• Please note that there are a few updates and clarifications made in the Audiokarma Rules, mostly relating to advertising and the addition of the new "Paying it Forward" & "Giving back" forums in the AudioKarma Audio Marketplace section.

Heat and cool cycles on amplifiers and receivers

Michiganamps

Active Member
Hello, sorry I know I have been asking some noobish questions lately, but another one I have is heat and cool or, on and off cycles on amps. I am an album guy and I really like my records! What I'll do is listen to a record 30-45 mins and then perhaps one more side then I'll leave, go smoke a cigarette ~10 minutes , come back in and turn on the amplifier and repeat. My question is it better to let the amp cool as I have been doing or better to leave it on to reduce turnoff and start up pulls on the tube filiments?
Thanks, John
 
Register to hide this ad
Now the transformer does get pretty hot after about 2 hours would it be advisable to turn it off for a period after say 3 hours??
 
Leave it on if you’re using it. The power transformer can run pretty hot once it gets fully warmed up. No need to let it cool. I always turn my tube amps off if I’m leaving the house or know I am finished listening for a good while. I don’t like leaving my tube gear unattended for long. An hour is probably my limit.

The output transformers on the other hand should not get hot or even very warm.
 
Leave it on if you’re using it. The power transformer can run pretty hot once it gets fully warmed up. No need to let it cool. I always turn my tube amps off if I’m leaving the house or know I am finished listening for a good while. I don’t like leaving my tube gear unattended for long. An hour is probably my limit.

The output transformers on the other hand should not get hot or even very warm.
The transformers on this unit can get pretty toasty I have a photo attached here. I feel like it's the proximity of the output tubes being so close to the transformer as it's only the left outout that gets warm and that's after about 2 hours and that's why I get a little cautious... Cautious enough to ask here on the forum :pIMG_20190604_125629347_LL.jpg
P.s. as a side note I know the left electolytic is bulging out it has been replaced :)
 
Output transformers never get warm, only the power transformer. The output transformers will get warm or hot by design as you said, by only being in close proximity to the tubes. That is all it is. Let's say that if you were to move the output transformers to another chassis, they would never get warm or hot. They will stay cool even if the amp was left on all day.

PS, that electrolytic is not bulging. That is just the plastic cap you see on top. It is the same as the capacitor to the right except it has no plastic outer case. They are metal cans and if you were to see one bulge, it would be around the sides and not the top. I have only seen one can capacitor bulging in my life time. Smaller capacitors have vent holes underneath or ridges across the top that are made to burst open if one were to let go.
 
Last edited:
Output transformers never get warm, only the power transformer. The output transformers will get warm or hot by design as you said, by only being in close proximity to the tubes. That is all it is. Let's say that if you were to move the output transformers to another chassis, they would never get warm or hot. They will stay cool even if the amp was left on all day.

PS, that electrolytic is not bulging. That is just the plastic cap you see on top. It is the same as the capacitor to the right except it has no plastic outer case. They are metal cans and if you were to see one bulge, it would be around the sides and not the top. I have only seen one can capacitor bulging in my life time. Smaller capacitors have vent holes underneath or ridges across the top that are made to burst open if one were to let go.
Oh okay that makes a lot more sense I actually might restuff that can because I kind of did it a little faster than I would've liked when I was changing out caps not really too bad but I j hooked some caps so I already want to go back in and put some shrink on some Leads that I did that on, because I just purchased a solder sucker when I had to replace some sockets that someone did a very sloppy job on in the past so the sockets would match
 
A large computer fan will properly cool the amplifier and prolong your tube's lifespan. The bigger the fan blades the higher the CFM at lower RPM, and thus the lower the noise. Heat kills tubes and electrolytic capacitors.

Each time you cycle the tube the heater (tungsten wire, like a light bulb, coated with alumina insulation, bent and wedged into the cathode sleeve) expands and contacts, which is damaging for the alumina insulation and it recrystlalizes the tungsten filament. The heating process is just like a light bulb and consider when a light bulb fails: upon initial turn-on. Once the filament is hot it equilibrates and no longer has current being forced through high-resistance portions by the lower-resistance segments since it's all more or less the same high resistance.

