kevin gilmore

wiring the transformer to the low voltage supplies with the center tap in the wrong place is definitely going to cause the caps to over voltage.

the joys of 35 year old gear with dirty switches... Finally got the higher voltage range working, but did not take pictures. Never did get the trace blanking working. 50 volts per division horizontal, .5ma per division vertical 20 ua base current per step. Horizontal and vertical flipped so it looks like a npn. The first 3 are known real http://gilmore.chem.northwestern.edu/dsc_0316.jpg http://gilmore.chem.northwestern.edu/dsc_0317.jpg http://gilmore.chem.northwestern.edu/dsc_0318.jpg the next 3 are http://gilmore.chem.northwestern.edu/dsc_0319.jpg http://gilmore.chem.northwestern.edu/dsc_0320.jpg http://gilmore.chem.northwestern.edu/dsc_0321.jpg So other than beta being lower, sure look to be the real thing. High voltage tests, all of them break down between 650 and 700v will measure capacitance later.

Definitely alive and very busy with work. Same with me. Hopefully this will change for both of us Soon

That's basically a BH Kerry has a slightly different design with the kgsshv input stage, The active battery and the t2 output stage.

you haven't said what you think about the sound... i'm assuming you like it??

you do have a seperate filament transformer winding for every top tube right?? if not, arcy sparky. and if those windings are not rated for 500V+ then also arcy sparky.

well there is the single ended version with 300B, (2 per stereo amplifier) and then there is the push pull version. (4 per amplifier) just in case you have any money left over after you build the first one.

I think that a curve tracer or a specifically designed test circuit is the way to go. What i am really looking for is a 500 series tektronix curve tracer, i will ask around in a few hours. If the curves are close and the BC capacitance is right, then even if they are fakes, they are usable. But the ROHM parts should be an exact substitute except they are surface mount only. The problem with the cheap curve tracers is that they don't go up to the high voltages. I put in a bid on a tek 577 thats local to me, so everyone stay away and we will all benefit.

Its pretty much self balancing unless the tubes are way off. initial testing on various tubes, the output voltage is +/-5V

Its a operational transconducance amp, and the load is 105 ohms. Increase to 500 ohms for 10db of gain. add pot after the 105 ohms and then a buffer.

here is another one i like, buildable with currently available parts http://gilmore.chem.northwestern.edu/fcnf.pdf input can handle 10 volts peak to peak, its another operational transconductance amp. input can also handle 2 volts of common mode or differential mode offset with 0 volts dc output, something just about no other circuit can do. No feedback, and distortion approaching .01%. Flat to about 500khz. Needs +/-45 volt power supplies, and plenty of heatsinks. Otherwise cheap to build. Add a unity gain output buffer (2 of them) and set the gain with 2 resistors (or a 3 position switch) and a constant impedance attenuator and have something very slick.

http://gilmore.chem.northwestern.edu/kgitsoch.pdf An even sweeter way to do unbalanced/balanced to balanced.

The input tube filaments for the T2 are also very close to the inputs. grounding one side of the filaments takes care of the problem. inside the tube the grid is a fraction of an inch from the filament anyway.

I paid $290 for that with shipping. I don't remember how much of it was shipping, but it sat in customs for 4 days, then took 8 days to go from NY to Chicago. Those white covered things are heavy. I'm giving some of them away as christmas presents, as there is nothing in chicago anything like this stuff. At least it got here in fresh condition. Due to lots of butter and booze. Last year birgir sent me 1 of each. That got me hooked. Birgir definitely knows how to increase sales. Birgir should start selling more of this stuff outside iceland.

35 of those are bigger than a mini fridge.

what we need to do is find a massive stock of 1000uf/25v blackgate non-polarized caps. The only place that has them in stock wants $80 each and we need 35 per board.

http://gilmore.chem.northwestern.edu/bernhoftsbakari1.jpg http://gilmore.chem.northwestern.edu/bernhoftsbakari2.jpg http://gilmore.chem.northwestern.edu/bernhoftsbakari3.jpg Next year i will definitely need more

Tom should have received his present today

updated schematic and picture, have a look. very pathetic BOM in progress includes some of the power supply 1000uf 35v ls=5 dia=13mm 871-B41041A7108M000 35 0.47 16.45 .1uf 100v ls=5mm 80-R82EC3100DQ70J 14 0.32 4.48 4.7uf 63v wima 505-MKS24.7/63/5 7 4.33 30.31 470pf 500v ls 5.9mm 5982-15-500V470 4 2.28 9.12 39pf 500v ls 5.9mm 598-CD17ED390JO3 2 1.35 2.7 220pf 500v ls 5.9mm 598-CD15FD221JO3F 2 1.22 2.44 220uf 50v ls 5mm dia 10mm 647-UPW1H221MPD 4 0.58 2.32 4700uf 50v ls10mm 35 dia 598-SLP472M050C5P3 2 3.2 6.4 mjf15030 6 1.63 9.78 mjf15031 4 1.63 6.52 heatsink amp board 513102b02500g 8 1 8 mpsa56 19 0.12 2.28 mpsa06 23 0.32 7.36 lsk389 2 relay 653-g5v-2-h1-dc24 2 2.76 5.52 op27 digikey 4 6.01 24.04 Its about $137 not including the resistors and some of the other power supply components. So definitely under $250 with transformer and everything. Unless you go very nuts on the resistors.

Thanks, got all of them. Not sure yet about q27 collector cap. The 4.7uf caps are at birgir's idea. (also kind of expensive) evidently no amp is complete without some wima's. will post again when i get home.

please someone (everyone) verify the board vs the schematic http://gilmore.chem.northwestern.edu/ksa5.pdf http://gilmore.chem.northwestern.edu/ksa5.jpg

The picture says 500R. will check the detailed pictures later.

same compatible pinout.

stereo board http://gilmore.chem.northwestern.edu/ksa5stereo.jpg may put an alpha pot on it depending on what people want. 10 x 7.5 inches. Slightly smaller than the original.

mikhail's just awful hack job removed and correctly replaced http://gilmore.chem.northwestern.edu/dsc_0234.jpg