Example .... https://www.digikey.com/en/products/detail/ni … 332MPD9/2598121 --> obsolete, click on substitutes... https://www.digikey.com/en/products/detail/ni … J332MHD/2598041

You need to use polymer capacitors if you want to match the ripple rating. My suggestion is to forget about capacitance -- low ESR and bulk ripple suppression is what's important for that VRM, not overall capacitance.

So focus on matching the diameter (10mm), and use 5V or higher parts for the VRM high (or 16v parts, if you want to use this mod), and 2.5V or higher for the VRM low.

giantclam wrote on 2023-10-16, 23:38:

Example .... https://www.digikey.com/en/products/detail/ni … 332MPD9/2598121 --> obsolete, click on substitutes... https://www.digikey.com/en/products/detail/ni … J332MHD/2598041

checked them all, it appears they are not good. Another person from the forum provided me the originals caps info:
rubycon MBZ 3300uf 6.3v (3300uf 10x23mm ... rated ripple current mA r.m.s/105C, 100kHz 2800 ... ESR mohms max/20C,100kHz 12 )

if i havet to follow:
1. = or higher ripple current then the originals
2. = or lower ESR/impendance then the originals.

from those "substitutes", :

this one https://www.digikey.com/en/products/detail/ni … J332MHD/2598204
has higher impedance (16) and lower ripple (2500)

this one https://www.digikey.com/en/products/detail/ni … J332MHD/2598041
has higher impedance (21) and lower ripple (2360)

now i undestand why its hard to find some old replacement capacitors.

Bump your search to 10vdc working

giantclam wrote on 2023-10-17, 01:46:

Bump your search to 10vdc working

still no good match

-The close one is this one from the screenshot below, but the ESR is 17 instead of 12
- When i go to 16v things start to match, but sadly , the diameter of the caps is 12.5 and not 10mm so they would not fit correctly on the motherboard.
- I also have some doubts on what should i check for ripple, should i be looking at the ripple current high frequency or low frequency? I am only making sure that the 100kHz ripple current HIGH frequency is bigger than the original cap.

You sure they won't fit? (they'll be right next to each other, but the silkscreening suggests 12.5mm is doable)

giantclam wrote on 2023-10-17, 02:44:

You sure they won't fit? (they'll be right next to each other, but the silkscreening suggests 12.5mm is doable)

12.5mm seems too much, i have a ruler at hand, might work with only 3 caps in a row, and if you move the ones on each side to make space. The problem is the top 5 caps in a row,
maybe 3dprinting some kind of base for some capacitors with holes , in order to gain more space so that some caps can be moved to bottom, up.

I'm not an electronics engineer, so take this advice for what it is.
1. Just match the capacitance and make sure you get >= voltage of the original. Then just try to get as low ESR are possible without laboring over it.
2. Stop quibbling over the other specs. These circuits are built with tolerance and, frankly, these characteristics vary with temperature and frequency anyway. They also change over their lifetime. The engineers that designed the power circuits likely built in a decent amount of tolerance.
3. If you absolutely have to hit some ESR target, I guess you can dig into the polymers

People being worried about capacitors is basically an Internet meme at this point.

smtkr wrote on 2023-10-17, 03:14:

I'm not an electronics engineer, so take this advice for what it is. 1. Just match the capacitance and make sure you get >= vol […]
Show full quote

I'm not an electronics engineer, so take this advice for what it is.
1. Just match the capacitance and make sure you get >= voltage of the original. Then just try to get as low ESR are possible without laboring over it.
2. Stop quibbling over the other specs. These circuits are built with tolerance and, frankly, these characteristics vary with temperature and frequency anyway. They also change over their lifetime. The engineers that designed the power circuits likely built in a decent amount of tolerance.
3. If you absolutely have to hit some ESR target, I guess you can dig into the polymers

People being worried about capacitors is basically an Internet meme at this point.

True that =)

Here you go:

https://www.digikey.com/en/products/detail/ke … AAE013/13420067

This is 10mm, has double the ripple capacity (7785mA) and the same ESR (13mOhm), but it's 2200uF. You do not need to match the capacitance. I would even recommend going lower than 2200uF.

I hesitate to recommend Kemet, so feel free to choose a better brand if you want.

I got interested on this topic and was mind blowing for me. I have read so many things about capacitors and people posting that if the motherboard used electrolytic then we should use electrolytic as replacements.
Also read about the power supply special thin caps that had no replacements available. I even remember a video from phils computer lab where he used polymer to recap a board and there were not really nice comments about that from people.

So reading the posts, now it is fine to use Polymer capacitors near the VRMs? , if i remember correctly on polymer you need to use the same voltage from the original cap. but i did not know that you could change a 3300uf electrolytic cap with a 2200uf polymer cap! because the UF value is a rule to follow when changing electrolytic capacitors!

