- Aug 25, 2001
- 56,450
- 10,119
- 126
I came across an interesting article on overclockers.com (of a technical nature, not one of Ed's biased political diatribes), about adding a capacitor inline in parallel with the molex connectors coming off of the PSU.
I would like to get the opinion of some of the tech-heads on AT, is this idea really, really smart, or really, really stupid? My first instinct upon reading the opening part of the article was, "a generic 300W for overclocking, and having problems? upgrade the PSU dummy!". Then I got to thinking, hmm, maybe this isn't a bad idea. It's basically analogous to the "reserve power" capacitors that are placed nearest the power/gnd pins of bigger ICs on circuit boards.
But I wonder what the real issue is here. I don't think that DVD drive spinup times are short enough, that adding a small additional cap would provide the additional power needed during that period for the drive. What I am suspecting is, that the generic PSU is actually supplying "noisy power", and that adding the cap is actually filtering out the noise moreso than briefly supplying the additional power needed during the spinup period.
I'm no EE, I'm a software guy, so if any hardware-heads want to comment, I'm all ears. Thanks.
I would like to get the opinion of some of the tech-heads on AT, is this idea really, really smart, or really, really stupid? My first instinct upon reading the opening part of the article was, "a generic 300W for overclocking, and having problems? upgrade the PSU dummy!". Then I got to thinking, hmm, maybe this isn't a bad idea. It's basically analogous to the "reserve power" capacitors that are placed nearest the power/gnd pins of bigger ICs on circuit boards.
But I wonder what the real issue is here. I don't think that DVD drive spinup times are short enough, that adding a small additional cap would provide the additional power needed during that period for the drive. What I am suspecting is, that the generic PSU is actually supplying "noisy power", and that adding the cap is actually filtering out the noise moreso than briefly supplying the additional power needed during the spinup period.
I'm no EE, I'm a software guy, so if any hardware-heads want to comment, I'm all ears. Thanks.