kevin gilmore

Somehow that is not the magazine i would not expect steve to be reading ??

3.73 watts for each of the 4 output devices .9 watts for both of the 3rd stage current sources (2ma) .9 watts for both of the 3rd stage amplifiers (2ma) About 20 watts for each amplifier.

I'm linking this one a lot. Needs a wood knob though. Now if customs would just pass thru the latest board sets from lil-knight i would be even more happy.

So i spent about 5 hours last night trying to find a matched set of tubes so i could have a listen. With about 100+ 5687's and an unlimited supply of 12ax7's, the best i could do was about 1 db differential and 1.5 db left to right channel. So matching is just about impossible. And after an hour of running, the numbers changed. Unlike the ES1/2 for which significant modifications can result in a pretty decent amp, for this mess, the surgery has to be extreme. Take the output preamp tubes, and turn them into input gain tubes. Take the original input gain tubes move them to after the volume knob, and apply local feedback from the output tubes. Then seperate the filaments for the output tubes and lift them to +300 volts. It might actually sound half way decent this way.

I am personally not going to spend any time whatsoever doing an A10/B52 board. Neither am i going to do an opa541 board because i have used my old power supply boards to test what it sounds like. The blackhole sounds way better. Especially with the alfet output devices. I will probably do a modification of the current kgsshv power supply board to do 600/300 volts for aristaeus/triode amplifiers. It definitely was worth the effort to reverse engineer ray's piece of crap. The fact that the stealth,raptor,B52 and now A10 are all exactly the same is quite entertaining.

I'm going to disagree with this, i have measured many a marshall amplifier. Unless you push them well into clipping, they typically measure .1 to .2% thd across the audio band. Besides which the job of the headphone amp is to accurately reproduce the distortions in the recorded sound. With something like a synthesizer, this is not as easy as it sounds due to rise and fall times that exceed what any musical instrument can do.

Fine, Dewey Cheatem and Howe it is. But later because i don't have the file at work.

So here is how this went down. Someone over there really pissed off Birgir. Its not nice to piss off the head of the stax mafia. I mean its REALLY NOT NICE to piss off the don of the stax mafia. So pictures were acquired. And reverse engineering proceeded. Starting at about 11am today. Then by about 4pm the stax mafia had an epiphany. And so while a slight bit more work is necessary... http://gilmore.chem....ern.edu/a10.pdf Now there is just so much wrong with this thing, i don't know where to start. OH wait, i do. http://gilmore.chem....raptorschem.jpg So Ray says the A10 is SRPP. Well go and look up the article written by Broskie... Its SRPP only when the load is resistive and close to impedance of the cathode resistors. Which by the way an electrostatic headphone definitely IS NOT. As usual ray does not have a fucking clue. Whats more the thing is completely open loop, and i just finished testing about 50 x 5687's in the output circuit, and guess what, a variance of about 4+ db. Reminds me of mikhail's ES1/2 disaster... And the same goes for the input section. What a piece of work. But let that not stop Moljinor(sp) audio... Who will be releasing a board that is both a B52 and a A10 at the same time. Add much bigger output caps for the dynamic thing, and put in a switch to lower the output section power supply voltage. Cheap as hell. Still the 5687 is being driven way over its cathode to filament specs, just like mikhail's disaster... By the way, just the output section of this thing has distortion specs that rival the Darkstar, >5% at even 200 vpp. And the power supply... What a clusterfuck. Not even worth the time to key that one in.