SETBLASTER wrote on 2023-10-17, 04:47:

So reading the posts, now it is fine to use Polymer capacitors near the VRMs? , if i remember correctly on polymer you need to use the same voltage from the original cap. but i did not know that you could change a 3300uf electrolytic cap with a 2200uf polymer cap! because the UF value is a rule to follow when changing electrolytic capacitors!

Polymers are pretty much the only choice for VRM caps on motherboards that used ultra-low ESR electrolytics (starting around the XP/S478 era or so), simply because electrolytics of that type like Rubycon MBZ are no longer made (and they were not very reliable to begin with either, especially in hot areas like the VRM -- in the best scenario they had a more limited lifetime and this applies even to good brands).

The "capacitance rule" is not necessarily per-capacitor, you have to look at the whole circuit. In the case of a motherboard, the VRM input and output have certain requirements such as a minimum total capacitance and a maximum total ESR. Manufacturers select capacitors to comply with those requirements. When using electrolytics, they had to go with much more total capacitance than required in order to meet the ESR requirement. This is because within the same series, larger caps have lower ESR (in fact, looking at datasheets you can easily tell there is a direct relationship between case size and ESR).

So, if you go with polymers, which provide much lower ESR than electrolytics at the same capacitance, you can use less total capacitance as long as you are above the minimum. Now, how to calculate the minimum is the tricky part and requires certain knowledge about the VRM circuit design which is way beyond my level. However a member of this board once provided a "guesstimate" that around half the original capacitance would be enough if using polymers. In that sense, YMMV.

Regarding the voltage rating, it's just an upper limit of what the capacitor can handle. So, on a 12V VRM input you will probably want to use 16V caps, and on the output more than whatever the max. CPU voltage supported by the motherboard is (after adding a certain safety margin of course).

Imito wrote on 2023-10-16, 21:14:

I would like to have those 8 replaced, or maybe the full motherboard caps replaced with high quality caps.

There is no point in blindly recapping stuff. Please describe the nature of instabilities you are experiencing with motherboard. Can it run say memtest86+ or 3DMark on loop for hour or two?

As mentioned, it's difficult/impossible to find newly manufactured liquid electrolytic caps that match the ESR used on early 2000s era motherboards, because everybody discontinued them.
I have some old stock Rubycon MBZs and such which I have occasionally used on my own boards, but I know using old caps is controversial.
Some newer series (with more conservative reliability standards) are getting closer to the levels of low ESR that those older motherboard caps had. But they don't equal them yet.

I've had good results with smaller polymers that had lower ESR than original, but I've only used them at CPU Vcore, memory, and on GPU voltages. I don't know if it's okay to lose capacitance in other locations like the supply rails. Since they often have lower voltage ratings, you might have to trace the connections to make sure what voltage is connected to each cap location. Soldering may also be harder - the polymer caps I've used have thicker leads so it's harder to clear the holes enough to get them in.

As far as ripple:
The ESR of the cap and the required ripple rating are related.
The ESR implies how much ripple current could potentially flow through the cap. More ESR => less ripple current can flow. So if the ESR is higher, then it's okay for the ripple rating to be lower in the same proportion.

I don't know how formally valid this is, but I tend to look at (ESR*ripple) due to this relationship.
The durability of a 12mOhm cap with 2800mA ripple would equal a 17mOhm cap at 1977mA ripple.

Higher ESR is not what you want, but if that's what you have to use, then it does mitigate how high the ripple rating needs to be.

How bad are the caps you found on the board? Higher ESR means the cap is less effective at filtering noise, which could stress the components on the board. "General purpose" caps are unsuitable because of this - their ESR is astronomically high compared to the originals, so they can't do their job remotely close to how the originals did.
But if they're just in the range of ~2-3x the original ESR, and the board is working, then I wouldn't worry much about it unless they're junk brands that you're worried about failing later.

