Three mobos, poly-modding help needed.

[racecar56]

Hello! I've got three motherboards that I would like to repair by replacing their bad, missing or risky electrolytic capacitors with polymers. Here are the motherboards and the capacitors on or missing from them that I would like to replace:

1. Biostar P4M800-M7: 3x TK ATWY 1800uF 16V 10mm, 4x TK ATWY 3300uF 6.3V 10mm
2. FIC AM37-L: 5x Nichicon HM 3300uF 6.3V 10mm, 3x UCC KZG 3300uF 6.3V 10mm
3. Acer G33T-AM: 3x TK ATWY 1800uF 16V 10mm, 5x TK ATWB 1800uF 6.3V 8mm

I want to use polymers since it seems to be anywhere from difficult to impossible to find suitable electrolytic replacements of the same sizes. I don't want to buy electrolytics too large to fit on the motherboards, and unless the polymers wind up being very expensive, I don't mind paying the extra price.

Since it seems like one does not simply replace electrolytics with polymers of the same capacitance, I would like some help picking good polymer replacements. I did look around before posting and found this thread. From that thread, I was able to come up with these as replacements:

UCC APSG160E821MH16S for all 1800uF
Nichicon RNU0J102MDN1 for all 3300uF

Could someone verify if these capacitor replacements are a good choice? If they're not, could I have better recommendations?

Thanks!

[Topcat]

Re: Three mobos, poly-modding help needed.

I've always considered polymodding in place of lytics to be very experimental....especially if the original lytics were known bad ones. Replacing the junk lytics with good quality lytics tends to be a better solution. I've done a lot of poly experiments, some were not successful....so if these machines serve a purpose, i'd recommend not polymodding unless you really know what you're doing or don't mind possibly some do-overs for stability issues.

[racecar56]

Re: Three mobos, poly-modding help needed.

Thanks for your response, founder of Badcaps.net.

Indeed, it's probably best that I stick with electrolytics. The only problem is that, after some capacitor searching, I've only come up with only one result for both 1800uF and 3300uF capacitors that are of the right diameter. They're of the Nichicon HW series. Unfortunately, both HW capacitors' ESR is 17mΩ, and I need 12mΩ. If it weren't for the fact that the Biostar board's VRM capacitors are right smack next to each other, I would go up in diameter.

Alternatively, I could buy NOS Nichicon HMs from eBay. These feature the exact specifications I need, they don't come from China, are hopefully from after 2005 (before purchasing, I'll contact him to be sure), and the seller has 100% positive feedback:
1800uF 3300uF

Should I steer clear of eBay for this project and deal with the higher ESR of the HWs, or should I purchase the HMs from eBay?

[Per Hansson]

Re: Three mobos, poly-modding help needed.

Why not support the site?

https://www.badcaps.net/store/produc...products_id=34

https://www.badcaps.net/store/produc...products_id=30

[racecar56]

Re: Three mobos, poly-modding help needed.

Unfortunately, the 1800uF ZLH only handles 2250mA and has an 18mΩ ESR. I need a ripple current rating of 2800mA+ and a 12mΩ or lower ESR rating. Otherwise, I'd be happy to get capacitors from here instead. Sorry about that.

[Topcat]

Re: Three mobos, poly-modding help needed.

Unfortunately, the 1800uF ZLH only handles 2250mA and has an 18mΩ ESR. I need a ripple current rating of 2800mA+ and a 12mΩ or lower ESR rating. Otherwise, I'd be happy to get capacitors from here instead. Sorry about that.

It won't pose any issues.....but best of luck with the fakes from ebay.

[kc8adu]

Re: Three mobos, poly-modding help needed.

while i have had 100% success polymodding it is indeed experimental.
btw many boards have unpopulated spots for vrm output caps.
you can use the zlh and fill out the empty spots if it makes you feel better.
most of the bad caps are shit anyway and the specs are not to be believed either.

[momaka]

Re: Three mobos, poly-modding help needed.

You can also use caps with slightly higher voltage rating (say 25V caps instead of 16V ones). This is an old "trick" when trying to use caps with slightly inferior specs. Reason being is that caps with a higher voltage rating tend to be physically bigger, and bigger can = lower ESR, higher ripple current (within the same series of cap, that is).

Another possibility is to go with slightly higher capacitance. In most cases, going up a notch or two on the capacitance won't cause any issues (i.e. going from 1800 uF to 2200 or 2700 uF). Again, the increase in capacitance means the capacitor will usually be physically bigger, and bigger cap can = lower ESR.

Finally, for the CPU VRM, you can also mix regular electrolytic caps with polymers. That way, the circuit will have very low ESR thanks to the polymers and capacitance much closer to the original (due to using electrolytic caps).

My preferred method is to make about half of the caps on the CPU VRM polymers (typically 820 uF caps) and mix them with electrolytic capacitors that have close enough specs to the originals and slightly higher capacitance. The increase in capacitance offsets the loss in capacitance from the poly caps, and the poly caps make up for the bulk of the combined low-ESR and ripple current specs.

The rest of the motherboard circuit should be okay with middle-low-ESR caps like Rubycon LZH/LZ/LZG and Nichicon HD/HW.