|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.138.53.32
In Reply to: RE: Less speculation, more facts for you posted by kurt s on June 09, 2011 at 00:17:38
since heating and cooling are not instant....shouldn't the device in question reach some kind of equalibrium state?
Too much is never enough
Follow Ups:
Figure, worst case, the tiny leads between the chip and the IC package's terminals. Play something through them and I'll bet you can almost see them moving.
I always try to use components a little bigger than necessary, to keep the inertia working for me.
-Kurt
Edits: 06/29/11 07/05/11
while cyclic application of energy will raise the temp, at some point, the temp difference becomes very small. That is during REAL signal conditions. As the energy applied becomes equal to the energy radiated or conducted over short time periods. While a very short 'burst' of applied energy will raise the temp, the rise over the ambient will be fairly small. Of course, you can overpower any resistor. I've changed out plenty which had been reduced to charcoal. Indeed, the key is the AVERAGE temp you mention. Just for the sake of numbers, let's say my house is 70f. The resistor will have an average temp in use of say....100f. During play, what will it vary? +-5f? or more or less?
Engineers always go for worst case, cost willing.
And yes, generally speaking, semiconductor cooling is a series of thermal resistances which can be usually boiled down to a single number. Don't forget that some amps simply sound better after being on for a while. My 'd' amp? Maybe 20 minutes. Some component in it likes to be warm.
But, short of immersion in LN2, what does this have to do with real world conditions? Designers like to keep stuff cool, but also compact and packagable.
Equipement generally does like better cooling, or at least lasts longer. We're all at the mercy of room ambient, listening habits and setup. The only group which has successfully fought thermal noise is the SETI folks who take extroidinary measures to cool equipment and run in a low noise (electronic) enviroment. I'd love a Faraday cage around my house which is fed by an isloation transformer this size of a bus.
Too much is never enough
Thermal time constants can only be ignored for far out-of-band signals like 'DC'. Since the discussion is about potential distortion products they are in play because they can cause IM.
Another area that can cause serious problems, especially for power resistors, is the termination where the metal with the deliberately 'high' resistance (and hence running hot and cold) transitions to the leads or tabs. I have some large resistors that I use for working on power supplies that have become virtually unusable from this problem.
Regards, Rick
I know stuff wears out. Everything.
But, will a resistor run within power rating given a 'conservative' design, be subject to early fatigue death? I've seen boards where the place under a 10 watt carbon has been heat charred...but this was in gear run 24/7. (semiconductor fabrication equipment)
Were the dead resistors you refer to always run within design limits or a little edgier?
And IM distortion? Short of LN2 immersion, what can be done? Every circuit element has resistance and therefore heats. I don't know how much a warm inductor changes, but certainly semiconductors...discrete power devices or even low level integration of OPAMPS or even hi level integration.....
You're saying resistors change, too. All well and good, but how much and to what real effect that you can address?
These kinds of effects make sense if you're looking for ETs and have a radio dish listening stuff millions of light years away and SNR is beyond critical.....
Aren't all these effects pretty much the 'cost of doing business' with electronics?
I guess my question would be....
What does it do to my listening experience that I can actually do something about?
Too much is never enough
-Kurt
Edits: 07/05/11
"Were the dead resistors you refer to always run within design limits or a little edgier?"
Well I was running them within limits, but they were old and were given to me to boot so who knows? I bet part of it was corrosion.
"And IM distortion? Short of LN2 immersion, what can be done? Every circuit element has resistance and therefore heats."
There are a lot of things that can be done to reduce these sort of effects, you mention op-amps, thermal feedback is a major problem with monolithic linears especially ones with significant output power so as you are probably well aware great pains are taken to use them differentially and to make the die symmetric so that the heat from the output stage couples equally to both sides canceling out the thermal terms. Mostly.
For resistors you reduce their dissipation, reduce their thermal resistance to air or increase their thermal mass to try and even out their highs and lows. The average power TC may cause a little compression or expansion but I'd guess it's not as audible as bass altering your high's levels when someone hits a drum. All that being said I'm not aware of ever having audible problems from the TC of resistors.
As far as sensitivity to these sorts of things go I think audio gear is right up there, our hearing can really eke out deep details.
Regards Rick
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: