some still seem to hold out for DA "linear distortion" being audible
Pease' Capacitor Soakage article gives a multiple term DA model of a 1.0 uF Mylar Cap
I put that model in a low pass with 1k Ohms ~ 160 Hz fc - not so different from a approximate RIAA EQ
then plot the AC ratio with a "perfect" Spice Cap
as I predicted the max deviation from "bad" Mylar Cap DA result is in the milli dB: 8.2 mDb @ 213 Hz, ~25 mDegrees max phase deviation, ~1/16" air propagation distance group delay deviation
(your results may vary since I had a New Holland Dragon's Milk and a Founder's Curmudgean while preparing/with dinner while working up the sim...)
Pease' Capacitor Soakage article gives a multiple term DA model of a 1.0 uF Mylar Cap
I put that model in a low pass with 1k Ohms ~ 160 Hz fc - not so different from a approximate RIAA EQ
then plot the AC ratio with a "perfect" Spice Cap
as I predicted the max deviation from "bad" Mylar Cap DA result is in the milli dB: 8.2 mDb @ 213 Hz, ~25 mDegrees max phase deviation, ~1/16" air propagation distance group delay deviation
(your results may vary since I had a New Holland Dragon's Milk and a Founder's Curmudgean while preparing/with dinner while working up the sim...)
Attachments
Last edited:
Besides high DA of 15.% for alum electrolytics, the THD is high (IM is measured to be worse)... two good reasons we suggested using films instead for audio.
View attachment THD of polar cap.pdf
-by Jung and Marsh
-THx-RNMarsh
View attachment THD of polar cap.pdf
-by Jung and Marsh
-THx-RNMarsh
Last edited:
Again SY, you just contradict Richard, providing NO datas to demonstrate your position. No measurements, no reference, zero, nada. Just your opinion. Words and noise, despite you were gently asked several time to provide some and teach people like me.
For the moment, on a pure rhetoric point of you, YOU are the one with "unsupported", "extraordinary" claim.
How do-you think that, us, unwashed people, could be convinced only by your huge authority and your fame in all the scientist world ?
You point seems obvious to you ? So, it should be easy to demonstrate !
The second thing is the inanity of this discussion. We don't build caps, we just use existing ones. So who care if non linearity of caps are issued from this or that physical aspects of various dielectrics or the name we, artificially, put on this or that phenomena.
The only thing that matters is how they behave. And the measurements Richard published 30 Years ago were very useful on this point.
Where were-you at this time, and what are-you trying exactly, insisting in such a way on this point of detail ? Just trying to bring discredit on his work ?
He had worked, spend a lot of time in "objective" measurements; What have-you done on this matter ?
Be careful with your challenges! I redid your weird 741 test some years ago.
It was in this very same thread, and I debunked your claim that there was
mysterious non-harmonic distortion. I even gave the intermodulation order by
stepping the signal source frequencies by a few Hz and watching the questionable
IMD move around by 9 times the step size or so.
And it would not have taken the spectrum analyzer and the Agilent synthesizers I used
to debunk that. Any soundcard would have served and everybody in the audience has one.
These armchair scientists here really remind me at the philosophers strike in Douglas
Adams "Hitchhikers Guide to the Galaxy", where the Deep Thought computer is constructed
to answer the LastQuestion. "We demand guaranteed areas of doubt and mystery; our
business model depends on it."
regards, Gerhard
Richard and John claim that DA causes nonlinearity. That's their claim, not mine. The data they present does not back up that claim. Re-asserting the claim doesn't make it any more true.
What is asserted without proof can be dismissed without proof.
And there is no proof, just vaguely converging irrelevances.
There is no mechanism, not even an assumed one, that
could be used to enrich the model or only to make weak predictions.
Correlation with price would also work somehow, a 0.5ct stinking
z5u can be expected to be worse both distortion- and DA-wise
than a fat Styroflex.
Last edited:
So, if you do not agree, demonstrate the contrary.
It is easy. Measure DA of different brands, measure distortions. Class the two results in any same order (let say ascending). If the two lists are in an identical order, there is a luck they is some correlation. If not, you had demonstrated the contrary, and both Richard and John Curl will change their minds.
Did-we all agree on this ?
Yes, i know, measurements are time consuming. That is what Richard and Jung had offered-us. Good education request we thank them for this gift. Correct engineering that, if any mistake, their work is criticized in an irrefutable way. Everything else just look like some kind of boring clan war.
For the moment, one party says "I believe this is blue", the other one "I believe not". Are we not turning in circles ?
Last edited:
Of course, the external signal seen when DA is present will depend on the external impedance present. This is because the DA in the cap is driving the external circuit, from a relatively high impedance. If you define DA to be what you see, rather than what is happening in the cap, then confusion is likely.john curl said:
The net result of the paper seems to be that if you put a low frequency square wave through a filter the output waveform is slightly different from the input waveform. Interestingly, it also admits that DA is "often essentially linear".
Here we go again! Why are people surprised when a filter modifies a waveform? Why do they then conclude that this must be audible and unwanted? Why do they call it 'distortion', and hope we don't ask too many questions about whether it was linear (i.e. no distortion) or nonlinear (i.e. distortion)?RNMarsh said:
Perhaps there should be a law forbidding people from buying or making sensitive test equipment unless they can first demonstrate that they have the understanding to correctly understand and interpret the results of their tests. Otherwise we will have a continual run of people frightening themselves with waveform 'distortions' which are actually just mild filtering.
My conclusion remains: DA does not cause distortion, but DA combined with asymmetric waveforms may expose existing nonlinear distortion by adding a bias thus providing the needed asymmetry to convert odd order nonlinearity to even order nonlinearity.
Why should people have to take time and effort to prove that yellow cars do not go faster just because they are yellow? The hypothesis is daft, so it is the people making it who need to prove it. Let them provide either experimental evidence or a plausible explanation that yellow paint speeds up whatever it coats. Then we can proceed by picking holes in their experimental method or their theory. A daft idea remains just a daft idea until it has some solid support; it doesn't need to be disproved.Esperado said:
Last edited:
Now this is easy. The culprit is the load. One of them must be low-grade.
Who knows ? Some of our spaceships use solar wind to accelerate...
Anyway, this "Why ?" "Because !" kindergarten war is just...boring, because it doesn't matter for us, DIYers.
https://www.youtube.com/watch?v=2nGw_vAnqPI
or
https://www.youtube.com/watch?v=chaP4MCXp4w&index=2&list=RD1NvgLkuEtkA
Last edited:
I'm not so sure about yellow paint, but that pulsating blue photon engine that some cars
have makes them much faster than ordinary Diesels or gas powered vehicles.
I can watch that daily. A drawback seems to be that this engine is quite loud.
42!
- Status
- Not open for further replies.