Phil Graham (@Roadstead) wrote in on twitter with a question and a comment. My reply was just a bit long for twitter, so I thought I'd answer it here.
Going to build my brother in law a Stella. Have you played with diode pairs or bridge between trim -> level or trim -> ground?
No I haven't! One of the things I did when I designed the front end was to put a little op amp distortion in there. Playing with diode distortion is on my list of things to do. I know a lot of theory behind it but I haven't played around with actual circuits before.
If you get something working that you like the sound of, let me know! Make a video of it if you can, and I'll post it here.
Also, you could add a cap in series with R1 to bleed off some of the higher harmonics like the miller capacitance does
I also have not played around with adding a tone circuit but if I did want a tone circuit, that's pretty much where I would put it. Except I would put it after the trim and level pots and before R1, or before the two trim and level pots entirely. I did experiment with a funky tone circuit that was integrated in to the entire backend, and it should have worked but it did absolutely nothing. So I got discouraged, but I should really dig that out and see if I can get it working. The idea was that it would cut the higher harmonics even if the back end amplifier was distorting, which sounded awesome on paper but in real life did not work at all.
John,
John,
I'm leaning towards an asymmetrical LED pair bypassed with a small cap, and with a small feed forward resistor to bleed a bit of the clean input. The goal, for me, is mild asymmetric clipping, not the high gain "jar of bees" sound. I'm not sure what color LEDs I'll use.
I'm thinking that putting the 2 diodes and a small cap in path to ground from trim should give the most control over the diode clipping onset, but i"ll probably try it in the gain path, too, to see how the character is different. I've also contemplated beefing up the DC isolation cap between the op amp and power amp, to lower the chances of undesirable IMD products.
Personally, I'll probably end up with single pole lowpass before the power amp, and possibly a larger Cp cap after the amp. This depending on how much of a lowpass I'll get from the 10" speaker.
Dr. Geddes and Lee's various work on harmonic and IMD distortion perception are a pretty interesting read. It certainly makes me mindful of how sensitivity grows for increasing distortion order: http://www.gedlee.com/distortion_perception.htm
Any chance of posting the fancy tone control circuit?
Cheers,
-Phil G
Member for
13 years 2 monthsSounds like a plan! The only
Sounds like a plan! The only thing I'll mention to keep in mind is that the distortion characteristics of the Stella Amp change depending on how much voltage you use to power it. Nine volts is a different amp than four volts.
Also the type of guitar you use has a surprising impact. It's hard to quantify, but there is an interaction between amp and guitar that I was not expecting. I suppose if I threw in an extra high impedance input buffer that wouldn't happen, but I didn't bother, because I don't find it objectionable.
So build it up stock (leaving yourself some extra wire at various points so you can add mods later) and see how it goes, and then start adding your extra stuff. Give the op amp a shot, I tried several of them out, and this one was the least harsh and most musical.
The general rule is, if you crank the gain pot, roll back on the trim, and vice versa. The trim pot is there to keep the whole amp from crazy oscillating when you crank the gain too much. But if you want a clean tone you roll the gain back and crank the trim up.
The guitar, cable, pickups,
The guitar, cable, pickups, pot, and input generally form an rlc conjugate filter, so not only can the guitar matter, so can the position of the volume pot. There's some guy on the internet, whose link escapes me, that does a very nice job of describing and calculating this effect.
One of the appeals, for me, was the "softness" of the psu, as it relates to potential tone. I'd even considered some sort of zener diode on the supply side to play around with the stiffness of the power to the opamp.
P.S.
P.S.
Peavey's primer on transistor-based amp design is really nice in helping explain the processes at play inside a tube amp:
http://www.peavey.com/support/technotes/hartley/chapter_3.pdf