Help! P4P800 Dlx won't boot

G

Guest

Guest
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

My screen froze up after plugging my scanner into the USB port (but I think
that was a coincidence). I hit the reset button to reboot and got a message
something like "overclocking settings failed". I hit F1, went into the bios
to check the default settings (I don't overclock). Everything looked OK so
I saved and exited the bios. After that, nothing. No post, blank monitor.
Prior to that all was good. I tried to clear the CMOS but with no effect.
I've completely powered down and restarted about twenty times. Drives spin,
fans whir but nothing. Anybody got any ideas?

P4P800 Dlx
P4 3.2 GHz
ATI 9800 Pro 128mb
1 Gig Corsair Ram
1 SATA + 2 ATA100 hard drives (no raid)
2 burners

Thanks in advance,
RH
 

Paul

Splendid
Mar 30, 2004
5,267
0
25,780
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <UNZ5e.3248$%c1.398@fed1read05>, "Roadhair"
<roadhair@hotmail.com> wrote:

> My screen froze up after plugging my scanner into the USB port (but I think
> that was a coincidence). I hit the reset button to reboot and got a message
> something like "overclocking settings failed". I hit F1, went into the bios
> to check the default settings (I don't overclock). Everything looked OK so
> I saved and exited the bios. After that, nothing. No post, blank monitor.
> Prior to that all was good. I tried to clear the CMOS but with no effect.
> I've completely powered down and restarted about twenty times. Drives spin,
> fans whir but nothing. Anybody got any ideas?
>
> P4P800 Dlx
> P4 3.2 GHz
> ATI 9800 Pro 128mb
> 1 Gig Corsair Ram
> 1 SATA + 2 ATA100 hard drives (no raid)
> 2 burners
>
> Thanks in advance,
> RH

See the posting by "Massimiliano" above. It would suggest your
Southbridge has been destroyed.

Was the USB port you used for the scanner a front mounted port ?
Perhaps a port which is missing a connection of the USB shield
ground connection to computer ground ? There have been a number
of failures recently, and they point to latchup caused by ESD
dumped into an electrically floating shield connection.

Someone else found this link for me:

http://tw.giga-byte.com/Motherboard/Support/FAQ/FAQ_456.htm

HTH,
Paul
 
G

Guest

Guest
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

Not what I wanted to hear. I added new FrontX USB connectors, 2 front & 2
rear to the motherboard headers. They use seperate 5-pin connectors for
each port but the header is only 9-pin and I don't think the 9th pin is even
connected to ground. The manual lists it as "NC". So far I haven't had any
problem hot plugging thumb drives but the scanner is powered and that may
have been the culprit. Such an ingenious design to have a feature you can't
use. Would this work if I pull the shield grounds from the MB connectors
and ground them directly to the chassis?

Thanks,
RH

"Paul" <nospam@needed.com> wrote in message
news:nospam-0904052147570001@192.168.1.178...
> In article <UNZ5e.3248$%c1.398@fed1read05>, "Roadhair"
> <roadhair@hotmail.com> wrote:
>
> > My screen froze up after plugging my scanner into the USB port (but I
think
> > that was a coincidence). I hit the reset button to reboot and got a
message
> > something like "overclocking settings failed". I hit F1, went into the
bios
> > to check the default settings (I don't overclock). Everything looked OK
so
> > I saved and exited the bios. After that, nothing. No post, blank
monitor.
> > Prior to that all was good. I tried to clear the CMOS but with no
effect.
> > I've completely powered down and restarted about twenty times. Drives
spin,
> > fans whir but nothing. Anybody got any ideas?
> >
> > P4P800 Dlx
> > P4 3.2 GHz
> > ATI 9800 Pro 128mb
> > 1 Gig Corsair Ram
> > 1 SATA + 2 ATA100 hard drives (no raid)
> > 2 burners
> >
> > Thanks in advance,
> > RH
>
> See the posting by "Massimiliano" above. It would suggest your
> Southbridge has been destroyed.
>
> Was the USB port you used for the scanner a front mounted port ?
> Perhaps a port which is missing a connection of the USB shield
> ground connection to computer ground ? There have been a number
> of failures recently, and they point to latchup caused by ESD
> dumped into an electrically floating shield connection.
>
> Someone else found this link for me:
>
> http://tw.giga-byte.com/Motherboard/Support/FAQ/FAQ_456.htm
>
> HTH,
> Paul
 

