On Jan 24, 11:51*am, Engineer wrote:
On Jan 24, 10:24*am, "Stephen Cowell"
wrote:
"Morris Slutsky" wrote
...
I don't know if it's original or not, in the context of a guitar amp,
but I think it's pretty neat.
http://img718.imageshack.us/img718/8...edfeedback.png
I like it! *Two thoughts:
1.) *If you replace one gain stage with this, you'll
* * invert the signal... this can cause problems sometimes.
2.) *Switching the cathode cap out also rolls off lows.
I'd like to hear it in action.
__
Steve
.
Looks good! *More thoughts... (for a flat frequency response
application)
-- use a much larger cap than 0.44 uF, e.g. 100 uF, for cathode
decouplng (and feedback) over a much wider frequency range
-- plate load on the 2nd triode is 100K//47K (at least when switch is
closed), or about 32K; this seems a bit low.
-- bypass the 2nd triode cathode resistor (and make it smaller, e,g.
2.7K) for more gain, then increase the above 47K for the right amount
of feedback
This design might serve to switch a single audio input from microphone
gain to AUX gain (not tried)
Cheers,
Roger
Hi Roger,
If I was to design a circuit for hi-fi work, I'd never want the
feedback loop to be broken at all. Perhaps I'd be swapping resistors
within the feedback loop, but I wouldn't want to break the loop
altogether. Only guitarists want the gain and distortion to go up at
the same time, along with a band-limited frequency response - that
really is just for this one specific application.
Yeah, I load that triode heavy. It's a tradeoff. The 12AX7 has a
pretty high rP, I think about 68K. But it has a ton of mu! The other
12A*7 tubes have lower rP but a lot less mu. It probably all works
out the same in the end, except for bias levels. But the 12AX7 is the
tube that I can conveniently buy, so I'm using it. Output impedance
would be expected to be 68K || 100K = 40K, so my 47K load does eat
about half the signal, about 3dB of gain. There's still plenty gain
there to go around, you just may have to turn the pot for the next
stage to 10 instead of 9, no big deal. And it is a 47K load under all
circumstances - with the loop open it's 47K to ground, with the loop
closed it's 47K to what amounts to a virtual ground.
Probably something very similar could work for hi-fi gain control, I
think it's probably pretty common in that context actually to do gain
by switching feedback resistors.
Thanks for your advice