A7V600-X install/stert-up problem

  • Thread starter Thread starter Jerry
  • Start date Start date
J

Jerry

Just replaced an Abit m/b which had mouse communication probs.

When I fire it up absolutely nothing happens. However, I have the
green LED lit up.

Other than that there is no activity at all. LED on from of box seem
to be dead too although they may be dependent upon the box on/off
switch being OK.

Tried clearing CMOS. I'm lost now. Can someone help or is the M/B
DOA?

I have the Mem, graphics card (ATI 8600) and a HD in place BTW.
 
Jerry said:
Just replaced an Abit m/b which had mouse communication probs.

When I fire it up absolutely nothing happens. However, I have the
green LED lit up.

Other than that there is no activity at all. LED on from of box seem
to be dead too although they may be dependent upon the box on/off
switch being OK.

Tried clearing CMOS. I'm lost now. Can someone help or is the M/B
DOA?

I have the Mem, graphics card (ATI 8600) and a HD in place BTW.

Is the power switch connected to the correct two pins ?

You don't have to connect everything up on the PANEL
header if you don't want. Connecting just the power
switch to its two pins, is enough to get things started.
You won't hear any beeps from the PC speaker, but if the
computer starts, maybe you won't need to listen to the
beeps.

Paul
 
Is the power switch connected to the correct two pins ?

You don't have to connect everything up on the PANEL
header if you don't want. Connecting just the power
switch to its two pins, is enough to get things started.
You won't hear any beeps from the PC speaker, but if the
computer starts, maybe you won't need to listen to the
beeps.

Paul

Thnx for your input :-)

Tried the power switch, checked it, even reversed polarity (although I
don't think that matter). Nothing!

Funnily enough a) I've just put a 939 Asus board in another machine -
worked like a dream and b) apart from the mouse issue the "old" board
worked fine.
 
Jerry said:
Thnx for your input :-)

Tried the power switch, checked it, even reversed polarity (although I
don't think that matter). Nothing!

Funnily enough a) I've just put a 939 Asus board in another machine -
worked like a dream and b) apart from the mouse issue the "old" board
worked fine.

Seeing the green LED on the motherboard, means the motherboard
is getting +5VSB. That is the only voltage that shows up
initially when you switch on, via the switch on the back of
the computer. One of the purposes of the green LED, is you
should not add or remove hardware components if the green LED
is lit.

The power switch goes to two pins on the PANEL connector. A
momentary contact on the switch, should be latched by a chip on
the motherboard. In turn, that drives the PS_ON# signal on the
ATX 20 pin power connector to a logic 0, which is supposed to
turn on the PSU. You have indicated neither a "twitch of the
fans" nor a short powered up interval after trying the power
button, so that means that the normal sequence of events has
been interrupted.

Some Asus boards have a circuit called AGP_warn. It is located
next to the AGP slot. In recent months, it looks like Asus has
been removing the components from that circuit, so they might
not be using it any more. What that circuit used to do, is if
a mis-keyed 3.3V only AGP card was stuffed in the AGP slot,
the motherboard would be prevented from powering up. Pretty
well your exact symptoms.

To test for that, you could try removing the AGP card, and
see if the system will power up. (A board should be able to
start without a video card connected, and you should get the
beep code that says the video card is missing.)

I think the next step for you, is to assemble the system
outside your computer case. That is referred to as the
"cardboard" test, because you rest the motherboard on
an insulator while testing. (I like to use a thick phone
book.) If the motherboard will start that way, then something
must have been shorting.

In terms of power connectors, there is only the main 20 pin ATX
power connector, so it doesn't look like any other power cable
is missing.

You might try disconnecting the reset switch. I'm not
really sure whether reset prevents the power switch from
working or not. If reset was stuck in the "ON" position,
the board would not POST, but I would think the PS_ON#
logic would likely still work. (Hardware designers normally
design a reset signal, so it overrides all other logic,
but motherboard designers don't seem to follow this
practice. My A7N8X-E for example, can get wedged in states
where the reset button won't do anything.)

The motherboard should have been functionally tested
at the factory. Unless this is a refurb, or the board
has been subjected to stresses while being shipped,
it really should work.

Tell me, has the PSU ever been tested with another
motherboard ? Do you know that the PSU is actually
capable of turning on ? Perhaps you could test the
new PSU on another computer. An alternative, is to
test it by connecting PS_ON# to an adjacent COM signal,
on the 20 pin connector (I don't like recommending this
to people, so my preference would be to have tested
the PSU on another motherboard. While there should be
no dire consequences from connecting PS_ON# to GND,
I don't have a guarantee in writing that such a thing
is completely safe to do. I use dummy resistor loads
on my PSU when operating the PSU manually like that.)

If the PSU has a 115/220 selector switch on it, check
to make sure it is in the correct position. If the PSU
has such a selector switch, don't play with it while the
PSU is plugged in! In fact, you should always unplug
the power, before working in the case. (One poster here
actually had a PSU that continued to run, after the
switch on the back of the computer was switched off.
Unplugging is the only way to guarantee there is
no power present.)

