Archived from groups: microsoft.public.windowsxp.hardware (
More info?)
Hi,
Sorry for the late reply.
I have checked the voltage read outs using my new metre. They are as follows.
+3.3 is reading 3.40
+5.5 is reading 5.15
+12 is reading 11.92
and the - is reading -11.90
If I remember, the hardware doctor is reading the +12 volts to be 10.98. Is
it right in saying that this reading is being taken from the BIOS, therefore
the seems to nothing wrong with the PSU and the problem lies with the
motherboard?
Once again, sorry for the late reply
thanks for your help
Ted
"w_tom" wrote:
> Meter has two leads. All voltages are 'electrical pressure'
> between two points. The previously posted procedures describe
> measuring voltage between two colored wires. IOW black lead
> touches PSU black wire (inside white nylon connector where it
> connects to motherboard). Red wire touches PSU's red, yellow,
> orange, green etc wires to measure voltage on that wire. Just
> set voltmeter selector to DC volts, touch probes, and read
> numbers on meter. Record voltage for each colored wire. Then
> compare to voltages on chart. 3.3, 5, and 12 volt wires
> should read in upper 3/4s of limits.
>
> Sidebar: Many 450 watt power supplies do not even output
> their rated power. Just another problem with so many power
> supplies dumped into America that don't even meet Intel
> specs. This web site demonstrates a problem too common with
> imported power supplies:
>
http://www6.tomshardware.com/howto/02q4/021021/index.html
>
>
http://www6.tomshardware.com/howto/02q4/021021/powersupplies-15.html
> > Time and time again, our lab measurements were unable to verify
> > the output figures represented on the model identification
> > sticker.
>
>
> Ted wrote:
> > No, the PSU is a 450 Watt, I know what you mean, I should be getting
> > the correct details instead of wild guessing. I will buy a metre on
> > Saturday as I work shifts. I am not sure how to use a metre or where
> > I should be putting it to test, sorry if I sound stupid!! But you've
> > gotta start somewhere. Help with this is much appreciated.
> >
> > "w_tom" wrote:
> >
> > > That 'swap PSU' test only implies that something else is
> > > wrong. However assumed (and not stated): is the new PSU
> > > exactly same? In the meantime, numbers are necessary to
> > > provide useful assistance. Considering all the work and all
> > > the added risk to both your system and his PSU, it makes more
> > > sense (and saves significant time) to first take those voltage
> > > readings.
> > >
> > > Those upper and lower limits are significantly outside of
> > > acceptable voltages. If you are getting alarms with limits
> > > that far too wide, then surprising the computer is still
> > > working. Winbond voltage readings are only as accurate as
> > > hardware on motherboard. Use the meter so we know what your
> > > system is really doing. Doing anything else will only make
> > > you and us more confused. Those multimeter numbers are
> > > essential. Without them, your responses will only be wild
> > > speculation.
> > >
> > > BTW what did those other information sources (Device
> > > Manager, system log, etc) report? Worry more about getting
> > > every piece of information requested. Being confused should
> > > have come much later.
> > >
> > > Ted wrote:
> > > > I have a friend that has let me borrow a PSU, when I check Winbond
> > > > hardware Doctor with the new PSU, instead of the alarm sounding now
> > > > and then, it's on constant.
> > > >
> > > > Does that make any more sense? Im so confused???
> > > >
> > > > "w_tom" wrote:
> > > >> Obtain that meter even in a screw driver store. Meter is as
> > > >> essential as a screw driver. So ubiquitous as to be sold even
> > > >> in Home Depot, Sears, and Lowes. So inexpensive that it is
> > > >> now a standard tool.
> > > >>
> > > >> BIOS measurement circuits can vary significantly. How
> > > >> accurate are they? An honest computer provider or motherboard
> > > >> manufactuer provides that numerical specification. However a
> > > >> 'motherboard voltmeter' must be accurate enough duplicate the
> > > >> accuracy of a 3.5 digit multimeter. See spec limits in
> > > >> previous posts. Voltages must reside in the upper 3/4s of
> > > >> those limits. Anything else can mean strange and intermittent
> > > >> failures. If the meter is less accurate, then the voltage
> > > >> readings must lie even further inside those limits.
> > > >>
> > > >> Limits provided below are too wide. For example, upper
> > > >> limit is 12.6. Below, that same limit is 13 volts - too
> > > >> high. A voltage below 4.85 is too low. And yet limits below
> > > >> are 4.5 - too low.
> > > >>
> > > >> Don't screw around. Don't waste good money and time by
> > > >> shotgunning - scam that an auto mechanic does when he does
> > > >> not have basic knowledge. Get the meter to confirm an
> > > >> essential component - the PSU. Then move on to other 'usual'
> > > >> suspects using standard information sources such as system
> > > >> (event) log, Device Manager, and comprehensive diagnostics
> > > >> provided free by any reponsible computer manufacturer.
> > >
>