BlownRecap: Ripple Current ?

[Fizzycapola]

Re: BlownRecap: Ripple Current ?

Monitoring voltages is key to knowing how bad your capacitors are. You replace bad because they supply the wrong voltage / unsteady voltage.

Monitoring in BIOS you can get an impression of how bad the issue is, but the load is constant. Monitoring in Windows where the load is constantly fluctuating is also worthwhile, seeing what effect spinning CD's, dis-engaging Idle mode or more advanced tests such as changing BUS power timing window (a technique for reducing heat on hot running systems without affecting performance at all).

On my first recap here, deciding to recap was based on my noticing over time of widening gaps in voltages,
https://www.badcaps.net/forum/showthread.php?t=2785

(*note if one wants to prevent cap tops from shorting on SMD parts of graphics cards (something that happened to me unfortunately despite being careful - the cause was replacement caps beint just a millimetre or so larger than original) I now recommend using electricians liquid tape instead of sellotape.)

After replacing bad capacitors I think you should go right in BIOS and see the difference, as in my recap, it was immediate, fluctuation was reduced from ~10% down to an admirable 2% even in Windows on an overnight test with mixed loads, any devitation over 2% was set to trip the alarm and wake me up and it stayed super stable thanks to the Rubycon ZL series.

[PCBONEZ]

Re: BlownRecap: Ripple Current ?

That's too many questions for one post..
You are mixing some things up and I'm not sure what-all because you are talking about too many things at once. Resistance relates to DC and that's not I think you are asking. ESR is AT SOME frequency.

First off realize that capacitors on motherboards don't all do the same thing at that makes answering you general questions hard. - You have VRM input (PSU noise), VRM output (MOSFET noise), POL regulator caps (point of load), the general scattered caps (which augment the PSU caps and compensate for 'wire run' stability problems), and caps on I/O ports which are typically smaller uF and higher volts and are there more for EMI than for PSU ripple.

>> Does less resistance automatically mean increased current?
Yes if the voltage across it is the same.
Bear in mind with caps this is voltage at some frequency.
The response will be different at other frequencies.

>> What do higher frequencies do to cap capacity?
Nothing, the capacitor and it's capacitance are the same.
The circuit will however respond differently at different frequencies.

>> What happens when too much current is run through a cap?
1- It gets hot.
2- The oxide layer may be damaged resulting in internal shorts.

>> Question is, what would've caused it/them to overheat?
Inadequate ventilation.
Too close to something hot (CPU, MOSFETS).
Excessive noise (ripple) from the PSU (often due to bad PSU caps.)
Other caps in parallel have failed so it's over worked.
Internal shorts due to, crappy electrolyte, crappy aluminum, poor construction, and/or deterioration.

Note: Internal shorts may only affect small areas of the surface (like pin hole leaks) and then grow or increase in numbers.

>> Confirming that longer leads wouldn't take that cap out of spec?
No, difference in the length of the lead is insignificant electrically.

>> What are the consequences of too little resistance (or too much) in the typical DC or filter setup ?
[Caps filter the DC by removing the AC. I'm taking resistance to mean ESR and ESR changes with the frequency. It is not the same across the entire spectrum.]
Too much and it won't filter out the ripple.
Zero would be a dead short *at that frequency* and as long as the cap could handle the ripple current it shouldn't be a problem. [I'm foggy on this last one. There may be some problem I'm not thinking of.]

That's enough from me for now. (It's late.)

.

[Alidor]

Re: BlownRecap: Ripple Current ?

I know months have gone by but I just recapped a board that had backwards cap markings on it. Could this be the case with yours? I noticed in your picture of the extended cap that the negative lead went to the white crescent.

My board had all of these on the positive side!
P4P800 Deluxe - I'm serious. (weirdos)

Just a thought. It sure would explain things.

[kc8adu]

Re: BlownRecap: Ripple Current ?

its an asus/assrock thing.
gotta watch out for that.
abit is marked correctly.

[Fizzycapola]

Re: BlownRecap: Ripple Current ?

Can you not simply tell positive negative with a multimeter though? than remember what mfg is tricky.