No boot, no POST, no beeps. Is my motherboard dead?

T

travisbicks

My system:-

Motherboard: (socket 939)
CPU: AMD 64 3700
Video card: Geforce 7600 GT (PCIe)
Power supply: SWEEX BA000110 650 Watt
Memory: 2 x 512MB ddr-400MHz PC 3200

I recently installed a new SATA hard drive and took the opportunity to
clean my computer. I used a can of compressed air to clean the
components and then placed a hoover nozzle inside the case to collect
the dust I’d disturbed with the compressed air.

After installing the drive I booted my computer and Win XP loaded as
normal. The new drive was recognised under “Computer Management” but I
didn’t initialise it. I used my computer for several hours that
evening and it appeared to be working as normal. The next morning I
booted the computer and downloaded some pod casts. I then inserted my
USB mp3 flash player to transfer the pod casts. When I looked at my
monitor a moment later it was displaying a Win XP blue screen error
message. Foolishly, being late for work, I didn’t read the message
properly, but I believe it mentioned an “exception error” and advised
me to remove any recently installed hardware. I turned the computer
off and thought it would be a simple matter of altering the BIOS or
removing the new drive when I returned home.

When I returned I attempted to boot my computer and found that the
monitor remained completely blank. There is power going to the
motherboard and other components as all power LEDs are lit and all
fans are operating. I then turned off and removed the new hard drive
with no change. There are no error beeps from the motherboard and I
noticed that there was no internal speaker connected to the mother
board. I therefore salvaged two internal speakers from old machines
and connected each of these in turn – still no error beeps (These
speakers had a four pin plug but with only the two outer pins wired.
The motherboard header for the internal speaker has a four pin plug
but only three pins. I am assuming that having only two wired pins on
the speakers should make no difference and they should still
function.)

To attempt to discover where the problem lay I took the following
actions:-

• I tested the monitor with another machine and found it to be
working.
• I tested my video card in another machine and found that it is
working.
• I tried a known working PCIe video card in my machine – no change.
• I disconnected all hard drives – no change.
• I reseated the CPU and heatsink – no change.
• I tried reseating both memory modules and testing each in turn – no
change
• I tried resetting the CMOS by both removal of motherboard battery
and motherboard jumper switch – no change.
• I removed all components attached to the motherboard apart from CPU
and heatsink, memory, video card, and case front panel – no change
• The ECS KN1 SLI motherboard had a “top hat flash” BIOS restore
feature. I attempted to use this with no effect.

I think I have eliminated all faults other than the motherboard, the
power supply or the CPU. As all fans are functioning, I believe the
power supply is unlikely to be faulty and, as the motherboard gives no
error beeps, I think the fault is most likely to lie with the
motherboard.

My questions are:-

1. Are there any other tests I should do to establish that the fault
is with the motherboard?
2. Could the fault be with the power supply despite the fact that
power appears to be supplied?
3. If the motherboard is faulty, is it possible or worth having it
repaired or would I be better to buy another 939 socket motherboard?
4. What might have been the cause of this fault? I can only speculate
that I might have caused damage by accidentally touching the
motherboard or through being over zealous with the air can. It seems
strange though that my computer was working for a time following the
hard drive installation and cleaning.

I apologise for the length of this post but wanted to give all
relevant information. I will be most grateful for any ideas or advice
people may have. Thank you.
 
J

John Doe

TVeblen said:
Yes, the power supply could be bad,

You need to figure out how to quote the text you're replying to.
Otherwise your reply is very difficult to follow.

Good luck.
 
T

travisbicks

News reader is set to indent the original message, but it didn't do it with
that one post. Strange.

Thanks to everyone for the help. I'll try a different power supply on
Monday and hope that's the problem.
Also, I neglected to give the model of my motherboard. It's a ECS KN1
SLI (939 socket)
 
W

w_tom

Thanks to everyone for the help. I'll try a differentpower supplyon
Monday and hope that's the problem.

In less than two minutes, you could have identified the problem or
elicited a useful response from the better informed by using a
multimeter and this procedure "When your computer dies without
warning....." starting 6 Feb 2007 in the newsgroup alt.windows-xp
at:
http://tinyurl.com/yvf9vh
Connector chart to locate each color:
http://www.hardwarebook.net/connector/power/atxpower.html

In your case, first important numbers would come from purple, green,
and gray wires both before and when power switch is pressed. Remember,
it is a power 'system' - more than just a power supply. And if the
power supply is replaced, take meter readings again to confirm the
power supply is good. Normal - a defective power supply boots the
computer. Worse, that defective power supply causes strange
intermittent failures and then complete failure months after the
warranty expired. Get the multimeter - a too so simple as to be sold
even in K-mart.
 
F

Franc Zabkar

News reader is set to indent the original message, but it didn't do it with
that one post. Strange.

OE doesn't know how to correctly quote messages that have a
"Content-Transfer-Encoding" type of "quoted-printable". Check the
headers of the messages you are replying to and you'll see what I
mean.

