One of my favorite boards.

Every single picture of every single revision of this board has them so i would place the caps if i were you....I don't know what they do but it would be a shame if the board fails because of them.

If it runs, then no, you shouldn't be worried about this unless there is a distinct solder bridge on those holes.

meljor wrote:

Every single picture of every single revision of this board has them so i would place the caps if i were you....I don't know what they do but it would be a shame if the board fails because of them.

Since it looks like they definitely used to be there, I tend to agree with you. Can someone who has this board please tell me what type of caps I should get to replace the missing ones?

The original caps were 1500uf 6.3v made by samxon or KZE. At least that's what my board (rev 1.06) has on it. Personally I'd advise you to replace them with rubycon, sanyo or panasonic caps - neither samxon or KZE are stellar brands.

The originals look like this:

They tend to go bad w/o any external bulging or leakage witch is annoying. I spent months on and off diagnosing two P5A boards, and was only able to fix one of the two after a complete recap. The other has a damaged northbridge (unfortunately quite common on ALi Aladdin V boards) and reports gate A20 error at boot.

Lots of rubycons are counterfeits, and panasonic caps can be expensive. These are a good replacement:

Thank you so much, kanecvr! I ordered some of the Sanyos from a USA based ebay seller!

KZEs are a Nippon Chemi Con series, so they're a reliable vendor. I'm not sure if KZE ever had batch failures as did KZG series. Still, caps failing on old hardware is just a reality. Higher quality caps just take longer to fail.

Sanyo (Suncon)/Nichicon/Rubycon/NCC/UCC are all excellent. Cornell Dubilier is good for bulk filtering but not for low ESR requirements like VRM sections. As always, know your vendor.

Do they always say if it's low esr? How can you tell?

meljor wrote:

Do they always say if it's low esr? How can you tell?

As far as I know, every manufacturer has their own color coding / markings to denote that they are low ESR (but take that with a grain of salt). The seller usually specifies. You can also make sure on the manufacturer's site.

Here's a little bit about ESR rating - what it is and why it's important : https://en.wikipedia.org/wiki/Equivalent_series_resistance

Thanx, i knew it's importance it's just that i do not always see it advertised. I didn't know about the color coding, but if i can't trust that it has no use.

meljor wrote:

Thanx, i knew it's importance it's just that i do not always see it advertised. I didn't know about the color coding, but if i can't trust that it has no use.

"Color coding" is wrong - it's not like it is for resistors, so it was the wrong expression for me to use... What I should have said is that low ESR caps are a different color or have a different color stripe in case of some manufacturers. A pertinent example is for panasonic capacitors - low ESR 1000uf caps are black / gold or black / silver, while regular caps used in stuff like audio equipment are blue / gold or blue / silver.

99% of the time a seller will tell you if the capacitor is low esr or not - or at least they will provide a product code that can be looked up on the manufacturer's website or catalog to see if the part is low ESR or not.

meljor wrote:

Do they always say if it's low esr? How can you tell?

Manufacturers (NCC for one) often mark their caps with a series label. You can then look at the datasheets to see if they're intended for low ESR situations or not.

ok, thank you guys.

gdjacobs wrote:

KZEs are a Nippon Chemi Con series, so they're a reliable vendor. I'm not sure if KZE ever had batch failures as did KZG series. Still, caps failing on old hardware is just a reality. Higher quality caps just take longer to fail.

Sanyo (Suncon)/Nichicon/Rubycon/NCC/UCC are all excellent. Cornell Dubilier is good for bulk filtering but not for low ESR requirements like VRM sections. As always, know your vendor.

Got a few examples of failed KZE series caps, but all of those were in far newer stuff than a P5A board. Might just be what you mention in this case, all electrolytic caps, no matter of quality, are to be considered perishable. At some point they all fail.