Very Hot Amp Sounds and 'Feels' Better ?

[HaroldBrooks]

I realize this is entirely subjective evaluation, but here it goes.

I've had occasion to run several tube amps over the years under extremely hot and poor operation conditions, as in out in the direct sun on a hot summers day, with high A/C voltages, and just plain biased too hot. Just about everything wrong including speakers that got really hot after playing like this, and I have to say a couple of those amps sound and "Felt" Fantastically smooth and compressed in a manner unachievable by an outboard compressor, but I always feared the amp would end up in smoke so shut down early to avoid a possible catastrophe.

What I notice when this happens is that the volume drops off noticeably over time, and the amp becomes much more compressed, and a bit less dynamic, but in an odd way that adds to the clarity of very high gain lead solo notes and sounds smoother and very liquid.

At first I attributed this all to my ears becoming numb or just some fancy regarding my playing, but I noticed that when I ran the same amps in another room and closed the door (to save my ears !) I got the same thing from widely different amps, albeit all being stressed temperature wise. As I stated, this happened on a wide variety of amps, and here is my short list from memory: Ampeg V-4, Marshall Plexi, Old Supro combo with a field coil speaker, National 1224, Gretsch 6156.

All of those amps have widely different circuits, but when run exceedingly hot, have the same relative changes to tone and response (compression). One thing is for sure, this NEVER happens at a lower volume, or in a very cool setting, or with modeling or a stomp box pretending to be a tube amp. This I believe is a too hot tube amp thing.

All of those amps have speakers that are low watt and just passable, so the magnet structure heats up a fair amount over time.

Question : Is it primarily the speakers, the power transformers, resistors, Tubes, or Output transformers ? A combination, or something else I am missing ? It might also be my imagination, but when it happens I try to think as rationally as I can to ward off any irrational exuberance.

Thanks for any input !

[J M Fahey]

There is some truth to that.

* hot biased power tubes are easy to drive (hint: peak signal needed is 0V, so lower bias voltage/higher bias current needs less actual voltage swing)

* tubes are non linear at lower currents, crossover distortion being a direct result of that.
Higher bias sends you into the more linear area.
Highest bias posible is, of course, Class A.

* power tubes compress when overdriven, because of grid rectification.
Lower bias voltage again makes that easier.

* overheated transformers have much increased DC resistance (copper is a PTC resistor) so increased voltage drop under current peaks.
Screens drop too, lower gain and contribute to compression.

Now add all the above mentioned effects .

[loudthud]

Question for Juan, how much (by percentage) could the resistance of a speaker voice coil go up under conditions the OP describes without permanent damage ? Is 25% possible ?

[pdf64]

Copper’s resistance temperature coefficient looks to be 0.393% / degree C, so an increase of nearly 64C would raise resistance 25%.
Not sure but it seems feasible that even old school paper and shellac construction should cope with 90C (a 70C rise above a normal room ambient of 20C)?

[Leo_Gnardo]

There is some truth to that. - - - >snip< - - - Now add all the above mentioned effects .

While keeping in mind that famous quote, attributed to Joe Walsh: "Sounded GREAT! Just before it blew up..."

[J M Fahey]

What PDF64 said

Just would be less optimistic on temperature calculations :
Old "speaker cement" was basically nitrocellulose dissolved in acetone + some kind of ether and other volatile solvents (think paint thinner) , same as old style Duco and Dope paints just honey thick.
It toasts, crackles and eventually burns above 100C , already bubbles above 90C (thatīs why speakers often scratch before full dying, bubbles touch gap walls) and above 80C start losing mechanical strength , often leading to loose turns buzzing and even full coil disassembly inside tha gap.

And being inside a tube amp, Iīd start with 45C ambient temperature , thatīs why old speakers often stood 8 to 10W ; 15W was relatively high power (Celestion Blue) and 20W (P/C 15N, Celestion G20) was the absolute maximum.

Hey, even huge coil Altec and James B Lansing were rated 30W RMS.

The prize there is that ultralight speakers (paper formers, almost invisible adhesive and thin paper cones) were incredibly efficient.

Modern speakers use specialized adhesives, mine stand up to 150C and some of those Kilowatt PA speakers up to 250C

The deal of course is that at 250C a speaker has mind wobbling 90% resistance increase ... so the "kW" speaker is actually standing little above 500W

This also explains why the full 8 x 12" stack with Greenbacks or G12H is unbeatable : power spread over 8 speakers means temperature rise is nil, and low moving weight means high efficiency to begin with.

While , say, modern "500W into 2 x 10" " Bass amp wonders barely compete with loud drummers.

Compare that to a 1969 SVT driving 16 x 10" CTS speakers.

In fact my personal opinion about the (by me unfounded) mystical Alnico "powers" **really** lies in the construction described above.

Just by sheer chance, change in dominant magnetic materials matched a change in speaker construction and materials which I think are more relevant. Think late 60īs early 70īs.

After all, a voice coil does not know or care about magnet material (which may be Alnico, Ceramic, Neodymium or even a Field Coil) , only the actual field strength surrounding it.

[tedmich]

Hmmm, I'm getting an idea for "The HotBox" amp oven/isolation cab!

Catchline "Spend a night in the box"!

[Leo_Gnardo]

Hmmm, I'm getting an idea for "The HotBox" amp oven/isolation cab!

Catchline "Spend a night in the box"!

HotBox oven/isolation cab, sounds like my house this time of year, no air cornditioning. Need to adapt an oldie but goodie: Gimme Swelter.