fdegrove said:
Nope, since not even you have yet responded to my last
post in that thread, I have found no reason to reconsider
my opinion for the moment. I may do so if somebody gives
a good argument against what I suggested.
Cheers
john curl said:
That is a good step forward..thanks
john curl said:
I'm surprised the aluminums didn't die sooner..I recall some kind of wearout mechanism related to chemistry. Were they all supply related, or were some in the signal path?
john curl said:
Perhaps the same mechanism you have to receive the airborne garbage is the same mechanism giving you all the harmonics, the ones Bruno does not get. It may be the physical design of the AP, with the I/O and cable layout.
It may also be that your rig is setting up airborne harmonics, and the various cable shield/resistivity combinations are the real culprit behind the cable dependence of your readouts. Have you modified the cable dress and re-tested? For example, re-dress a test cable to alter it's loop profile?
Ah, the list of things to test is long..Hope you really desire getting to the bottom of it..
Cheers, John
Re: APS2
What the hell's it doing at your feet!?!? What, did it fall off the test bench? 😀
Thank you for the kind offer.
Perhaps John would like to see Bruno's measurements repeated by someone else. If so, I could ask Bruno if he could send you the cables I'll be sending him on to you when he's finished.
How 'bout it, John? Is there something you'd like CGL to measure?
se
-CGL- said:
What the hell's it doing at your feet!?!? What, did it fall off the test bench? 😀
Thank you for the kind offer.
Perhaps John would like to see Bruno's measurements repeated by someone else. If so, I could ask Bruno if he could send you the cables I'll be sending him on to you when he's finished.
How 'bout it, John? Is there something you'd like CGL to measure?
se
Re: APS2
Hi,
Well CGL, I don’t think you need actually Bruno’s cables. Just measure proper cables, with solid copper core and foamed PE insulation for instance like the cheap CATV cable I used. And on the other hand measure a really cheap cable with PVC insulation. PVC has a relative high loss factor. If dielectric hysteresis shows up it will probably do on PVC. But to show up you need to drive the cable with a high impedance I think to avoid shorting the parallel cap formed by the dielectric of the cable.
Cheers 😉
-CGL- said:
Steve Eddy said:
Hi,
Well CGL, I don’t think you need actually Bruno’s cables. Just measure proper cables, with solid copper core and foamed PE insulation for instance like the cheap CATV cable I used. And on the other hand measure a really cheap cable with PVC insulation. PVC has a relative high loss factor. If dielectric hysteresis shows up it will probably do on PVC. But to show up you need to drive the cable with a high impedance I think to avoid shorting the parallel cap formed by the dielectric of the cable.
Cheers 😉
Re: Re: APS2
I would recommend the same cables seeing as Bruno's measurements had already been dismissed because he wasn't measuring the same cables as John.
DA would show up in the time domain. A group delay measurement would be best here.
se
Pjotr said:
I would recommend the same cables seeing as Bruno's measurements had already been dismissed because he wasn't measuring the same cables as John.
DA would show up in the time domain. A group delay measurement would be best here.
se
Re: Re: Re: APS2
Nope, if distortion shows up it will do with other cables too. It is only dismissed by John and not by others as far as I remember this lengthy thread. And after all not many people are USING Bruno’s cables in their set-up.
Mentioned “dielectric hysteresis” and not particular DA. These are not necessarily the same.
Cheers
Steve Eddy said:
Nope, if distortion shows up it will do with other cables too. It is only dismissed by John and not by others as far as I remember this lengthy thread. And after all not many people are USING Bruno’s cables in their set-up.
Mentioned “dielectric hysteresis” and not particular DA. These are not necessarily the same.
Cheers
CGL, yes you could try it yourself.
This is the test:
You get some audio interconnects, they can be any length.
If they are shielded, so much the better. If they all have RCA jacks, that makes it easier to change them around, but BNC to BNC connectors will do OK as well. You can experiment with level, but 30 mV AC, 1-5KHz should be good for you. I have found that I get more effect, in general with a load on the cable. At present, I am using a 2K load, with a 600 ohm drive source. When you first make your linear measurement, without any digital processing, you will always see noise at the scope output. This is because ANY residual distortion is below the noise, both equipment or DUT. When you do a spectral analysis with the digital part of the system, then the noise will be highly reduced by the effective bandwidth of the spectrum analysis. In my case it is machine set at about 100Hz. Yours will probably be narrower.
I resort to signal averaging to get my final results (100) but you will probably need much less if any.
I look for changes in harmonic generation from the 5th to the 9th harmonic.
