The equivalent circuit doubtless has some aspect of reality, and microphony can't be discounted, but we would suspect rattling coils as being far worse offenders. It's really not possible to make a 10uF capacitor out of pure gold foil and polyethylene without it being the size of a car battery though...
I'll settle for fat yellow 400V polypropylenes at £5 a throw. No snake oil for me!
Perhaps the crossovers should be in the next room though? I used to do that with my turntable.
I'll settle for fat yellow 400V polypropylenes at £5 a throw. No snake oil for me!
Perhaps the crossovers should be in the next room though? I used to do that with my turntable.
Attachments
There's a very interesting and simple apparatus you could make to test capacitors sound quality, in fact I'm surprised no-one has done it yet!
You use the thermocouple idea to balance a system of two capacitors:
You then apply DC or maybe a 3KHz sine wave to this system. The loudspeaker (which may need a little balancing with potentiometers or varicaps to get exactly balanced) should then produce NO OUTPUT.
If the capacitors are microphonic, tapping one should produce an output on the speaker exactly like the distortion a capacitor might make. In fact you could place capacitors of different types (maybe even an electrolytic) in each arm, and the difference will be audible. Oh, I amaze myself sometimes!
You use the thermocouple idea to balance a system of two capacitors:
An externally hosted image should be here but it was not working when we last tested it.
You then apply DC or maybe a 3KHz sine wave to this system. The loudspeaker (which may need a little balancing with potentiometers or varicaps to get exactly balanced) should then produce NO OUTPUT.
If the capacitors are microphonic, tapping one should produce an output on the speaker exactly like the distortion a capacitor might make. In fact you could place capacitors of different types (maybe even an electrolytic) in each arm, and the difference will be audible. Oh, I amaze myself sometimes!
Attachments
There's a very interesting and simple apparatus you could make to test capacitors sound quality, in fact I'm surprised no-one has done it yet!
You use the thermocouple idea to balance a system of two capacitors:
An externally hosted image should be here but it was not working when we last tested it.
You then apply DC or maybe a 3KHz sine wave to this system. The loudspeaker (which may need a little balancing with potentiometers or varicaps to get exactly balanced) should then produce NO OUTPUT.
If the capacitors are microphonic, tapping one should produce an output on the speaker exactly like the distortion a capacitor might make. In fact you could place capacitors of different types (maybe even an electrolytic) in each arm, and the difference will be audible. Oh, I amaze myself sometimes!
Well.........one problem with the above is that if the two caps being compared are not exactly the same value then of course you will get some sound out of the speaker. If the capacitance and ESR differ then yeah you will get sound out of the speaker. For the tapping test it sounds like a good idea.I would think a much better "test" would be doing a SBT or a DBT if you think one cap sounds different from another.
No idea what a SBT or DBT is my friend. Sometimes I think people here speak a special form of techno-gibberish!
It's just the old null-impedance idea familiar to all radio engineers. It assumes you make some effort to balance the bridge, but the surprising thing is it can be done in the resistive arms which is rather straightforward.
AC bridge circuits : AC METERING CIRCUITS
What makes it particularly powerful, is you can balance a series and parallel LCR circuit, for instance, which is useful to equalise a loudspeaker impedance or fit a impedance matching LC circuit.
FWIW, Troels Gravesen noticed Jantzen Superior Z cap did rather better than standard polyprops in a midrange crossover, "night and day!" he said, so there's something in it.
http://www.troelsgravesen.dk/OBL11.htm
Last edited:
Haven't compared these but have found the Supreme S/O to be too bright for my taste. In comparison, I found the Multicap RTX to be more listenable. Not that I settled on the RTXs after extensive cap comparisons. I just happened to have them to compare with the S/O, and much preferred them.
No idea what a SBT or DBT is my friend. Sometimes I think people here speak a special form of techno-gibberish!
FWIW, Troels Gravesen noticed Jantzen Superior Z cap did rather better than standard polyprops in a midrange crossover, "night and day!" he said, so there's something in it.
OBL11
You're kidding right? SBT=Single Blind Test DBT= Double Blind Test but then again it's not surprising that most "audiophiles" have no idea what those are even when spelled out.
"Night and day"? Then it should be obvious in a frequency measurement and the two caps compared should measure differently also hmmmmm? Strange how he didn't bother to do that. Also he says..."Anything from 80-90 uF is fine for midrange series capacitor"...so he thinks using the boutique caps makes an audible difference yet I'm doubting he matched cap values when he replaced the standard cap.Looks sort of sloppy in the details.
Last edited:
SBT equals people being offered an unmarked cup of coke and a cup of pepsi and being asked which they prefer, eh? OK, enough of subjective, lets do science.
Typical power factor (tan d) of capacitors A 1kHz:
Electrolytic non-polar: 0.15
Jantzen standard MKP polypropylene: 0.0002
Jantzen Superior Z : 0.00002
Jantzen Silver/Gold or whatever: 0.00002
Very little difference between Mundorf and Jantzen that I can see.
Jantzen Audio Superior Z-cap
Claims of thicker film, exotic film and exotic metal leads. All a bit bogus for the most part since we know silver (or gold) is not really a better conductor than copper or aluminium or tin. Silver corrodes quite badly too. I'm inclined to think the main factor is the thickness of the metallised film, because the better MKP capacitors do have thicker film.
