6800uF 42V Substitution: Heat vs. Ripple

[Spenser]

Yes. Methinks some funny switching business in the power amp section may be going on. We shall see how it performs after replacement of the diodes in the power supply and electrolytic capacitors throughout

I am so glad you added the final sentence, "...would be changing them no matter what." Thank you thank you.

Last but not least, thank you for your tag line, or whatever we calls these in AK. Yours, "Fighting the Dunning-Kruger Effect." Oh my word. Looked it up. Absolutely made my day.

 

[c.coyle]

Yinz are welcome. I'm a Burgh expat.

 

[Spenser]

Replacing did the trick -- the two zener diodes, and the three switching diodes -- fixed the problem with the power supply. One of the zeners I'm pretty sure was bad. Also replaced the two big filter capacitors and the smaller two electrolytic ones. One of the big filter caps I'm pretty sure was bad. Left well enough alone the two ceramic caps, resistors, and bridge rectifier. BTW went with the 10000uf50V Nichicon KW. Dropped right in.

http://www.mouser.com/ProductDetail/Nichicon/UKW1H103MRD/?qs=kArNe9LFxXlXU/0N8Vy%2byg==

 

[triode17]

Not to add to the confusion, but I would have gone with 647-UKW1J682MRD. Do they sound any better being "audio grade"? Who knows, but we can try can't we.

As for the 85C vs 105C rating, I've never worked on any amp where the main caps have shown signs of getting hot. Smaller caps around regulated power supplies, that's a different story, and I tend to use PW series here.
Lee.

Oh yeah? Ever work on a tube amp. Sometimes the caps. have to be placed near the rectifier tube and/or vertical power resistors. They can get to about 60 deg.C.

 

[Spenser]

PS ... not to derail, but I do miss the old terminal tab caps for power rails. A lot of them are those silly snap terminals now. One shortcut that makes it a whole lot easier is to just snip the spades off the old cans and solder those to the snap terminals on the new cans.

No desoldering, rewrapping, and hoping the wires are still long enough.

I felt compelled to re-post this above idea I came across. Very cool.

Your patience please as I hijack my own thread to mention I'm currently pondering how to proceed with main filter capacitor substitution-component-choice;

AND how to handle switching over the connections;

Kenwood KA-5500

Will require minimum 10000uF63V and has a bit of a rats nest of connections across the existing big filter capacitors that have the "forked tab terminals" or whatever we call them..

I can post a picture if anyone is interested. Might be a candidate for variant of the sKiZo move above. IDK..

 

[Spenser]

Kenwood KA-5500 big filter capacitor connections (bit of a rats nest). Reference Post #25.

 

[Spenser]

I would look at some Cornell Dublier 381LX/383LX Series with a 3000 hour rating.

my experience tells me usually the old caps (at least the rugged caps used in most older amps) are still above specified capacitance and new caps seem below, so I go next value capacitance up

Wanted you to know you imprinted on my feeble brain.

Two-year delay.

Really like the looks of a 12,000uF 63V Cornell Dubilier 380LX at $7.80 apiece from Mouser -- for my Kenwood KA-5500 (part spec 10,000uF 56V) -- a project started back then (earlier completed other capacitor replacements and trim pots), and finally now dusting off and placing in the queue for near future

Surprised (and only now seeing through my fog), the two-thirds shorter lifetime rating of the slightly more expensive, slightly lower ripple and higher ESR Nichicon KG which I've used for some other power supplies during this time.

The rugged old Elna capacitors in the KA-5500 are probably fine or maybe even better but who knows and I cannot help myself.