Paul

Splendid
Mar 30, 2004
5,267
0
25,780
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <a826e.3270$%c1.270@fed1read05>, "Roadhair"
<roadhair@hotmail.com> wrote:

> Not what I wanted to hear. I added new FrontX USB connectors, 2 front & 2
> rear to the motherboard headers. They use seperate 5-pin connectors for
> each port but the header is only 9-pin and I don't think the 9th pin is even
> connected to ground. The manual lists it as "NC". So far I haven't had any
> problem hot plugging thumb drives but the scanner is powered and that may
> have been the culprit. Such an ingenious design to have a feature you can't
> use. Would this work if I pull the shield grounds from the MB connectors
> and ground them directly to the chassis?
>
> Thanks,
> RH

One FrontX part is the "Dual USB Type A Internal"

http://www.frontx.com/cpx108_2.html

The pinout is:

VCC X X VCC
D- X X D-
D+ X X D+
GND X X GND
X S-GND

When that mates to the Asus header, the Asus "NC" no-connector
is connected to the S-GND shield ground. That means the plastic
FrontX assembly has no DC path to ground, and the shield metal
on the FrontX metal shield is floating. It is waiting to be
zapped by static electricity, killing the Southbridge. At
least, that is what I'm assuming has happened here.

I'm surprised that the reset button worked, after the
computer froze. And that the Southbridge was usable for a few
moments, to allow the flash BIOS chip to be read via LPC.
Have a look at the Southbridge and see if there is any
sign of heat damage. Latchup does not have to produce a visible
result, to destroy a part, but other posters have been observing
the "meltdown" flavor of latchup.

Once you get the motherboard repaired under warranty (RMA), if
you want to connect the FrontX connector again, connect
the S-GND to a ground somewhere. Since you plan on using the
two 2x5 USB headers, that means there aren't any spare ground
pins around. Perhaps you can get a crimpable lug from the hardware
store, some wire, and fasten the lug to a PCI faceplate screw.
The other end of the wire has to go to the S-GND on each
FrontX assembly. If you buy 1/4 watt or 1/8 watt resistors at
Radio Shack, the ends of one of them might fit into the opening
on the FrontX wire. (The FrontX wire will have to be released
from its housing to do this.) Or, you could just cut the end off
the FrontX assembly and splice the end of the wire to your
home made ground connection.

The third prong on a North American ATX power supply, provides a
safety ground to the computer. For it to function properly, your
wall outlet should have a matching safety ground. Safety ground
should be connected to the chassis of the power supply, and when
the power supply makes contact with the computer case, the metal
chassis of the computer should be grounded too. Running a wire
to the chassis, as a means of finding a safety ground, will work
as long as you verify that there is ground continuity back to the
ground on the wall socket. If won't help to set up a gimmick ground
like this, if the chassis or PSU metal are in fact floating. At
least one European poster to this group got a shock from his
computer, and safety ground was the missing item.

Really, once you have suffered a "latchup meltdown", you should be
shopping for a cheap USB2 PCI card, and making the connections
to that instead. You could still destroy the PCI card with static
electricity, but it will be easier to replace that card, than
to RMA the motherboard again. One would hope the PCI USB2 card
is a bit more robust than the Intel Southbridge.

