Heat and electrolytic caps - proximity to power tubes an issue?

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
MattG
Posts: 5
Joined: Tue Oct 19, 2021 2:42 pm

Heat and electrolytic caps - proximity to power tubes an issue?

Post by MattG »

Hi all, first post. I've dabbled in DIY audio electronics for a while, and built my first tube amp (from a kit) a couple years ago. I am under the impression that, as a general rule, high heat will age electrolytic capacitors more quickly (compared to low/no heat above ambient). It seems that many tube amps (at least those for guitar) have the big filter caps somewhat close to the tubes. Is this something to be concerned about? I assume it's not a serious problem, or the layout of tube amps would have changed dramatically by now... but I was unable to find any previous discussion on this topic (here or anywhere else), so thought I'd try to get some opinions.

For what it's worth - what got me thinking about this is the amp I built, a Trinity Triwatt. It's basically a 70s Hiwatt DR103/504 preamp with a more modest output section, switchable between a pair of 6V6 or KT66 tubes. I recently swapped in KT66 tubes (had previously used only 6V6s), and the enclosure felt noticeably warmer to the touch in the area around the chassis. This isn't terribly surprising, the KT66s are more powerful tubes. I built mine into a 1x12 combo, which intuitively seems like it would have worse airflow than the same in a dedicated head cabinet (which typically have vents on the back and top). That combined with the fact that one of the power tube sockets is about two inches from one of the big filter caps made me wonder if the life of the cap(s) will be shortened considerably from tube heat?

To be clear, everything works fine, I love the amp. But as it's my baby, I can't help but worry about it like I do my human babies! :)

Thanks for any thoughts or feedback!
User avatar
pompeiisneaks
Site Admin
Posts: 4222
Joined: Sat Jan 14, 2017 4:36 pm
Location: Washington State, USA
Contact:

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by pompeiisneaks »

the trickiest part of the point you're making, although it is a good one, is that you also want the power filtering as far away as possible from the preamp sections of the amp. There are some design decisions that tend to make this worse that even I still keep doing like tubes down builds where the heat from the tubes rises up to the chassis heating them even more. A tubes up build means the heat goes away from the amp more than towards it. EIther way, though, 105C rated caps are more than plenty for what kind of heat tube amps see on a regular basis inside the chassis. The tubes heat has to radiate through the air, then the metal enclosure, etc, so it dissipates some of that heat. I've seen videos of tube amps that ran with electrolytics for 50 years before someone replaced them, and they were still functional and working moderately well, just a bit noisy due to losing some capacitance over all that time. Ultimately I think it would be better that as we improve the smaller size of the non electrolytic caps we'd see them last mostly indefinitely instead. I just built an amp with only metal film resistors in the power filtering section and it sounds great to me. This amp should never need a person to touch any part again, so long as nothing goes wrong like a power tube shorting etc.

~Phil
tUber Nerd!
User avatar
sonicmojo
Posts: 669
Joined: Thu Mar 03, 2011 3:41 am
Location: Oahu, HI

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by sonicmojo »

The most common 40/20/20/20 CE can caps made today are only 55C/131F rated. I’ve seen many Princeton Reverbs fail with “not that old” new caps. Try to find caps with higher ratings.
---------
Bryan
Charlie Wilson
Posts: 1104
Joined: Thu Jan 23, 2014 7:32 pm
Location: Laguna Niguel, California

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by Charlie Wilson »

I had a friend/customer play his amp at a gig in the desert and the temperature outside was 112 degrees. His amp was on all day and it is already kind of pushing voltages to the limit. He said after the gig the amp starting blowing fuses and didn't sound that great. I didn't even have to look to know he had cooked his filter caps.
CW
R.G.
Posts: 1234
Joined: Tue Dec 02, 2014 9:01 pm

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by R.G. »

Good intuition. For electronic components as used in tube amps, the general rule is the hotter they get, the shorter their life. Any way you can keep your caps (and tubes!) a little cooler will prolong their working life.

Mild air flow from natural heat-rises convection is what most amps use, but as you guessed, this takes some careful layout. Tubes-up and tubes-down are the easiest and what nearly always gets used, but tubes-horizontal is the best for natural convection.

