Capacitor Sound Quality

I have been somewhat astonished by the difference between a 2200pf polystyrene (unknown brand, but not cheap nowadays) and a Wima FKP1 pp film caps used in an output first order RC filter.

The filter sits between the secondary of a Sowter transformer with 2.4k secondary that I was experimenting with following the I/V stage of a Sony SCDXB940 SACD player, the idea being to remove two opamps elements and a coupling cap from the signal path, and reduce digital hash.

I was originally using a 100kHz filter with an FKP1 and the sound was quite satisfying except it was a touch sharp in the treble. So I reduced the freq to 50kHz with the polystyrene cap. The sound deteriorated very considerably, with a dry treble and narrower sound stage. I then replaced this with an FKP1 of the same value. Lo and behold, the sound became warmer and an improvement over the original 100kHz filter.

Now my polysyrene caps measured well and experience tells me that they shouldn't have sounded that bad. Any comment or similar experience?
 
Questions like these I go to myself: I am maybe capable to tell a difference ONLY if I can switch fast between the alternatives. The ears forget very fast unless there is a huge difference.

When I tested my DAC vs. the built-in in my DENON DCD-1520 I wanted very very very much that my DAC would sound better and yes it does under very special circumstances. I can't tell the difference if I can't compare fast enough. The mind is powerfull when we want to believe. When I tested the DAC I could switch fast between the signals because I used a preamp with many inputs.

But I realize also the same people have gifts. Some people can tell if a tone is off key simply by hearing this particular tone.

fmak, no offence, can you comfirm your statements in a blindtest?

My point of view in this matter is that capacitors adds very little to the sound in most applications if they are proberly chosen. My designs rules are to avoid caps in the signalpath but this is more of technical reasons. It's a challenge to make amps with few caps.

Since we must be nice here, everyone is entitled to an opion and my is as I have written above.

Since we are (I, at least) about believing, I like polyprobulene the most. Styrene is OK but they are hard to get and have also no-good shape for industrial use. 63V polyester aren't either so bad, rather good actually. There are new materials for SMD caps poly???etylen??falat?? which I have no experience of.
 
Well, I always stuck with polypropylene since you can't really go wrong. But a little while back I tried experimenting with different polystyrene caps, and cheap ones sound bad, expensive ones sound really good. That's my conclusion. Maybe a different brand (better) polystyrene would sound even better. But since they're hard to come by, and a pain to find good ones, I just stick to polypropylene. I didn't even try to get some teflons to try out..
 

Nelson Pass

The one and only
Paid Member
2001-03-29 12:38 am
I find that polycarbonates go best with white wines,
such as a good Chardonnay, whereas polystyrenes
require sweetening with some trokenbierenauslese.

My real preference is a good Bordeaux; it really improves
those electrolytics, but you need to get a head start.
 
Questions like these I go to myself: I am maybe capable to tell a difference ONLY if I can switch fast between the alternatives. The ears forget very fast unless there is a huge difference.
Absolutly this confirms my own experience!
I always wonder how folks can describe differences in sound from single parts as capacitors or whatever so easily.
To make valided comparisons would require a huge effort of the test procedure. For example telling about the difference in sound of different capacitors in a circuit would require to switch every single capacitor to another. As this switching activity might introduce more difference itself than the DUT this is impractical in many (if not most) cases and an otherwise identical circuit for a true comparison has to be build.
I doubt that many DIY`ers are going so far when making their comparisons and therefore always I find it somewhat curious hearing others telling about sonic differencies without mention HOW they did the comparison setup actually.

A note to blind tests:
Confirmation of sonic differences is valid only if it can be verified to at least 80-90%.
This means one has to identify the DUT at least 8 to 9 times from 10 comparisons and wether or not some white wines or trokenbierenauslese is involved or one stood up early in the morning or did it after a hard working day.
Remember: a result of only 50% is just a hit by chance.

whereas polystyrenes require sweetening with some trokenbierenauslese
How much trokenbierenauslese is necessary for sweetening a 47nF polystyrene??
I wonder if I can stand to equalize the 8x 47nF in my active crossover:p
 
cocolino, well just to let you know, I didn't change components. I built a second with an identical pcb, just used different caps. I had to build a preamp for my father so I built mine and his with different caps to see the difference. Sure the difference of sound could be atributed to any other subtle change in the construction. But the caps were the only change I meant to do.
 
rljones said:
fmak,

I'm interested in what you are doing, and have a few questions.

