Signal valves with grid current.

[Michael Maurice]

I've been recently servicing a Bush Radiogram where both EL84's had positive volts on the grids but I've also found it with the hexode side of the ECH81.

I've replaced the valve which didn't make any difference but I thought it was only power valves that did this.

Could the ECH81 have carried on like this and I presume I was right to replace it.

[ukcol]

I have a Murphy A372 that I am just starting on and the 10C1 FC valve shows a control grid leak, in the mixer section, on my Mullard E7600/4 tester. I borrowed another 10C1 from a different model on my "to do" pile and that had the same fault but not as bad. I ordered a couple of NOS valves and they are OK so it is not a problem with the tester. I haven't powered the set up yet so I don't know if the fault manifests itself as positive grid volts.

[G8HQP Dave]

All sorts of valves can go gassy. The symptoms depend on how the valve is used. Output valves usually overheat. Frequency changers may disrupt AGC action if, as is usually the case in a broadcast set, the FC has AGC.

[GrimJosef]

It can be hard to tell between grid current caused by gas and grid current caused by a small amount of surface conduction, either across a mica or across the glass where the leads feed through to the outside. The conductor might be getter material which has ended up in the wrong place or even, in the case of grids which have overheated, re-deposited grid material. I wonder if 'crowded' valves like triode-hexodes and triode-heptodes where the electrodes and their connections must be close together might be particularly vulnerable to such leakage ?

Cheers,

GJ

[Boater Sam]

This could explain why some ECH81 valves seem to be less "sensitive" than others when the emision is OK on test.
I have never considered grid leakage in other than output valves to be a problem. Live and learn.

[Michael Maurice]

I should have said that replacing the ECH81 didnt make any or much difference to the performance but after replacing it the grid voltage went to 0V

[ukcol]

Originally Posted by ukcol

I haven't powered the set up yet so I don't know if the fault manifests itself as positive grid volts.

I have just got to the stage with the Murphy that allows me to power it up.

With the faulty 10C1 in place the AGC voltage feeding the mixer is about 1.5 volts positive under no signal conditions. With a good 10C1 (or with the 10C1 removed and the heater chain kept going with a suitable resistor) the AGC voltage drops to about 0.3 volts positive.

Also with the bad 10C1 in place the volume control has to be nearly at the bottom of the track to get a normal listening level on a strong signal, although there is no distortion. With a good 10C1 the AGC works well and the volume control works normally.

That +0.3 volts positive on the AGC line is a fault but I haven't discovered the cause yet and it doesn't seem to be causing any symptoms.

So in the case of this Murphy and its 10C1 FC valve, grid current does indeed cause a problem.

[Sideband]

I've had IF amplifiers showing grid current and also diodes in multiple valves. In my Philips 470A once I'd got the set to the stage of testing, it would run for about 5 minutes and then start to fade. I found the AGC was going positive. The AGC diode is within the output valve (PEN4DD) and it was faulty (proved by temporarily wiring in a germanium diode). A new valve sorted that out.

I had an IF amplifier (don't remember the number now but it was an old B7 base) and that was going into grid current after a few minutes. Once again a new valve sorted it.

[G8HQP Dave]

In some sets a positive AGC rail voltage can reduce the gain, because the increase in valve gain is more than overcome by the grids loading the tuned circuits. Tuning in a mid-strength signal can thus increase the gain by pushing the AGC a bit negative and removing the loading. At first it can feel like the set is latching between two different gain positions. I once had this with a KB; I forget whether it was the 6K7 or 6K8 which needed replacing. Puzzling until I realised what was happening.

[ukcol]

Originally Posted by G8HQP Dave

In some sets a positive AGC rail voltage can reduce the gain..........

I think what is happening in the case of my Murphy is that on strong signals the positive voltage from the faulty valve is partly offsetting the negative voltage from the AGC system resulting in too much gain.

I haven't checked Dave but given what you say the Murphy probably has reduced gain on weak signals! A sort of inverted AGC!!!