HTH,
Paul
 
On Thu, 28 Apr 2005 14:51:12 -0400, (e-mail address removed) (Paul) wrote:

snip

Thnx for your input :-) AGAIN
You have indicated neither a "twitch of the
fans" nor a short powered up interval after trying the power
button, so that means that the normal sequence of events has
been interrupted.

There was one ocassion when the fan twiched - only the briefest twitch
moved the fan maybe just half a revolution. Happened once in around
20 or so attempts.
Some Asus boards have a circuit called AGP_warn. It is located
next to the AGP slot. In recent months, it looks like Asus has
been removing the components from that circuit, so they might
not be using it any more. What that circuit used to do, is if
a mis-keyed 3.3V only AGP card was stuffed in the AGP slot,
the motherboard would be prevented from powering up. Pretty
well your exact symptoms.

To test for that, you could try removing the AGP card, and
see if the system will power up. (A board should be able to
start without a video card connected, and you should get the
beep code that says the video card is missing.)

Removed. Still no joy
I think the next step for you, is to assemble the system
outside your computer case. That is referred to as the
"cardboard" test, because you rest the motherboard on
an insulator while testing. (I like to use a thick phone
book.) If the motherboard will start that way, then something
must have been shorting.

Try that later.
You might try disconnecting the reset switch. I'm not
really sure whether reset prevents the power switch from
working or not. If reset was stuck in the "ON" position,
the board would not POST, but I would think the PS_ON#
logic would likely still work.

Tried with both reset in place and disconnected.
The motherboard should have been functionally tested
at the factory. Unless this is a refurb, or the board
has been subjected to stresses while being shipped,
it really should work.

Tell me, has the PSU ever been tested with another
motherboard ?<

Yes it worked with the old m/b IMMEDIATELY prior to exchange
Do you know that the PSU is actually
capable of turning on ?

Not sure what you mean, but see above.
 
Jerry said:
On Thu, 28 Apr 2005 14:51:12 -0400, (e-mail address removed) (Paul) wrote:

snip

Thnx for your input :-) AGAIN


There was one ocassion when the fan twiched - only the briefest twitch
moved the fan maybe just half a revolution. Happened once in around
20 or so attempts.

That almost sounds like a power supply problem. It could be a
short somewhere is doing it. If you can get a twitch out of it,
that means the power switch path is working. I think the
"cardboard test" is the next step, followed by a PSU swap if
it still doesn't work.

Paul
 
That almost sounds like a power supply problem. It could be a
short somewhere is doing it. If you can get a twitch out of it,
that means the power switch path is working. I think the
"cardboard test" is the next step, followed by a PSU swap if
it still doesn't work.

Well the cardboard test (I just got around to it) produced the same
result.

You say PSU swap now but i don't have one so I'd have to buy one which
would be more than the m/b.

Paul, the PSU is only around 12 months old. It worked fine
immediately before the new m/b perfectly. The chance of it having
broken in the time it takes to switch the m/b are extremely slim, no?

What does this mean then? The m/b is NEW BTW.
 
Well the cardboard test (I just got around to it) produced the same
result.

You say PSU swap now but i don't have one so I'd have to buy one which
would be more than the m/b.

Paul, the PSU is only around 12 months old. It worked fine
immediately before the new m/b perfectly. The chance of it having
broken in the time it takes to switch the m/b are extremely slim, no?

What does this mean then? The m/b is NEW BTW.

I don't know if you are still here, Paul..............in answer to my
own query..........

I had a lab down the oroad test the m/b with the CPU and memory in
situ. Apparently it works fine.

So I guess it is the PSU or the connector!! Don't see what else it
can be now.
 
Jerry said:
I don't know if you are still here, Paul..............in answer to my
own query..........

I had a lab down the oroad test the m/b with the CPU and memory in
situ. Apparently it works fine.

So I guess it is the PSU or the connector!! Don't see what else it
can be now.

Could the lab down the road test your PSU for you ? That
is a tougher thing to do, because a proper test required
a programmable load box, if you want to do the job right
(i.e. test at rated load).

If you cannot borrow any more of that lab's time, you
can start a PSU by itself, by connecting PS_ON# to
one of the COM pins (there are COM pins very near to
PS_ON#). The PSU fan should start to spin, in response.
All that this test tells you, is that the control circuit
still has some response in it.

I do not recommend opening up the power supply under any
circumstances. The two main electrolytic caps store an
unbelievable number of joules of energy, and are not to
be trifled with. Even though the caps likely have bleeder
resistors across them, bleeders can open up, and putting
shunts across caps like this could deafen you (think
fire cracker loud). As a result, there is really no point
of opening up the lid of the supply. Using your nose
is as good an indicator of health as any, and if you smell
something burnt in the supply, you have all the info you
need about your supply.

