• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

What is the purpose of the g2 bypass cap?

leadbelly

Disabled Account
2002-12-22 2:13 am
Calgary, Alberta
Just wondering what the purpose of the bypass cap between g2 and ground is on a pentode stage? Morgan Jones does say what the goal is, to keep the the screen grid near AC ground potential, and goes into how to calculate its size. However I was wondering what the difference in operation is between having one and not having one, and also why they are not used on pentode output stages? TIA
 
G2 is a grid and the voltage on this grid will affect the current through the tube. The original intention of G2 was to accelerate electrons toward the plate and reduce the G1 to plate capacitance (thats why it is called a screen). For these effects to work as intended G2 needs to have a constant (with respect to the cathode) DC voltage (pentode operation).

It didn't take long for clever engineers to realize that signal or feedback could be applied to G2. Applying an AC voltage with the same polarity as the plate voltage results in a reduction in gain and distortion (negative feedback). Applying 100% feedback, connecting G2 to the plate, results in triode like characteristics. Applying a percentage of the plate signal to G2 results in a partial triode emulation. This partial plate signal is usually obtained from a tap on the OPT (UL operation). Some of us have realized that the signal can be applied to G2 instead of G1 (screen drive).

Mullard published a paper on the EL34 that stated that a small resistance in series with G2 can lower the overall distortion. Typical values are 100 to 1000 ohms. This resistor will also lower the G2 dissipation when the amp is driven to clipping. It is recommended in guitar amps for this reason.

However I was wondering what the difference in operation is between having one and not having one,

An unbypassed small resistor will slightly lower the stage gain and MAY reduce distortion (depends on the tube). An unbypassed large resistor will usually raise the distortion due to the non linear nature of G2 current in large signal situations.

and also why they are not used on pentode output stages?

They are. There may be a 100 to 1 k ohm (possibly higher in a guitar amp) resistor in series with it, and the "bypass" may be one of the power supply caps. In other words G2 may be connected directly or through a small resistor to one of the B+ supplies.
 
leadbelly said:
Just wondering what the purpose of the bypass cap between g2 and ground is on a pentode stage? Morgan Jones does say what the goal is, to keep the the screen grid near AC ground potential, and goes into how to calculate its size.

This is basically true, but not always. You can connect the screen bypass capacitor to ground if the cathode is similarly connected to ground. If it isn't, then the screen is bypassed to the cathode. The important consideration is the voltage differential between the screen and cathode. The Vsgk is what you want to hold constant. If it isn't, you get reduced gain, which may or may not be a bad thing since this gives an ultralinear type operation that can improve linearity.

However I was wondering what the difference in operation is between having one and not having one, and also why they are not used on pentode output stages? TIA

In my designs, I use active screen voltage regulation for pentode finals. In this case, the active regulator has a very low AC impedance. That Lo-Z to ground makes screen bypassing totally unnecessary. With some types like the 6AQ5, 6V6, 50C5 the screen runs at the same DC voltage as the plate. In that case, the output filter capacitor provides the screen AC grounding. Also, screen bypassing would be most counterproductive when doing an ultralinear design. Some designs that don't use active regulation, or an OPT with an ultralinear tap, may run an unbypassed screen to get that ultralinear effect.
 

Tubby23

Member
2020-05-13 5:19 am
London
Bump!

I'm modding my front end into a Cascode configuration.
In his hi-fi Preamplifier book, Blencowe mentions that by omitting the screen grid bypass cap on the lower resistor of the potential divider we can allow PS noise into the screen grid as a NFB, which ends up cancelling this out.
Anyone tried this?
I can't find this hack mentioned anywhere, only his book.
His valve wizard site omits this hack in its otherwise copied cascode chapter.