EL34 reliability mod

[lowell]

Anyone have an explanation of how changing 220k bias feed resistors to 100k increases reliability? I understand that this will increase the current of the negative bias, although how does this work as far as treating the tube more kind?

[Wakculloch]

Would the bias current not be set by the potential divider formed by the feed resistor and the bias adjusting pot (in most cases) , and the two 220k resistors from the grids reference these grids to that voltage, but not draw current.
The Mullard data sheet says they should not exceed 500k.
If they were lowered in value would they not start to shunt the PI load resistors, dropping the gain of that stage.
This is all conjecture, cos i've never had problems that i thought could be caused by these resistors.
A quick look at a few EL34 circuits seem to suggest that keeping the value higher than the PI anode resistors is the thing.
Someone with more experience should come in on this one cos i dunno.

[Mars Amp Repair]

yep,
Who posted that this was a mod of any sort? Gotta be certain the one making the post really has any credibility. I also don't see any way this would change much of anything to increase tube life & agree that at some point it would begin to shunt the drive signal. glen

[lowell]

In Gerald Weber's first book pg 183 in the Trainwreck section by Ken Fischer.

[Bruce / Mission Amps]

oops

[Bruce / Mission Amps]

In Gerald Weber's first book pg 183 in the Trainwreck section by Ken Fischer.

Kenny F told me that he thought the grid load values of around, or less then, 150K would make the EL34s available today a little more stable... why I'm not sure but if you look around you'll find a few amps using EL34s with lower then the 220K value grid load resistors found in amps with 6V6s, 6L6s, KT66s, 6550 etc etc...
With smaller valued coupling caps and lower the grid load, the less likely of grid blocking.
It will have no effect on the bias voltage though as that is merely a static voltage applied to the grid with no bias current flowing anywhere to cause a drop across the grid load resistors.
That is basic Ohm's law and another novice electronics point that seems to have been missed by GW.

[lowell]

wait wouldn't -50V applied to grids w/ 220k have less current (-50V/220,000) than say -50V applied to grids w/ 100k (-50V/100,000)?? There IS current flowing withing the bias right? I don't understand how there is NO current flowing to the grids. Please explain if you would.

[Steve Conner]

with no bias current flowing anywhere to cause a drop across the grid load resistors.

This mod is supposed to guard against thermal runaway caused by grid current. The current is caused by overheating of the control grid, or gas, or grid contamination with cathode material, and it does indeed cause a drop across the resistors. This drop tends to bias the tube even hotter, which increases the grid current, and so on until the tube is completely turned on and redplating. In a 4 tube amp, it can take its partner with it, since they share a bias feed resistor and coupling cap.

EL34s are especially prone to this because they have a big, high powered cathode with the control grid spaced very close to it, in order to give the tube its high gain. I think the Philips data sheet recommended 100k max for the resistor.

IMO, the best reliability mod for EL34s is 1.5k screen resistors, or a lowered screen voltage.

[Enzo]

Lowell, draw out JUST the grid circuit and plot the course of your current. Bias voltage from the supply, up through the grid resistor, to the grid, then where?

The bias supply puts a potential, a voltage, on the grid. But for current to flow, a complete circuit must exist. Look at a 9v battery. There is +9v on one terminal. COnnect a 100k resistor to it, just the one end. How much curent flows through that 100k resistor? None, because there is no circuit.

The grid of a tube is a wire sitting in open space in a vacuum inside the tube. ANy voltage you put on it will essentually sit there. its job is to repel electrons. Electrons are negatively charges things. They are repelled by negative voltage - they are repelled by the grid. Until the grid becomes more positive that the cathode (which is the source of the electrons), no el;ectrons will flow to it. That means no current.

The only time current flows is when the grid starts conducting. That is unusual. And unless that is happening, there is no current through the grid resistor.

Under component failures, things change, but that is not the case day to day.

[Alex/Tubewonder]

What Steve said. Thermal drift due to control grid leakage current. This is also the reason why data sheets specify different max values of resistance between control grid and ground for fixed and cathode bias configuration.

Why shouldn't there be no control grid leakage current in power tubes when there's in signal tubes AND is often used for "grid leak bias" via 5 to 20 MOhm resistor with cathode grounded.

[jrfrond]

...With smaller valued coupling caps and lower the grid load, the less likely of grid blocking. It will have no effect on the bias voltage though as that is merely a static voltage applied to the grid with no bias current flowing anywhere to cause a drop across the grid load resistors...

It WILL affect low-frequency response though. Dropping the resistor value will lower the time constant, so the coupling cap value must me increased to compensate. The risk of grid blocking remains the same, but the safe operating parameters will have been addressed.

With all of the talk of reliability, we have to keep tone in mind as well.

[lowell]

Enzo,
yes that makes perfect sense. I guess since cathode, grid, and anode are all never in physical contact yet current flows through the tube, I would think that just cause the grid isn't making a physical circuit doesn't mean that current does not leave it to the plate. I guess this brings up the ? of why not? If current is flowing from the cathode to plate why not the grid to plate. Is this because the grid is NOT grounded and has no SOURCE of electrons to pull from? In cathode biased amps the bias feed resistor IS grounded, how does this change things?

JRfrond yes thanks for pointing that out!

[Enzo]

The grid is - hopefully - not hot and so is not emitting electrons.

Leakage currents are exceedingly small. That's why grid leak stages have grid resistors of large size, like 5 megs.