"Zero bias"/ driving the grid positive

[Jeb-D.]

I know that driving the grid positive is for the most part uncommon. But I have seen some designs where a driver tube is configured like a self biased stage, but with no cathode resistor.

Is driving the grid positive a no-no under all circumstances?

I'm doing something with a 12ax7 and need about 10 more volts peak of swing for the amp to reach full output power potential.

[Alastair E]

Originally posted by Jeb-D.
I know that driving the grid positive is for the most part uncommon. But I have seen some designs where a driver tube is configured like a self biased stage, but with no cathode resistor.

Is driving the grid positive a no-no under all circumstances?

I'm doing something with a 12ax7 and need about 10 more volts peak of swing for the amp to reach full output power potential.

It really all depends on the tube that you want to 'drive positive' the grids on...!

Some are made for this service, while others are not. Some others while not intended for this service will operate reasonably well here, while others suffer badly/perform poorly. Im guessing the 12AX7 is a pre/driver tube supplying the output tube/s ? that you want to drive positive grids...?

When driving the grids of power tubes Positive, the Driver-stage should be made as low impedance as possible if you are to avoid distortion, as it will have to source current the moment the grid becomes positive With some power-tubes, the amount of current that the driver needs to supply can be really pretty high..... In some cases, a MOSFET drive such as Tubelab's Power-drive scheme is really to be recommended, as it supplies all the current you need, while retaining linearity.

[Jeb-D.]

I'll be driving a 12AX7(voltage amplification stage) slightly positive at full volume, The output tube isn't going to be driven positive. Just the output stage is a cathode follower that requires a large swing. It's about 8V away from it's potential I want to get that extra supply swing without having to increase rail voltage.

But thanks for your input, based on what you said it will take very little current to drive the 12ax7 positive. I'm just worried if it will cause the grid to burn out or anything like that.

[Alastair E]

Cant you increase the plate-load resistor somewhat, and maybe alter the bias slightly to reduce the current through the valve....

That way you'll get your extra 8V, without resorting to the distortions usually associated when the grid goes positive. Its the sudden change in load (From very little load,when the grid is negative, to comparatively high load as the grid turns positive) that the preceeding stage 'sees' as the grid goes positive that can result in distortions....

[bear]

Someone can correct me if I am wrong... I am unaware of any tube that can not be "driven positive."

But I think we're confusing something here.

Driving a grid positive usually refers to running in class A2 or AB2 or B2 - ie. the grid draws current.

In the case of a "zero bias class B tube" obviously almost all the driving signal is "positive", but it may or may not draw grid current depending on how the grid is connected to the driving source - coupled with a cap, no grid current is drawn.

The way that is usually used with cathode follower output stages to get sufficient swing is to use "bootstrapped" drive - the driver tube gets to float on the output swing. A form of positive feedback.

If you look carefully at some of the larger McIntosh amp schematics, you'll see that they did just this. There are other circuits that used "bootstrapped" drivers as well.

If it's a unity gain output stage, then getting an additional 8volts is hardly worth the effort?? What's the point/benefit?? Alternately, dump in an interstage transformer for a simple "fix" if you absolutely want to clip the output stage.

IF you are out of swing in the driver stage - ur hitting the max level of the plate voltage that you can, increasing the gain as was suggested will do nothing for you. Increase the B+ going to the driver or decrease the B+ going to the outputs. Fwiw, this is a common thing in solid state amps where the output is a follower and the driver stage can not drive the ouput to full bore... they typically run the driver stage(s) with a slightly higher voltage supply to get full swing.

_-_-bear

[Jeb-D.]

Bootstrapping, I had never thought of using that in a tube amp, but have seen it in many SS amps. Interesting idea for future projects.

Yeah it's pretty stupid for me to stress about 10ish Vpk even though the output load is only 300ohms. It will only get me another .33Wpk .66Wpk-pk. I'll probably just leave my amp as-is. I was curious about driving grids positive in general, more so than my specific application.

As far as raising the gain, the stage is already constant current sourced so gain is maxed out. I could lower the idle current and bias the VGS more negative but don't really want to go too skimpy on the current supply for the proceeding stage.

[Dave Cigna]

Originally posted by bear
In the case of a "zero bias class B tube" obviously almost all the driving signal is "positive", but it may or may not draw grid current depending on how the grid is connected to the driving source - coupled with a cap, no grid current is drawn.

With a coupling cap you'll still draw grid current, but it will charge the cap, thereby changing the grid voltage ... Then you have to wait around for the cap to discharge before bias returns to normal ...

Trust me, if you want to drive a grid positive then you must be able to supply the current that it requires. One way to look at it is that the load impedance presented by the grid drops drastically above Vgk = 0, and continues to drop drastically as the grid goes more and more positive. In order for the driver to 'keep up' it must have a low enough output impedance. If not, it will get loaded down resulting in distortion. How low it needs to be depends on how positive you want to drive the grid. Just a little might not be too much of a challenge.

Cap coupling will not work. Direct coupling or transformers are required. There's no way around it. Going just a little positive with a coupling cap leads to the 'blocking' described in my first paragraph. I don't think anyone finds blocking acceptable in any degree.

-- Dave