Machine won't Power up

[Mel]

Was fine last night. My wife doesn't remember powering it off, as we
usually don't but when I came home from work it was off. It won't
start again. When you hit the power button, I can see the fan on the
heatsync spin a few times and that is it. When you try again you get
nothing. I replaced the power supply with a known good one with same
results. I removed and reseated the memory. Same thing. Any
suggestions.

Mel

 

[Vanguard]

Was fine last night. My wife doesn't remember powering it off, as we
usually don't but when I came home from work it was off. It won't
start again. When you hit the power button, I can see the fan on the
heatsync spin a few times and that is it. When you try again you get
nothing. I replaced the power supply with a known good one with same
results. I removed and reseated the memory. Same thing. Any
suggestions.

Mel

Unplug everything from the power supply except:

- Power cord to wall outlet, UPS, or surge protector.
- One hard drive.

Have you tried yanking the power cord to the computer, pressing the
power button, and then reconnecting the power cord and pressing the
power button again? Have you checked the line voltage using a
multimeter to make sure you are not having a prolonged brownout?

When you say you see the fan rotate a couple times on the "heatsink", is
this the heatsink on the processor or inside the PSU (power supply
unit)? If it is the processor's fan, maybe it went bad and won't spin
up. Was the fan noisy before by making a lot of buzzing noise (i.e.,
the bearings had worn)? The BIOS may see the CPU fan isn't spinning, or
isn't spinning fast enough, and immediately drop power to protect the
CPU from overheating. You might have to replace the CPU fan. Do you
have a voltage controller on the CPU fan (to drop its voltage to drop
its RPM and reduce noise)? If so, you may have dropped the voltage too
far (i.e., under 7V) so the fan won't start to spin, or its RPM is too
low and the BIOS figures the fan is bad. So either remove the voltage
controller or reduce how much voltage it drops (i.e., up the fan speed).

Maybe you had a surge and it blew your motherboard, CPU, memory, or
multiple components. Even if you have the computer attached to a surge
protector, that surge protector might already have been fried by prior
surges. You also need to make sure that everything you connect to the
computer is upstream of the same surge protector. Two surge protectors,
even on the same wall outlet, that have 5-foot cords means there is 10
feet between those outlets on the surge protectors, and the impedance of
that 10 feet for a lightning strike could result in 400V, or more,
potential difference between them. You can have multiple power strips
for your equipment but they all need to be upstream of the same surge
protector. If you have a modem (internal or external), get a surge
protector with jacks for the phone line, too. A whole-home surge
arrestor is better but not always doable, especially if it isn't your
house.

A UPS is not a surge protector/arrestor (unless it has an isolation
transformer which means it has to produce the output AC itself,
preferably sinusoidal, or unless the UPS incorporates surge protection
but most don't).

 

[Henry Padilla]

Was fine last night. My wife doesn't remember powering it off, as we
usually don't but when I came home from work it was off. It won't
start again. When you hit the power button, I can see the fan on the
heatsync spin a few times and that is it. When you try again you get
nothing. I replaced the power supply with a known good one with same
results. I removed and reseated the memory. Same thing. Any
suggestions.

Mel

I just had the same thing happen to me! I went out and got a power supply
and MB. I tried the power supply first but nothing changed. I swapped out
the mainboard and now the power comes on but nothing else happens. I mean
nothing. I don't get a BIOS message on screen, nothing.

I think I'm gonna try a CPU next.

But what about the memory? If it's not in the right banks to use
Dual-channel will the PC still boot up? My old mainboard never did use the
memory at 400FSB dual-channel but it still came up.

Any ideas anyone?

Tom P.

SYSTEM SPECS
MB - 400FSB, Dual-Channel DDR, Socket A (Currently MSI KT8800 Delta)
Power Supply - 450w
CPU - AMD Athlon XP 3000+
Video - nVidia 5900 Ultra
Mem- 2 x 512 Dual-Channel DDR
HDD - 4 various sizes
DVD - Plextor 716A

 

[Vanguard]

Unplug everything from the power supply except:

- Power cord to wall outlet, UPS, or surge protector.
- One hard drive.

Have you tried yanking the power cord to the computer, pressing the
power button, and then reconnecting the power cord and pressing the
power button again? Have you checked the line voltage using a
multimeter to make sure you are not having a prolonged brownout?

