Motherboard cannot turn on the PSU

[Pobyms]

It is a msi K7N2 Delta-ILSR

And my problem is

-I press the ON button and NOTHING happens(No fans spinning, n
lights)
-I have checked the PSU and it works perfectl
-The buttons in the case and battery in the mobo are OK as wel
(reseted the bios as well

-I have tested with a multimeter the voltages at the PSU connector an
they are OK (the 5VSB is around 5V
-The problem comes when I press the ON button. At this time the PS_O
voltage in the PSU (green cable) must change from around 5V to 0V. I
my PC this voltage does not change at all. In other words, the puls
at the ON switch does not reach to the PSU

What should I check

PS: If I force the PSU to be on (shorting PS_ON and GND) the PC work
pretty well so that I guess the processor, memory, graphics card..
are not a proble

Now my mobo is out of the case and I have connected only a basi
graphics card, one stick of memory and a HD (to reduce the powe
consumption I have no DVD or CD-ROM devices

I would like to know the path of the ON signal from the switch to th
PSU in order to check it with the multimeter. It could be a dea
component (capacitors look OK, though)

Thanks in advanc

 

[old jon]

It is a msi K7N2 Delta-ILSR.

And my problem is:

-I press the ON button and NOTHING happens(No fans spinning, no
lights).
-I have checked the PSU and it works perfectly
-The buttons in the case and battery in the mobo are OK as well
(reseted the bios as well )

-I have tested with a multimeter the voltages at the PSU connector and
they are OK (the 5VSB is around 5V)
-The problem comes when I press the ON button. At this time the PS_ON
voltage in the PSU (green cable) must change from around 5V to 0V. In
my PC this voltage does not change at all. In other words, the pulse
at the ON switch does not reach to the PSU.


What should I check?

PS: If I force the PSU to be on (shorting PS_ON and GND) the PC works
pretty well so that I guess the processor, memory, graphics card...
are not a problem

Now my mobo is out of the case and I have connected only a basic
graphics card, one stick of memory and a HD (to reduce the power
consumption I have no DVD or CD-ROM devices)

I would like to know the path of the ON signal from the switch to the
PSU in order to check it with the multimeter. It could be a dead
component (capacitors look OK, though).

Thanks in advance

Short the 2 pins on the Mboard. PO\Grnd. if the PSU fires up then the fault
is in the front switch\or wires. If not then, the fault is in the Mboard
tracks.
bw..OJ

 

[David Maynard]

It is a msi K7N2 Delta-ILSR.

And my problem is:

-I press the ON button and NOTHING happens(No fans spinning, no
lights).
-I have checked the PSU and it works perfectly
-The buttons in the case and battery in the mobo are OK as well
(reseted the bios as well )

-I have tested with a multimeter the voltages at the PSU connector and
they are OK (the 5VSB is around 5V)
-The problem comes when I press the ON button. At this time the PS_ON
voltage in the PSU (green cable) must change from around 5V to 0V. In
my PC this voltage does not change at all. In other words, the pulse
at the ON switch does not reach to the PSU.


What should I check?

PS: If I force the PSU to be on (shorting PS_ON and GND) the PC works
pretty well so that I guess the processor, memory, graphics card...
are not a problem

Now my mobo is out of the case and I have connected only a basic
graphics card, one stick of memory and a HD (to reduce the power
consumption I have no DVD or CD-ROM devices)

I would like to know the path of the ON signal from the switch to the
PSU in order to check it with the multimeter. It could be a dead
component (capacitors look OK, though).

Frankly, I think the problem is from the switch to being plugged into the
wrong pins on the header.

 

[Pobyms]

I have checked the power switch connection many times. I have even
tried to turn it on by using a flat screwdriver to short the power
switch pins. The problem is not there (unfortunately).

 

[George Pontis]

I have checked the power switch connection many times. I have even
tried to turn it on by using a flat screwdriver to short the power
switch pins. The problem is not there (unfortunately).

This is a long shot: have you checked the battery on the motherboard ?

--

 

[Guy]

Try switching out mouse and keyboard, or disconnecting one at a time
if you do not have spares. If either is bad, the system will not
power on.

If these are not the problem, you need to pull all accessories out and
go down to a bare system -- check RAM, video card, etc., until you can
isolate the fault.

Good Luck!

 

[Pobyms]

Guy, I will disconnect the mouse, keyboard, HD and graphics card. They
are the only things connected now (and a stick of ram).

George Pontis, yes, I checked the battery and it is OK (around
2.95V).

Thanks for the ideas.

 

[old jon]

Guy, I will disconnect the mouse, keyboard, HD and graphics card. They
are the only things connected now (and a stick of ram).

George Pontis, yes, I checked the battery and it is OK (around
2.95V).

Under load ?. Or out of it`s holder ?.

 

[JAD]

what power supply are you using and are ALL the connectors to the main board
connected?

