Why do some amps sound better when fully warmed up?

I'm sure I'll get technical answers that I won't fully understand, but I'm interested anyway so.....

All my valve amps have sounded better once fully warmed up (i.e. maybe up to 2 hours or so after switching on), and my current Copland valve integrated more noticeably so than others I've had. My Sugden A21a was also the same, so it's not just a valve thing. Previous amps like MF B200 and Arcam Alpha 2 not so.

So, why some better when warm?

 

A guess is that:
- When warmed, transisters, capacitors and similar temporary change their crystaline structure, and thus how they conduct electricity. Picture it as being the opposite of supercooling a conductor to create a superconducting material.
- Whilst warming infers that a conductor will not operate quite as effeciently, it's probably more important to the audio designer that it acts in a predictable and consistent manner. So I'm guessing that they voice kit at operating temperatures and not from cold.

 

I remember someone asking HFC (the mag not the forum) the same question and the answer was that it's because that's how they are designed.

As you say, if they voiced it to sound best when cold, at higher temps it can only sound worse.

 

Its a steady state thing.....when all the components have stopped charging, warming up, expanding etc less enery is lost to the system attempting to get to its steady state, I think the scientific term is equilibrium, and it means more than just in balance as the name suggests. Its a bit like a car they always run better once warmed up .

 

My Son of Heffa tube amp takes about 30 minutes to really play sweetly. It's not bad before that but it really sings once warm.

I guess the solid state stuff I have (CDP) takes the same.
I've never noticed the speakers needing a warm up - but then how would I?

 

Conductors change their resistance relative to increase or decrease in temperature. This change is described by the temperature coefficient of the conductor relative to measured resistance usually at 20 or 25 degrees C. The same also applies to the measured capacitance of a capacitor. Not necessarily a full answer but part way there.

(now wait for someone to post a tweak about loudspeaker cables and ice cubes / hairdryers )

 

Fair enough, that all makes sense to me, thanks for the replies!

I'll go and take the electric blanket off my speakers now, they've had about 45 mins, should be just about ready...