When you say you see the fan rotate a couple times on the "heatsink",
is this the heatsink on the processor or inside the PSU (power supply
unit)? If it is the processor's fan, maybe it went bad and won't spin
up. Was the fan noisy before by making a lot of buzzing noise (i.e.,
the bearings had worn)? The BIOS may see the CPU fan isn't spinning,
or isn't spinning fast enough, and immediately drop power to protect
the CPU from overheating. You might have to replace the CPU fan. Do
you have a voltage controller on the CPU fan (to drop its voltage to
drop its RPM and reduce noise)? If so, you may have dropped the
voltage too far (i.e., under 7V) so the fan won't start to spin, or
its RPM is too low and the BIOS figures the fan is bad. So either
remove the voltage controller or reduce how much voltage it drops
(i.e., up the fan speed).

Maybe you had a surge and it blew your motherboard, CPU, memory, or
multiple components. Even if you have the computer attached to a
surge protector, that surge protector might already have been fried by
prior surges. You also need to make sure that everything you connect
to the computer is upstream of the same surge protector. Two surge
protectors, even on the same wall outlet, that have 5-foot cords means
there is 10 feet between those outlets on the surge protectors, and
the impedance of that 10 feet for a lightning strike could result in
400V, or more, potential difference between them. You can have
multiple power strips for your equipment but they all need to be
upstream of the same surge protector. If you have a modem (internal
or external), get a surge protector with jacks for the phone line,
too. A whole-home surge arrestor is better but not always doable,
especially if it isn't your house.

A UPS is not a surge protector/arrestor (unless it has an isolation
transformer which means it has to produce the output AC itself,
preferably sinusoidal, or unless the UPS incorporates surge protection
but most don't).

Oops, the first part of my post "Unplug ..." and the 2 bullets should
have been deleted. That was when I was going to describe how to test
the PSU (which I then deleted except for those first couple of lines)
but you said that you tried another working PSU and the problem
remained, so there would be no point in testing the PSU.

 

[Pelysma]

But what about the memory? If it's not in the right banks to use
Dual-channel will the PC still boot up? My old mainboard never did use
the memory at 400FSB dual-channel but it still came up.

If it's not in the right banks to use Dual Channel, it won't use Dual
Channel, that's all. The difference there will be subtle but may be
noticeable to those who need the extra little bit of performance.

Some mainboards won't power-up unless there is RAM in a particular slot,
typically Bank 0. And a slightly unseated memory stick or video card or a
drive with a short in it can cause this symptom.

My (perfectly normal) system briefly spins the CPU fan when plugged in, and
the LED on the mobo comes on to warn technicians that it's powered, but the
computer doesn't start until the soft power pins near the corner of the
board are shorted (normally, by hitting the button on the front). Once in a
while that tiny little pushbutton switch goes bad, sorta like the old story
of a quarter-million-dollar racecar that won't run because of a 20-cent fuel
pump fuse.

I seem to be seeing a lot of people posting this same problem in a matter of
a few days. Makes me wonder what's going on.

 

[w_tom]

Swapping parts is not a 'fastest' or conclusive solution.
First, the power supply 'system' is more than just the power
supply. Second, using a meter, one can identify a defective
supply OR confirm its integrity in less than 2 minutes.
Symptoms of other problems also could be identified in that 2
minutes.

General procedures to confirm the power supply 'system' are
summarized in two posts. For Mel, the numbers on purple,
green, and grey wire before, during, and after power button is
pressed would have been most informative. Also the numbers
(what happens) to red, orange, and yellow wires in the seconds
after power button is pressed also provides useful information
currently not available:
"Computer doesnt start at all" in alt.comp.hardware on 10
Jan 2004 at
http://tinyurl.com/2t69q and
"I think my power supply is dead" in alt.comp.hardware on 5
Feb 2004 at
http://www.tinyurl.com/2musa .
Pictures that demonstrate equipment and measurement points
are:
http://techrepublic.com.com/5102-10586-5566528.html
www.ochardware.com/articles/psuvolt/psuvolt.html

With facts as provided, useful solutions are not possible.
Insufficient information was provided. Either the power
supply 'system' is defective or it is good. Either we get the
complete answer first time and move on; or keep swapping parts
in blind speculation that something will be learned. Get the
numbers in two minutes for decisive answers.

