I think I like the sound of cold valves

ecc83

"On the other hand, cymbals definitely do break in. Go to any drum shop with a decent range and compare a brand new in-the-box cymbal with its sister that's been out on display for a few years."

The work-hardening properties of copper/brass have been known for millenia. I wonder if a shop would let you put a torch on the demo crash?!!!!

Dave.

ICBM

monquixote said:

I'm thinking this might be like the perpetual drummer myth that cymbals "Warm up" when actually all that happens is by halfway through the gig they've punished their ears so much they've lost a load of top end.

It definitely isn't a myth. When the sound engineer insists that you've turned your amp up during the set when you can't have done because it was on full volume in the first place , you know the amp really has got louder.

imalone

ecc83 said:

"Does temperature affect the strength of a magnet? "

Temperature certainly DOES affect magnets! But only in a ***t or bust way AFAIK. That is at the "Curie Point" they loose all magnetic properties completely but I think this is several hundred degrees C for mild steel*, the material I think most pole pieces are made from? In any case the vast majority of guitar speakers are ferrite powered and the actual magnet is cms away from the hot voice coil, on the back, a'radiatin'!

Just SOMETIMES I wish I were rich enough to properly investigate these tales!

*Yup, well over 600C for iron.

Dave.

The thing to bear in mind is that a transformer core is not a permanent magnet (the speakers as you point out are, and getting them hot enough to demagnetize would be bad). The hysteresis is the thing you care about for the core and materials for ferrite cores may have fairly low Curie points. It's not really my area, so I don't have any good references to point out (there must be loads of data on this somewhere), a quick google turns up this which looks like there can be some variation in the 0-100C range (unfortunately might not be open access, but may be accessible on ResarchGate or somewhere):
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=05257370 "Modeling the Temperature Dependence of Hysteresis
Based on Jiles–Atherton Theory" A. Raghunathan et al.

Perhaps more useful might be this article on selecting core materials http://www.mag-inc.com/File Library/Product Literature/Strip Wound Core Literature/twc-s2.pdf which shows properties for some different core materials and also talks about the (relatively obvious need) to take into account expected operating temperature. So transformer temperature could affect efficiency depending what material is used, but it'll take a while for a core to warm up, other components probably warm much faster. On ecc83's point about low thermal mass of speaker coils, it's true they'll quickly get back to the amp ambient, but if the whole amp is warming slowly then the temperature they're at is going to be changing at that pace. Thinking about it the resistance of a big coil of wire might be one of the more temperature dependent things, something you should see more in combos than separate head+cab then?

ecc83

imalone said:

ecc83 said:

"Does temperature affect the strength of a magnet? "

Temperature certainly DOES affect magnets! But only in a ***t or bust way AFAIK. That is at the "Curie Point" they loose all magnetic properties completely but I think this is several hundred degrees C for mild steel*, the material I think most pole pieces are made from? In any case the vast majority of guitar speakers are ferrite powered and the actual magnet is cms away from the hot voice coil, on the back, a'radiatin'!

Just SOMETIMES I wish I were rich enough to properly investigate these tales!

*Yup, well over 600C for iron.

Dave.

The thing to bear in mind is that a transformer core is not a permanent magnet (the speakers as you point out are, and getting them hot enough to demagnetize would be bad). The hysteresis is the thing you care about for the core and materials for ferrite cores may have fairly low Curie points. It's not really my area, so I don't have any good references to point out (there must be loads of data on this somewhere), a quick google turns up this which looks like there can be some variation in the 0-100C range (unfortunately might not be open access, but may be accessible on ResarchGate or somewhere):
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=05257370 "Modeling the Temperature Dependence of Hysteresis
Based on Jiles–Atherton Theory" A. Raghunathan et al.

Perhaps more useful might be this article on selecting core materials http://www.mag-inc.com/File Library/Product Literature/Strip Wound Core Literature/twc-s2.pdf which shows properties for some different core materials and also talks about the (relatively obvious need) to take into account expected operating temperature. So transformer temperature could affect efficiency depending what material is used, but it'll take a while for a core to warm up, other components probably warm much faster. On ecc83's point about low thermal mass of speaker coils, it's true they'll quickly get back to the amp ambient, but if the whole amp is warming slowly then the temperature they're at is going to be changing at that pace. Thinking about it the resistance of a big coil of wire might be one of the more temperature dependent things, something you should see more in combos than separate head+cab then?

One of those links did not work and the other threw up a stern warning! Just got back from a fresh install of OS so forgive my cowardice!

Mains transformers. They are not in fact "linear" devices. The rectifier actually takes pulses of current only mSecs long to "top up" the caps (or choke if one is fitted) so, since the mains traff is effectively "disconnected" from the amp for 70-90% of the time it is hard to see it making any audible contribution, hot OR cold!

Output transformers are of course linear (fairly!) but in my experience they never get as hot as mains traffs in the same time frame, after 4 hours say? I find mains Ts might be at a skin peeling 55C but the OPT at a pleasantish 40? But of course  there is infinite variety in the construction of amps, ventilation, orientation of chassis and sizes (even for nominally same power claim) of transformers. 

