Power supply caps: They're not all equal

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm doing a partial recap to my Nakamichi TA-2A receiver. It requires repair so now is the time.

I entertained the idea of replacing the big power supply caps (although I won't this time around) so I browsed big capacitors. Of course there's low esr and low impedance caps, but there is a spec listed for power supply caps that I hadn't scrutinized before: ripple current. It varies widely for a given value cap.

Here's 4700 uF, 63 volt caps. Aluminum Electrolytic Capacitors | Mouser

Panasonic cap, 4.86 A ripple current, 3000 hour life. ECO-S1JP472CA Panasonic | Mouser

Nichicon cap, 3.2 A ripple current, 2000 hour life. UVR1J472MRD6 Nichicon | Mouser

Nichicon cap "audio grade", 3.4 A ripple current, 2000 hour life. UFW1J472MRD Nichicon | Mouser

For smaller caps (like 470 uF which I need for the low voltage power supply board) the discrepancies are even wider.

I would think that a cap with higher "ripple current" specification would be a better choice for a power supply application. Does anybody know how they arrive at this specification? Is this specification really that relevant? I suppose that it is related to esr; am I correct?

Thank you for your comments.
 
Last edited:
Member
Joined 2011
Paid Member
I treat it as "You are allowed to operate our capacitor with This Much ripple current or less. Any more than that and all bets are off. The capacitor could fail, catch fire, or shoot out a 50 foot long stream of boiling acid if you exceed the ripple current spec. "
 
I treat it as "You are allowed to operate our capacitor with This Much ripple current or less. Any more than that and all bets are off. The capacitor could fail, catch fire, or shoot out a 50 foot long stream of boiling acid if you exceed the ripple current spec. "

Did you ever blow up an electrolytic before? I have; on purpose and by accident.

It isn't easy stuffing all that ribbon back into the can. :p
 
The ripple current rating refers to heating. It is related to ESR but not exactly the same thing. Two caps with the same ESR can have different ripple current limits if they have different effectiveness at getting rid of heat. Like most other things with electrolytics, the closer you are to the limit the shorter the cap life. It tells you nothing whatsoever about how the cap will 'sound'.
 
I would think that a cap with higher "ripple current" specification would be a better choice for a power supply application. Does anybody know how they arrive at this specification? Is this specification really that relevant? I suppose that it is related to esr; am I correct?
Max ripple current is like any abs max rating: it is a fact, a given of the problem.
You should stay within the limits, period, but going further and deducing quality level from that data is probably not a good idea: a prudent manufacturer will give a conservative figure, whereas an eager newcomer will tend to give a very optimistic estimation.
In the end, an apparently "inferior" product might in fact be better. An example of that is the Siemens capacitors of the seventies: general purpose capacitors from that firm were better than low esr/high ripple parts from practically all competitors.
But the datasheet didn't show anything obvious, the opposite in fact
 
Thanks DF and Elvee.

"Inferior" depends to an extent on the application and desired circuit performance; can we agree? Datasheet parameters like temperature operating range and estimated service life are relevant parameters for some applications.

In some ways, "better" is sometimes subjective though. I noticed that some of the "audio grade" ;) electrolytics have ten times the leakage current and half (or less) the service life of more general purpose capacitors. And do they really "sound" better? Who knows; but with the leakage current I can tell you that they won't work better in some circuits!

I still say that a properly designed circuit will not have any use for "audio grade" electrolytics.
 
I would shop for good name brands 1st
higher temp ratings second
low esr third
ripple ratings and operating hours fourth


The ripple current
heats the capacitor and the maximum permitted ripple current is
set by how much can be permitted and still meet the capacitor’s
load life specification. Too much temperature rise will cause the
capacitor to exceed its maximum permitted core temperature
and fail quickly, but operation close to the maximum permitted
core temperature dramatically shortens expected life. The load
life specifications for aluminum electrolytic capacitors operating
at maximum permitted core temperature are typically 1000
to 10000 hours. That’s a load life of six weeks to a year and a

seventh and not long enough time for most applications.
 
And by the way, the capacitors in question are on the low voltage power supply board. One of them is intermittent (tapping it with your fingernail causes it to cut out and come back in) so I will replace all of them on that board and the 7812 regulator too.

The audio capacitors are Luxman brand orange case (" audio grade capacitors made by Elna" according to google). I am contemplating replacing some of them; especially on the tone control board since the tone controls suck on that receiver (I use my own humble preamp with the receiver and it sounds way, way better). The switch on that board has been bad for a long time too so I will try to massage it or else just wire around it. So what do you want to bet that if I replace them with rank and file long life Nichicon or Panasonic capacitors that it won't change the sound? :cool:
 
Last edited:
I would shop for good name brands 1st
higher temp ratings second
low esr third
ripple ratings and operating hours fourth


The ripple current
heats the capacitor and the maximum permitted ripple current is
set by how much can be permitted and still meet the capacitor’s
load life specification. Too much temperature rise will cause the
capacitor to exceed its maximum permitted core temperature
and fail quickly, but operation close to the maximum permitted
core temperature dramatically shortens expected life. The load
life specifications for aluminum electrolytic capacitors operating
at maximum permitted core temperature are typically 1000
to 10000 hours. That’s a load life of six weeks to a year and a

seventh and not long enough time for most applications.

Thanks. This is about what I figured.
 
They usually last a lot of years.
Probably because they are run at far cooler than the rated maximum temperature.

The internal heat generating mechanism is the ESR times the ripple current.
The external heat source is the Ta.
There is a further short term heat source. Leakage current times ESR.

Reduce the ripple current and reduce the Ta and the capacitors will generally last 10 times to 1000 times the lifetime quoted in the specification.
 
trusted vendors + higher temp rating 105*C + lower esr = better long lasting capacitors full stop ,
the ripple current is just a numbers game.

the Big Ones usually last much longer than the little ones.
next up how to buy Automotive Batteries LOL
 
Last edited:
higher temp rating 105*C + lower esr = better long lasting capacitors full stop ,
the ripple current is just a numbers game.

If you really believe that, you may place a nice 105°C, low esr cap of sufficient voltage, and a beefy diode across a beefy transformer secondary while you sit on that cap for a few minutes = all specs met except ripple current, that's just a numbers game.

I'd like to see your lower back region after that experiment. :eek:

Rundmaus
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.