Help needed with fault

[John]

A friend of mine is having trouble with his computer, which is why I'm
posting this message for him. The motherboard is a Matsonic MS8188E -
details here:

http://www.matsonic.com/mb/productsdata/ms8188e.htm

When the power button is pressed, nothing happens - you just hear a very
slight click but the hard drive doesn't spin up and none of the fans (case
fan, CPU fan or PSU fan) operate at all.

The motherboard and PSU have the usual 20-pin ATX connector and also the
supplemental 4-pin connector that some (processors?) require. I haven't got
a spare PSU to try on it but if I disconnect the 4-pin connector, the hard
drive spins up and the case fan spins, but the CPU fan doesn't. Plug in the
4-pin connector again and nothing works.

Is this pointing to a PSU problem or mobo/cpu? Obviously, I don't want to
tell him to go and buy a PSU if it's the mobo/cpu or vice versa so all help
gratefully received.

Thanks in advance,

John.

 

[Rod Speed]

A friend of mine is having trouble with his computer, which is why I'm posting this message for
him. The motherboard is a Matsonic MS8188E - details here:

When the power button is pressed, nothing happens - you just hear a very slight click but the hard
drive doesn't spin up and none of the fans (case fan, CPU fan or PSU fan) operate at all.

That and the below indicate that the power supply is shutting down
straight away because it doesnt like the load on one of the output rails.

The motherboard and PSU have the usual 20-pin ATX connector and also the supplemental 4-pin
connector that some (processors?) require. I haven't got a spare PSU to try on it but if I
disconnect the 4-pin
connector, the hard drive spins up and the case fan spins,

That implys that its the load on that rail that's the problem.

but the CPU fan doesn't.

Likely the cpu fan is powered from that connection and the others arent.

Plug in the 4-pin connector again and nothing works.

Is this pointing to a PSU problem or mobo/cpu?

More likely to be the motherboard, very unlikely to be the cpu.

Its possible that the problem is with the PSU over current sensing,
the current isnt too high and the PSU thinks it is anyway.

Obviously, I don't want to tell him to go and buy a PSU if it's the mobo/cpu or vice versa so all
help gratefully received.

Check the motherboard for bad caps. These are the usually blue
or black plastic covered post like things sticking up vertically from
the motherboard surface. The tops should be flat. If any have bulged
or have leaked, thats a bad cap and is likely the cause of the problem.
Those often do fail in the regulator that regulates the 12V rail on that
4 pin connector down to the Vcore voltage for the cpu when those
caps with the bad electrolyte design were used.

If you have a bad cap or bad caps, they should be covered by the
motherboard warranty, even if the motherboard is technically out
of warranty, under the principle of goods not of merchantable
quality or fit for purpose which doesnt have that stated time limit.

Even if you cant see any bad caps, they can be bad but not visibily bad.

If you cant see any bad caps, it would be worth trying a new power
supply from an operation that allows you to return it for a refund.

 

[John]

That and the below indicate that the power supply is shutting down
straight away because it doesnt like the load on one of the output rails.


That implys that its the load on that rail that's the problem.


Likely the cpu fan is powered from that connection and the others arent.



More likely to be the motherboard, very unlikely to be the cpu.

Its possible that the problem is with the PSU over current sensing,
the current isnt too high and the PSU thinks it is anyway.


Check the motherboard for bad caps. These are the usually blue
or black plastic covered post like things sticking up vertically from
the motherboard surface. The tops should be flat. If any have bulged
or have leaked, thats a bad cap and is likely the cause of the problem.
Those often do fail in the regulator that regulates the 12V rail on that
4 pin connector down to the Vcore voltage for the cpu when those
caps with the bad electrolyte design were used.

If you have a bad cap or bad caps, they should be covered by the
motherboard warranty, even if the motherboard is technically out
of warranty, under the principle of goods not of merchantable
quality or fit for purpose which doesnt have that stated time limit.