Speakers in combos will barely get above ambient overall since all valve combos known to me are open backed? (and as I said. The sealed up Stage 60 did not get much warmer!) 

The voice coil must be self limiting if the speaker is power matched to the amp and so it will get to operating temperature pretty quickly and cool off between thrashes. If driven very hard, thermal compression will set in. 

In any case, if speakers were the cause of the effect it would show up with similarly rated solid state designs and the evidence, such a it is, suggests it doesn't to anything like the same degree (Boom! Boom!)

Dave.

imalone

ecc83 said:

One of those links did not work and the other threw up a stern warning! Just got back from a fresh install of OS so forgive my cowardice!

Can PM you the pdfs if you want (possibly, actually I don't know if attachments can PMed). Neither is terribly useful though. There must be some engineering textbook packed with information on this. Anyway, it looks like with some materials you might see differences by the time you get near 100C.

Mains transformers. They are not in fact "linear" devices. The rectifier actually takes pulses of current only mSecs long to "top up" the caps (or choke if one is fitted) so, since the mains traff is effectively "disconnected" from the amp for 70-90% of the time it is hard to see it making any audible contribution, hot OR cold!

Hadn't even been considering the power supply, though if the efficiency did drop you'd see the average level drop too, since it's going to take longer to get that charge across. But can imagine the efficiency is good up to the operating temperature for a power supply, since that's what it's designed for.

Output transformers are of course linear (fairly!) but in my experience they never get as hot as mains traffs in the same time frame, after 4 hours say? I find mains Ts might be at a skin peeling 55C but the OPT at a pleasantish 40? But of course  there is infinite variety in the construction of amps, ventilation, orientation of chassis and sizes (even for nominally same power claim) of transformers. 

So at 20 degrees above room temp probably no difference. ICBM's Mesa at 5000C or whatever it's reaching (personally, I'd remove the oven element...) might be heading into a region where the output transformer is going to get warm enough to be affected, depending what it's made of.

Speakers in combos will barely get above ambient overall since all valve combos known to me are open backed? (and as I said. The sealed up Stage 60 did not get much warmer!)

Point. (Aside: the Tubemeister combos seem to only have small gaps at the rear, the 36 looks completely closed with a vent at the top. But they're not the amps people have mentioned here, so not really relevant, just I've been toying with getting one. so they came to mind.)

monquixote

ICBM said:

monquixote said:

I'm thinking this might be like the perpetual drummer myth that cymbals "Warm up" when actually all that happens is by halfway through the gig they've punished their ears so much they've lost a load of top end.

It definitely isn't a myth. When the sound engineer insists that you've turned your amp up during the set when you can't have done because it was on full volume in the first place , you know the amp really has got louder.

I don't doubt what you say, but I do know that playing at high stage volumes for an hour will alter your perception of high frequencies.

ICBM

monquixote said:

I don't doubt what you say, but I do know that playing at high stage volumes for an hour will alter your perception of high frequencies.

Absolutely - but that doesn't explain why a valve amp appears to get louder in a mix relative to a solid-state amp producing essentially the same sound, or relative to other instruments (eg bass) where the highs aren't the important factor.

PolarityMan

In this instance I also noticed it as soon as i started playing. I was like..hmm this is sounding more beasting than usual. Then an hour or so later I definitely felt that it had lost a bit of the aggressive edge to the sound. 

Cirrus

I wonder if guitars change their sound when they warm up?

PolarityMan

Interesting idea, certainly affects tuning and tuning stability.

ecc83

ICBM said:

monquixote said:

I don't doubt what you say, but I do know that playing at high stage volumes for an hour will alter your perception of high frequencies.

Absolutely - but that doesn't explain why a valve amp appears to get louder in a mix relative to a solid-state amp producing essentially the same sound, or relative to other instruments (eg bass) where the highs aren't the important factor.

That valve amps beat transistors for SPL for nominally equal power is well known and there are good, technical reasons for it.

It comes as no surprise to me then that the frequency/power distribution would also be different? If you A/B two nominally identical valve amps there WILL be differences! Mainly due to the vast tolerance of pots and them not giving the same resistance for the same visual setting..I think! It would be hard therefore to even compare  "hot" and "cold" amps of the same model..EVEN! If they were 1 ser# digit apart, factory fresh and had all new valves in them!  And who among us could do that? 

Dave.

ICBM

Yes of course - but none of that explains why when you're playing a gig with a valve amp, the other guitarist in the band is using a solid-state amp, and you set them to balance nicely in the mix at the soundcheck, by the end of the gig the valve amp is noticeably louder, and neither of you touched the settings.

This is what proves to me that the effect is real and something to do with the amp changing as it heats up.

Cacofonix

I read somewhere that resistors change resistance in the cold - hence superconductors use liquid nitrogen for cooling.

So the speakers' impedance will be different in a cold room - more current passing, so more efficient and louder (?), as will the resistances in the amp itself.

I'm not sure if the movement of the speaker coil and/or use over time will alter its temperature without more, though.