How critical is ripple current for audio caps?

[lti]

I recapped a set of computer speakers about a year ago that used Fujicon RA series caps and a TDA1517 amplifier IC. I ended up using the cheapest 105°C caps I could find, which were Panasonic EB series for the output coupling caps and Nichicon VZ series for the other caps. At the time, I couldn't find the datasheet for the original caps. I finally managed to find the datasheet and discovered that the replacement caps have much lower ripple current values (the Panasonics have almost half the ripple current of the original Fujicons). Will this cause the new caps to fail soon?

Pictures of the amp before recapping can be found here. One of the new caps had to be installed at an angle or it would touch the heatsink. This amp gets very warm.

[PCBONEZ]

Re: How critical is ripple current for audio caps?

Could but not necessarily. Stock caps may have been over rated or over-kill.

For the OP couplers calculate you amps at max output.
Compare that to the ripple current rating to see if you're safe.
[It's ball-park but should be close enough to get some idea.]

The power looks to be DC so there shouldn't be much ripple there.
Those caps are probably more for EMI.
.

[kaboom]

Re: How critical is ripple current for audio caps?

discovered that the replacement caps have much lower ripple current values (the Panasonics have almost half the ripple current of the original Fujicons).

More like the fake caps had inflated ratings on their datasheet, as usual. That way when Mr. Beancounter does a 'cost-reduction' study before things go into production, the following happens:
(reads datasheets)
Panasonic- FC, 2200u 16v, 16x20mm, 2205mA ripple, 29 milliohm ESR, 5000 HR, 35 cents apiece per 10,000
Fuhjjjyoooohoo- TNR or some such with 'oddly similar specs' (presented in such a way as to be "appealing" to beancounter propellerheads), 5 cents apiece per 10,000

And then Mr. Hero, of the accounting office, goes, "A-Hyuck, them datasheets show the cheep caps gotz the same specs as those more expensive cap-citorz! Look how much we'll save!" A-Hyuck indeed, Bubba.


The older TVs with the hot-running STR linear regulators used that series or their predecessor, among others. Obviously, they weren't up against the heatsink, but you could easily get 10+ years out of them in such conditions.

I'll bet they'll last long enough in that amp until you get tired of it. Those are 'honest' caps, 105 degree as well, versus 85 degree fakes.

[kc8adu]

Re: How critical is ripple current for audio caps?

if no hum or distortion just use it.unlikely to give you any more trouble.

[999999999]

Re: How critical is ripple current for audio caps?

The answer is that ripple current rating is not important for audio, so far as capacitor life is concerned unless it's in certain parts of a class D amp circuit or part of a switching power supply for the amp.

You'll tend to have generic caps fail in them because they weren't stable, or weren't suitable for the high temperatures regardless of their temperature rating.

However, purists would want high ripple rated low ESR capacitors for a "potential" sonic benefit.

[kaboom]

Re: How critical is ripple current for audio caps?

The answer is that ripple current rating is not important for audio

Unless we're talking about speaker coupling caps such as what a Sansui 4000 would have.

There, you'd want a ripple rating at least as high as the speaker current.

[lti]

Re: How critical is ripple current for audio caps?

How do I determine the output current of the amp? This amp is powered by a 12V unregulated power supply and 4 ohm speakers are being used.

[Th3_uN1Qu3]

Re: How critical is ripple current for audio caps?

Use Ohm's Law. The TDA1517 is a 6W per channel IC. I = sqrt (P/R). So 6W over 4 ohms is 1.22A. This will be the ripple current the caps will be seeing when the amplifier is ran at maximum volume.

[999999999]

Re: How critical is ripple current for audio caps?

^ the calculation isn't nearly that simple. Speakers are a reactive load, peak current is at low frequency below nominal 4 ohm rating, the chip can't produce 6W until it hits 10% THD at 14.4V power rail, AND the power supply is 12V and probably some economized 1A to 2A thing split between two amp channels, which would have rail collapse if peak output were for an extended enough period that the capacitor ESR (at a higher, 100Hz rating than audio band) put it outside an acceptable range.

If I had to guess the guess would be a little under 1A peak ripple current but in the spec sheet the likely candidate caps are rated up at 100KHz, not < ~ 22KHz and rated for RMS current not instantaneous... it's a dissipation/heating issue. Just about any sensibly sized capacitor would handle this job so long as it's not a Chinese kamikaze brand that would kill itself no matter what. (8mm dia., 10V, 1000uF as per datasheet).

[Th3_uN1Qu3]

Re: How critical is ripple current for audio caps?

IME you always have to consider worst case scenario... take two modules of the same kind, one where all of the amp's manufacturer considerations have been respected, and one where most of the parts run close to their absolute maximum ratings. Which one will last longer?

Also, for obvious reasons, small amps tend to get pushed harder than larger ones. Other than that i agree with what you said, if the caps don't get hot when the amp is pushed at full volume then they're going to be fine.

[Logistics]

Re: How critical is ripple current for audio caps?

I have a set of these HK-206's that I just got at a electronics surplus place for $3. I opened it up and it's full of bright, blue Samxon caps. Time for some audio skullduggery on these babies.

[ben7]

Re: How critical is ripple current for audio caps?

The max repetitive output current for the TDA1517 is 2.5A according to the datasheet.

[lti]

Re: How critical is ripple current for audio caps?

I opened it up and it's full of bright, blue Samxon caps.

What series are they?