partial feedback

[dave dove]

after numerous problems i am knocking partial feedback on the head
here's the articles that can explain the theory much better than me:

http://www.tubecad.com/march2001/

and olsher's t rex:

http://www.enjoythemusic.com/magazin.../0105/trex.htm

for some reason i keep getting a voltage on the 300b grid
it is a gradually increasing phenomenon getting up to over 150V
and popping the cathode bypass cap
there is a distortion which increases with time
presumably as the voltage on the grid rises
takes a few weeks

i thought at first it was the 2uf cap
in a short loop.. 300b anode to grid
-as i'd done it the t.rex way-
but i've tried with different caps
and now i've now tried a 100K feedback resistor
to driver anode and the same happens

on one occasion i could follow the measured voltages
when switching 300bs which suggest a valve problem
but when i remove feedback the problem is gone

i don't know why this is
i can post some thoughts though:

the valves are faulty
but why only when feeback applied?
is it because of the aikido driver configuration?
or the resultant low driver output impedence?
whatever
i've got all these half realised
half understood ideas about this
but it doesn't work in this house

[James D]

Dave,

As you know I am completely puzzled by this...

I guess the 300B goes into A2 and draws grid current but what delivers that current if the 2u2 cap isn't short circuit? The anode could and would provide the drive to take the valve into A2 but not at dc through the cap and a music signal averages close to zero volts ac wise...

I just don't understand this. Mind you I don't understand JRBs values for the feedback loop either but I'm not up to arguing against him! But I would use quite different values for the feedback resistor. I would use 75k for the grid resistor and 150-220k for the feedback resistor...

If anyone has any ideas on this please throw them into the pot - we are really stuck on this one...

James

[A Stuart]

Originally Posted by James D

Dave,I guess the 300B goes into A2 and draws grid current
James

Doesn't grid current involve electrons from the cathode landing on the grid, so surely making it Negative, not positive? My understanding of valve grids is extremely elementary so don't anyone take my word for this please.
Once I tried to build a regenerative radio where feedback (positive in phase) is used to increase the Q-factor of the tuned circuit. Positive feedback should be totally irrelevant to this circuit, but the behaviour seems remarkably similar to what happened when it went into oscillation. The active device (FET) drew a large current but didn't heat up. I think it was switching in class-C. Do you have an oscilloscope so you could check if oscillation is involved.
I regret I can't work out complex currents and feedback well enough to see whether this circuit would allow positive feedback at any frequency. I presume you could still, up to a point hear audio on top of an RF signal, (a bit like the ac bias when recording on tape) until the RF got so large as to cause clipping.
This doesn't answer the problem, but hopefully spawns an explanation from someone more knowledgeable.
Alastair.

[Paul Barker]

I can't see any other culprit than the cap conducting dc.

grid current pulls a voltage drop across the grid's reference to ground, so making the overall valve bias more positive, so for valves known to go into grid current the dc resistance of this mechanism is minimised.

[A Stuart]

So is Grid current electrons from the cathode landing on the grid, or electrons somehow freeing themselves from the grid and travelling to the anode? Sorry to hijack the thread slightly, but it does seem to be relevant to the explanations. If simply DC leakage through the capacitor is being suggested, grid current surely needn't be invoked as an explanation for grid going positive.
Alastair.
edit - from my very un-knowledgeable standpoint, given my query about inaudible RF oscillation above, would the insertion of a grid-stopper distinguish this as the cause, if an oscilloscope is not to hand.
Back to the grid current question, surely a leaky 2.2uF capacitor would simply draw current through the 200k resistor to develop the positive voltage.

[James D]

Hi Alistair,

A2 operation (i.e. grid current flowing) is current flowing into the grid. The grid being positive with respect to the cathode releases current into the valve which terminates on the cathode and flows through the cathode circuit ulimately back to the grid to complete the current loop. SO the grid current contributes to the cathode current and in triode A2 operation cathode current is equal to grid current plus anode current.

Now in a practical circuit with the bottom of the cathode circuit grounded and the grid resistor grounded.. for current to flow into the grid it has to come from somewhere more positive than the grid (which is positive w.r.t. the cathode) so in this circuit where is that current path? It must ulitmately be from the B+ line as that is the only positive current source in the circuit... so it has to be through either of the caps connecting to the grid and the most likely candidate is the 2u2 cap connecting in the feedback circuit... particularly as the circuit behaves itself once this is disconnected - only trouble is Dave finds that this cap is OK when tested...

so is it leakage current when the cap is in circuit??? I don't know...

Hope this helps a bit. Any other ideas? Anything at all that we can think through and discuss? You never know what might trigger the right ideas for this circuit????

James

[IslandPink]

Dave,

I think I asked you this once..but anyway :
Have you any other caps you can substitute for the anode-grid feedback cap , even if they are the wrong value ( but enough voltage rating obviously ) . The 2.2's may be leaking at high voltage but not at low (test) voltage .

Mark

[John Caswell]

Hi all,
The fact that the amp is OK without feedback but not with feedback, begins to smell of HF instability.
Dave are you able to look at the 300b grid with a scope to see what is there with and without feedback.

John aka Dr John

[dave dove]

thanks guys

i have tried two different sets of caps in the short loop
an old pair of big PIOs 600V
a pair NOS russian K75 500V
but i've also tried it using just a resistor
connect just before the 0.47uf coupling cap
which is rated at 1200V

so unless all my caps are faulty.....

also

all this doesn't explain this:

i have told james about this before
i had a cathode bypass cap popping
so i turned it over and measured:
0V on one 300B grid
188V on the other
i swapped the 300Bs around
and the odd voltage moved with them
but it disappeared as soon as i disconnected the loop

i do have a scope
but i don't know if it works properly yet
and i don't have any complete probes
it is on my very very long to do list

dave

[Global]

Originally Posted by dave dove

i swapped the 300Bs around
and the odd voltage moved with them

Valve fault which emerges during high current demand due to HF oscillation?

Mark