Yes. Because you specifically and unequivocally referred to harmonic distortion (which is inherently non-linear) and just as specifically and unequivocally attributed it to DA, going so far as to say the models needed to be revisited.
The issue has never been distortion in capacitors in general, the question has always been how you were able to so unequivocally attribute it specifically to DA when even the most complex and accurate models have never shown there to be a non-linear component to DA.
Without your adamant attribution to DA, the question never would have been asked in the first place and we wouldn't have had to go through all this.
Which plot are you referring to exactly? I went back about six or seven pages and didn't see it.
se
Well, aside from the surrounding noise, I think that Stuart and Richard have narrowed things enough to actually test. Richard, it is acceptable for you to not to want to dive back into something you noticed informally years ago. You have provided a jumping off point for those who can SIM or, even better, build something up and test. But that test expands beyond the narrow scope we started with and can encompass other defects and capacitor types.
Folks who aren't doing any more than reading along, give those who are doing something time to get through experimenting. There are things to build and perhaps other test situations to dream up. but postulating and guessing is just a whole lot of noise that no one really needs right now. Everyone doing something deserves respect for their knowledge and effort. SIM pilots should be experienced in electronics and well trained in running a SIM before objecting to a physical test. Math and physics explain everything well - but only if all factors are known. Physical testing takes that into account.
-Chris
Folks who aren't doing any more than reading along, give those who are doing something time to get through experimenting. There are things to build and perhaps other test situations to dream up. but postulating and guessing is just a whole lot of noise that no one really needs right now. Everyone doing something deserves respect for their knowledge and effort. SIM pilots should be experienced in electronics and well trained in running a SIM before objecting to a physical test. Math and physics explain everything well - but only if all factors are known. Physical testing takes that into account.
-Chris
Any non-linear distortion in a polar (or film), appears to me to be coming from the dielectric material. DA is a sensitive test when Not using symmetrical waveforms. It can be used to select caps based on DA. Non-linear distortion of polar caps is very likely due to DA. IMO. Its just a smoking gun sort of thing to me.
But I dont recall giving a plot of non-linear distortion here as SY says I did..... That is a confuser. I didnt make such tests myself.... but others have. C.Bateman measured high distortion of polar caps in his writings.
THx-RNMarsh
You said in your measurements you got distortion. Specifically you said "H2." So did you actually mean hydrogen and not second harmonic distortion? When you say "H2," you are saying "non-linear distortion."
se
At the heart of the matter is whether there is a change of effective capacitance for any reason during operation - and how do you measure such a change? Here is a review of approaches, Low Value Capacitance Measurements for Capacitive Sensors - A Review - The novel results in the field of measurements of very small capacitance changes are presented. The... - Sensors & Transducers.
Already linked to by Joachim Gerhard, marce, etc, elsewhere, but possibly worthwhile doing so here again, for easy future reference, Capacitors: Types, Specs, notes on Use "Capacitors: A Field Guide to Types and Habitats" - by Harry R. Bissell Jr..
Polar capacitors
I am sorry but I am not able to find harmonic distortion in a standard, 10uF/35V electrolytic capacitor. Two in series, no bias, resistor load, pls see the image with circuit shown. That 0.00042% is a sound card distortion.
P.S.: input AC voltage is 1.55Vrms
I am sorry but I am not able to find harmonic distortion in a standard, 10uF/35V electrolytic capacitor. Two in series, no bias, resistor load, pls see the image with circuit shown. That 0.00042% is a sound card distortion.
P.S.: input AC voltage is 1.55Vrms
Attachments
Last edited:
Yes but he specifically noted that DA does not cause audio distortion. I posted that article fragment a few days ago.
Also, note that when Bateman talks about polar caps he does not mean electrolytic polarised caps but capacitors build with polar dielectric. Completely different.
It occurred t me that if you calculate the transfer function of a DA capacitor model, you are facing a bunch of R's and C's. It looks impossible that in such a calculation (which I have not done) you can get anything that leads to additional harmonics like sin2*omega or cos3*omega.