Here's an excerpt from what I elsewhere wrote on the subject:

A filament or heater is made from tungsten wire, just like the filament in a lightbulb. Tungsten has a Positive Temperature Coefficient (PTC). That means as it heats the resistance goes from zero (cold) to a very high value (hot). If it had the opposite property, i.e. Negative Temperature Coefficient (NTC) like a common thermistor (although not all are NTC devices), as it heated the resistance would continually drop towards zero and the current would increase until it burned out. This has the opposite effect, because as resistance increases with temperature, the current drops.

What happens is that the tungsten has defects in it and has variable resistance. So certain spots, because of bends, welds, manufacturing defects (tungsten wire is tricky to fabricate), alloying with nickel migrating through the alumina insulator, etc. have higher resistance. The lower surrounding resistance forces current through those high-resistance spots (the total resistance of the wire is what matters for the aggregate current flow, not the local maxima or minima), making them hotter and hotter, with higher and higher resistance, until they glow or flash. This is a thermal runaway process which damps out before it burns out the filament because the entirety of the wire becomes high resistance.

This is why lightbulbs typically fail on turn-on. The massive inrush current burns out a high-resistance portion.
 
Curious on this...also a noobish...
I've been breaking in new tubes (cause one other burned out). I meant to be playing streaming music all night. Just that I didn't do it right and the amp was just left on.
Is that bad - how bad.

Funny that it is in a room without a heater, so kinda makes the office warm (70F vs 65F or abouts) in the morning.
 
Don't worry about it.

For the future, what damages tubes is a combination of heat and cutoff. Here's the short version of a few of the factors.

Temperature liberates the absorbed/adsorbed gas in the metallic structures and envelope. While such gas was eliminated to a large extent by cleaning stages and pump-down, some remains and as temperature increases the energetic gas increasingly escapes from its metallic lattice prison. The unreacted cathode carbonates (alkali metals are pyrophoric in air, so the cathode is assembled in air using intert carbonates which is activate during pump-down) also break down and liberate gas. The water in the mica is decomposed, liberating gas.

In addition, when a tube is held in cutoff for long periods of time the silicon in the cathode (used to activate it and re-activate during use) reacts with the alkali metals to form orthosilicates. This non-conductive layer interferes with cathode function and also acts like a capacitor. (Metals separated by an insulator form a capacitor.)
 
Part II. Many of these amps/tubes have a "break-in". I kinda rolled my eyes but after playing my new tubes several days, I am almost almost broken in. They do sound different.

I hear pink noise, or music. What does just leaving the amp on do?
 
Part II. Many of these amps/tubes have a "break-in". I kinda rolled my eyes but after playing my new tubes several days, I am almost almost broken in. They do sound different. I hear pink noise, or music. What does just leaving the amp on do?

Leaving it on has exactly the same effect. Don't listen to audiophoolery, it will only fill your head with nonsense.

The whole concept is, at best, highly dubious, and, at worst, proof of confirmation bias.

It is possible that some tubes may not have been fully activated by the manufacturer. The heat from the filament would, of course, fully activate any cathode material not yet activated, complete various chemical reactions, drive out any absorbed/adsorbed gas or water to final activation with whatever getter material remains, and generally finish what the activation process began. It isn't a burn-in or break-in, it's simply the tail end of a curve which already happened.

I don't believe that anything would be required. If it were, the manufacturers would have aged those tubes longer than the manufacturing process did prior to packaging and shipping, both to fully activate the tube and to weed out any quality control issues.

Assuming any issues existed, and likely none do, tens of minutes to a a few hours should be sufficient to resolve issues for any ordinary signal tubes. Transmitting tubes have a more complex process involving gradual voltage increases, but that does not here apply.

But, again, this is just oft-repeated dogma not having any proof. It's always "to my ears" which really means "to my adjusted brain".

