Jump to content

kevin gilmore

High Rollers
  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by kevin gilmore

  1. kevin gilmore

    current feedback electrostatic amp

    so sorenb really wanted this, so here it is http://gilmore.chem.northwestern.edu/cfaelectrostat.pdf the outputs are 300v darlington transistors just need to find a suitable npn match for the stn9360, and this can be done all surface mount tiny, fully complementary, fully dc coupled, decent input impedance (none of this 75 ohm nonsense) push pull output... edit: PBHV2160ZX looks great for the npn. so you can add one more set of outputs and run up to 500v power supplies balanced input only, and you need 4 of them
  2. updated Megatron schematic. If you manage to figure out what the stax mafia is doing, please let us know. We are having a hard time keeping track of it all
  3. I'm happy to report that the goldenreference low voltage dual power supply works as designed, and tests less than 1 microvolt of noise. Going to have to take it to work and put it in a screen room to really measure the noise. http://gilmore.chem.northwestern.edu/goldenreferencedual.jpg however there is one silkscreen error seen by the extra + near the 47uf cap on the right side. I will update the files shortly. All 4 of the 47uf caps need to have their plus sides pointing to the left. very happy with this
  4. been working on this one a while, with all the popular transistors disappearing, I hope that this one won't go obsolete before the boards are out. I don't think so because Fairchild has said that the bc parts should be available for at least 5 years but onsemi said the same thing. This one is a diamond transistor input set, used as a voltage to current converter. A pair of Wilson current mirrors with input and output cascade as the VAS and a triple darlington output stage. http://gilmore.chem.northwestern.edu/cfa2.pdf if someone knows of better surface mount parts to replace the bc series, I might as well do that from the start. ft >150mhz, hfe >100 Vceo >50 etc
  5. kevin gilmore

    Why no KG class AB ?

    long time ago there was a Yamaha integrated amp that had an A/AB switch. like about 1978. really nice thing, would make a perfect headphone amp if you can find one that still works. in fact the CFA amp is a A/B amp with the bias between 150 and 200ma. These days that kind of bias into just about any headphone results in pure class A operation. The uberamp2 is also a A/B amp with the bias between 2 and 3 amperes. about 100 watts into 8 ohms balanced.
  6. kevin gilmore

    grounded grid

    kgsshv-carbon with the sic-fets turned into 6ca7 was going to be next year
  7. kevin gilmore

    krell ksa5 klone

    Both Birgir and I have wanted one of these for quite a while, and they never seem to show up. So with a little help from one of the members here we were able to reverse engineer this one. http://gilmore.chem....rn.edu/ksa5.pdf http://gilmore.chem....rn.edu/ksa5.jpg still have work to do on the power supply. All the parts are still available, but identical and compatible parts also available. mpsw06 and mpsw56 and lsk389 I think this one is fairly cheap to do. And easy to assemble. Single channel is 10 x 3.65 inches. Single ended not really a HE6 killer, but balanced, a HE6 heavyweight for sure. If you replace the servo opamp with opa445, then you can go up to +/-35v rails. Then its definitely a HE6 killer single ended. picture of the original http://gilmore.chem....du/ksa5orig.jpg single ended with +/-21 power supplies, it does 10 vrms into 4 ohms or more. balanced 20 vrms into 8 ohms or more. single ended with +/-35 power supplies, it does 16 vrms into 4 ohms or more. balanced 32 vrms into 8 ohms or more.
  8. a number of lab power supplies have an adjustable current limit, so one of those set appropriately would be identical to the smps supply without the extra noise. but it would be easier and a bit cheaper to build a diy version with grlv and grhv supplies, a real transformer, in a standard heatsink chassis, with the outputs transistors on the heatsinks etc. then it would last more than 3 years. Those smps supplies when they get old, the main cap dries out, and usually bad things happen, especially to the output voltage.
  9. Unlikely to be a transient but only Alex knows at this point. Something with the clever way the filament and high voltage power supplies start up. You would think that rudimentary real world testing would have found this before a batch of them were made. then again on the liquid carbon version 1, using balanced and unbalanced headphones at the same time, and the amp blows up, something else that should have been caught before a large batch were made.
  10. so here is a reason not to buy the LP. if you turn off the switching power supply (or a power hit, or unplugging it) with the amplifier power switch on, a mosfet in the amplifier power supply fries. another of the many reasons why rail splitter power supplies are a bad idea.
  11. a number of people are complaining that there is no gain switch and the default gain is too high. kind of what you would expect from a current mirror feeding a tube. evidently its dc coupled but no pictures good enough to reverse engineer. If you go by the price of the cth x 2, this should be about $450 max.
  12. kevin gilmore

    New JDS Labs Atom Headphone Amp

    lets see jds atom $99 opamp and output buffer, lots of feedback, dual power supply, probably the lowest thd of the bunch. made in usa schiit magni3 $99 discrete current feedback amp, minimal feedback, dual power supply, probably the highest thd of the bunch. made in usa monoprice liquid spark $99.99 voltage feedback amp, fair bit of feedback, rail splitter power supply, in the middle for thd.made in china sounds like a big pile of same same
  13. kevin gilmore

