RGD1800 Line Drive Question

[IJK2008]

Hello

I am working on this set, second TV and complete beginner. I have no line whistle but have changed C35 and other wax caps in this part of the circuit and checked resistor values. All seem OK and I double checked my wiring to the schematic.

Checked voltage on grid of line output valve and it is 12V dc +ve. With meter connected the line whistle appeared.

Should the grid not be at a negative voltage? Also I don't understand what the connection is for between the cathode and grid via R57?


Thanks

Ian

[Steve_P]

The voltage at the grid of the EL38 should be -ve. The first thing I would do is change C35. It is very much like THAT capacitor in a radio. Check your wiring.

The connection via R57 is to the width control via C39.

Might also be worth checking/changing the other capacitors in the area as well, as they will all be past their best.

There should be 250v on the screen grid of the EL38. It's worth having a spare EL38 to try as well. There appears to be two EY51's in this set, these are always worth checking too. Are they lit?

Also check that the resistors are within specification too.

Cheers,

Steve P.

[murphyv310]

Hi Ian.
If you have replaced C35 and you have a positive voltage on the grid, R57 could be o/c or very high value, the EL38 may have a leak causing the positive grid voltage.
If you are stuck for a valve I have a few good EL38 "pulls".

[IJK2008]

Steve

Thanks for your advice. I have changed C35 butI'll check it again. The only cap not changed is C33 0.0033uf. This is mica shaped but wax covered and marked as 5%. Could this be a culprit?

By the way I found your oldtellys.co.uk site good reading.

Ian

[IJK2008]

Trevor

I have changed C35 and found that R57 was about 920k rather than the 820k it should be. This doesn't seem to be that far out to me but I'll change it. Thanks for the valve offer.

Can't do anything till later tonight now.

Ian

[Mike Phelan]

Ian

The grid leak resistor is not that critical.

What sort of meter are you using, and is the line whistle only present when you put the meter on the grid?

Is the width control pot O/C?

[murphyv310]

The grid leak at 920K will be fine, it does of course serve two purposes in a line o/p stage it also will act as wave shaping with the coupling cap but it would need to be very very high before you would notice a difference.
Sounds like the EL38!

[DangerMan]

The circuit is slightly unusual as the grid isn't returned to chassis but to the junction of R59/R60.
Therefore, any voltage dropped across R60 due to the cathode current will appear at the grid as a positive voltage.
The valve generates automatic bias across R50, just like an ordinary class A stage, and the pot effectively reduces the HT provided to the stage to reduce the width by allowing the cathode to "float" positive.
If you set the pot to minimum resistance (zero ohms = max width) the positive voltage should disappear. Alternatively you could measure between grid and the junction of the resistors, and again no positive voltage should be on the grid.
Even if the line oscillator stops there is a cathode bias resistor (R59) to stop the valve going into meltdown but its presence will also reduce the negative reading you'd expect at an ordinary LOP grid.
Pete

[Steve_P]

Even with C39 in circuit?

Cheers,

Steve P.

[murphyv310]

The Circuit is very similar to the Dyanatron I have restored, but even on minimum width there is -5v on the grid and -24v on full width.
A line output stage is not class A as the valve is driven to cut off on flyback hence the overall negative drive.
A better test of course is using a scope to examine the waveform also a true DC level can be ascertained which often is incorrect using a meter.
If you look at the original post (1) Ian says that putting his meter on the grid the line stage runs this is the reason I think the EL38 has a leak. The thyratron line osc should be driving the line o/p valve with about 120v p-p so really this should be well high enough to overcome any leaks in the valve and still give a negative reading on the grid, so either the valve is duff or possibly the cathode resistor or width control are way high.

[Heatercathodeshort]

Hello Ian,
The line oscillator in this receiver is a gas Thyratron and I presume you have changed C54 in the cathode circuit. If you have a scope check the waveform/drive on the anode of the 6K25 to confirm oscillation. C38, C40 and C41 are .001uf high voltage and my guess is that they are Visconols made by TCC. These will be very leaky unless you are very lucky and will heavily damp the line timebase. They form part of the voltage doubler circuit from the flyback pulse available at the anode of the EL38. No transformer EHT overwind is required with this circuit. I would suggest you diconnect the single wire end from V12 EY51 and also the connection to C38 and power the set. If you get a line whistle and a pulse voltage on the top cap of the EL38 I reckon the Visconols are faulty. Don't do the test with any form of meter. 'Scrape' a well insulated screwdriver on the top cap and you should see a small arc. The EY51's should light under these conditions. Regards, John.