Of course, if nothing is there, no matter how deep you go by signal averaging, then that cable is free of this effect. It is best to try different cables, even cables in your lab, to see if you can measure anything.
The cost or external appearance of the interconnect does not usually give an indication of whether it will measure higher order distortion. Some very expensive and well made cables may have a great deal, others none at all. My reference cables are now a copper VDH video cable that someone made up for me to try, and the JPS cable which is essentially copper-aluminum tubing made for CATV. They measure virtually a flat line on my system residual. Also, very heavily used cables tend to measure very low distortion, all else being equal. Give it a shot.
This is the test:
You get some audio interconnects, they can be any length.
If they are shielded, so much the better. If they all have RCA jacks, that makes it easier to change them around, but BNC to BNC connectors will do OK as well. You can experiment with level, but 30 mV AC, 1-5KHz should be good for you. I have found that I get more effect, in general with a load on the cable. At present, I am using a 2K load, with a 600 ohm drive source. When you first make your linear measurement, without any digital processing, you will always see noise at the scope output. This is because ANY residual distortion is below the noise, both equipment or DUT. When you do a spectral analysis with the digital part of the system, then the noise will be highly reduced by the effective bandwidth of the spectrum analysis. In my case it is machine set at about 100Hz. Yours will probably be narrower.
I resort to signal averaging to get my final results (100) but you will probably need much less if any.
I look for changes in harmonic generation from the 5th to the 9th harmonic.
Of course, if nothing is there, no matter how deep you go by signal averaging, then that cable is free of this effect. It is best to try different cables, even cables in your lab, to see if you can measure anything.
The cost or external appearance of the interconnect does not usually give an indication of whether it will measure higher order distortion. Some very expensive and well made cables may have a great deal, others none at all. My reference cables are now a copper VDH video cable that someone made up for me to try, and the JPS cable which is essentially copper-aluminum tubing made for CATV. They measure virtually a flat line on my system residual. Also, very heavily used cables tend to measure very low distortion, all else being equal. Give it a shot.
Re: Re: Re: Re: APS2
They weren't dismissed here, but on another forum.
According to the glossary at FaradNet they are (DA, DH and soakage are synonmymous). And I've yet to see any evidence to the contrary.
I tried to get an answer out of Frank but to no avail, so perhaps you could explain what the difference is between DA and dielectric hysteresis?
se
Pjotr said:
They weren't dismissed here, but on another forum.
According to the glossary at FaradNet they are (DA, DH and soakage are synonmymous). And I've yet to see any evidence to the contrary.
I tried to get an answer out of Frank but to no avail, so perhaps you could explain what the difference is between DA and dielectric hysteresis?
se
Dielectric hysteresis certainly does exist in insulating materials. Cheap and popular ceramic capacitors have plenty. To accurately measure it is difficult, however. I first saw it in 1974 with a modification to a TEK 577 curve tracer, that can measure this directly. TEK invented the modification, but they did not release it as a product accessory. When you see a cap not return to its starting point, you gain understanding of this. Now, how much is in a cable? Who knows, I was measuring maybe 10% deviation with ceramic caps, and maybe cheap, high DA cable material has some too. I would not be surprised. This is the main reason that I don't recommend ceramic caps for audio applications.
Hi,
Some people like their privacy the way it is.
While I've been a personal friend for many years I also come to respect his need for this privacy and am not inclined to pass this on without his go ahead.
If you want to get in touch, try contacting the university where he's appointed as a professor.
As far as I know he's still active in the "Materials Research" department.
I did give you a pointer to another site, didn't I. If you can't tell the difference I can only repeat what I said about it before and others seem to confirm it.
Hi John,
Absolutely.
This is what I've tried to pass on all along. The measurements were done about ten years ago and to the best of my knowledge at Aalt Jouk van den Huls' request.
I doubt he made them himself at the company's lab as I doubt that at point that company had the means to do such tests.
As I have explained to Steve Eddy and the rest of the audience these test were also repeated with all kinds of wires and conclusions were drawn. As the test results are intelectual property of my former company and others I fail to see why I should publicise details here or anywhere else but cables can and do distort.
Some are noisier than others and some, even very simple ones are downright fabulously quiet and show exemplary linearity.
One thing was obvious from the test and I don't mind passing this on, solid core and Litz configurations didn't show any self-noise and were impervious to microphony tests unlike most multistrand wire.
Cheers,😉
Some people like their privacy the way it is.
While I've been a personal friend for many years I also come to respect his need for this privacy and am not inclined to pass this on without his go ahead.
If you want to get in touch, try contacting the university where he's appointed as a professor.