I was thinking of doing something similar. Find some real good epoxy, and glue it together and put it in a press overnight at say 1000 lbs. Then drill and attach the metal mount with flat bottom wood screws and more epoxy. It will definitely hold together so that i can machine it.

justin's knob and the tool i'm going to use to turn it to the final dimension http://gilmore.chem.northwestern.edu/justinknob.jpg

http://gilmore.chem....du/woodies3.jpg so i took some lacquer, dilluted it 5:1 then applied it with the lathe running. Came out like glass. And now the mounting thing for doing BHSE knobs (not cleaned up yet) Justin is getting a cocobolo knob. The smaller diameter goes inside the knob so everything is flush. http://gilmore.chem....u/bhsemount.jpg

Current d'agostino products are similar to the krell sustaned plateau bias stuff. So the meter is showing the amount of class A bias, as a cpu inside monitors the input signal and calculates the minimum amount of bias to maintain class A operation. At least until the thermal limits kick in. Real Vu meters are logarithmic, not linear. They would be hooked up directly to the output terminals. These days there are fancy ways of doing stuff like this with hall effect detectors on the outputs to measure real current, combine with the voltage, run it all into a cpu and calculate real power.

I bought the samsung 512mb ssd for my new computer, actually tests over 5.5 GB/S at the moment. In the future with die shrinks, the amount of times you can rewrite a sector is going to go down a bunch. So buy the current intel's or samsungs now, because in the future, something is going to have to be done with the operating systems to prevent rewrites. And always put in way more memory than you need, and add a regular hard drive, and put the swap file on it.

But they are a heavily inductive load, and that definitely pisses off some power amps.

Power supply can supply as much current as you want. Transistors can take up to 4 times as much bias if you can get rid of the heat. Best done on the off board heatsink versions.

No other changes. One version ago, lil-knight had me make the mounting holes for the heatsinks slightly larger.

There is a new version of both of the amp boards with jumpers for local or global feedback for the next round of boards when needed. I still like the global feedback version better.

I have no trouble soldering the heatsinks with my weller, but i'm using 50+ year old eutectic solder. And plan on doing so forever, or until my 30+ lbs of stock runs out.

Pretty sure that the input for maximum output is 2.8 volts peak to peak. Will try and find the datasheet later. You can always add a 3 or 6 db input attenuator.

B&O USA is about 3 miles from my house, i wandered over there one day and they let me buy a pair as spare parts for their speakers.

Boy are those speakers old. I would look carefully at the surround on the woofer before spending any money on amps to drive them. The B&O icepower things sound even better, and you can get them in a box with a power supply up to about 300 watts. Just wait till the batteries in the RWA amps start cracking and leaking acid jello due to their quality charging circuitry. Anyone that owns one and doesn't change the batteries every 1.5 years is looking for trouble.

Preamp board is 12.4 x 6 inches multi amp is 6.1 x 3.4 inches black hole is 3.5 x 3.62 inches

Absolutely i an do silver. The regular stuff is kind of soft, so i would have to be careful. There is some german silver that is drop forged, weighs about 30% more per unit volume and that stuff is really hard. I have some of that, but the form factor is wrong. If you don't coat the silver with something, then it is going to oxidize very fast, possibly with a rainbow. Might look cool that way. Denatured alcohol around these parts is sold in the drug stores, and is absolutely clear.

Its gonna be a busy spring and summer.

2 x 2 x 1.5 is the absolute minimum. Which is why i'm having trouble finding snakewood that is big enough in the "desired" direction. I do this on a monarch engine lathe with very sharp carbide tool bits. Step 1, use a bandsaw to make a roughly round blank. Step 2 use a pressure plate and a live center, and make the blank completely round. (pick from standard size that will fit perfectly into a collet) Step 3 insert round blank in collet, drill the hole for the shaft piece and use a boring bar to make the mounting area. Step 4,5,6 make the shaft piece which takes 3 steps on the lathe and then 3 more steps on a vertical mill. (center drill, drill, flip and countersink) drill holes thru the shaft piece into the wood and attach the wood screws. insert back into the lathe, and machine to proper dimensions insert into the NC lathe and do precision micro machining on the front and side. The knobs come out that shiny without any oil, varnish, or any finish of any kind. Takes about 3.5 hours per knob. Eventually i will make the shaft pieces completely NC and that will shorten stuff about 1 hour. There is no way to NC the prep work. The harder and more dense the wood is, the easier this is to do.