• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Best caps to replace with?

Mat86

Mat86

Active Member
So I have a headphone tube amp I've been tweaking and noticed 4 of the 6 larger electrolytic caps were bloating. It's about 12 years old now and comes stock with Rubycon 470uF 250Vs.

While I'm sure I can get away with just leaving it a few more years, many recommend recapping after 10 years or at signs of bloat/bulging.

What's my best option for replacing these? Trying to upgrade sound quality if possible, not necessarily just prolong life
 
Register to hide this ad
I'd go up in voltage if size allows it.

Cornell Dubilier (no affiliation) has some 105º, high ripple, long life capacitors at good prices, like the 381LX series.

There isn't a "best option". The "best" capacitor doesn't exist. There are thousands of different capacitors for different applications.

I agree with redk9258, a Rubycon 12 yeas old capacitor is probably OK if not pushed too hard (working voltage not too close to the rated 250V)
 
elnaldo said:
I'd go up in voltage if size allows it.

Cornell Dubilier (no affiliation) has some 105º, high ripple, long life capacitors at good prices, like the 381LX series.

There isn't a "best option". The "best" capacitor doesn't exist. There are thousands of different capacitors for different applications.

I agree with redk9258, a Rubycon 12 yeas old capacitor is probably OK if not pushed too hard (working voltage not too close to the rated 250V)
Click to expand...

Why go up in voltage?

And sorry my phrasing was poor about a 'best' on something so subjective, but moreso meant for my application in the specs I need to narrow it down since I'm starting from scratch on this topic
 
Capacitors last longer when used at lower voltages. There is usually a table at the datasheets, detailing a multiplier factor when you run the capacitor at lower voltages. A capacitor rated for 3000hs at 250V, will multiply that number when used at lower voltages.

Quoting a capacitor manufacturer paper:
"
. For medium and large sizes (Snap-in and screw terminal types) the influence of actually applied voltage gains some impact on lifetime, because operating voltages below the rated voltage
cause less stress to the dielectric layer. The closer the operating voltage approaches the rated
voltage, the more of the electrolyte is consumed for the self-healing of small flaws within the
dielectric layer. The self-healing (and thus the electrolyte consumption) is also exponentially
depending on temperature. Vice versa, a lower operating voltage than rated voltage may
extend the lifetime of elcaps significantly [4 "
 
Last edited:
So if I'm looking this all up correctly, my options are the CD 381LX (315v, or 450v) for $15, the Kemet ALT22 series (385v) for $13, and Nichicon KX (250v) for $4.50

anything else I need to consider in specs? And are they in a similar "audiophile" league or do you get what you pay for?
 
Last edited:
Without a schematic, just guessing by the capacitor value, I think in this case "a capacitor is a capacitor". A tube headphone amp shouldn't be a problem for any of the capacitors you found. As mentioned, higher voltage capacitors should last longer, I'd choose a 105º capacitor in a tube amp because heat also reduces the life of the capacitor, and that's all.

Reading the datasheets, the CD is a "better" capacitor, longer life (3000hs at rated voltage and 105ºC vs 2000hs of the Nichicon KX), and better ripple specs, and that's probably why it costs more.

The Kemet is a 85ºC capacitor with a very long life (12,000 to 26,000hs at 85ºC), but the CD , if I'm reading the graphs the right way, being a 105º cap, will last 20,000hs at the rated voltage and 85ºC.

But again, a tube amp is not a very demanding application (like a switching mode power supply, for example). I'd choose the one you like better, the one with easier and cheaper shipping costs, or the one you can afford. I bet money you won't notice any difference. I think I'd get the Nichicon KX, or compare it to the Cornell Dubilier 250V, to see a real price difference. Now you are comparing a 250V cap vs a 450V, which is more expensive
 
^^ agreed, ripple is at double your mains frequency so 100 or 120, nothing nasty like you get out of a switching supply. No audio is passing through them, so all you really need are basic general purpose caps. For this I'd probably be sorting by what physically fits best first, and then by whatever other parameter you like second. Color, price, cool name, etc.

if you want to know if it makes sense to bump the voltage, measure the voltage across them. If its not more than 80% or so of the marked voltage, its probably sufficient as-is.
 
Okay well that simplifies it if there's no 'audiophile' reason to be paying more for this component this time. I checked and my other headphone amp uses Nichicon KG "Fine Golds" but they're lower capacitance. KX series in a higher voltage is the same price as the Kemets & CDs so I guess go by cost and size then? Nichicon it is (even though I was willing to spend whatever if it could improve things slightly)

At least that takes the $45 Mundorf's off the table ><
 
I understand going with higher voltage ratings in caps can be advantageous, but does any one know how much the capacitance can be safely varied from the original in cases where the original value is no longer available? I'm planning to recap the power supply in my McIntosh 2505 power amp I purchased back in 1969 (yup, I'm an old guy). For example, could a 10,000 uf be ok to replace with an original 9300 uf? Thanks to anyone who can help!
 
Ron Skwier said:
could a 10,000 uf be ok to replace with an original 9300 uf?
Click to expand...
Yes, easily. The tolerance on the original cap would allow more of a range than that.The way caps are produced now, they are likely to be on the lower side of the tolerance instead of high like your originals. I would consider going up 20%. ...this is for the filter caps.
 
Thanks redk9258. This really helps. Is there another workable variance in capacitance that would apply to capacitors in other parts of the circuit? Thanks
 
You must log in or register to reply here.
Back
Top Bottom