|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
192.94.94.105
In Reply to: Sorry, no bangs on reversing 'lytic cap to DC. (long) posted by cheap-Jack on February 26, 2007 at 12:44:05:
If your battery voltage dropped, you simply didn't have enough energy to cause the cap to fail in a dramatic way. You need a lower source impedance. Try a car battery and a backwards tantalum cap. Wear oven mits :)I've personally had the pleasure of having molten tantalum melt it's way through my shirt; strangely, somtimes the cap doesn't explode and catch fire, but rather shoots moltem metal at you. Also have had the pleasure of containg a fire in a burn-in oven caused by backwards Ta caps. Not to mention many PCB's with a lump of charcoal with leads on it that smell bad. Trust me, they will catch fire if there is enough energy available to heat them up.
Oh, BTW, some modern Ta caps have a current limiting fuse in series so that they just open up instead of blowing up. Darn UL ruins all the fun...
My experiences with Aluminum caps aren't as dramatic. Put 450-500V across a 300V snap-in electrolytic and it gets hot, then pops the seam on the top, spewing hot goo. I've heard that they too can explode but haven't had this happen to me.
I think many Al caps will exist just fine with reverse voltage to a point. Past some voltage they will start leaking current and heating up. Also I believe that the dielectric breaks down over time, so even though it may be happy for a few hours, eventually the dielectric will break down (dissolve) and the cap will short.
Follow Ups:
Hi.Never I want to give an impression here hooking up a 'lytic cap with reversed polarity is a matter of course. Whoever would get such impression is an "idiot" indeed.
As I stated clearly to Tre' John Curl presented this cross-parallel tant caps for interstage coupling idea in his IEEE paper as "Omitted factors" draws my attention to this god-forbidden so called 'idiotic' doing. But he stated the condition & substaniated with measured data.
I believe the effort he made deserves an "idiot" like me to pursue.
Again, why you guys shot up the wrong tree? The whole issue is not, repeat not, about powering up a 'lytic cap with polarity reversd.
Even an "idiot" knows this can cause trouble, like sticking one's finger into the AC wall socket. Is this so so obvious? You guys all stirred up such a flame so needlessly.So far, only Russ & Tre come up with something sensibly up to the point. Why all these irrelevant stirrup around here.
So my conclusion is you guys are not better than I am, in coming with some thoughful thinking on how to make it works as John Curl suggested it in his papers decades ago.
Enough said.
c-J
Why you elected to inquire per a low-voltage SS application @ a specialized high-voltage vacuum tube forum is beyond me. You got replys accordingly. If you have prior build experience & an EE degree, self-designing a new pcb that could accomodate being populated with modern parts should be a breeze as well as solve your conundrum. Curls' paper is decades old, surely if there were merit someone wiser than you would have already exploited it.
Hi.What I am working on is only a SS phonostage upgrade. I don't want to rebuild the whole thing even if I had the time & patience. This is not a technical competence issue, bud.
I hope you are not suggesting readers here can't handle SS 'cause they only know tubes.
Apparently all these irrelevant stirrup is only a reading comprehension issue.
the forum was divided into numerous sections for a reason! :)
Hi.But there are always gray areas hard to split hairs somehow. 'Cause
speaker cables, speakers constitute a load for an audio amp, tube or SS regardless.That's why I just wanted to focus on 'lytic caps for coupling use per John Curl's paper. It is so obvious that only SS amp use 'lytic caps for coupling as a matter of course.
It is up to the moderators to decide to retain or delete those sorta
off forum topics.c-J
Jack, the intent of your posts isn't as clear as it could be. In the case of this ying-yang DC blocking cap I can't tell if you are:1) Asking if you think it is a good way to reduce a blocking caps inherent distortion
2) Are trying to tell us we are doing blocking caps wrong (or at least spending too much on them) and John Curl has proven it
3) Trying to prove us wrong when we stated before that such a connection would go bang if it saw DC voltage
4) I could go on
So if you took a little more time and used a few more words and made it clear just what answers you were looking for you would get relevant answers. Some of your posts you try to "lead" and people don't want to guess.