Which DAC and which Sowter transformer are you using? Do you know the resistance of the primary? Also, are you doing anything about the DC offset on the primary side, or are you running the output of the DAC directly into the transformer?

Thanks,

Robert
----------------------------------------------------------------

Its the Sony balanced current pulse CX??? and Sowter 9360, modified 8540. See dddac.de 777 tweak 1. Primary 600R.

The output goes from I/V opamp output. Lead dressing is important in this noisy environmrnt.
 
[
fmak, no offence, can you comfirm your statements in a blindtest?

My point of view in this matter is that capacitors adds very little to the sound in most applications if they are proberly chosen. My designs rules are to avoid caps in the signalpath but this is more of technical reasons. It's a challenge to make amps with few caps.

---------------------------------------------------------
There is no need to go blind if the difference is so pronounced as I have found. There is plenty of evidence since the Jung article years ago that capacitor sound IS THERE. See also Colloms on the latest HFN.

If you can't hear differences, why bother to improve designs? What is the yardstick other than theoretical advantages or no measurable difference??

Just to show I am not a nut, I am professionally qualified in acoustics and control and instrumentation. However, I do trust my ears, having done the mesurements.:)
 
fmak said:

There is no need to go blind if the difference is so pronounced as I have found. There is plenty of evidence since the Jung article years ago that capacitor sound IS THERE. See also Colloms on the latest HFN.

I go for the same line as cocolino, can't you tell in a blind test by 80-90%, you have no real difference (for the program material you've listened to, I must add), but a MINOR difference maybe.


fmak said:

If you can't hear differences, why bother to improve designs? What is the yardstick other than theoretical advantages or no measurable difference??

There you hit the hammer on the nail (is it right in english? ) "Där slog du huvudet på spiken" if you talk swedish. Why improve CD when <0,1% of all recordings need only vinyl or cassette quality? Why do amatuers need 192 kSp when pros feel 96 kSp is sufficient?
I have the answer,

some people want to make money and then you have to come up with something NEW!

I suspect that you have more problems with 192 kSp because of emission and problems with speed in analog circuits. I don't know that but I suspect that.

fmak said:

Just to show I am not a nut, I am professionally qualified in acoustics and control and instrumentation. However, I do trust my ears, having done the mesurements.:)

I don't regard you as a nut. It's refreshing with peolpe who feels something or have point of views.
 
peranders said:


some people want to make money and then you have to come up with something NEW!

I suspect that you have more problems with 192 kSp because of emission and problems with speed in analog circuits. I don't know that but I suspect that.

-----------------------------------------------------------------
There are some who want to stay with the present. With an open mind, you will find 192 DVDA and SACD way better than CD with the brickwall filter. Hence my attempts to improve SACD. Many will agree with me and many with you, I suppose.

I have learnt over the years to trust value judgements, not imperfect equations and deductions.
 
fmak said:

There are some who want to stay with the present. With an open mind, you will find 192 DVDA and SACD way better than CD with the brickwall filter. Hence my attempts to improve SACD. Many will agree with me and many with you, I suppose.

I have learnt over the years to trust value judgements, not imperfect equations and deductions.

I am keen on new cool things but also, some things aren't better or worth paying for. If you talk new techology, is MP3 a good thing?

Don't you agree that many many recordnings are medium to really bad? Why do we need 192 if the recordning is bad(note I didn't say the music)? In my collection of 700 CD fewer than 50 have really high audio quality. No offence americans but _very_ few of them are american! Lot's of analog tapes with high noise and distorsion. Danish, german, english and swedish recordnings have usually high quality. I hope my last statement will be good fuel for a debate.
 
Oh, I've learnt some new words ...

Trockenbeerenauslese. German In the Qualitätswein mit Prädikat category of German white wines, as described by German wine law, this wine is made from the ripest grapes and so is the most rare, rich and expensive. It is made from the selected harvest (auslese) of individually picked grapes (beeren) that have been dried up (trocken) by noble rot, resulting in a wine that is richly sweet and deep gold in colour. The difficulty and risk of producing these wines, along with their exceptional quality, make them among the most expensive and highly prized wines in the world.

enology. The study of wine and the making of wine; viticulture. [Greek oinos, wine + –LOGY.]

Davis. A city of central California west of Sacramento?
 
phil said:
There is an interesting artical in July 2002 of Electronics world on the subject of capacitor sound.
zapel.
------------------------------

Yes, but this is actually about capacitor distortion. The interesting aspect is that some samples of ceramics distort badly even though others of the same don't. There is no reference to the memory effect which is significant.