As far as I know he's still active in the "Materials Research" department.
I did give you a pointer to another site, didn't I. If you can't tell the difference I can only repeat what I said about it before and others seem to confirm it.
Hi John,
Absolutely.
This is what I've tried to pass on all along. The measurements were done about ten years ago and to the best of my knowledge at Aalt Jouk van den Huls' request.
I doubt he made them himself at the company's lab as I doubt that at point that company had the means to do such tests.
As I have explained to Steve Eddy and the rest of the audience these test were also repeated with all kinds of wires and conclusions were drawn. As the test results are intelectual property of my former company and others I fail to see why I should publicise details here or anywhere else but cables can and do distort.
Some are noisier than others and some, even very simple ones are downright fabulously quiet and show exemplary linearity.
One thing was obvious from the test and I don't mind passing this on, solid core and Litz configurations didn't show any self-noise and were impervious to microphony tests unlike most multistrand wire.
Cheers,😉
fdegrove said:
Then his first name is hardly any big secret now is it?
What, that site that didn't have anything on it differentiating DA from DH?
And what's this "other seem to confirm it" stuff? Confirm what? Confirm that some are of the notion that DA and DH are different things? So what? So far, while I've seen a number of people say they think DA and DH are two separate things, I have yet to see a single explanation as to what these differences are.
If the two are not one and the same, why is it that no one can seem to explain what the differences are?
Hmmm. The way Curl tells it (or rather the way he tells what he says van den Hul told him), van den Hul did the cable distortion measurements on his own some 20 years ago using some surplus test equipment from Philips.
se
fdegrove said:
Frank, John (C),
I am reminded of an experience I had many years ago with an amp I built. The two channels distorted differently (one was OK, the other higher than expected). Also, the distorting channel had for some reason about 6dB less gain on music (estimated lower level) while measuring exact correct gain on the bench. This came out while testing at a friend of mine (Peter van Willenswaard, the wizard behind Audio Magic in Rotterdam, the Netherlands, for the cognoscenti).
We finally tracked it down to the 6 inch shielded wire between the input RCA and the amp board: the one in the bad channel was not only distorting, but had a very low resistance, and this showed up in listening tests because the tube preamp we used had a rather large output impedance. I don't remember the input cable brand, but it was a well-reputed one which was given to me to try. We replaced it and that solved it. I now regret that I didn't take the time to get to the bottom of it. But yes, cables can distort, although I still think in this case it was a defective lot.
Jan Didden
To John Curl:
By the way, is there any specific reason to do these tests at 30mV signal? I may have missed the reason in the many pages of posts above, and if so I apologise for asking the obvious. But wouldn't the test have better resolution related to the noise floor at a higher test level? Or is the distortion we are looking for expected to be constant in amplitude rather than constant in %?
Jan Didden
By the way, is there any specific reason to do these tests at 30mV signal? I may have missed the reason in the many pages of posts above, and if so I apologise for asking the obvious. But wouldn't the test have better resolution related to the noise floor at a higher test level? Or is the distortion we are looking for expected to be constant in amplitude rather than constant in %?
Jan Didden
I'll skip the quotes.
DA and DH:
1) I commented on the paper Frank linked to (unless he has
linked to other papers too that I have missed) and said that
as far as I could see, but I could very well be wrong, what was
claimed as DH in that paper was just one of the several different
dipole mechanisms contributing to DA according to Kunderts
very detailed paper. That is, DH as defined in "Frank's paper"
seems not to be a separate phenomenon. That must have been
at least a week ago and I still haven't seen any response from
Frank about this. Maybe he is still busy reading Kundert? 😉
2) Johns measurements on ceramics caps seem to give the
expected results, but do they indicate a separete DH
phenomenon? Don't think so. Ceramic caps are inherently
non-linear, that's elementary text book knowledge. Throw in
some DA to get hysteresis and there is nothing strange with
the measurements that requires yet another concept fo explain
them, as far as I can see (but I've only had two cups of coffee
yet).
Cable measurements:
1) I am happy to see that John C. seems to have positive attitude
now towards joining forces with some other people having
access to appropriate measuring equipment. That's the way to
do it. Try to see if you can get som reasonably consistent
repeatability of experiments, and then you might be onto
something. Might even be scientifically groundbreaking if
you succeed. But do it soon John, the Nobel committee usually
don't hand out awards until some 20 to 30 years after the
research is done. 🙂
2) It is interesting to hear that similar results seems to/might
have been established by others, as Frank indicates.
Unfortunately, it is of little use to us since it is commercially
classified and we have no access to the results or methods
used.