There are a lot of ALI chipset USB cards on Newegg, and this
one is pretty cheap - cheap enough to chuck if it doesn't work
well -

http://www.newegg.com/app/viewproductdesc.asp?description=15-201-206&DEPA=0

If Asus is reading this, it is time to change that "NC" pin
into a "GND" on the 2x5 USB header. Or, another option, would be
to solder some single GND stake pins to the motherboard, for
grounding things. The single stake pins could be kept close to
the header, but far enough away that a 2x5 assembly will not
bump into them. Some spare grounds are occasionally needed for
front panel audio, so some GND stake pins near FP_AUDIO would
help too.

Paul
 
G

Guest

Guest
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

"Paul" <nospam@needed.com> wrote in message
news:nospam-1004050415030001@192.168.1.178...
> In article <a826e.3270$%c1.270@fed1read05>, "Roadhair"
> <roadhair@hotmail.com> wrote:
>
> > Not what I wanted to hear. I added new FrontX USB connectors, 2 front &
2
> > rear to the motherboard headers. They use seperate 5-pin connectors for
> > each port but the header is only 9-pin and I don't think the 9th pin is
even
> > connected to ground. The manual lists it as "NC". So far I haven't had
any
> > problem hot plugging thumb drives but the scanner is powered and that
may
> > have been the culprit. Such an ingenious design to have a feature you
can't
> > use. Would this work if I pull the shield grounds from the MB
connectors
> > and ground them directly to the chassis?
> >
> > Thanks,
> > RH
>
> One FrontX part is the "Dual USB Type A Internal"
>
> http://www.frontx.com/cpx108_2.html
>
> The pinout is:
>
> VCC X X VCC
> D- X X D-
> D+ X X D+
> GND X X GND
> X S-GND
>
> When that mates to the Asus header, the Asus "NC" no-connector
> is connected to the S-GND shield ground. That means the plastic
> FrontX assembly has no DC path to ground, and the shield metal
> on the FrontX metal shield is floating. It is waiting to be
> zapped by static electricity, killing the Southbridge. At
> least, that is what I'm assuming has happened here.
>
> I'm surprised that the reset button worked, after the
> computer froze. And that the Southbridge was usable for a few
> moments, to allow the flash BIOS chip to be read via LPC.
> Have a look at the Southbridge and see if there is any
> sign of heat damage. Latchup does not have to produce a visible
> result, to destroy a part, but other posters have been observing
> the "meltdown" flavor of latchup.
>
> Once you get the motherboard repaired under warranty (RMA), if
> you want to connect the FrontX connector again, connect
> the S-GND to a ground somewhere. Since you plan on using the
> two 2x5 USB headers, that means there aren't any spare ground
> pins around. Perhaps you can get a crimpable lug from the hardware
> store, some wire, and fasten the lug to a PCI faceplate screw.
> The other end of the wire has to go to the S-GND on each
> FrontX assembly. If you buy 1/4 watt or 1/8 watt resistors at
> Radio Shack, the ends of one of them might fit into the opening
> on the FrontX wire. (The FrontX wire will have to be released
> from its housing to do this.) Or, you could just cut the end off
> the FrontX assembly and splice the end of the wire to your
> home made ground connection.
>
> The third prong on a North American ATX power supply, provides a
> safety ground to the computer. For it to function properly, your
> wall outlet should have a matching safety ground. Safety ground
> should be connected to the chassis of the power supply, and when
> the power supply makes contact with the computer case, the metal
> chassis of the computer should be grounded too. Running a wire
> to the chassis, as a means of finding a safety ground, will work
> as long as you verify that there is ground continuity back to the
> ground on the wall socket. If won't help to set up a gimmick ground
> like this, if the chassis or PSU metal are in fact floating. At
> least one European poster to this group got a shock from his
> computer, and safety ground was the missing item.
>
> Really, once you have suffered a "latchup meltdown", you should be
> shopping for a cheap USB2 PCI card, and making the connections
> to that instead. You could still destroy the PCI card with static
> electricity, but it will be easier to replace that card, than
> to RMA the motherboard again. One would hope the PCI USB2 card
> is a bit more robust than the Intel Southbridge.
>
> There are a lot of ALI chipset USB cards on Newegg, and this
> one is pretty cheap - cheap enough to chuck if it doesn't work
> well -
>
>
http://www.newegg.com/app/viewproductdesc.asp?description=15-201-206&DEPA=0
>
> If Asus is reading this, it is time to change that "NC" pin
> into a "GND" on the 2x5 USB header. Or, another option, would be
> to solder some single GND stake pins to the motherboard, for
> grounding things. The single stake pins could be kept close to
> the header, but far enough away that a 2x5 assembly will not
> bump into them. Some spare grounds are occasionally needed for
> front panel audio, so some GND stake pins near FP_AUDIO would
> help too.
>
> Paul