Get a new mattress. Experience same effect. Did the mattress change? No, your brain and musculature adapted.

This is not the same as breaking in a mechanical device, like an internal combusion engine or a loudspeaker surround.
 
I would say to read about the principal of metallurgy and understand the process of how Metal turns brittle due to the heating and cooling process. The metal in the circuits gets brittle due to the heating and cooling process. The only way to stop the process of circuitry becoming brittle is to avoid the process. What happens to the sound after circuitry gets brittle.....you will find out eventually.

Personal preference has nothing to do with the process.....
 
Thank you guys! I enjoy learning stuff from this forum, as no one pushes stuff down your throat, such as use these caps or do this burn in. It's all just facts and I honestly really respect the fact no one is trying to sell me pipe dreams :) thank you all :p
 
I would say to read about the principal of metallurgy and understand the process of how Metal turns brittle due to the heating and cooling process. The metal in the circuits gets brittle due to the heating and cooling process. The only way to stop the process of circuitry becoming brittle is to avoid the process. What happens to the sound after circuitry gets brittle.....you will find out eventually.

The only embrittlement is to the filament from contaminants (nickel and aluminum) migrating into the tungsten because of the direction of the heater-to-cathode potential performs metal migration. That plus Miller-Larson recrystallization will eventually fracture the filament. This does not, however, change the sound, except to the extent that when the filament fractures and no longer heats the tube all of the sound stops. It's not exactly a function of thermal cycling, it's more electrochemistry in terms of poisoning which then causes thermal fracturing.

The reasons why a tube fails, outside of heaters, is electrochemistry, not metallurgy. In normal operation the cathode eventually dies because it is gradually poisoned by orthosilicates, because it has burned out all of its alkali metals, or because excessive current has left burned regions all over it. The tube's Gm eventually falls too low to function because the grid ends up covered with ablated cathode and heater material and becomes less sensitive so Gm drops. The grid and screen may also be damaged from excessive current flow. The plate is damaged from excessive current, etc. etc. etc.

This is not metallurgy, per se, except to the extent that high current melts or burns holes in metals and alkali coatings.
 
The only embrittlement is to the filament from contaminants (nickel and aluminum) migrating into the tungsten because of the direction of the heater-to-cathode potential performs metal migration. That plus Miller-Larson recrystallization will eventually fracture the filament. This does not, however, change the sound, except to the extent that when the filament fractures and no longer heats the tube all of the sound stops. It's not exactly a function of thermal cycling, it's more electrochemistry in terms of poisoning which then causes thermal fracturing.

The reasons why a tube fails, outside of heaters, is electrochemistry, not metallurgy. In normal operation the cathode eventually dies because it is gradually poisoned by orthosilicates, because it has burned out all of its alkali metals, or because excessive current has left burned regions all over it. The tube's Gm eventually falls too low to function because the grid ends up covered with ablated cathode and heater material and becomes less sensitive so Gm drops. The grid and screen may also be damaged from excessive current flow. The plate is damaged from excessive current, etc. etc. etc.

This is not metallurgy, per se, except to the extent that high current melts or burns holes in metals and alkali coatings.
Okay, so basically when a filament goes it's pretty bad for the whole tube. Kind of like a chain reaction. That's actually pretty cool! That's why I think that chemistry and physics are really interesting, as I feel like they go hand in hand with tube equipment!
 
Okay, so basically when a filament goes it's pretty bad for the whole tube. Kind of like a chain reaction. That's actually pretty cool! That's why I think that chemistry and physics are really interesting, as I feel like they go hand in hand with tube equipment!

Yes, it's all chemistry and physics, nicely and intricately tied together. A lot of tube weirdness originates with the physics and electrochemistry, and understanding such factors can explain a lot about why circuits are the way they are.

A vacuum tube starts with a pile of chemistry and maintains that over its lifespan. If you want to further read about it, I suggest some of the books about fabricating vacuum tubes, such as the volumes produced by RCA and the older books about fabrication techniques, like those by Reich, Spangenberg, etc.
 
Back
Top Bottom