For sure, on a revealing system you can easily hear the improvement of MKP over electrolytic in crossovers. Maybe the Superiors are even better too, since they measure 10 times better. But the final step into expensive Silver and Gold looks questionable to me.
Typical power factor (tan d) of capacitors A 1kHz:
Electrolytic non-polar: 0.15
Jantzen standard MKP polypropylene: 0.0002
Jantzen Superior Z : 0.00002
Jantzen Silver/Gold or whatever: 0.00002
Very little difference between Mundorf and Jantzen that I can see.
Jantzen Audio Superior Z-cap
Claims of thicker film, exotic film and exotic metal leads. All a bit bogus for the most part since we know silver (or gold) is not really a better conductor than copper or aluminium or tin. Silver corrodes quite badly too. I'm inclined to think the main factor is the thickness of the metallised film, because the better MKP capacitors do have thicker film.
For sure, on a revealing system you can easily hear the improvement of MKP over electrolytic in crossovers. Maybe the Superiors are even better too, since they measure 10 times better. But the final step into expensive Silver and Gold looks questionable to me.
SBT equals people being offered an unmarked cup of coke and a cup of pepsi and being asked which they prefer, eh? OK, enough of subjective, lets do science.
.
Uhhhhhh not quite. You ask them to mark down which they think is in circuit, A or B then do the statistics math and see if there is enough of a result that is better than just chance. Preference doesn't come into the picture.
Also as I stated above he didn't bother to test each cap for correct value nor did he bother to test if replacing the caps caused a measurable frequency difference.
EXACTLY
Last edited:
While I have no particular view that exotic capacitors will make my own low-resolution system better, because my current amplifier has enough electrolytics in the signal path to swamp any improvements in the speaker components, there IS enough measurable difference in the various grades of MKP polypropylenes to say there could be something in it and do some real engineering tests such as that bridge technique I mentioned, perhaps including radio frequencies to nail some stray capacitance issues too.
Look at it again:
Typical power factor (tan d) of capacitors @ 1kHz (* marks my best estimate):
Electrolytic non-polar*: 0.15
Typical MKT polyester* : 0.002
Jantzen standard MKP polypropylene: 0.0002
Jantzen Superior Z : 0.00002
Jantzen Silver/Gold exotic or whatever: 0.00002
Here's something else to consider when testing. Capacitor values in high pass filters are most critical in the stop band, and probably at the -3dB point and below is where you will see most effect. In the pass band they have almost no voltage across them so microphonic, non-linearity, power factor and stray capacitance and mutual interaction effects play less part. All things being equal, the Superior MKP are 10 times better than the regular MKP at whatever it is they do. It's undeniable. Based on the science, I can't see any improvement in the "Gold" and "Silver" exotics though.
Look at it again:
Typical power factor (tan d) of capacitors @ 1kHz (* marks my best estimate):
Electrolytic non-polar*: 0.15
Typical MKT polyester* : 0.002
Jantzen standard MKP polypropylene: 0.0002
Jantzen Superior Z : 0.00002
Jantzen Silver/Gold exotic or whatever: 0.00002
Here's something else to consider when testing. Capacitor values in high pass filters are most critical in the stop band, and probably at the -3dB point and below is where you will see most effect. In the pass band they have almost no voltage across them so microphonic, non-linearity, power factor and stray capacitance and mutual interaction effects play less part. All things being equal, the Superior MKP are 10 times better than the regular MKP at whatever it is they do. It's undeniable. Based on the science, I can't see any improvement in the "Gold" and "Silver" exotics though.
You are determined to poo poo any audible difference here. I have a more open mind. If, for instance, microphonic effects occur in the loud interior of a cabinet, I have already suggested a simple old radio-technique known as the AC Bridge.:
An externally hosted image should be here but it was not working when we last tested it.
Let's reiterate. The Superior Z caps have 10 times lower power factor over standard polypylenes. If you want to make a capacitative microphone, essentially you polarise a capacitor with a voltage, and then squeeze the plates together. It's also not hard to visualise that the mechanical forces rattling a loudspeaker have an equal component in the forces within capacitors and coils in a connected crossover at the -3dB point. The more I think about it, the more interesting it is.
It´s not surprising that most advocates of DBTs are not aware of their limitations, nor do they know how to correctly conduct them, and finally how to interpret them. If you compare how DBTs usually are done in the audio field to those done e.g. for medicine or biology then you are quickly aware that this is "pseudo science".
Have you noticed how some of the words used to describe capacitors (aside from the ones that resemble microphonics) suggest simply harmonic distortion? Words like dark (as in blanketed) or nervous (like a car with its ignition advanced too far). I've noticed these traits and what I believe is 5HD, which is quite distinct and annoying.
Intriguingly I've heard excellent clarity via examples of polyethylene, polypropylene, silvered mica, polystyrene, paper and even electrolytic capacitors.
Intriguingly I've heard excellent clarity via examples of polyethylene, polypropylene, silvered mica, polystyrene, paper and even electrolytic capacitors.
Oh really? Please show us some examples and describe how they were done "wrong" in your opinion. "Pseudo science"? Come on, those that spout the glories of subjective listening are the ones that believe in fairy dust. not the objective ones that actually do some real testing. I rarely even see SBT done because I suppose people are too lazy to even TRY.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.