[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [Sheflug] PC Problem
On 06-Aug-05 bones wrote:
>
> A friend has recently given me a PC. At first the computer would
> just say reset, and now it seems completely dead.......it won't turn
> on. Could the power supply unit have packed in? Or can I reset the
> motherboard or something? There is no noise or whirring or nothing.
>
> I plan on putting FC4 on the PC.
>
> Cheers
>
> Doug.
On the assumption that power is not getting through somehow, you may
be dealing with one of two configurations here.
A) The power-on button is connected directly to the PSU by a mechanical
linkage, so that when you push the button the linkage physically
operates an on-off switch in the PSU itself.
B) There is an indirect connection via the motherboard. In this, the
on-off button is connected to the motherboard by a slim 2-wire cable
with a small connector at the other end, which pushes on to a couple
of tiny pins on the motherboard. The PSU is also similarly connected
to the motherboard. The two connections may be separated on the
motherboard by quite a large distance (e.g. even at diagonally
opposite corners -- it depends on the M/board). The modus operandi
of this arrangement is that when the PSU is connected to the mains
supply, an circuit internal to the PSU can detect the completion of
the circuit via the M/board when the on-off button is pressed. This
initiates a switch-on via a relay within the PSU.
In both cases the power supply to the computer is communicated to
the motherboard via sets of cables one of which (with multiple
contacts) plugs into the motherboard, and others are connected
to disk drives etc.
You can tell which of (A) or (B) applies in your case by opening
the box and simply looking.
However, further diagnostics depends on which it is.
In case (A) it is relatively simple. Disconnect the cables leading
from the PSU to anything on the computer. Connect the PSU to the
mains. Switch it on. Then use a voltmeter to test for the presence
of voltages between various leads (one of which is "Earth", others
usually being +/-5v and +/-12v). If you get the right results at
this stage then the PSU should be OK and the problem is elsewhere.
If you're not getting the voltages, then there are basically two
sources of problems. The linkage may be broken (I've met cases
where it had become detached somewhere along the line so that
pushing the button did not operate the PSU switch). You should
be able to check this by tracing the linkage to the PSU switch
and manually operating the latter yourself. You can also try
checking that the connections from the PSU switch to the transformer
etc have not become detached or broken; if so, put this right.
If none of this works then your PSU is bust.
In case (B) it gets more problematic. You should start by checking
that the two connections (from on-off switch to motherboard, and
the similar connection from PSU to motherboard) are connected,
and to the right sets of pins on the motherboard, and the right way
round. If not, then put this right and try again.
If they were properly connected, then unplug the on-off button
connection (making a note of where it came from), connect a couple
of wires with small crocodile clips to the m/board pins (care!),
connect the PSU to the mains, and complete the "on-off" connection
by touching the two wires. If the computer then starts, the problem
was in the circuit-completion via the on-off button. Test this
by pushing a couple of pins into the holes on the on-off connector,
attach your voltmeter in "resistance" mode to the pins with the
crocodile clips, push the "on-off" button, and see if you get a
circuit.
If the "on-off" button completes the circuit OK, but the computer
does not start up, then it gets more obscure. Go back to the
crocodile clips on the two pins on the m/board where the "on-off"
connector pushes on. Connect the mains to the PSU, and use the
voltmeter to check for voltage between these two pins. If absent,
you may have a broken motherboard.
At this point I run out of reliable advice (though others may
be able to take it further).
The issue with case (B) is that most recent motherboards have
the capability to switch on the computer on detection of a variety
of events, only one of which is the pushing of the on-off button.
Other include things like receiving a signal from a network
connection. This is achieved by the motherboard in fact being
partially "live" (via the thin cable connector from the PSU)
provided the PSU is connected to the mains: part of the m/baord
circuitry is devoted to detecting such events, which then send
a signal to the PSU which "switches on" to provide power to the
rest of the motherboard and the other components.
I hope this helps, but it is incomplete for case (B) and I hope
others can provide the details I'm not able to be specific about!
Best wishes,
Ted.
--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding [at] nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 08-Aug-05 Time: 13:06:12
------------------------------ XFMail ------------------------------
___________________________________________________________________
Sheffield Linux User's Group -
http://www.sheflug.co.uk/mailfaq.html
GNU the choice of a complete generation.