My opinion, without being an electronic engineer or anything similar and having a great lack of technical knowledge about electronics, is that there may be more than a hundred threads on capacitors in this forum and more or less the same thing as in this thread are always repeated.
Personally, I don't like spending a lot of money on old hardware, so I always use polymer capacitors that I get from motherboards that I can't repair.
My rule is half the capacitance and I check the voltage that the capacitor has to withstand in that place, for example the VRM capacitors of the processor, if they are Socket 462/370/478 or newer, do not exceed 2V unless you do a high CPU overvolting, so I use 2.5V capacitors, and sometimes I add an MLCC on the opposite side in parallel, in the case of boards that do not have the 12V auxiliary connector, almost everywhere the 6v capacitors or 6.3v are valid.
This post from PCBONEZ clarified many things for me, and he explained it a lot better than I can do it. Re: Capacitors of Asus TUV4x and Gigabyte GA-7VT600 1394
I put some example images.
In the case of the A7N8X-X the three 470uf and 16V capacitors get quite hot, possibly because at 5v they have to withstand a fairly high current and the capacitance is very low, the originals were 1200uf I think I remember, maybe more, but it is designed for a future mod and use an auxiliary 12v connector so I imagine that the current they have to withstand will be less.
The other two motherboards do not present any temperature or stability problems, in fact, the temperatures are lower than with electrolytic capacitors.
I repeat that I am not an electronic engineer and I lack a lot of knowledge and surely there will be those who say that it is crazy and that one day I will cause a fire but in years it has not happened yet.
Nobody pay attention to me, I am not responsible for anything.

Hoping wrote on 2023-10-17, 14:15:

<snip>

Hey, how is that Aopen AK73? Do you like it? I have one and I was thinking about re-capping it. Anything special about it?

mockingbird wrote on 2023-10-18, 00:31:

Hey, how is that Aopen AK73? Do you like it? I have one and I was thinking about re-capping it. Anything special about it?

It is an AK73(A) that is, KT133A+686B. With the latest version of BIOS it does not start with the Athlon XP, which forces you to get one of the few T-Birds that exist with a 266 bus if you want to take advantage of it. I tried an XP 1500+ and it does not start with the bus at 133 but with the bus at 100 it does, but it does not recognize it as Athlon XP. I have a Soltek SL-75KAV that does recognize the Athlon XP and uses the same chipset. I did this today because I didn't remember anything about this motherboard to respond. The truth is I don't find anything special apart from the option to enable Sound Blaster emulation in the bios for MSDOS, for me this is not very useful but for someone it could be.
I did the tests with a T-bird 1000Mhz, with the 100Mhz bus I don't find any problems, but with the 133Mhz bus (1333Mhz) it does strange things, it seems stable for a random time, it doesn't fail in memtest.....etc, but suddenly it fails in some weird way. It does not have CPU voltage control.
I found it curious to do the poly-mod because the motherboard is black and using those capacitors makes it look much newer or something like many modern retro motherboards but without being truly modern.

Hoping wrote on 2023-10-18, 18:34:

It is an AK73(A) that is, KT133A+686B. With the latest version of BIOS it does not start with the Athlon XP, which forces you to get one of the few T-Birds that exist with a 266 bus if you want to take advantage of it. I tried an XP 1500+ and it does not start with the bus at 133 but with the bus at 100 it does, but it does not recognize it as Athlon XP. I have a Soltek SL-75KAV that does recognize the Athlon XP and uses the same chipset. I did this today because I didn't remember anything about this motherboard to respond. The truth is I don't find anything special apart from the option to enable Sound Blaster emulation in the bios for MSDOS, for me this is not very useful but for someone it could be.
I did the tests with a T-bird 1000Mhz, with the 100Mhz bus I don't find any problems, but with the 133Mhz bus (1333Mhz) it does strange things, it seems stable for a random time, it doesn't fail in memtest.....etc, but suddenly it fails in some weird way. It does not have CPU voltage control.
I found it curious to do the poly-mod because the motherboard is black and using those capacitors makes it look much newer or something like many modern retro motherboards but without being truly modern.

Ah, my mistake, it is the AK77-333 I own... You mean T-Bred, not T-Bird, big difference.

But I have some advice for you nevertheless:

1) Get *any* thoroughbred you want, doesn't matter which, and mod it to a Mobile. I've done this, works perfectly fine.
2) Use bios patcher on your BIOS. Only do this if you have an external programmer so you can revert back. Use the switch that disables the backup feature or you might have issues.
3) Use the pin mod guide to figure out the multiplier you want and then solder the wires on the back of the PCB

With this method you get great flexibility... Forget about 133Mhz FSB, that doesn't offer any benefit with the measly SDRAM. Just do something like 100 * 18 for 1.8Ghz on something like a AXDA2200DKV3C which is widely available for next to nothing.

But you need the skills for this...

Hoping wrote on 2023-10-17, 14:15:

I repeat that I am not an electronic engineer

Well I see some mistakes. On yellow boards, capacitors with blue markings do not look good - only with red and black ones.
On blue ones - everything except with green markings.
And everything is fine. 😀
Because everything else is "urban legends" about the instability of the voltage regulator.

PS. My gigabyte S754 had instability in the memory power regulator. There was instability with polymer capacitors. There was instability with electrolytic capacitors, there was instability without capacitors.
Because the designer added extra transistors for S1/S3 energy saving states.
It was the extra ones, I had to be simulated its in Microcap.