I see no physical damage to the southbridge nor does it smell cooked at all.
I was also surprised that it rebooted long enough to post an error message
especially one claiming the system was over-clocked. It was not. I'm just
pissed I went to all the trouble installing the front USB connectors in a
case that had no original provision for it. I also installed a firewire
port as well but I hadn't used it yet. Naturally I'm also pissed that my PC
is dead and I'm making due with my old P3 550 running win98. Well, I guess
I got some time to figure it all out while I wait for a replacement.
Speaking of which, anyone have some tips on the fastest way to request an
RMA. I already submitted a help/RMA request to the ASUS online tech
support. And can an end user do an advance RMA?

Again, thanks for all the support
RH
 

Paul

Splendid
Mar 30, 2004
5,267
0
25,780
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <a826e.3270$%c1.270@fed1read05>, "Roadhair"
<roadhair@hotmail.com> wrote:

> Not what I wanted to hear. I added new FrontX USB connectors, 2 front & 2
> rear to the motherboard headers. They use seperate 5-pin connectors for
> each port but the header is only 9-pin and I don't think the 9th pin is even
> connected to ground. The manual lists it as "NC". So far I haven't had any
> problem hot plugging thumb drives but the scanner is powered and that may
> have been the culprit. Such an ingenious design to have a feature you can't
> use. Would this work if I pull the shield grounds from the MB connectors
> and ground them directly to the chassis?
>
> Thanks,
> RH
>

If you check the earlier thread, we are now at the point that
I can only recommend a USB2 PCI card as a solution. Forget
about touching the Intel Southbridge USB ports, whether they
are on the back of the computer, or via the 2x5 headers. A
poster has reported a failure based on a rear USB port, and
that means grounding the shield is simply not enough.

That means RMA the motherboard, and buy a $10 USB2 PCI card.
Only connect to the USB2 PCI card, still taking precautions
to drain static before making connections.

I wonder what Intel has to say about this, and if Intel is
supplying free replacement ICH4's and ICH5's to Asus ?

Paul
 

Jimbo

Distinguished
Feb 20, 2004
390
0
18,780
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

"Paul" <nospam@needed.com> wrote in message
news:nospam-1004050948540001@192.168.1.178...
> In article <a826e.3270$%c1.270@fed1read05>, "Roadhair"
> <roadhair@hotmail.com> wrote:
>
>> Not what I wanted to hear. I added new FrontX USB connectors, 2 front &
>> 2
>> rear to the motherboard headers. They use seperate 5-pin connectors for
>> each port but the header is only 9-pin and I don't think the 9th pin is
>> even
>> connected to ground. The manual lists it as "NC". So far I haven't had
>> any
>> problem hot plugging thumb drives but the scanner is powered and that may
>> have been the culprit. Such an ingenious design to have a feature you
>> can't
>> use. Would this work if I pull the shield grounds from the MB connectors
>> and ground them directly to the chassis?
>>
>> Thanks,
>> RH
>>
>
> If you check the earlier thread, we are now at the point that
> I can only recommend a USB2 PCI card as a solution. Forget
> about touching the Intel Southbridge USB ports, whether they
> are on the back of the computer, or via the 2x5 headers. A
> poster has reported a failure based on a rear USB port, and
> that means grounding the shield is simply not enough.
>
> That means RMA the motherboard, and buy a $10 USB2 PCI card.
> Only connect to the USB2 PCI card, still taking precautions
> to drain static before making connections.
>
> I wonder what Intel has to say about this, and if Intel is
> supplying free replacement ICH4's and ICH5's to Asus ?
>
> Paul

