Windows XP help - Dell Optiplex SX260 Computer Won't Boot Up

wrote:
Hi,

Here is an update:

Using a "lo-pwr ohms" (low voltage) B&K meter, I measured about 5.5
ohms across the bank of 2200/6.3 caps numbers C146, C147, C148, etc. Note:
This reading is not caused by charging. I switched to "hi-pwr ohms" and the
reading dropped to 4.4 ohms. Note: 412mV measured (via a DMM) across B&K
meter probes during "hi-pwr" ohms reading.

It is possible that one or more of these caps is the cause of this low
ohms reading (not normal). There are about 9 or 10 caps in parallel.

I will put this on the side for now, and wait for more "freedback".

Thanks in advance, John

As a reference, try this doc. This is the regulator on
my P4C800-E Deluxe motherboard. Flip to PDF page 11 for
Figure 4.

http://web.archive.org/web/200403310...5ADP3180_0.pdf

I probed the (+) side of C1 and got 39 ohms roughly. That's the
resistance seen on the input side caps (C1 through C6). The input side
caps are put there, to "eat switching ripple" so it doesn't go back
into the ATX supply.

I probed a pad on the inductor L2, which gives me a reading
across the output caps C21 through C26. I got 5.5 ohms there.
The processor VCore is also in parallel with C21 through C26.
I was using the 200 ohm range on my multimeter.

I can't say whether this is normal or of diagnostic interest.
If I were to pursue this angle (i.e. take apart my working PC :-) ),
I would remove the processor from the socket (to remove the load
on the output side). I would unplug the ATX12V cable (which is
in parallel with the input capacitors C1 through C6. There's still
no guarantee though, that I would have an "aha" moment, after doing
that.

If you think about it, both the input side and the output side
can easily drive a resistance of that value. The output side
(as shown in my example VCore design), is rated to drive 65 amps.
And ~1 volt over 4.4 ohms would be about 0.25 amps, so negligible
compared to the capability of 65 amps. On the input side, the
ATX supply could have a rating of 18 amps max on 12V, and again,
12V divided by 4 ohms or so, would give 3 amps, which the ATX
supply could likely handle with ease. So I don't think the
resistance value is low enough to be considered a "short".
If you were measuring output side capacitors, our two measured
values are relatively close to one another.

HTH,
Paul