If that is the case (Scott?) isn't that clear proof that DA cannot lead to nonlinear/harmonic distortion based on the current model?
(Not specifically aimed at you Dick).
Jan
Last edited:
See C.Bateman ----
View attachment Bateman_Notes_Cap_Sound_5.pdf
Duplicate his tests and see if you get same results. Then go argue with him about it.
THx-RNMarsh
View attachment Bateman_Notes_Cap_Sound_5.pdf
Duplicate his tests and see if you get same results. Then go argue with him about it.
THx-RNMarsh
It is his reason for what other than the DA model which I would be interested in. And, also his reason for what it is that does cause the distortion he measures. In the ref by bateman I gave -- it is about electrolytics.
Mylar film caps have similar polar-dielectric properties btw and they do measure worse than say ps or pp. Many variations in film caps to consider.
Also, the sine wave is a very insensitive test compared to pulses.... it seems there isnt yet enough appreciation for sym vs asyn testing.
Do the tests the way i did it and experiment with the polarities etc. and you will get a different picture than sine wave -- regardless of the model.
But there is no doubt the non-linearities are real and measureable in polar (electrolytics) and that is what we can also hang our hat on if you want the lowest thd in your designs..... which i do.
THx-RNMarsh
Last edited:
I know his papers and I also know your papers very well. I will NOT repeat his test. I believe that different cap production technologies may lead to some differences, reflected in distortion e.g.
I am not a fool to use tantalum capacitors for audio.
I always test the capacitor if it is suitable for the application where it is used.
But I have never been able to find harmonic distortion in a coupling electrolytic capacitor. Despite this fact, I do not use electrolytic capacitors in a signal path.
We had big troubles with DA in a series damped HV capacitive dividers, R-C-R-C- ...... topology, in HV impulse measurements. That's where I first seen DA in a real life, 35 years ago. You and audio guys are not the only one who have experienced the capacitor DA issue.
Last edited:
Dick I read this paper, and it is all about distortion mechanisms in various types of caps. The only ref to DA I found is where he says:
'The main difference I found which clearly does relate to dielectric absorption, is the magnitude of the second harmonic. This increases with applied voltage, especially so with electrolytic capacitors.'
I understand that as saying that if there is 2nd harmonic distortion, DA has an impact on the level; NOT that DA causes this 2nd harmonic. Bateman is a very clear and consice writer.
That unfortunately is no longer possible.
Jan
Last edited:
I fully agree with that. And Bateman gives an extensive explanation of the mechanisms that cause his distortion, and DA is not among them, other than the impact I quoted before.
Jan
BTW Isn't it past bedtime at your side??
yes it is.....
Check again.... Bateman .. copy of his page on DA comments. I take it that he doesnt see the descriptions listeners have given to electro with the tests he has done. That then is a fair statement. He hasnt tested it with asym waveforms which then begin to look like the comments I was hearing from listeners, then and now.
Attached are his words...... see where he says.... " The main difference I found which clearly does relate to dielectric absorption, is the magnitude of the second harmonic. This increases with applied voltage, especially so with electrolytic capacitors".
Which was my comparison. It interesting that we found same behaviour (2H) but from different tests.... mine finding from measuring the Vda. But I think the unique test I did rather than thd/fft with sine waves has benefit in finding out how it affects transient waveforms and that 'seems' to be heard as well.
View attachment DA.pdf
Ok. yes, it is past my bed time.
THx-RNMarsh
Last edited:
yes. BUT that wasnt done and still isnt done univerally.... look at most any reciever... polar caps for coupling.... with zero bias on them. I have a new Marantz I looked at because it use CFA topolgy.... polar electrolytic coupling caps...... still. And 2-4 of them in series (Bal/unbal on I/O of opamps).
THx-RNMarsh
Last edited:
- Status
- Not open for further replies.