It would be really unusual to have any sort of multiplier with a small monochrome CRT, the EHT is only 9-12kV in most cases.

Agreed, this is still low for a CRT at 15KV (I'm seeing 8KV) but I've seen quite a few small mono CRT displays without a LOPT and use a small transformer or SMPS that drive the EHT with a doubler/tripler unit.

[fzabkar]

If the EHT is meant to be 15kV but is actually only 8kV, then I would have expected the blooming effect to be massive. Are you sure your measurement device is accurate? Are you watching the screen when measuring the EHT?

[TheBay]

If the EHT is meant to be 15kV but is actually only 8kV, then I would have expected the blooming effect to be massive. Are you sure your measurement device is accurate? Are you watching the screen when measuring the EHT?

My probe is 1000x has an input resistance of 1,000MΩ and was going in to my Fluke 87V which is 10/11MΩ impedance depending on what range is used.
It's a Tenma 72-3040 which is now called a TT-HVP 40 from Farnell, it is pretty similar to a Fluke 80K-40.

There is no difference to the picture when the probe is connected to the EHT.

I'll take a video from cold tomorrow to show the resize as it warms up and also the "blooming" when I active relays etc.

[james_s]

Agreed, this is still low for a CRT at 15KV (I'm seeing 8KV) but I've seen quite a few small mono CRT displays without a LOPT and use a small transformer or SMPS that drive the EHT with a doubler/tripler unit.

Are you sure it's supposed to be 15kV? 8kV doesn't sound too out of line for a small tube like that. I've measured a few 5" B&W TVs that were under 5kV.

[fzabkar]

Agreed, this is still low for a CRT at 15KV (I'm seeing 8KV) but I've seen quite a few small mono CRT displays without a LOPT and use a small transformer or SMPS that drive the EHT with a doubler/tripler unit.

Are you sure it's supposed to be 15kV? 8kV doesn't sound too out of line for a small tube like that. I've measured a few 5" B&W TVs that were under 5kV.

The circuit diagram has 15kV at the CRT anode. I agree it does sound like a lot for a small tube.