DA and DH:
1) I commented on the paper Frank linked to (unless he has
linked to other papers too that I have missed) and said that
as far as I could see, but I could very well be wrong, what was
claimed as DH in that paper was just one of the several different
dipole mechanisms contributing to DA according to Kunderts
very detailed paper. That is, DH as defined in "Frank's paper"
seems not to be a separate phenomenon. That must have been
at least a week ago and I still haven't seen any response from
Frank about this. Maybe he is still busy reading Kundert? 😉
2) Johns measurements on ceramics caps seem to give the
expected results, but do they indicate a separete DH
phenomenon? Don't think so. Ceramic caps are inherently
non-linear, that's elementary text book knowledge. Throw in
some DA to get hysteresis and there is nothing strange with
the measurements that requires yet another concept fo explain
them, as far as I can see (but I've only had two cups of coffee
yet).
Cable measurements:
1) I am happy to see that John C. seems to have positive attitude
now towards joining forces with some other people having
access to appropriate measuring equipment. That's the way to
do it. Try to see if you can get som reasonably consistent
repeatability of experiments, and then you might be onto
something. Might even be scientifically groundbreaking if
you succeed. But do it soon John, the Nobel committee usually
don't hand out awards until some 20 to 30 years after the
research is done. 🙂
2) It is interesting to hear that similar results seems to/might
have been established by others, as Frank indicates.
Unfortunately, it is of little use to us since it is commercially
classified and we have no access to the results or methods
used.
Hmm, I think the DA/DH debate goes rounds and rounds. Hysteresis by itself is a non linear function with something what you can call a “snap” action. Like the B-H loop of magnetic materials. Just a time lag function is not what I consider hysteresis. Although it is most time the case, hysteresis by itself does not need necessarily be time dependant.
I do not say it will exists in cables. It was just a suggestion to explain a possible cause of the harmonics as seen in John’s measurements due to the nature of the harmonic profile.
Cheers 😉
I do not say it will exists in cables. It was just a suggestion to explain a possible cause of the harmonics as seen in John’s measurements due to the nature of the harmonic profile.
Cheers 😉
DA AND DH.
Hi,
Maybe not enough for some....
Dielectric hysteresis decreases as the loss angle of the dielectric increases.
Hopefully that's sufficiently clear too see that DA and DH are not one and the same .
Cheers,😉
Hi,
Maybe not enough for some....
Dielectric hysteresis decreases as the loss angle of the dielectric increases.
Hopefully that's sufficiently clear too see that DA and DH are not one and the same .
Cheers,😉
Re: DA AND DH.
I assume you are referring to this document you linked to
earlier in the thread,
http://www.tpub.com/neets/book2/3a.htm
since looking through the thread again I couldn't find any
other relevant documents you have referred to. First, this
paper never mentions DA at all, so the paper by itself does
not make it possible to tell if DA and DH are different phenomena.
Second, I could not find any claim like the one you made above,
but maybe I missed it despite rereading it twice?
fdegrove said:
I assume you are referring to this document you linked to
earlier in the thread,
http://www.tpub.com/neets/book2/3a.htm
since looking through the thread again I couldn't find any
other relevant documents you have referred to. First, this
paper never mentions DA at all, so the paper by itself does
not make it possible to tell if DA and DH are different phenomena.
Second, I could not find any claim like the one you made above,
but maybe I missed it despite rereading it twice?
Hi,
That's because that paper talks about dielectric contants which refers to the insulators DA.
Wrong assumption.
If you do a Google search on "hysteresis in dielectric" it will unearth alot more than just one paper mentioning it.
Cheers,😉
That's because that paper talks about dielectric contants which refers to the insulators DA.
Wrong assumption.
If you do a Google search on "hysteresis in dielectric" it will unearth alot more than just one paper mentioning it.
Cheers,😉
fdegrove said:
Obviously. You referred to that document earlier when I or
someone else asked about the difference, so I assumed that
was still the one referred to.
Yes, and it reveals more than Altavista, which only gave me
two references on Geophysics. Google also gave me references
to the petroleum industry and two very specialized physical
papers on dielectric properties of nitrogenium molecules in
its solid phase and of Copper-formate Tetrahydrate. Ah, yes,
there were also two papers on the dielectric properties of
porous materials exposed to wetting and non-wetting liquids,
respectively, but these seemed only to refer to various types
of earth and sandstone and were for the purpose of finding oil by
ground-penetrating radar. Maybe it could be relevant for
paper-in-oil capacitors??? 😉
Isn't the internet fantastic sometimes? 🙂
- Status
- Not open for further replies.