Author Topic: Capacitor Stuffing  (Read 3302 times)

0 Members and 1 Guest are viewing this topic.

Offline Tim TTopic starter

  • Contributor
  • Posts: 32
  • Country: us
Capacitor Stuffing
« on: August 05, 2016, 10:33:33 pm »
Over the past few months I have picked up a couple of 30-40 year old HP bench power supplies where big radial electrolytic caps had to be replaced, and as the Spragues/Mallorys are no longer readily available (at least cheaply) this has introduced me to the practice of replacing the guts of the original cap body with a modern electrolytic this way keeping the original appearance and connection points (some of these caps have 3-4 pins).

Now, there is a lot of emotion surrounding cap stuffing (e.g. vintage radio collectors) and I really don’t want to go there (though the big aluminum cans do look cool) but I have to yet see a rational discussion around the technical consequences. That is, is the life span of that new cap anyway compromised by being placed into a sealed can? Does it run hotter, negating some of the advantage of upgrading that 85C cap?

Thanks,

Tim (Resisting the urge to splice blue glowing vacuum tubes into my HP 6113a)
 

Offline dacman

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: us
Re: Capacitor Stuffing
« Reply #1 on: August 05, 2016, 11:10:07 pm »
If the capacitors run near ambient temperature, then their life should not be affected much.  If they run hot, then their life would most likely be shortened.  (Also, a failure mode, which is rare with HP gear, is that if the fluid blows the cap, it will not be as noticeable if the cap is covered.)  Take a look at this data sheet:
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12855
Re: Capacitor Stuffing
« Reply #2 on: August 05, 2016, 11:15:35 pm »
See http://www.rubycon.co.jp/en/products/alumi/pdf/Life.pdf
As a first approximation, 'ambient' for the inner cap will be the temperature of the outer can.  If the dissipation is enough to raise its temperature more than a few degrees above true ambient, you potentially have a problem.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16607
  • Country: us
  • DavidH
Re: Capacitor Stuffing
« Reply #3 on: August 06, 2016, 06:25:26 am »
And to add to the problem, the newer modern capacitors have a smaller surface area leading to a higher thermal resistance.

On the other hand, those old can capacitors had a terribly high thermal resistance because the wound up capacitor was just stuffed loosely into the can so this is nothing to worry about.
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 9057
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: Capacitor Stuffing
« Reply #4 on: August 07, 2016, 05:04:55 am »
I've contemplated this as well since I also like the look of the big ol' metal cans in vintage gear. If you're not sure of the thermal behavior of the various capacitors that are being replaced, install them unstuffed and run the gear. Measure the component temperatures to determine which ones might be safe to stuff into the old housings.

If heat build up is a potential issue, you still want the aesthetics of the old can, and don't mind it being not-quite-original, add some ventilation holes or slots to the can.
TEA is the way. | TEA Time channel
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12855
Re: Capacitor Stuffing
« Reply #5 on: August 07, 2016, 06:49:19 am »
The ultimate solution would be to arrange for an effective heat transfer medium between the caps and the outer can.  Although immersion in a liquid would work, the prospects of having to remake the hermetic can seal, and chemical compatibility issues with the seals would make any sane person run screaming for the hills.

Probably the best option would be an aluminium block heat spreader machined to be a snug sliding fit in the old can, and bored for each replacement cap, + one vent hole to ease insertion.. Strip the jackets from the replacement caps for better heat transfer, and use heatsink compound.  Due to the far greater surface area of the original can, that should run cooler than the replacement caps would, grouped in free air.  If one was mass-producing re-stuffed caps, then setting up to form heavy Aluminium strip in a C shape round each cap with the ends bent back to fit snugly against the can wall would be a possibility instead of a machined heat spreader, but it would be a P.I.T.A. developing the process and tooling.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16276
  • Country: za
Re: Capacitor Stuffing
« Reply #6 on: August 07, 2016, 09:00:26 am »
What is wrong with either a thermpad on the base of the can, with the bare aluminium caps of the filling capacitors in good contact with it. Or just fill the body of the stuffed unit with something that is a better thermal conductor than air, like very high centipoise grease, or just some electronics grade silicone sealer as a blob in the base you push the new caps into.

In most older equipment the capacitors run cool enough, if it was something that had significant ripple current in the capacitor then it would be in a can with screw terminals, which are still available new and the same size, with a low ESR as well.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 12855
Re: Capacitor Stuffing
« Reply #7 on: August 07, 2016, 09:19:45 am »
Yes, a self-adhesive thermpad would be a good solution if the new cap ends are flat enough, but that's probably going to be a real P.I.T.A to stuff if you want it all internal.  If you drill the endcap next to each terminal tag so the leads can be threaded through and soldered externally to the tags after refitting the end cap, it should go pretty easily.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf