EBERGER@B.PSC.EDU (04/10/90)
This Weekend I had the chance to audition 2 more brands of polystyrene capacitors in my test setup: Seacor & RelCap. I received the Seacor's as samples from the company. These are from their more expensive line and are yellow tubular style. Sonically these have a slightly dark & warm coloration, not bad. These appear to be more resolving then most of the other brands I've tried. Their colorations are mostly inoffensive... These would probably sound right at home in a 2yr old conrad-johnson preamp. I'll give them third place. (minimum order appears to be 1000 & cost ~$2each for 1%@630V) The RelCaps were part of someone's update for a Theta A preamp, which my friend bought with a used circuit board through AudioMart. One of these was an axial black cube style, the other tubular yellow style. These are Excellent capacitors. They have very good detail and pass through dynamics very well. They don't seem to resolve the upper octaves as well as my favored Siemens. I can imagine where they could be preferrable though. I'll give them second place. Several mail-order places list these in their ads, I'll have to check on what values are available... My favorite capacitor for small values is still the SIEMENS B33062 series. These are little clear tubular capacitors with a red endcap on one side and a white endcap on the other. There are no markings denoting them as SIEMENS. If you see these guys anywhere, send me a note... -Ed Berger eberger@b.psc.edu
EBERGER@B.PSC.EDU (04/11/90)
The circuit I use for evaluating these capacitors ~750pf & ~3000pf is a simple NE531 ic-opamp based phono preamp with regulated powersupply. This circuit uses an active RIAA eq very similar to the Audio Research SP-6 series preamplifiers. The circuit is built with 1% resista resistors and WonderCap power supply bypasses. It uses an electrolytic (yuck!) in the feedback path to ground so that it can be direct coupled to the line in of a modified SP-6 (exhaustive parts changes, India 12AX7s). Speakers/Amps are modified Acoustat X electrostatics/servo-charge amps on stands. Input source is a Supex SDX-1000 mc (modified& still running after 8yrs!) pre-preamplified by a Thorens PPA-990 headamp (modified). Cables are all Livewire DB-1 litz interconnect, which aren't made anymore, but really should be, since they ARE my favorites :-) Most of the products are very picky about ancillary components, so chances are that nobody except for me and few friends will appreciat them, especially the Dual 731q direct drive turntable (Anyone have one for sale?) The SP-6 already has SIEMENS caps in the RIAA stage, so I made the test setup to try other things like different ICs & eq caps. I like the NE531 IC for its dynamics & bass, but it is a quirky chip, and was used by Rappaport as digging into a dead lump of epoxy & wires revealed... -Ed Berger eberger@b.psc.edu
eacj@tcgould.TN.CORNELL.EDU (Julian Vrieslander) (04/11/90)
Thanks for the reply. I have been out of the tweaking business for a few years now, but I used to do a lot of playing with op-amps and passive components in modified HiFi gear. I did a heavily rebuilt Dyna PAT-5 back in about '76 that used some teflon caps in the RIAA EQ network. I got them as freebies by begging for samples from some mil spec cap vendor (can't remember who). I'm not sure if they would sound as good as the film caps favored by audiophiles these days, but they sure looked sexy - little stainless steel and hermetic glass tubes, with 1% tolerance markings and gold leads! The PAT-5 sounded pretty good, too. It embrassed quite a few SP-3a's, etc. before I sold it to a friend. The most effective A/B demo I ever did involved drag racing the Dyna against a Crown IC-150A. The Crown had better specs than a PAT-5 (at least the stock PAT-5, I never measured the modified one). For source material, I used a recording of Julian Bream and his consort. I first played a selection through the Crown - what came out was a very nice, smooth solo guitar sound. When I played it on the Dyna, jaws would drop. You could hear that there was a harpsichord playing in unison behind the guitar - the Crown completely eliminated it. I wish I could restage this demo for the experts on the net who contune to claim that audible differences between components exist only in the imaginations of audio fetishists. I am currently thinking about doing some upgrade mods on my CD player (Philips CD-680X). Most likely I will go with most of the recommendations made by Walt Jung in his "POOGE" series from Audio Amateur, and possibly an upgrade to the IC voltage regulators. Jung seems to recommend the Panasonic film and electrolytics, probably because they are cheap and easily available from places like DigiKey. I'm not sure if your recommendations (Siemens, etc.) are relevant to this application. The Philips players seem to benefit from better electrolytics in the power supply filters and local bypass locations, and good film caps for bypassing the electrolytics and DAC taps, and for the final output coupling caps. But there is not much space on the board. The DAC tap bypasses have to be really compact, since you are replacing surface mount ceramic caps. The kit from Audio Amateur/Old Colony provides 0.82 microfarad Panasonic stacked film caps for that application, but they are still rather big. For some Philips/Maggotbox decks you have to cut clearance holes in the bottom of the case (plastic) to clear the caps, since they mount on the underside of the circuit board. I have requested capacitor catalogs from Panasonic, Roederstein, and (at your suggestion) Siemens. I notice that a lot of high end audio products use WIMA caps. Do you know who makes or distributes these? I can't find WIMA in the EE Master. And have you figured out who makes "WonderCaps?" I suspect that the markup on those puppies must be enormous. There are several other audiophile capacitors floating around (Solen, Sidereal, etc.) that I am likewise suspicious of. Maybe I'm too much the skeptic, but I find it hard to believe that the tiny companies advertising in the back pages of Audio are manufacturing their own components. - Julian