The ESR of caps in linear power supplies is very seldom an issue, an additional benefit to these is their price and the fact that they are rated for ~22 yrs of service at 1h a day.
They represent a much better deal than Atom axials (TVA1620-E3) which are only rated to 85C, with no lifetime data, for 550% more, $16.53!
Those are 100uF 350V with a 290mA ripple current rating. An Atom TVA1718 100uf 450V is 800mA. The ripple current rating is very important and if exceeded the core temperature of the cap rises and failure soon follows. At the very least life is dramatically reduced. The ripple current in the cap is typically more than twice the average DC load current in a linear full-wave power supply.
Last edited by nickb; 02-11-2016, 09:21 PM.
Reason: 350V not 35V!!
Experience is something you get, just after you really needed it.
Those are 100uF 350V with a 290mA ripple current rating. An Atom TVA1718 100uf 450V is 800mA. The ripple current rating is very important and if exceeded the core temperature of the cap rises and failure soon follows. At the very least life is dramatically reduced. The ripple current in the cap is typically more than twice the average DC load current in a linear full-wave power supply.
I don't often consider the ripple current rating, but perhaps I should. (Most of my amp stuff is derivative of existing designs.)
Since the Twin Reverb uses two caps in series at the first position, do we get to double the ripple current rating, or are we not so lucky?
I don't often consider the ripple current rating, but perhaps I should. (Most of my amp stuff is derivative of existing designs.)
Since the Twin Reverb uses two caps in series at the first position, do we get to double the ripple current rating, or are we not so lucky?
The current is the same in both caps. Sorry, you didn't get lucky.
But as Tedm points out, they are cheap so you can put an extra one in parallel with each. The current is now shared so the ripple current rating of the pair is 2x that of one, you also get half the voltage ripple and still have money left over for a couple of beers. So you do still win after all
Experience is something you get, just after you really needed it.
Yes ripple current can be important! https://maaplibrary.files.wordpress....chapter-58.pdf
Most modern caps should easily handle guitar amp ripple currents (especially if they use tube rectification ) but the calculations should be made to be sure.
NickB: I was looking for the specs on the Atoms. Where did you find that?
This is all I have. Actually, this is all they have since it's what I got when I asked them directly.
As a rough guide, if the cap is getting more than just warm, the ripple current is too high. A cap's ripple current rating is closely related to how well it can dissipate the heat the current generates. The lowest temps will be on the surface but the core will be much hotter. The life halves with every 10C rise in temp.
The relationship between working voltage, current and lifetime is the way cheap axial caps (Ruby?) get to claim 500VDC ratings without any lifetime data. All caps can be run at higher than conservative voltage ratings, the trade off is lifetime. The same with ripple current, a 47uF cap rated at 500VDC with 1A ripple current is relatively easy to make with a <1000 hr lifetime... https://www.parts-express.com/ruby-g...citor--020-670
no such thing as a free lunch etc...
Thank you for that, Nick my man. I'm a bit suspicious of those boys at Vishay because, as TedMich alludes, the specification is incomplete without a lifetime rating. Then there's the 85º rating on the Atom. Comparing that to the Nichicon above with 105º and 8,000 hrs. !! I dunno.
Of course, either one of these would probably outlive me in this amplifier ... I just want to put something different in there to, you know, rattle the cages of my buddies with their insistence on using the "proper" "vintage-correct" components.
I think I'm going to try the Nichicons for the fun of it (and I'm cheap — I should mention that).
... I just want to put something different in there to, you know, rattle the cages of my buddies with their insistence on using the "proper" "vintage-correct" components...
Those buddies will be horrified to know that a small modern chunk of capacitor guts is inside the "vintage-correct" size Atom cap case.
Comment