diyAudio (
-   Tubes / Valves (
-   -   Power-tube grid-leak bias (

Jeb-D. 16th April 2008 05:34 PM

Power-tube grid-leak bias
Out of pure curiosity, has anyone tried grid-leak bias with an output tube?

Poindexter 16th April 2008 05:54 PM

Sure, but you need a device that can take a substantial positive swing on the grid; e.g. a transmitting tube.  In the old Glass Audio, Eric Barbour (when he was a tech consultant for Svetlana) designed a very cool SET with the SV811-10, minimizing the rp (and thus output transformer primary Z) by running it low and hot without a cathode resistor.  He cut a couple of other corners in that design that I didn't particularly agree with, to keep cost down, but the basic idea was brilliant.



Miles Prower 17th April 2008 04:13 AM

Only place I've ever seen that done is with RF finals. It works there since these run Class C with grids driven positive on each RF cycle. Be pretty useless for an audio final.

Shoog 17th April 2008 12:33 PM

I have successfully used grid leak on inputs, but cant imagine its a recipe for long and happy life of output tubes.


gingertube 18th April 2008 06:31 AM

Shoog has it right but understated the problem.

Grid leak bias is NOT suitable for the general range of audio output tubes - instead it is a sure "recipe for cooking tubes".

Don't do it.

Joe_CA 6th September 2008 12:18 AM

The horizontal output tubes in all vintage tvs operated with grid leak bias. Pull the horizontal oscillator tube (if parallel filiment strings) and watch the output tube melt.

As mentioned below, not much use for an audio stage but the question was 'has anyone.....?'

Wavebourn 6th September 2008 01:16 AM

5M1 in a mic pre, 6J32P, grounded cathode. Can't imagine a grid bias in an output tube. Speaking of transmitters, there driving signal rectified on the grid causes a negative bias, but it is not a grid leak bias, strictly speaking.

Tubelab_com 6th September 2008 01:20 AM


The horizontal output tubes in all vintage tvs operated with grid leak bias. Pull the horizontal oscillator tube (if parallel filiment strings) and watch the output tube melt.
Back in 1970 I worked in a TV repair shop and did exactly that. The HOT went into nuclear meltdown while I was busy in the next room stuffing the oscillator tube into the tube tester and answering the phone. When I came back the circuit breaker on the TV had tripped, and the HOT had shrink wrapped itself. The glass had melted and was sucked in around the plate. It looked cool. Unfortunately the HOT (6KD6?) had also soldered itself into the socket, and I broke it trying to extract it. I have tried several times to intentionally repeat this without success.

The Horizontal Output Tube in a TV set does indeed operate with what appears to be grid leak bias. In reality the drive voltage switches the output tube from cutoff (retrace) to full on which is a constant current in a pentode (trace). The constant current is charging the flyback transformer resulting in a sawtooth waveform at the plate.

A class C RF amplifier works in a similar fashion, except that the RF drive is being rectified by the grid cathode diode resulting in a high negative voltage on the grid. Plate current only flows on positive drive peaks.

Grid leak bias is not feasible with output tubes operating in a linear mode. Modern production tubes, and many old ones that have been stored for 50+ years simply draw too much grid current. The grid current is also highly unpredictable and varies from tube to tube, and with the operating temperature.

In an ideal world the grid will collect electrons from the cathode's space charge resulting in a negative voltage on the grid. A perfect tube will bias itself into cutoff if the control grid is left unconnected. A high value resistor is connected from grid to ground to "leak" off enough electrons to establish the correct negative voltage on the grid.

In the real world the grid gets hot just from being near the cathode. It gets hot enough to emit some of those recently collected electrons. Add some "gas" (imperfect vacuum) inside the tube and more electrons can be emitted than collected resulting in a positive grid voltage and a runaway condition. Output tubes operate at a higher temperature than small signal tubes, and modern tubes have tighter internal spacing than the oldies. This is why you will see a maximum grid resistance spec on most every modern tube.

It should be noted that zero bias operation is not the same as grid leak bias. Several tubes (the 811A, 3-500Z, 838) were designed to operate at zero bias voltage over a narrow range of plate voltages. The maximum grid circuit resistance must still be respected to avoid the bright red glow.

All times are GMT. The time now is 04:23 PM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2