Even if you cant see any bad caps, they can be bad but not visibily bad.

If you cant see any bad caps, it would be worth trying a new power
supply from an operation that allows you to return it for a refund.

Thanks for that info Rod, that's brilliant. I'll have a good look at the
mobo tomorrow and hopefully sort it one way or the other.

Cheers mate,

John.

 

[w_tom]

We can speculate forever. We can look at this and look at that. Or
in but two minutes, we can follow the evidence to identify a suspect.
Little that is visual will be helpful. Your only alternative is to
shotgun - keep buying new parts until something works.

Get the 3.5 digit mulitmeter - a tool so simple and so ubiquitous as
to be sold even in K-mart (or Lowes, Radio Shack, Home Depot, Sears,
etc).for $20. Two minute procedure is posted as "When your computer
dies without warning....." starting 6 Feb 2007 in the newsgroup
alt.windows-xp or at:
http://tinyurl.com/yvf9vh

Number posted here may then result in further useful information.

 

[John]

We can speculate forever. We can look at this and look at that. Or
in but two minutes, we can follow the evidence to identify a suspect.
Little that is visual will be helpful. Your only alternative is to
shotgun - keep buying new parts until something works.

Get the 3.5 digit mulitmeter - a tool so simple and so ubiquitous as
to be sold even in K-mart (or Lowes, Radio Shack, Home Depot, Sears,
etc).for $20. Two minute procedure is posted as "When your computer
dies without warning....." starting 6 Feb 2007 in the newsgroup
alt.windows-xp or at:
http://tinyurl.com/yvf9vh

Number posted here may then result in further useful information.

Thanks very much w_tom. I'll get a multimeter (tonight if I can - good
excuse to get it now) and get back to the group.

John.

 

[Rod Speed]

We can speculate forever. We can look at this and look at that. Or
in but two minutes, we can follow the evidence to identify a suspect.

Just another pig ignorant fantasy from this fool who cant even manage to work
out how multimeters work, or what the ATX specs say about voltage levels either.

Little that is visual will be helpful.

Just another of your silly little pig ignorant fantasys, most obviously with bad caps.

Your only alternative is to shotgun - keep buying new parts until something works.

Or you can analyse the symptoms and work out what has likely failed.

Get the 3.5 digit mulitmeter - a tool so simple and so ubiquitous as
to be sold even in K-mart (or Lowes, Radio Shack, Home Depot, Sears,
etc).for $20. Two minute procedure is posted as "When your computer
dies without warning....." starting 6 Feb 2007 in the newsgroup
alt.windows-xp or at:
http://tinyurl.com/yvf9vh

Pity that just flaunts the fact that you havent got a clue about how multimeters work.

Number posted here may then result in further useful information.

Not a chance with your mindless shit.

 

[Rod Speed]

Thanks very much w_tom. I'll get a multimeter (tonight if I can - good
excuse to get it now) and get back to the group.

Ignore his stupid pig ignorant claims about what the minimum
voltages on the rails need to be, he cant even manage to read
and comprehend what the ATX specs actually say on that.

And a multimeter cant tell you about unacceptible ripple levels on the rails either.

 

[kony]

A friend of mine is having trouble with his computer, which is why I'm
posting this message for him. The motherboard is a Matsonic MS8188E -
details here:

http://www.matsonic.com/mb/productsdata/ms8188e.htm

When the power button is pressed, nothing happens - you just hear a very
slight click but the hard drive doesn't spin up and none of the fans (case
fan, CPU fan or PSU fan) operate at all.

The motherboard and PSU have the usual 20-pin ATX connector and also the
supplemental 4-pin connector that some (processors?) require. I haven't got
a spare PSU to try on it but if I disconnect the 4-pin connector, the hard
drive spins up and the case fan spins, but the CPU fan doesn't. Plug in the
4-pin connector again and nothing works.

