external caps close to the power tubes. what's the optimum for the dry up effect?

[pietro_moog]

hi guys. i'm building an amp, but the chassis i choose is not that big. i came to have my caps ( JJ radial caps on the outside) close to the power tubes.
they are like one inch of distance (maybe a little more, but not 2 inches).

how far should they be for the power tubes to have not an effect on them? i mean for not drying them up.
what 's the optimal distance to minimize the heat- dry up effect?

[Chuck H]

Heat doesn't actually dry them up. The gooey material inside crystalizes like honey when the cap isn't in service. In fact, it's heat that's used to rejuvenate some older caps (not recommended, but it sort of works) This, entirely by coincidence, is also like honey. The thing you need to worry about is voltage derate with temperature. If the can is rated for temp X but hits temp Y that near the power tubes then the voltage spec drops. It's probably not a problem. A lot of amps have been designed as you describe. But keep looking at the design and see if you can't relocate the caps as far as possible/practical from any heat.

[pietro_moog]

thanks

[pdf64]

Bear in mind that radial ecaps are available that are far more space efficient than the traditional can caps used in guitar amps.
So it may be possible to mount such caps inside the chassis.

[Mick Bailey]

Capcitors are rated (among other things) with service hours at a given temperature - usually 85c or 105c. Either of these feels pretty hot to the touch and you need to measure the temperature to be sure. Roughly, for every 10 degree reduction in temperature you get double the service life.

Surface temperature of the caps will depend on tube type, but you could interpose a piece of aluminium sheet midway between the tubes and caps both to sink and reflect some heat.

I sometimes see amps where a bypass cap is strapped right across a power resistor and the whole thing runs very hot to the touch - sometimes like this for many years before the cap dies.

[Chuck H]

I sometimes see amps where a bypass cap is strapped right across a power resistor and the whole thing runs very hot to the touch - sometimes like this for many years before the cap dies.

Like every cathode biased Fender amp made before 1990

[mhuss]

Another option is you're worried about it is to put in a metal heat shield as Hiwatt did (in those amps, the caps were lined up next to the power tubes).
Hiwatt Internals

[R.G.]

The heat shield is a good idea if you can't move them to a cooler place.

Capacitors generate their own internal heat, from the RMS value of the ripple current that charges and discharges the cap. The RMS ripple current squared times the ESR is generated as heat. So are the various dielectric hysteresis losses. Add more heat to them from outside and they can get really hot inside.

Caps degrade not only by boiling the internal goop, but also by chemically degrading it. Chemical degradation is generally thought to increase by the Arrhenius equation, with reactions doubling in speed for each 10C rise in temperature. We now know that's not fixed, but it's a good guide. So the insides not only boil, but at sub-boil, chemically degrade.

Keep them cool. They're temp rated for X hours at Y temp. Y is usually 85C for non-special caps, and may be 105C or higher for premium high temp caps. The received wisdom is that if a cap is rated for 10,000 power on hours at 85C internal, it should last around 20,000 at an internal of 75C, and 5000 at an internal of 95C.

[Gregg]

Bear in mind that radial ecaps are available that are far more space efficient than the traditional can caps used in guitar amps.
So it may be possible to mount such caps inside the chassis.

+1. If you're building a new amp go for the much better, smaller and with longer life 105C radial caps.