kevin gilmore

updated ksa5 small board and golden reference changes from the original, on board heatsinks, darlington outputs and lt1021 as the reference, 10V which makes a difference too. http://gilmore.chem.northwestern.edu/goldenreference.jpg http://gilmore.chem.northwestern.edu/boards/goldenreference.zip

actually the golden reference regulator does perform slightly better than the krell supply with a couple of changes. (lt1021 and dn2540 replacing fet and lm336) krell supply is about 2 microvolts of noise at 500ma, the golden reference is 90 nanovolts of noise at 500ma. if people want, I can turn the krell supply into the golden reference, but still with the on board heatsinks. definitely will cost a fair bit more than the Chinese version.

very nice wood chassis. are the white circuit boards the latest in trendy?

http://gilmore.chem.northwestern.edu/boards/cfp3rs.zip someone really needs to check this hard to make sure when I changed the transistors everything still goes to the right place.

http://gilmore.chem.northwestern.edu/boards/ksa5psgsmall.zip old board was 6.27 x 4.8 new board is 5.5 x 4.37 changed mpsw to pzta

The alternate krell supply is bigger than it needs to be, because I made it the original size, will look into shrinking it a bunch.

this power supply performs no better than the fully regulated alternate krell ksa5 power supply so it may be easier for people to do that.

looks like all the parts are available to build that. 2sk170GR, GR only. A board to do that would be pretty trivial if we can't buy in the usa. Still going to cost more than a standard linear I think.

where did you get the power supply board?

the 5pf cap needs to be 1000v not sure about the other one

I use a very small machinists vice like this https://www.southern-tool.com/store/wilton-super-precision-tool-makers-steel.php mine is about 40 years old, and even better built, no idea where I got it

the fact is that unless i use piles of NC machines to make stuff (in which case i'm not really making it) birgir and justin do a way better job of assembly. I just don't have the patience.

so i have a blind yorkie and a friend has a blind dachsund will that work. seriously my 3 remaining dogs, turn on the classical music and they sit on the floor and don't move, wagging tails.

actually the top does not come off, the boards are bolted to it. the bottom comes off. (paging steve...) its Purdy on the inside, birgir will have to post pictures.

lets call it solid state grounded grid. and yes this is the first one with a silicon carbide fet.

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

I like this idea, will simulate soon actually there is a voltage reference, its ground. looks good.

you do have the wood grounded.... its important

i used the taller ones on the krell klone, i liked them better because they only had one hole

if that is one of the 2 sections in one section things, those are awful.

some future boards, probably the dynaloss would be 3 or 4 layers if i go completely surface mount. not really possible on the kgsshv boards

it is a dc amp, however the servo's will cause issues measuring dc. of the 5 pots i bought for this, i was only able to use 2 of them the others had problems either at low volumes and high volumes

its the same heatsink as the krell klone. (can't find it at the moment) the 2sc4686a parts are insulated, so just thermal washers both sides, screw and nut

those are insulated parts, so just regular thermal washer,screw and nut.

yep, how I did that I do not know. it was still possible to build that board, you had to bend the pins and solder all the parts that were not the output devices standing up and backwards. a lot of clean up and various traces on both sides the same exact length just for extra fun. and some ground plane on the bottom. new version http://gilmore.chem.northwestern.edu/kgdynalobal7.jpg