Is this pointing to a PSU problem or mobo/cpu? Obviously, I don't want to
tell him to go and buy a PSU if it's the mobo/cpu or vice versa so all help
gratefully received.

Thanks in advance,

John.

The click is likely the PSU overcurrent or voltage
protection circuit kicking in, turning the PSU off.

What remains is to determine if the PSU is turning off from
a fault within itself, or an external (like motherboard)
part failure.

Unplugging the connector to a part tends to implicate that
part, and your board is likely to have poor capacitors that
may fail - so I'd check those board capacitors for venting,
BUT unplugging that particular plug also drastically
decreaes the amount of 12V current the PSU tries to provide,
and if this subcircuit in the PSU had failed (progressively
failing still), it could have a similar result in the
significantly higher load causing the PSU to shut off.

Inspect these two components, and if you're so inclined,
unplug the PSU from AC for a few minutes then inspect it too
(leaving disconnected from AC).

It is theoretically possible for a CPU failure to cause
this, but not likely.

 

[John]

The click is likely the PSU overcurrent or voltage
protection circuit kicking in, turning the PSU off.

What remains is to determine if the PSU is turning off from
a fault within itself, or an external (like motherboard)
part failure.

Unplugging the connector to a part tends to implicate that
part, and your board is likely to have poor capacitors that
may fail - so I'd check those board capacitors for venting,
BUT unplugging that particular plug also drastically
decreaes the amount of 12V current the PSU tries to provide,
and if this subcircuit in the PSU had failed (progressively
failing still), it could have a similar result in the
significantly higher load causing the PSU to shut off.

Inspect these two components, and if you're so inclined,
unplug the PSU from AC for a few minutes then inspect it too
(leaving disconnected from AC).

It is theoretically possible for a CPU failure to cause
this, but not likely.

Thanks for your input guys, much appreciated. Just a quickie, regarding the
capacitors. If you look at these two photos (I know it needs a good
vacuuming out to get rid of the dust) do you think the caps to the left of
the CPU have blown? There's no leakage as far as I can see but they
definitely look to have "domed" tops compared to any other caps on the
board - the rest of them have very definite flat tops.

http://www.prestoncwu.co.uk/dvd/caps1.jpg
http://www.prestoncwu.co.uk/dvd/caps2.jpg

Thanks, John.

 

[kony]

Thanks for your input guys, much appreciated. Just a quickie, regarding the
capacitors. If you look at these two photos (I know it needs a good
vacuuming out to get rid of the dust) do you think the caps to the left of
the CPU have blown? There's no leakage as far as I can see but they
definitely look to have "domed" tops compared to any other caps on the
board - the rest of them have very definite flat tops.

http://www.prestoncwu.co.uk/dvd/caps1.jpg
http://www.prestoncwu.co.uk/dvd/caps2.jpg

Thanks, John.

Yes, at least 5 of the caps, to the left of the CPU are
failing. They aren't particularly good brands/models of
caps but not the worst either, in this case the somewhat
broad area of discoloration on the board makes it look as
though the area had overheated long term.

Having the heatsink cleaner would have allowed more airflow
to cool them, or even better a heatsink that had a far
larger fan and a PSU that was on top with a bottom intake
area and a rear chassis fan. This latter combination
might've completely prevented their demise, or at least a
few more years out of the board if nothing else failed
next/first.

 

[John]

Yes, at least 5 of the caps, to the left of the CPU are
failing. They aren't particularly good brands/models of
caps but not the worst either, in this case the somewhat
broad area of discoloration on the board makes it look as
though the area had overheated long term.

Having the heatsink cleaner would have allowed more airflow
to cool them, or even better a heatsink that had a far
larger fan and a PSU that was on top with a bottom intake
area and a rear chassis fan. This latter combination
might've completely prevented their demise, or at least a
few more years out of the board if nothing else failed
next/first.

Ah, thanks very much Kony. I'll tell him the bad news.

Thanks again,

John.