 

[Henry Padilla]

Swapping parts is not a 'fastest' or conclusive solution.
First, the power supply 'system' is more than just the power
supply. Second, using a meter, one can identify a defective
supply OR confirm its integrity in less than 2 minutes.
Symptoms of other problems also could be identified in that 2
minutes.

General procedures to confirm the power supply 'system' are
summarized in two posts. For Mel, the numbers on purple,
green, and grey wire before, during, and after power button is
pressed would have been most informative. Also the numbers
(what happens) to red, orange, and yellow wires in the seconds
after power button is pressed also provides useful information
currently not available:
"Computer doesnt start at all" in alt.comp.hardware on 10
Jan 2004 at
http://tinyurl.com/2t69q and
"I think my power supply is dead" in alt.comp.hardware on 5
Feb 2004 at
http://www.tinyurl.com/2musa .
Pictures that demonstrate equipment and measurement points
are:
http://techrepublic.com.com/5102-10586-5566528.html
www.ochardware.com/articles/psuvolt/psuvolt.html

With facts as provided, useful solutions are not possible.
Insufficient information was provided. Either the power
supply 'system' is defective or it is good. Either we get the
complete answer first time and move on; or keep swapping parts
in blind speculation that something will be learned. Get the
numbers in two minutes for decisive answers.

And I'm glad there are people out there like you that know this stuff. All
I see are little canisters and little rolls of stuff and circuit board
material.

I can use a mulitmeter to tell if a battery has gone bad or if a wire has
been broken, I've got no idea what the other stuff means.

I've tried. I took a semester of EE in college and thankfully got a C-. It
took several students and HowStuffWorks.com to get me to understand a
transistor. I just don't get it.

I'm stuck in the school of "start somewhere, replace one piece, try again".

Tom P.
BTW, It was the mainboard but while I was fooling around I dropped the heat
sink on the CPU and had to get a new one of those too.

 

[Bob]

I'm stuck in the school of "start somewhere, replace one piece, try again".

Why not just buy a new computer.


--

Map of the Vast Right Wing Conspiracy
http://home.houston.rr.com/rkba/vrwc.html

"Whatever crushes individuality is despotism."
--John Stuart Mill, "On Liberty"

 

[w_tom]

We are not there. And you cannot see electrons. By taking
numbers using the multimeter, one provides necessary
information that empower a useful response. Once a multimeter
is placed in DC voltage mode, then the meter cannot harm the
computer. Just take and report the numbers that the meter
displays. Then others here can provide useful answers without
a 'shotgun' solution.

 

[CBFalconer]

We are not there. And you cannot see electrons. By taking
numbers using the multimeter, one provides necessary
information that empower a useful response. Once a multimeter
is placed in DC voltage mode, then the meter cannot harm the
computer. Just take and report the numbers that the meter
displays. Then others here can provide useful answers without
a 'shotgun' solution.

Please don't toppost. Fixed this one. However hamfistedness with
the meter probe can cause damage. What you want is a probe which
is mostly insulated and has a small tip. Some of the best come on
scopes, and allow you to check the ripple and noise level
simultaneously.


--
Some informative links:
http://www.geocities.com/nnqweb/

http://www.caliburn.nl/topposting.html
http://www.netmeister.org/news/learn2quote.html

 

[Pelysma]

Was fine last night. My wife doesn't remember powering it off, as we
usually don't but when I came home from work it was off. It won't
start again. When you hit the power button, I can see the fan on the
heatsync spin a few times and that is it. When you try again you get
nothing. I replaced the power supply with a known good one with same
results. I removed and reseated the memory. Same thing. Any
suggestions.

Mel

After all these posts about computers that won't power up, a neighbor handed
me one to work with, a ~3yr old HP with P4 1.3MHz and lotsa nice stuff.
Suddenly it wouldn't turn on. Hit the button, nothing visible happened
except a brief squawk from the CPU fan. She had been hearing fan noise for
a while, so we looked there first.

Removed the fan, blew the dust off of it with WD-40 (pardon the reference to
another previous thread, but this time I used it on purpose) and reinstalled
it. The fan, aided by my greased forefinger, ran for about ten seconds
before wobbling off its bearings again, but in that ten seconds the computer
started to boot. Might have continued, but I shut it off because I didn't
want to sit there with my finger on the top of the fan any longer.

So we're putting in a new fan tomorrow. HP seems to have shopped around for
the cheapest and chintziest fan that could be found in bulk, so it shouldn't
be hard to find one at least as good.