[murphyv310]

C39 is just to decouple the AC derived at the cathode, the width control reduces the current through the valve and will give a reduction in gain in the stage thus reducing the width and EHT and vice versa.

[murphyv310]

John has made some very good points and another thing to remember Thyratrons actually wear out, the ones used in line stages wear out much quicker than the frame oscillator ones.
It is possible the line osc is not running and when touching the grid of the o/p valve noise is fedback to the osc and its starting, so thats another thing to check!

[Heatercathodeshort]

Hello Ian,
If you require a 6K25 I have 30 brand new ones, enough to last me till the Olympics in 3012. If you require one please let me know and I can forward one FOC. I am doing a mass posting backlog on Monday! J.
Just a quick point Ian, as you are new to this carry out one action at a time switching on after each test. Following this procedure way you will not introduce faults which will create confusion and loss of esteem. We ALL make these mistakes after 40 years so if you do a hash up don't feel too bad about it!

[DangerMan]

To clarify my post, it's just an explanation why there might be a positive voltage without a duff valve or duff a leaky C35.

If the line drive is either low or non-existant, then the line output stage will be biased in class A by cathode resistor R59 (which stops it melting and does reduce its DC gain). In low drive case, there will be very little -ve bias between Cathode and grid 1.

Any positive voltage (DC or sawtooth) dropped across R60 (due either to DC fault current or real scan current) will appear at the grid if you measure to chassis.

The offsetting of -ve bias due to voltage drop across the width pot is seen in the readings at grid 1 given by Trevor for min and max width. (Hence advice to try at max setting... as it will give a clearer idea of oscillator output.)

In normal operation, no, the valve won't be in class A mode (although we don't know what's wrong yet).
If the thyratron was kaput or C33 were leaky that could cause inadequate drive... already said in previous posts.

Likewise, any loads on/in the transfomer would increase the positive positive voltage at the junction of R50/60 offsetting the negative from the valve grid when you measure this point with respect to chassis, even with correct drive.

HKS cuts to the chase ?... very probably... again !! ??

Pete

[IJK2008]

Hello All

I must thank everyone for the assistance so far. From the posts above I had quite a list of things to try. I started with easiest first - advancing the line width pot. The voltage on the gird went to -4.5V dc. And a raster appeared! I am now trying to tune in the Aurora. I've got a nice loud tone and a watery looking raster. Will try and improve this and report back.

Regards

Ian

[IJK2008]

Test card C achieved but the frame height pot is reading low and has no effect on picture. I have some pots in the attic so will have a look - hopefully the right value is there. Tomorrow before I can do any more.

Ian

[IJK2008]

Hello Again

Last night I focused the test card and although set at high brilliance the picture is really quite good although the width is low

I am awaiting a new line osc valve courtesy of HKshort (thanks) and a NOS EL38 so perhaps these will bring even more improvement.

In the meantime I have another problem - the frame height will not reduce! The height pot is confirmed OK after test out of circuit. I have checked resistors around this part of the circuit and they seem OK. But the original wiring is not to the schematic - the red markups show the actual state. What difference would this make?

Ian

[murphyv310]

Hi.
R98 & C71 are just to give a little damping on the primary of the frame transformer, it would not make any real difference whether the "earthy" end was returned to HT or Chassis with of course the exception that the cap will have DC potential across it when returned to chassis.

The Height control problem though is odd, if the control is working and wired ok the height should work, try measuring the cathode volts on the 6P25 when you adjust the control, it should vary if not have a good look about for bad wiring etc. What is the resistance reading in ohms from the cathode to chassis (set off) and does it alter when you adjust the control?

[IJK2008]

Trevor

I checked the resistance and it varied between 148 and 1.2k ohms with the control pot. Seems right looking at the drawing. So i took out the output valves and gave the sockets another dose of servisol. Result was proper control of height and starting to get a not too terrible test card. Shoved a DVD through the aurora and picture but no sound. That surely should be OK to sort though.

Ian