Grid bias with Class A SE ?
 

Has this been done and is it practical? The obvious benefits would be less heat generated in the amp due to elimination of the cathode resistor, as well as no resistor and capacitor to introduce time constants and affect the bias during transients.

If I wanted to take a 6bq5 and apply around -12v of C- to the grid through a 470k resistor and tie the cathode to ground, would this give me any obvious advantage over the typical cathode bias scheme?


I have a second 6.3v tap in my amp which I could FWD to about -16v and then use an LM337 or 7812 ( the negative version at least) to regulate it, with a pot on the output to adjust the voltage to the grid.
Sugguestions?

Hi

It surely can be done, and you could probably find schematics were it is done! An alternative for SE biasing I have in the back of my mind is the one presented by Broskie here

Scroll down to the 4th and 5th picture. An auto-bias using the LM337 or LM337+MJE350 for higher voltages. It is limited to class A operation, but that is no problem for a SE design which operates is class A anyway.

Erik

This is SE fixed bias. It is a guitar amp, but it will give you the idea.

http://satanda.aspweb.cz/schem/atomicse.png

Hey thanks, that CCS looks nice. How does the sound compare to cathode bias? Better bass? better transient responce?

My experience with this type of CCS bias is that overall the sound is slightly tighter and cleaner sounding than with the unbypasssed self bias. The thing to bare in mind is that if you are bypassing the cathode resistor then that should dominate the result and there should be nearly no difference in the end result.

Shoog

Back to the original question about using a bias voltage instead
of a cathode resistor. I just looked in the RCA tube manual for the 6bq5. If you use use cathode bias the max grid leak resistance is 1 meg ohm. If you use a fixed bias the max grid leak resistance is 300k ohm. So you shouldn't use a 470k ohm for your bias supply unless you unless you used a 470k from grid to ground also but then you would need a 24v supply. Or just use a 250k resistor with your 12v supply.

Forgot about that, thanks for pointing it out.
What would be the reccomended bias voltage to achieve? shouln't It be really the same as what would normally be dropped by the cathode resistor?

My experience is that the voltages with CCS bias, LED bias and resistor bias all settle slightly differently. I think this has something to do with DC feedback over the bias element.

Shoog

Does using Neg. Bias give me an advantage over cathode bias in an SE amp, other than the obvious benefit of not using a heat-generating power resistor/ capacitor in the signal path?

The three main advantages are that you can use much smaller decoupling caps on the output of the bias supply, so can use a cheaper high quality film cap as the final. Secondly, you can readily play with your operating point. You may find the sweet spot for your system and taste. Third advantage is that you get to use your full B+ voltage.

The disadvantage is that you need a negative supply. BTW, unless you are regulating your B+, it's not a good idea to regulate the bias supply. If your line voltage increases, you want your bias to go more negative, so the output tube won't want to run away. Better to use a simple series of RC sections in this case.


Sheldon