kevin gilmore

Sure. Lower noise

I use 10k pots for everything. makes it simpler

http://gilmore.chem.northwestern.edu/currentamps.jpg

the stax mafia are definitely planning on a multibit dac. Problem is finding switches good enough (and fast enough) to do 24 bits, 1/2 lsb

http://gilmore.chem.northwestern.edu/g9f21766.jpg

https://en.wikipedia.org/wiki/Joe_Alaskey

I got my hands on the old Tektronix all tube curve tracer, the high voltage version goes to 1700 volts. I tested about 50 6ca7 tubes, no 2 indentical. not even close. did the same thing with c2m1000, and to the resolution of the screen, fairly easy to get bunches of matched quads but there is still some variation.

Is that how it works. I've been doing it wrong all these years.

its not like a voltage amp, so a balanced version is still the same output current. other than the cast output circuit i'm unaware of any balanced transconductance amps.

alcohol AND diy, especially electrostatic power amps. at the same time. usually entertaining, and possibly deadly.

overkill? ME? NO, never. the resistors i picked are actually 1.5 inches long, the layout is 1.8 inches. So you can use the concrete resistors. i don't like the sound of protection circuits. So this amp is capable of about 30 amps into .1 ohm. Its just so much simpler this way.

you are buying the 10m90 from mouser right?

susy is low impedance input, so you can't use a 50k pot. 10k pot max!

the attenuator resistors are not actually 10k, they depend on attenuation per switch and total impedance.

partial schematic suitable for checking http://gilmore.chem.northwestern.edu/attenuatorv2.pdf board pictures http://gilmore.chem.northwestern.edu/attenuatorsmtv42flipground7.pdf

payment sent

I think it's closer to a 6U as a stereo balanced the board is 4.35 inches high and 11.55 wide the transistors take up a bit of extra space past the board so say 10 inches X 15 inch heatsinks

if people really want, I can probably flip the tubes to the other side to make stuff like this easier.

if you want a fully balanced amp, then you need 2 amp boards per channel plus one input board plus 2 goldenreference boards one set for 40v and the other set for 30v plus a very large transformer, a pair of 25 amp diode bridges and some really honking large power caps and mass quantities of heatsinks. if you ever saw the original levinson monoblock class A amps, those were 18 watts (really about 30) This is 100. you can also build the unbalanced version which is still pretty silly and less parts. And still stable into 1 ohm.

if you don't already have the boards, there are new shorter versions of the boards that make doing the c2m1000 much easier

the opamp blows up at +/-40V would have to check the other parts, but +/-30V should absolutely work speaking of power... unbalanced/balanced/cast to balanced input block for all sorts of things, but mainly the uberamp, just need to find a way to get it into the box will probably involve splitting off the driver section. Uses GR power supply +/-30V for the front end and another GR power supply +/-40V to run the regulated output voltages to the output amp. If i add zeners to the servo amp, can easily go to +/-50V and as shown, its 100 watts of pure class A, and something north of 225 watts A/B front end NO feedback of any kind, .0025% thd at 20khz, 2nd harmonic. 3rd harmonic not measurable. output amps are current feedback. i think i have created a monster

It does about 1.5 amps there will be a higher voltage one soon that does 6 amps

not a liquid.

pure ethylene glycol. I use it by the gallon to cool the xray cameras. slightly toxic.

the tidbits i post over there are an attempt to get people to turn their fucking brains on. It rarely if ever works. Like the bit about which of my children (amp designs) i like more. They clearly don't fucking get it. and isn't it chang 3.0?