Re: [Nolug] AGP / DRI diagnostics problem solved

On Sunday 07 December 2003 11:05 pm, John Tiedeman wrote:
> On Sun, 7 Dec 2003 19:54:53 -0600
> "Andrew S. Johnson" <andy@asjohnson.com> wrote:
> >On Thursday 20 November 2003 08:29 pm, Andrew S. Johnson wrote:
> >> My computer has been locking up frequently lately. I've determined
> >> through trial and error that by not loading the agpgart and radeon
> >> modules I can use X. Everything has worked fine for the last few
> >> years, so something in the hardware is going out. I don't know if
> >> it's the video card or the motherboard. I made a quick google for
> >> diags, but I didn't find anything.
> >>
> >> Does anyone know a way to determine which it is, assuming I don't
> >> have another video card or motherboard to try the parts?
> >>
> >> Andy Johnson
> >
> >Problem solved. Since I don't want to burden everyone with a
> >long explaination and pictures, the interested may go here for
> >the whole story:
> >
> >http://www.asjohnson.com/~asjohn1/bad-caps.html
> >
> >Andy Johnson
> >
> >___________________
> >Nolug mailing list
> >nolug@nolug.org
>
> ===========================================================================
> AFAK, the voltage rating on a capacitor merely indicates the maximum voltage
> that won't damage it. So if it's rated at a higher voltage than you have,
> that's just a little harmless overkill. The really critical figure is
> microfarads (&#956;f), which, in simple terms, is how much of a charge the
> capacitor will accumulate before discharging. If this value doesn't match,
> the capacitor probably won't do the job you intend for it.

The problem wasn't the voltage rating per se, since I only need to meet the minimum.
A 2200 microfarad capacitor isn't going to be used for any timing in a fast system,
so they are obviously used for power regulation, or more specifically, ripple reduction.
In this case, a capacitor of a higher voltage rating is a problem because it is physically
larger. This is because the dielectric has to be thicker to withstand the higher voltage,
which in turn reduces the capacitance per unit area, thus making the whole larger for
a given capacity. The capacitor near the northbridge was alone as it were, and thus
it was easy. The one near the DIMM slot is a snug fit, and the one near the CPU has
to lean just because it was too close to it's neighbor. If size wasn't an issue, I might
have gotten some <grin> 20 kilofarad </grin> ones instead. If I ever got them fully
charged, I'd never have to worry about a UPS.

Andy Johnson

P.S. Hmm.... I wonder how the cost of a kilofarad (if there is such a thing) compares
to the cost of a kilobyte.

___________________
Nolug mailing list
nolug@nolug.org
Received on 12/08/03