    The Headcase Stax thread

    yep, pretty much that schematic. pull up resistor is 1 meg. false advertising, shows a rca to mini, not supplied, shows 1/4 jack not supplied. comes with usb cable. bias resistor is 20 meg. power supply is fully regulated, something the sr-001 was certainly not.output stage bias is 200 microamps. I think that this needs a lot more input voltage than it should. also no volume control, so designed to hook to something with a volume knob. made in Taiwan. no names anywhere. after a few hours of warmup, still sounds like absolute crap. No way is this worth $700. or the $1k+ the guy on ebay is selling it for.
  14. kevin gilmore

    The Headcase Stax thread

    it has arrived, going to wait for it to warm up before listening N1,N2,N3,N4 are labeled XG (2sk209) q1,q2,q3,q4 are labeled lkw (pmbta45) opamps are jrc401f single high voltage of +440V. opamps get +/-17 this is very similar to the original stax sr-001 amp
  15. kevin gilmore

    Massdrop Cavalli Tube Hybrid

    so it turns out this one is a bit different than the rest of them. Input is 2sk2145 dual n-channel jfet, driven by current mirror on top, and current sink on bottom. Definitely a voltage feedback amp, lots of feedback. liquid gold, liquid carbon,massdrop liquid carbon, massdrop cth are all current feedback amplifiers.
  16. kevin gilmore

    The Headcase Stax thread

    If there is a diaphram arc due to very large excursion, the extra stored energy will cause damage to the diaphram
  17. kevin gilmore


    its like that monster 2 box singlepower thing I have. keep the case and front panel, and gut the insides.
  18. kevin gilmore

    Massdrop Cavalli Tube Hybrid

    cavalli liquid carbon massdrop version rail splitter cth and spark are similar railsplitter.PDF
  19. kevin gilmore

    19V Linear PSU for PC

    you use either of the 2 references not both, and you have to adjust the resistor string for desired voltage jp1 jumpers pin 8 of ic2 to ground. definitely not for the lt1021 or ref01/02.
  20. kevin gilmore

    19V Linear PSU for PC

    ic2 is replacement for ic1, ref01,ref02,lt1021 etc lps200.PDF
  21. kevin gilmore

    19V Linear PSU for PC

    this is already very similar to the grlv. missing the double noise inverting current source, but that is about it. ic2 is an optional and higher quality alternate reference.
  22. kevin gilmore

    Mr. Speakers Voce

    these headphones stopped by in Chicago on their way to Iceland, I listened for a couple of hours. My initial comments was that they were very sterile, but really I think it was the complete lack of imaging that is the problem. removing the foam probably fixed a bunch of that. The stax plug is a major fuckup. And I don't see why. Everyone else including stax,hifiman,kingsound,chinsettawong,birgir and Justin are capable of making plugs that are the right dimensions, no reason why this has to be a piece of crap. Yes they are about 6db less efficient. And the case is some kind stupid. I did not measure the capacitance of the cable, but i'm going to bet that its twice that of the stax made cables. birgir will hopefully do that. Next up the monoprice things once they actually start shipping.
  23. kevin gilmore

    Massdrop Cavalli Tube Hybrid

    Mismatched very high cacitance vishay siliconix output mosfets. And ground splitter. Otherwise identical current feedback amplifier same as liquid gold,liquid carbon,Cth, liquid platinum etc. at least it's better assembled than schiit.
  24. kevin gilmore

    introducing the black hole

    that was ray's darkstar amp. the blackhole is something different. the blackhole uses a national semiconductor driver chip lme49830 and a pair of alf16P/alf16N semisouth fets for the outputs. You can probably substitute something else for those parts, including bipolars, as the board has options for that too, but I never tested that way. the darkstar sounds absolutely miserable, and these days has stupidly large voltage swings guaranteed to set just about anything on fire. the blackhole is suitable as a small speaker amp.
  25. kevin gilmore

    introducing the black hole

    So it turns out that the lme49830 is pretty much a dynafet on a chip, with diamond buffers driving the fets. SO... combine with the semelab-tt lateral audio mosfets, and the result is something super sweet, and easy to build, about $50 in parts per channel. Combine 4 channels with my balanced preamp, and have one serious kickass thing guaranteed to drive anything, he6's and k1000's included. At about 2 orders of magnitude less distortion than that other thing. And more than 10 times the available output current. at... 20hz to 20khz... .1mw .05% thd (most of which is noise) 1mw .01% thd 100mw .005% thd 1w .0003% thd many watts <.0001% thd http://gilmore.chem....u/blackhole.jpg Works great at +/- 30 volt supplies, But also works all the way up to +/-100 volt supplies. 30 volt supplies highly recommended especially balanced, because lets face it, enough is enough. Suitable for driving say ten thousand pairs of sr-009 at the same time (400 volts peak to peak) A bit more work to do, but getting close now. 3.2 x 3.6 inches plus required external heatsinks.