This *may* be a possible fix for OE:
http://home.in.tum.de/~jain/software/oe-quotefix/

Here is one MS document that explains why OE is designed to
(mis)behave the way that it does:
http://support.microsoft.com/kb/q168779/#appliesto

- Franc Zabkar
 
T

travisbicks

Yes, the power supply could be bad, even though the fans run. A PS supplies
different voltages and if the 5v is bad the 12v could still work. Test it..
You need to try a different Power Supply.

I tested the motherboard with new power supply and no change. The lack
of POST error beeps leads me to believe it’s a motherboard fault
rather than CPU. There’s no visible damage but I suppose it’s unlikely
there would be. Oh well, just have to find myself an old 9393 board.
In less than two minutes, you could have identified the problem or
elicited a useful response from the better informed by using a
multimeter and this procedure "When your computer dies without
warning....." starting 6 Feb 2007 in the newsgroup alt.windows-xp
at:
http://tinyurl.com/yvf9vh
Connector chart to locate each color:
http://www.hardwarebook.net/connector/power/atxpower.html

That’s very interesting. I’ve tested using a new power supply but I’ll
certainly perform these multi-meter tests if only to satisfy my
curiosity and so that I’ll know how to perform them next time.

Once again, thanks to everyone (especially TVeblen for the in depth
reply) for the advice and help :)
 
W

w_tom

That’s very interesting. I’ve tested using a newpower supplybut I’ll
certainly perform these multi-meter tests if only to satisfy my
curiosity and so that I’ll know how to perform them next time.

Then post those numbers here. Numbers may contain information that
only more knowledgeable can see. Your replies will only be as useful
as the information you provide. Just another reason to use the
multimeter.

What results in those replies? Learn how the entire power supply
'system' works including why the 'power supply controller' halts or
permits motherboard to operate.
 
T

travisbicks

  Then post those numbers here.  Numbers may contain information that
only more knowledgeable can see.  Your replies will only be as useful
as the information you provide.  Just another reason to use the
multimeter.

  What results in those replies?  Learn how the entire power supply
'system' works including why the 'power supply controller' halts or
permits motherboard to operate.

I have now tested my PSU with a multi-meter and I obtained the
following results:-

Upon powering up the board:-
Purple was stable at 5.14
Green decreased from 5.13 to 0.05
Gray rose from 0 to 4.69
Orange rose from 0 to 3.32
Red rose form 0.34 to 5.08
Yellow rose from 0.25 to 11.98
 
W

w_tom

I have now tested my PSU with a multi-meter and I obtained the
following results:-

Upon powering up the board:-
Purple was stable at 5.14
Green decreased from 5.13 to 0.05
Gray rose from 0 to 4.69
Orange rose from 0 to 3.32
Red rose form 0.34 to 5.08
Yellow rose from 0.25 to 11.98

No problem exists with any component in that power supply 'system'.
All numbers are well in spec. Replacing a power supply to fix a
failure would only be wasted time, wasted money, and may even add more
defects to the system. Move on to other suspects.

What those numbers says: Purple wire voltage to power supply
controller is stable. Power supply controller saw power switch
pressed; a safety lockout circuit was not activated; so controller
told power supply to power on - green wire. Power supply applied
voltage to orange, red, and yellow wires. Since those voltages were
stable in seconds, then gray wire told power supply controller to
maintain power on, and told CPU to start executing. Using about 20
times less labor, we even know the power switch and AC power cord are
OK. We know a power supply system is OK and that shotgunning a power
supply would be wasted time and money. Others would recommend
shotgunning or using a power supply tester. Neither can discover what
a multimeter has just reported.
 
T

travisbicks

  No problem exists with any component in that power supply 'system'.
All numbers are well in spec.  Replacing a power supply to fix a
failure would only be wasted time, wasted money, and may even add more
defects to the system.  Move on to other suspects.

  What those numbers says:  Purple wire voltage to power supply
controller is stable.  Power supply controller saw power switch
pressed; a safety lockout circuit was not activated; so controller
told power supply to power on - green wire.  Power supply applied
voltage to orange, red, and yellow wires.  Since those voltages were
stable in seconds, then gray wire told power supply controller to
maintain power on, and told CPU to start executing.  Using about 20
times less labor, we even know the power switch and AC power cord are
OK.  We know a power supply system is OK and that shotgunning a power
supply would be wasted time and money.  Others would recommend
shotgunning or using a power supply tester.  Neither can discover what
a multimeter has just reported.

Glad to know the PSU is not faulty. Thank you for your help. The
process was interesting and I now know how to test any suspect PSUs in
future.
 
T

travisbicks

On Mon, 28 Jul 2008 17:07:11 -0700 (PDT), travisbicks

Let us all know what you figure out.  I'll put money on the power
supply, myself.

Actually turned out to be the motherboard. I got hold of another
socket 939 board and it works fine with the same PSU, CPU, case and
memory.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads


Top