Negative Gate Bias

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

LevinGuitar

Well-known member
Joined
Jun 13, 2019
Messages
394
I wonder if a Fet like 2N3819 (or any other similar) can be biased with negative voltage on Gate and grounded Source. I know Fets and Tubes are different animals, but is it possible to do?

With tubes, going negative grid "a la C12" result in a lower impedance and easier work for the transformer (less distortion), right? Don't know if there is anything more involved rather than impedance change but for now my ears like it more than the cathode bias.

Would I achieve similar result with a Fet?

In case it's possible, there are some mics that have negative voltage on the voltage amp for figure 8, could I try get the voltage from somewhere there?

Thanks and Happy Miclove! :)
 
A lower... output impedance, you mean? I would've figured that's largely influence by the plate / drain resistor, but i could very well be wrong. 🤷‍♂️
 
I think with the source grounded and a fixed negative voltage on the gate, there might be a risk of thermal drift.
No idea if this will be a problem in reality!
When you use a soure resistor, the system is more or less self compensating.
If the drain (and source) current gets higher, the voltage drop over the source resistor will increase.
This will cause a more negative voltage on the gate and as a result of this, the current will get lower.
 
A lower... output impedance, you mean? I would've figured that's largely influence by the plate / drain resistor, but i could very well be wrong. 🤷‍♂️

I guess I got it wrong, it was time ago after reading (in Fox Audio web and Dave's comments in AA forum) an about matching transformer ratio for a EF86 tube in a U67 circuit where negative feedback was what lowered the output impedance, not the biasing.

So what makes the audio difference between biasings? A presence of a negative feedback (cathode degeneration) because of cathode resistor or just the propieties of the capacitor in paralel that tries to avoid it?

In the link below also says that this feedback increases gain, will it not affect impedance after all?


https://www.ampbooks.com/mobile/amplifier-calculators/cathode-capacitor/
Are Fets free from that side effect of using source resistor?
 
Last edited:
I wonder if a Fet like 2N3819 (or any other similar) can be biased with negative voltage on Gate and grounded Source. I know Fets and Tubes are different animals, but is it possible to do?
Of course it is possible. As Ruud mentioned, with tubes anf FET's, cathode or source bias is more stable than grid or gate bias. That's the reason why it was often called "automatic bias".
However, fixed bias works well, but often needs to be adjusted for the individual characteristics of the tube or FET.
Particularly with FET's, where the tolerance is much larger than tubes.
Even with source bias, the source resistor in Neumanns that use the 2N3819 needs to be AOT (Adjusted On Test).
If the tube or FET operates in class A, the difference between automatic and fixed bias is negligible, as long as the decoupling capacitor is of large enough value.
In case it's possible, there are some mics that have negative voltage on the voltage amp for figure 8, could I try get the voltage from somewhere there?
Of course. The bias current of the FET is so small it will take a small capacitor and a large resistor to filter whatever noise is on the negative rail. Just make sure that the voltage divider is of very large resistance as not to overload the negative rail.
Alternatively you could use a combination of LED and photodiode to generate negative voltage.
 

Latest posts

Back
Top