You have heard my thoughts, namely that if I needed DC protection than I want it to be functional. If it was needed, say a DC coupled solid state amp, it could be okay to replace with wire. An example might be if the preamp had an output transformer and was thereby incapable of passing DC. Another is if it if the preamp output is cap coupled and you can insure the preamp goes on first and off last.
If instead the real question is the best cheap way to have a huge output coupling cap, in something like a CD player or active crossover, where room is a problem and uF must be that large.....well then I would tell you today's cap performance is so much better that I'd bet the extra cap with its extra lead length and physical size might make matter worse and isn't likely to make matters better. In other words go get a bi-polar black gate and be done. They are only a few bucks in those sizes/voltages and are way better than ying yang tantalum. Don't take my word ask John Curl what he would use now.
You also seemed to miss or discount the important work in the paper. But maybe you felt the little devoted to caps was the only thing of importance?
Hi.As I posted repeatedly, a practical cap is not a pure capacitance device. It is built up of a complex LCR network: its DCR, self indutance, capacitance C, dielectric loss R & insulation loss Rin.
Unless today's caps are built of some space-shuttle techologies I am not aware of, its LCR equivalence should be the same. Maybe modern designs will improve the magnitude of the individual L, C or R elements, to deliver better performance, but the basics remain same.
Correct me if I am incorrect.
Hi.Tre' & cororosin seem to understand more what I wanted to say.
Thanks for their invaluable sensible inputs.(1) Yes, it is a good way to reduce harmonic distortion of ANY caps,
be them polarized or non-polarized, by paralleling them. John
Curl has proven by it by ying-yang parallelling a pair of tant.
caps in his paper.I have repeatedly posted, even yesterday, the technical reasons
why it is a win win situation to do so due to the LCR equivalent
network structure of any caps.(2) No, it is not wrong, but it can be done better by parallelling the caps instead of using only one "blocking" cap for coupling use. This is one of the "Omitted factors" he wanted to raise in his paper, IMO. At least this is how I understand it.
It's is not a money issue. It is an important distortion reduction issue that so many many tend to "omit". He only brought out 'lytic cap as a typical case for SS or zero potential situation, IMO.
(3) This is a reading comprehension issue.
I was extremely skeptical with John Curl' theory in ying-yang parallelling tant. cap for coupling use although he stated clearly in his paper his measurement condition: zero potential across the caps terminals.
IMO, this seldom exists normally for active stage coupling. But as I posted repeatedly even in most most I/P coupling situation, the driving end is electriclally open circuit & normally zero potential (except accidents like DC inrush as you suggested), & the the other end of the cap connected to very very low DC potential, some decimal DC volts for a bi-polasr base or a tube grid.
So the danger of bang up a coupling polarized cap would be not as serious as applying a HV DIRECTLY to a reversed 'lytic caps as most of you guys jumped on without first reading carefully what I was trying to tell.
Hence I tried out my experiment using 15V hooked up DIRECTLY to the
pair of 4.75uF25V tant caps to see what would happen. This surely simulated a much much worse situaiton than noraml I/P coupling.There was no harm done after my 5 minutes hooked up this reversed way. This would lead me to believe for I/P coupling only, it should work given normally zero potential from the driving end & decimal DC volt at the other end of the cap.
The whole exercise is to try to improve the inherent harmonic distortion of a god-damned 'lytic caps in the signal path without need to resorting for costly & BULKY film caps which can not fit on the original tightly fitted PCB.
Wrong, John' paper is mainly on phono cartridge subsonic & supersonic distortion beyong the conventional 20-20KHz spectrum. I had some lengthy discussion with him on this cartridge issue in PHP as I am a die-hard LP lover.
I don't want to touch phono cartridge in a tube DIY forum here for obvious reason you should know. Not that I wanted to "discount" his good research on subsonic & supersonic distortion on phono cartridges which were commonly "omitted factors" back then.
I repeat, it is NOT a money issue. It is the issue of how to improve the inherent harmonic distortion of a 'lytic cap given sticky situation where better caps can't phyiscally fit it.
Hopefully I am more understood by now. Thanks for letting me clarify it.
c-J
This post is made possible by the generous support of people like you and our sponsors: