Help Diagnose Failure

  • Thread starter Thread starter Ken Hall
  • Start date Start date
K

Ken Hall

I've put together several computers, but they just plug together and
I've had almost no trouble/problems, so I'm not real good at
diagnosing failures -- particularly when the system does nothing.

Sometime today while my wife's computer was sitting idle it died.
Everything was working this morning. When she came back this after
noon it was dead. It does nothing. Turning it off and back on has no
effect.

The power lamp on the front glows amber. It seems like it should be
green but I'm not sure any more. There is no other sign of life. If
I turn off the power switch on the back the lamp goes out. When I
turn the switch back on the lamp lights immediately -- without me
pressing the power button. Normally when it's on the optical mouse's
butt would glow -- it doesn't glow now.

It's a SOYO K7VTA Pro motherboard. I built the system about 3 years
ago and it's been used every day since.

Before I open it up without knowing what to look for, has anybody got
any suggestion on how to isolate the problem?

Ken
 
My diagnosis is that the power supply unit has failed. I would recommend
getting a hold of a working PSU and install it and see what happens.
 
Ken Hall said:
I've put together several computers, but they just plug together and
I've had almost no trouble/problems, so I'm not real good at
diagnosing failures -- particularly when the system does nothing.

Sometime today while my wife's computer was sitting idle it died.
Everything was working this morning. When she came back this after
noon it was dead. It does nothing. Turning it off and back on has no
effect.

The power lamp on the front glows amber. It seems like it should be
green but I'm not sure any more. There is no other sign of life. If
I turn off the power switch on the back the lamp goes out. When I
turn the switch back on the lamp lights immediately -- without me
pressing the power button. Normally when it's on the optical mouse's
butt would glow -- it doesn't glow now.

It's a SOYO K7VTA Pro motherboard. I built the system about 3 years
ago and it's been used every day since.

Before I open it up without knowing what to look for, has anybody got
any suggestion on how to isolate the problem?
Click to expand...

My 1st guess would also be the PSU. But I'd take the old one out and test it
before trying a new one. There are loads of posts here about testing the
PSU. I think it goes something like; remove PSU, connect a single load
(typically a spare HD), then power on and measure the voltages on the PSU
connector(s).
Does the PSU fan spin?
 
My 1st guess would also be the PSU. But I'd take the old one out and test it
before trying a new one. There are loads of posts here about testing the
PSU. I think it goes something like; remove PSU, connect a single load
(typically a spare HD), then power on and measure the voltages on the PSU
connector(s).
Does the PSU fan spin?
Click to expand...

To whom it may concern, it was the PSU. I replaced it with a cheapo
from an old case and away she went. I'm considering buying a better
PSU.

Graeme -- I would certainly have done the test you suggest if I had
seen your post in time. No, the fan did not run.

Thanks to all who replied.

BTW, can a poor quality PSU harm anything or does it just cause data
problems?

Ken
 
On Fri, 14 Jan 2005 11:02:47 -0600, Ken Hall

| On Fri, 14 Jan 2005 15:45:47 -0000, "Graeme"
|
| >My 1st guess would also be the PSU. But I'd take the old one out and test it
| >before trying a new one. There are loads of posts here about testing the
| >PSU. I think it goes something like; remove PSU, connect a single load
| >(typically a spare HD), then power on and measure the voltages on the PSU
| >connector(s).
| >Does the PSU fan spin?
|
| To whom it may concern, it was the PSU. I replaced it with a cheapo
| from an old case and away she went. I'm considering buying a better
| PSU.
|
| Graeme -- I would certainly have done the test you suggest if I had
| seen your post in time. No, the fan did not run.
|
| Thanks to all who replied.
|
| BTW, can a poor quality PSU harm anything or does it just cause data
| problems?

Usually not, although an event that causes a PSU to fail can sometimes
affect other system parts as well.

Larc



§§§ - Change planet to earth to reply by email - §§§
 
Graeme's test takes too long, requires too much work, does
not always detect the failure - leaves some problems
undiscovered. There is no faster solution to a defect supply
than using a 3.5 digit meter test. Your problems still could
be due to marginal conditions from, for example, a motherboard
power supply controller. A different supply was not a
sensitive to that marginal signal and therefore worked.

Functions that were defacto standard even 30 years ago are
missing in many power supplies dumped into N America. Asian
manufacturers have discovered a ripe market of computer
assemblers who do not even know how electricity works.

Tom's Hardware demonstrated only one example of the problem:
http://www6.tomshardware.com/howto/02q4/021021/index.html

Numbers so essential to any minimally acceptable power supply
are demonstrated by abridged specs that should be provided
with each minimally acceptable supply:
Specification compliance: ATX 2.03 & ATX12V v1.1
Acoustics noise 25.8dBA typical at 70w, 30cm
Short circuit protection on all outputs
Over voltage protection
Over power protection
100% hi-pot test
100% burn in, high temperature cycled on/off
PFC harmonics compliance: EN61000-3-2 + A1 + A2
EMI/RFI compliance: CE, CISPR22 & FCC part 15 class B
Safety compliance: VDE, TUV, D, N, S, Fi, UL, C-UL & CB
Hold up time, full load: 16ms. typical
Efficiency; 100-120VAC and full range: >65%
Dielectric withstand, input to frame/ground: 1800VAC, 1sec.
Dielectric withstand, input to output: 1800VAC, 1sec.
Ripple/noise: 1%
MTBF, full load @ 25°C amb.: >100k hrs

How many of those supplies provide such information. Only
some. For example one sold in CompUSA violates those
necessary requirements. 'Forgetting to include specs' is
necessary to dump inferior $25 supplies into the market.
Missing specs immediately should raise suspicions.

Your original question: Yes, many power supplies are
missing essential functions that can damage other computer
components. Even if a power supply fails catastrophically,
other computer hardware still must never be damaged. Again,
that was a defacto requirement 30 years ago and is required by
Intel standards.

To include essential functions, a minimally acceptable power
supply retails for about $65. Price does not prove quality.
But a price too low is another symptom of a defective power
supply.
 
w_tom said:
Graeme's test takes too long, requires too much work, does
not always detect the failure - leaves some problems
undiscovered. There is no faster solution to a defect supply
than using a 3.5 digit meter test.
Click to expand...

Well, the only difference between my test and yours, is that I suggested
disconnecting everything except a single load. Which isn't really likely to
increase the length of a test by more than a minute. An overloaded PSU may
shutdown, showing symptoms that it is dead.
I was suggsting that it is sensible to determine the cause, without
resorting to a blind replacement.
 
Back
Top