solid state amp warm-up time, myth or fact?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Disabled Account
Joined 2003
well, as long as a device's "peformance", be it electric or physical, is tied to its temperature, warming up will have an impact on its performance.

The question is, for a modern (analog) amp with a lot of feedback, is the performance change audiable? Take Nelson's Class A amp for example. It has considerably less feedback (low gain as well) so it is natural to expect that one is more likely to hear the difference (not that one can). That's assuming the the device performs differently at different frequency.

For a class B amp, or a digital amp, I find it hard to believe that one can hear a 5% drop in beta, etc.
 
Need for heat sinks.........

Are we to conclude then, that monster heatsinks are bad for sound then because they stop the amp from warming up very much????

Heat sinks of any size are used to keep the operating temperature as low as possible. This is especially critical for (most) class A amps that dissipate lots of heat under normal conditions. Large heatsinks are used also in high power ss amps. The purpose is to keep the temperature down. Apart from avoiding any thermal runaway , it also increases the life of the components. The hotter it gets the shorter it life.

So reaching a stable operating point is not determined by how hot it gets but by achieving a stable operating current / voltage in the circuit. With a large heatsink this could occur at a lower operating temperature !
Cheers.
 
Hmm-Nelson, do I understand you correct that the audio signal is in fact "disturbing" the optimal setting of an amplifier? It implicates that this coping with the audio signal "disturbance" is a quality parameter of a design. I learned that the insensitivity of a design to changes in component values,current settings, temperatures etc. made a good design. We have to broaden that perspective to insensitivity towards the inputsignal. Hmm- interesting!:scratch:

Ward
 
For a given amp, the input signal will have a direct impact on what percentage of the bias is dissipated as heat and what percentage of it gets to the speakers. So, yes the input signal has an effect on the thermal equilibrium of the amplifer, but over any appreciable period of time, it would probably average out quite nicely.

Most components (both active and passive) do change as they warm up and capacitors need time to 'form' when they are first used.
 
planet10 said:


This has been coined memory distortion. A French fellow (name something like Lavardian) has done some research and there are some amps whose claim to fame is to pay special attention to this distortion.

dave

Yes, I read a little bit about that. Actually, my comment was
meant rather as a general one, not specifically regarding
thermal effects, since someone seemed so surprised that the
actual signal could affect the performance.
 
Is there anyone having experience with spectrum analysis-, distortion- or stepresponse measurements with amplifiers just after switching on and say an hour later? Any differences that might explain a this "better" sound?

Strangely enough it seems that the the sound always "improves" I never heard that it became worse.

It is not just the amplifier floating to it correct setting causing less perfect sound during the process. All settings should be more or less on the expected levels directly after switching on (1 minute). If these small differences in settings in the first hour after switching on cause that audible difference I believe a redesign might be appropriate.

Or are we talking about an audio "decanting" ritual that must be performed to appreciate the full audio bouquet?:D

Ward
 
I don't claim to know the reason(s) for the improvement.
However, although most components are likely to heat up
quickly from their quiescent power dissipation (componenets
mounted to heatsinks will heat up slower, of course), don't
forget that components also affect each other thermally.
Nearby components may heat each other through radiation, but
probably more importantly, the temperature of the air inside
the case will increase and this is likely to cause a much slower
additional heating of all components.

I think it is common that various lab measurement equipment
must be powered on say 30 min before being used to stabilize,
and I would suppose the manufacturers have a reason for
prescribing that procedure.
 
Interesting topic...

I have built several P3A (A/B) amps and without exception they sound at their best when the bias was set to maintain an output tranny heatsink temperature of approximately 50 deg centigrade. This was not a quick judgement, as I tried many, many bias settings over a period of months to arrive at what provided the best sound.

Although the answer remains elusive, I did discover polyester film capacitor (what I was using for the input decoupling cap) dissipation is at its lowest around 50deg, actually two to three times lower than at 25 deg, depending on the manufacturer. >50 and the dissipation begins to rise. This is a characteristic of polyester only.

So I suspect this, along with transistor gain vs temperature changes are responsible for at least some of the change in sound quality.

Cheers,

M
 
Re EDUM's last comment:

It does, indeed seem logical that if there can be an audible improvement in performance after an hour of power up, there could just as well be a degredation. This is a simple application of the Murphy phenomonon.

This is somewhat analogous to a discussion I read once about "break in" of loudspeakers. After a lot of the usual comments, a poster who claimed a prior association with a speaker manufacturer wrote that "reverse break in" was not that uncommon. After all, you want you product to sound it's best when new so as to favorable impress the magazine reviewers.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.