This idea is nearly always viewed as an abomination, but one of today's low speed and nearly silent computer fans will drop the air temp in an amp down to nearly ambient. Running a 12V "silent" computer fan on 6-7Vdc can make it so quiet you have to stick your ear near the openings in the amp to hear it. And your caps will live a long, long time if nothing else kills them.

To cut off the inevitable next comment, yes, you'll get some dust inside that needs to be removed every so often. It's really not that bad, no worse than your computer.
Stevem
Posts: 4576
Joined: Fri Jan 24, 2014 3:01 pm
Location: 1/3rd the way out one of the arms of the Milkyway.

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by Stevem »

Good reason to install a Peltier device or two with a heat sink on the filter can
When I die, I want to go like my Grandfather did, peacefully in his sleep.
Not screaming like the passengers in his car!😊

Cutting out a man's tongue does not mean he’s a liar, but it does show that you fear the truth he might speak about you!
User avatar
mhuss
Posts: 490
Joined: Wed Jan 19, 2005 3:09 am
Location: SE PA, USA
Contact:

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by mhuss »

I think this is part of why DR put the 'aluminum wall' between the power tubes and the can caps in Hiwatt amps - to keep radiated heat away from the latter.
MattG
Posts: 5
Joined: Tue Oct 19, 2021 2:42 pm

Re: Heat and electrolytic caps - proximity to power tubes an issue?

Post by MattG »

pompeiisneaks wrote: Tue Oct 19, 2021 7:55 pm EIther way, though, 105C rated caps are more than plenty for what kind of heat tube amps see on a regular basis inside the chassis.
In my particular case, the cap I'm most concerned about is the 50-50uF / 500v PSU filter cap which is really close to one of the power tubes. Here are pics of my build. In the pic, I can't see the label for those caps, and I looked up another guy's build pics, I can't see the label on his either. I don't have easy access to the amp at the moment, so not sure if they really aren't labeled, or I'm just unlucky with the pics. Either way, it seems those high voltage "multiple caps in one can" are not particularly common. Though, I was able to find a handful of different 50-50/500v caps; some didn't have any temperature rating, and others fell between 55C and 85C. Definitely couldn't find any 105C rated caps in this style.

However, with a quick Mouser search, I could find 500V 47uF and 68uF "single" caps with 105C temp rating. I think I'll start with the fan idea; but with some mounting creativity, it seems possible to install more robust caps.

sonicmojo wrote: Tue Oct 19, 2021 11:29 pm The most common 40/20/20/20 CE can caps made today are only 55C/131F rated. I’ve seen many Princeton Reverbs fail with “not that old” new caps. Try to find caps with higher ratings.
How old is "not that old"? :)

R.G. wrote: Wed Oct 20, 2021 3:14 pm Mild air flow from natural heat-rises convection is what most amps use, but as you guessed, this takes some careful layout. Tubes-up and tubes-down are the easiest and what nearly always gets used, but tubes-horizontal is the best for natural convection.
I think you would call my combo "tubes-down". I feel like my combo's arrangement is such that there really is no where for the heat to go, except to "pool" in the little compartment created by the cabinet's rear panel. (You can see that rear panel in the pics I linked above; it goes right across where the tubes hang down from the chassis.)

R.G. wrote: Wed Oct 20, 2021 3:14 pm This idea is nearly always viewed as an abomination, but one of today's low speed and nearly silent computer fans will drop the air temp in an amp down to nearly ambient. Running a 12V "silent" computer fan on 6-7Vdc can make it so quiet you have to stick your ear near the openings in the amp to hear it. And your caps will live a long, long time if nothing else kills them.
I've been building my own PCs for many many years now, I have no shortage of silent 12VDC computer fans. :) I think I'll modify that back piece of wood such that I actually hollow most of it out (i.e. kind of turn it into a "frame"), and the void can be filled with that metal mesh I've seen on other amp heads and combos (even my cheapie Monoprice amp has this). That will keep the tube area protected, but also allow some ventilation from the fan I plan to install.

pdf64 wrote: Wed Oct 20, 2021 5:18 pm http://www.blueglow.de/FenFan.html
Thanks for that, very helpful! That's actually what I was thinking when R.G. mentioned installing a fan, the low voltage secondaries for the filaments seem ideal for powering a fan.

Thank you everyone for the helpful feedback! Much appreciated!
Post Reply