I test my PSUs here before I put them in a computer. I
have some power resistors I got from an electronics
store, plus some Minifit Jr pins to connect to the
power supply connector. I actually load all five rails
on the power supply. What I cannot do with this setup,
is draw full rated load from the power supply - instead
I draw only about 1 amp or so from the major outputs, and
maybe 0.5 amps from the others. That way, the resistors
do not get too warm (I still have a fan blowing on them).
Getting decent dummy load resistors means too much bulky
stuff to store, when I'm not using the tester. It is possible
to design "electronic loads", using power transistors and
the like, but you still need to fabricate a cooling solution.
(No matter how you build a load, the heat has to go somewhere.)
I suppose $200 worth of water cooling stuff could cool it.
In any case, for my purposes, there isn't enough benefit
for the amount of effort required.

One thing to remember about your old motherboard, and your
new motherboard, is they likely present entirely different
levels of electrical loading to the PSU. It could be that
the PSU is weak enough, that it just cannot handle the new
board. For example, some people have a supply "tip over"
when they add something innocuous, like a CD drive. PSUs
these days, no matter what brand, are marginal at best.
Many of them cannot run at full rated load - some of
the ones with 400 or 450W written on the side, are just
last year's 350W with a new label pasted on it. Picking
good supplies is a tough job.

If you are short of money, the only alternative I can
see, is to buy a super-cheap supply. There was a web site
a while back, where the author weighed ATX power supplies,
and evaluated what was inside. He found that quality was
directly proportional to weight, so if you are at a large
computer supply place, compare the weight of the supplies
and take the heaviest of the cheap ones. (Fortunately, the
Chinese don't think we select supplies by weight, as
otherwise they would add lead plates to the case :-) )
That is about as scientific a selection algorithm as I
can give you, for the Deere, Powmax, Allied, etc supplies
of the world.

To work out what size supply you need, try this site:
http://takaman.jp/D/?english

Good luck,
Paul
 
Paul4 said:
Could the lab down the road test your PSU for you ? That
is a tougher thing to do, because a proper test required
a programmable load box, if you want to do the job right
(i.e. test at rated load).

If you cannot borrow any more of that lab's time, you
can start a PSU by itself, by connecting PS_ON# to
one of the COM pins (there are COM pins very near to
PS_ON#). The PSU fan should start to spin, in response.
All that this test tells you, is that the control circuit
still has some response in it.

I do not recommend opening up the power supply under any
circumstances. The two main electrolytic caps store an
unbelievable number of joules of energy, and are not to
be trifled with. Even though the caps likely have bleeder
resistors across them, bleeders can open up, and putting
shunts across caps like this could deafen you (think
fire cracker loud). As a result, there is really no point
of opening up the lid of the supply. Using your nose
is as good an indicator of health as any, and if you smell
something burnt in the supply, you have all the info you
need about your supply.

I test my PSUs here before I put them in a computer. I
have some power resistors I got from an electronics
store, plus some Minifit Jr pins to connect to the
power supply connector. I actually load all five rails
on the power supply. What I cannot do with this setup,
is draw full rated load from the power supply - instead
I draw only about 1 amp or so from the major outputs, and
maybe 0.5 amps from the others. That way, the resistors
do not get too warm (I still have a fan blowing on them).
Getting decent dummy load resistors means too much bulky
stuff to store, when I'm not using the tester. It is possible
to design "electronic loads", using power transistors and
the like, but you still need to fabricate a cooling solution.
(No matter how you build a load, the heat has to go
somewhere.)
I suppose $200 worth of water cooling stuff could cool it.
In any case, for my purposes, there isn't enough benefit
for the amount of effort required.

One thing to remember about your old motherboard, and your
new motherboard, is they likely present entirely different
levels of electrical loading to the PSU. It could be that
the PSU is weak enough, that it just cannot handle the new
board. For example, some people have a supply "tip over"
when they add something innocuous, like a CD drive. PSUs
these days, no matter what brand, are marginal at best.
Many of them cannot run at full rated load - some of
the ones with 400 or 450W written on the side, are just
last year's 350W with a new label pasted on it. Picking
good supplies is a tough job.

If you are short of money, the only alternative I can
see, is to buy a super-cheap supply. There was a web site
a while back, where the author weighed ATX power supplies,
and evaluated what was inside. He found that quality was
directly proportional to weight, so if you are at a large
computer supply place, compare the weight of the supplies
and take the heaviest of the cheap ones. (Fortunately, the
Chinese don't think we select supplies by weight, as
otherwise they would add lead plates to the case :-) )
That is about as scientific a selection algorithm as I
can give you, for the Deere, Powmax, Allied, etc supplies
of the world.

To work out what size supply you need, try this site:
http://takaman.jp/D/?english

Good luck,
Paul

Have you updated BIOS?
 
Could the lab down the road test your PSU for you ?
snip

To work out what size supply you need, try this site:
http://takaman.jp/D/?english

Good luck,
Paul

Paul, I am greatly indebted to you. I have learned a lot from this
highly frustrating problem.

I asked the Lab to test the PSU. It was a faulty PSU. The m/b is now
installed and working (except for wireless networking :-( but thats
another story)

I had no idea a failing PSU could just afgfect mouse performance to
start with. I assumed it was the m/b and replaced it. Probably need
not have done. Ah well.
 
Back
Top