 

[Guy]

I was thinking more about this.... This is an Nforce2 board; what CPU
do you have, and what kind of heatsink / fan?

If the Heatsink is not properly seated on the CPU, this will also
prevent AMD chips from starting. The fan should also be plugged into
the board.

HTH!

 

[Pobyms]

old jon,
I tested it out of the mobo . Anyway I replaced it by a new
one. Today I will check its voltage. I guess the botton of the battery
(the -) is connected to ground, isnt it? I will meassure the voltage
between the top part of the battery (the +) and GND. Should it be
around 3V or it is much lower when loaded?

Guy,
The CPU is an AMD XP 2800+ and the fan is connected to the mobo. The
heatsink is a generic one, nothing special. I think it is well seated
(removed it, changed the thermal paste and placed it again). I will
remove and place it again.

When I force the PSU to be on, the temperature at the CPU when working
with XP and office or similar is around 40-45ºC. I think the heatsink
is OK, but who knows...

Thanks

 

[Guy]

No need to reseat the heatsink; just visually check to see that the
recessed (lapped) edge of the heatsink is lined up with the edge of
the socket, and not touching.

BTW: 40 C is really hot for that chip; but still within perfomance
specs. You should be able to target 28-30 C.

 

[dawg]

Did you just replace the motherboard? If so,I would check to see if it's
grounding on something. I had this same problem. I had to re-install the
mobo.

 

[w_tom]

The power switch connects to electronics on motherboard.
Electronics inside one motherboard 'chipset IC' requires a
constant +5VSB inspec. One function inside that controller is
a safety lockout. It will lockout power supply - ignore
switch - until reset by a complete power cord disconnect.

'About 5 volts' is not sufficient for the 5VSB. What
exactly is that voltage? The number is important.

Obviously, if the power switch contacts are shorted and the
Green power on signal does not drop below 0.8 volts, then
control electronics on motherboard is defective. To better
answer your question, one requires data sheets for the
'motherboard chipset' IC that controls power on.

Without disconnecting power supply connection to
motherboard, short green wire to black wire. If power supply
and its load is ok, then this short will power up the PC.
Testing this way will provide more information than just doing
same test with power supply "out of the mobo".

Temperature has nothing to do with your problem. Clearly
the problem centers around the power supply controller on
motherboard. The only remaining question is whether the
controller is defective or whether it is acting in response to
some input or the safety lockout circuit.

 

[Pobyms]

Yesterday I checked some voltages in the mobo. Today I am checking th
heatsink.

I dont remember the exact value of +5VSB but I have it at home, I wil
post it here

When I short PS_ON and GND and connect it to the movo, the PC work
pretty well

Now I have to find if the problem comes from the chipset. Where can
find the datasheet? and what kind of information can I get from it

 

[Guy]

Check the NVidia website for the specs, or Microstar.

 

[Pobyms]

After checing it all, I am thinking about buying a new mobo

It is strange that pressing the on button has no answer, but once th
PC is on (with the short in the PSU) the reset button resets the PC

 

[w_tom]

Not strange for one second - assuming the number on that
+5VSB (purple wire) is what I suspect it might be. You did
not post the number, so I can only speculate. That
speculation says what you are seeing is normal due to a
particular problem that a +5VSB number would confirm.

 

[Pobyms]

I checked it again some days ago and it was around 5.03V (I don´
remember the exact voltage but the value is over 5V and very close t
those 5V). I will post the exact value as soon as possible

what are you guessing, w_tom

In 3-4 days I will receive a new PSU (it is an 380W antec PSU, with m
new case, Antec Sonata). I think it is powerful enough for my PC (X
2800+, FX5200, DVD burner and ATA 160Gb HD). If the problem is i
that voltage my mobo should work with this PSU

 

[w_tom]

Many would consider a value of 4.85 at OK. That would be
too low and could indicate a problem with power supply.
Around 5.03 means power supply is good. The other half of
that control system is therefore suspect - on motherboard.

Once started, the motherboard circuits could be a full +5
volts from the +5 supply line. This would explain why the
power supply controller cannot turn power supply on but works
OK once +5V is obtained from another source. By process of
elimination, motherboard based power supply controller circuit
is the only remaining suspect.

Appreciate why I could not provide an answer on a subjective
"+5VSB looks OK". Without the specific number, I still did
not know if +5VSB was OK. Numbers from a meter are also
required for voltages on red, yellow, and orange wires to
confirm power supply is not telling motherboard controller to
'shut down due to a bad voltage'. Also that message of bad
power would be seen on gray wire - power good declared some
seconds later when the gray wire rises to over 2.4 volts.

If power supply does not tell motherboard controler that
voltages are OK after a few seconds, then motherboard
controller will shutdown power supply. Meter can display
that.

One need not buy a supply to test. One uses the meter to
either prove a supply fully sufficient or define that supply
as insufficient. Two minutes with numbers from critical
voltages could say, "Power supply is good. Move on to other
usual suspects." A definitive answer without speculation.