Capacitors and double-blind listening tests
Over the past 20 years or so, many articles have been written proving beyond any doubt that there are quality differences between capacitors. For example, capacitors produce distortion which depends strongly on their dielectric and construction. A recent series in Electronics World shows that class II and III ceramics are the very worst, unipolar electrolytic capacitors are better, bipolar electrolytics are better than unipolar ones and so on, the very best types being polystyrene, polypropylene and class I ceramic capacitors. Further, other articles show that dielectric absorption produces measurable differences in the small-signal step response (or equivalently in the magnitude and/or phase response) of an RC high-pass compared to an ideal first-order high-pass.
However, except for the worst types, capacitor distortion is extremely small and the differences in small-signal response between a high-pass with a dielectrically absorbing capacitor and a perfect first-order high-pass is very small compared to the difference between an ideal first-order high-pass and a straight wire (no high-pass at all). Therefore, it seems logical to assume that as long as all RC time constants are high enough and as long as you don't use class II or III ceramic capacitors, it doesn't matter much what type of capacitor you use for DC blocking.
Nevertheless, most audiophiles claim to hear clear differences between capacitor types. Has anyone seen any articles about double-blind listening tests supporting these claims? If not, is there anyone in the neighbourhood of Haarlem willing to participate in a simple double-blind test, using a home-made ABX switch box?
Thanks in advance for your answers and/or insults,
Marcel van de Gevel
I'm sorry that nobody seems interested in participating in this discussion. For what it's worth, I'd like to share some of my thoughts on caps.
EW articles are written very seriously and with expected results, of which about 90% are applicable in practice IMHO. However, testing conditions (DC offset, current thru cap, etc.) and working condition can differ. Some people claim that caps behave differently in extreme conditions (mechanical vibration, magnetic field), which may account for the remaining 10% of the results. Also, total distortion for most capacitor types is so very low that although I've seen many regular elcos in several consoles, I've never been able to hear any distortion.
If there are any differences between capacitor types, I don't think it is possible to hear them at first listening. For this reason, I think that double-blind tests are pretty useless. I regret that I'm not in your neighbourhood to come to your test, though. After a few cans of beer I bet I would be able to hear a lot of distortion coming from those little black ones.:drink:
A while back, I changed the caps in cheap 'Whu Flung Dhung' DVD player.
I had a second example as a reference (both brand new), and I first changed the DSP pcb audio signal electros to Hitano EXR low esr caps.
The sonic difference was clearly audible wrt to the virgin one.
Next I changed the supply bypass caps on the DSP pcb to the same Hitanos.
Again quite a difference to the stock unit.
Next I changed the SMPS secondary side caps to Hitano EXR and again quite a difference wrt the stock unit.
The end result was that the modded item sounded extremely good indeed.
Overall distortion and IMD clouding were dramatically reduced, and music extended both lower and higher.
Detail and depth imaging was much improved, whilst more relaxing sounding overall.
For about $5.00 and minimal effort, this produced a very nice to live with sound.
I have also changed cheap looking coupling caps in pro gear to Hitano EXR, and I get nice improvement every time.
I say that electros can be strongly audible, and in my experience the Hitano EXR ones are nicely pleasant sounding - they have very little self sound really, but their sound is always less than, and different to the originals in my experience.
Re: Capacitors and double-blind listening tests
I would love to participate in your double blind testing if that makes you happy.
I usually live a while with the modded cap. Double blind testing is too complicated for me. If in doubt I sometimes switch back to the old situation. I am not doing scientific research and luckyly I don't have to prove anything!
The audibility of changes to the sound is very dependent on the installation used especially the speakers.
I remember a friend of mine trying to sell a expensive moving coil cartridge to a guy with a rather cheap set. When the cartridge was installed nobody heard any difference. But on Klipschorns the differences were clearly audible.
My most recent experiments indicate most polypropyle film caps ring like hel......Or does the CD ring....? Still not sure........
Polycarbonate seem better in this respect but leaves out some information, perhaps.
At the DAC output with some 6.5V DC riding on the signal some old polycarbonate ERO KC1854 sounded better than ERO MKC1862.
The screaming quality on some bad CD's was reduced a lot compared to MKP caps like Audyn or Audyn Plus.
As the non-metalised ERO's sounded better than the metallised I am inclined to try non-metallised polypropylene like Hovland, Ero KP1832 as sold NOS by Steinmusic Germany or Arcotronics from www.rswww.com in UK.
The latter two suggestions courtesy of Thorsten Loesch.
As concerns mesurements I doubt a meaningfull test is available.
I have a capacitance tester that can measure the Q or dissipation factor of the cap.
Polypropylenes usually have a Q > 999.(limit of measurement)
Polyester about 180
Metallised paper from AR3a loudspeaker: 16
Oscon about 10
Blackgate about 5 and dropping! Value keeps dropping during several minutes. Burn in phenomenon???
Ordinary electrolytic 4.3
Please note the measurement frequency is 1kHz.
How these values correlate with the listening experience? I have no idea:confused: :confused:
Although I am not really in your area, I would be interested to spend the time to come up to Haarlem from Maastricht for this.
I don't think I am an experienced golden ear, but would like to try. I have one or two friends that I think DO have golden ears, who might be interested as well. Please keep me informed of your plans.
PS I do enjoy your articles in the journals!
Dear Elso and Jan,
I guess we better discuss the details by e-mail rather than on this forum. I haven't really thought about the exact procedure yet. The equipment, listening room and procedure used have to be such that the capacitor audibility is optimised according to the test persons believing in capacitor audibility. I guess I need your input for that.
By the way, Jan, when you wrote "PS I do enjoy your articles in the journals!", did you mean the series of articles by Cyril Bateman in Electronics World that I referred to in the first message in this thread? Or did you mean the few articles I have written? In either case, I'm glad you enjoyed them.
I have often used these bi-polar caps with good result!
(but never compared w. Black Gates - bi-polar).
Anyway, they are cheap!
Three big cheers for the simple truth, Marcel, and a couple more for the support, Moamps.
I also believe that some people hear differences between quality capacitors that have been changed in an audio circuit. But, I believe that the real differences are more due to changes in the solder joint or other effects of heat then the capacitors. There are also cases where the different physical configurations of different capacators result in a different response to RF frequencies. This can change the "mud" level in an audio circuit.
Except for relative frequency (relative pitch) and phase differences (sound source location) hearing is one of the least precise human sense and the one most easily miss-analyzed by the mind.
A wise man once said; "One mans passion is often another mans poison" and there by hangs the best of arguments.
The laws of physics always apply.
The real question is;
which ones this time?
Paper Caps in Xovers
My school-age kids have excellent hearing, and I rely on their opinions about how my diy speakers and amps sound. For speakers they usually prefer the sound that comes with good paper caps in the crossover (right, the ones sold by Tony and Gintaras on eBay). For the electronics...well, they like 2nd order harmonics...that's a tube design question, not a cap question. Oh, interestingly, they love the AR3a sound for rock.
|All times are GMT. The time now is 10:19 PM.|
Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio