Re: Replacing 8 Caps on motherboard and 2 on PSU - which ones are good replacements?
Well, the mixing was meant for the caps around the CPU. Basically, you mix a few high-capacity (but still fairly low ESR) electrolytics with a few polymers. The polymers basically do the majority of the work as they have the lower ESR of the two, and the high-capacity caps do a bit less work, but they keep the circuit closer to what the controller feedback loop was designed for.
So for the 3300 uF KZG, for example, if you were to use all polymers, you'd probably find it hard to get polymers larger than 1500-1800 uF (if even that.) Thus, the overall capacity in the circuit will end up being about half. Most CPU VRM circuits will work fine with that, but a very few older ones with low-switching frequency (typically employing a TL494/DBL494/KA7500 PWM controller) may not be too happy. I'm almost certain your board won't have an issue with polymers either. But again, for the sake of keeping things closer to how they were designed, it would be better if the capacity was a little higher.
This is where polyer+electrolytic mixing can come helpful. If you replace the above KZG with, say, 2-3x 3300-3900 electrolytics (Panasonic FR, FM, of FS perhaps) and 2-3x 1500 uF polymers, the overall capacity won't drop to half, but only about 22-33% lower than the original... which is a lot better than 50% lower capacity.
I've been doing it on a lot of my own motherboards (after seeing a few OEM boards done that way) and it's always worked fine.
Anyways, it's OK that you are going all with Panasonic FR/FM for the motherboard. KZG aren't that much better in terms of ESR and RC (especially not when they bulge/go bad.)
Originally posted by Macieyerk
View Post
So for the 3300 uF KZG, for example, if you were to use all polymers, you'd probably find it hard to get polymers larger than 1500-1800 uF (if even that.) Thus, the overall capacity in the circuit will end up being about half. Most CPU VRM circuits will work fine with that, but a very few older ones with low-switching frequency (typically employing a TL494/DBL494/KA7500 PWM controller) may not be too happy. I'm almost certain your board won't have an issue with polymers either. But again, for the sake of keeping things closer to how they were designed, it would be better if the capacity was a little higher.
This is where polyer+electrolytic mixing can come helpful. If you replace the above KZG with, say, 2-3x 3300-3900 electrolytics (Panasonic FR, FM, of FS perhaps) and 2-3x 1500 uF polymers, the overall capacity won't drop to half, but only about 22-33% lower than the original... which is a lot better than 50% lower capacity.
I've been doing it on a lot of my own motherboards (after seeing a few OEM boards done that way) and it's always worked fine.
Anyways, it's OK that you are going all with Panasonic FR/FM for the motherboard. KZG aren't that much better in terms of ESR and RC (especially not when they bulge/go bad.)
Comment