Power Supply Filter Capacitor Sizing

I have a Marantz SR4000, the 1980's version (there should be a law about re-using model numbers :\ ). I replaced the power supply filter capacitors several years ago. I used some ebay Rubycon 10,000uf 50V, which is to spec for the receiver, but they were physically much smaller than the originals. These turned out to be fakes or trash, I recently opened up the receiver and saw the tops bulging up and after pulling them out, they rattled around inside. Opened one up and it fizzed like a soda can and had a much smaller element inside. So off to Mouser this time and I thought I would fix it once and for all with some Nippon-Chem-Con 22,000uf 63V capacitors. Larger than the originals physically and electrically. Installed them, and things work fine, perhaps an extra half second for the speaker relay to click ON, as the larger capacitors take a bit longer to charge up.

But then I do some homework (after the fact) on how good an idea it is to up the capacitance on the power supply filter capacitors. Seems that too big a capacitor puts more of a load on the transformer on startup. The transformer is one part that is not possible to replace other than a used one on ebay.

The 22,000uf measures about 18,800uf whereas I seem to recall the originals measured around 8,000uf out of the 10,000uf they were supposed to be. For reference, this is a 50watt amplifier running about 36volt rails. When I got it, the bridge rectifier diodes were toasted, so replaced those with some stronger diodes back then.

So... Did I over do it with the 22,000uf replacing the 10,000uf?
 
Member
Joined 2014
Paid Member
Yes - a good value would have been 12K or 15K - modern capacitors almost always are on the low side of the tolerance percentage (got to save a few cents). Those are probably +- 20% so your measured values are easily within spec - just low.

Hal
 
  • Like
Reactions: 1 user
A good minimum value would be 2x 4.700-10.000uF for such an amp with 36V rails. Marantz was spot on. Power transformers do not age, they can work for 100 years theoretically, if not heated up too much. A short draw of it's maximum current is perfectly OK, as the resistance of the primary winding limit's the current anyway. The only critical parts are the rectifier diodes. They can age and blow up. You replaced them with stronger ones, fine. Larger amps have inrush current limiter, not for protection of the amp, but for conserving the AC line fuse.
22.000uF may improve the low end of the amp, compared to a small capacitor of 2.200 or 4.700uF. 10.000 to 22.000 is hardly audible, maybe with very low load.
As very large capacitors do not allways have lower ESR, I would prefer 4x4.700 to 1x22.000 if I wanted to improve it. Anyway, I see no need at all to change anything on your amp now. Enjoy it with your "cap tuning".
If anyone tells you a larger capacitors will degrade your amp, maybe just ignore him. Capacitors have been very expensive in the past, so any large scale manufacturer put a smaller one inside than the cirquit designer would have picked.
 
  • Like
Reactions: 1 users
Thank you for the replies. I think I can rest easier with your info. The ESR on the 22,000uf capacitors is listed as 11 mOhms. I should have just concentrated on the higher voltage rating as a way to get better longevity out of them. I don't listen very loud (trying to hang on to what is left of my hearing), so not interested in trying to get more power out of it. My speakers are not all that efficient at 85db, but if the Watt Meters on the receiver are to be beloved, 4 watts covers 99% of my listening. Still plenty easy to drive at low levels within 3ft of my ears. It is all a glorified PC speaker system/vintage audio system/learning to fix things by breaking, them kind of thing.
 
  • Like
Reactions: 1 user