Paul,
If Intel provides Asus with replacement Southbridge's, do you know if Asus
repairs the motherbard by soldering in the new ICH4 or 5's or do they just
throw out the mb and use a new one to rma back to the customer?

Jimbo
 

Paul

Splendid
Mar 30, 2004
5,267
0
25,780
Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <xZadnQkpWJIY2MTfRVn-pg@comcast.com>, "Jimbo"
<jimmy7forever@yahoo.com> wrote:

> "Paul" <nospam@needed.com> wrote in message
> news:nospam-1004050948540001@192.168.1.178...
> > In article <a826e.3270$%c1.270@fed1read05>, "Roadhair"
> > <roadhair@hotmail.com> wrote:
> >
> >> Not what I wanted to hear. I added new FrontX USB connectors, 2 front &
> >> 2
> >> rear to the motherboard headers. They use seperate 5-pin connectors for
> >> each port but the header is only 9-pin and I don't think the 9th pin is
> >> even
> >> connected to ground. The manual lists it as "NC". So far I haven't had
> >> any
> >> problem hot plugging thumb drives but the scanner is powered and that may
> >> have been the culprit. Such an ingenious design to have a feature you
> >> can't
> >> use. Would this work if I pull the shield grounds from the MB connectors
> >> and ground them directly to the chassis?
> >>
> >> Thanks,
> >> RH
> >>
> >
> > If you check the earlier thread, we are now at the point that
> > I can only recommend a USB2 PCI card as a solution. Forget
> > about touching the Intel Southbridge USB ports, whether they
> > are on the back of the computer, or via the 2x5 headers. A
> > poster has reported a failure based on a rear USB port, and
> > that means grounding the shield is simply not enough.
> >
> > That means RMA the motherboard, and buy a $10 USB2 PCI card.
> > Only connect to the USB2 PCI card, still taking precautions
> > to drain static before making connections.
> >
> > I wonder what Intel has to say about this, and if Intel is
> > supplying free replacement ICH4's and ICH5's to Asus ?
> >
> > Paul
>
> Paul,
> If Intel provides Asus with replacement Southbridge's, do you know if Asus
> repairs the motherbard by soldering in the new ICH4 or 5's or do they just
> throw out the mb and use a new one to rma back to the customer?
>
> Jimbo

With BGA chips, it is possible to do three replacement operations,
before the PCB is considered unusable (those were the rules at my
previous employer, but the quality of our circuit boards was much
higher than on motherboards). It means a tool is used to heat up
the chip, until all the solder is molten, and then the chip can
be removed. Where I live, a local company will perform that service
for $1000 roughly a chip - that would include X-ray inspection.
I'm willing to bet the labor charge in Taiwan is a lot less :)
So, the motherboards can be recycled, assuming the customer has
not visibly damaged the motherboard. I expect many more boards
are rejected for cosmetic reasons, than for an inability
to repair them.

Whether a motherboard is worth repairing, really depends on the
cost of labor and the cost of test equipment time to find the
problem. With the rash of failed Southbridges, it probably isn't
hard for the repair people to recognize the need for a new
Southbridge.

There will be a wide disparity between practices in North America
and Taiwan, so it is hard to guess as to how much work they will
do before giving up on a motherboard. When a company makes two
or three million circuit boards a month, it doesn't take much of
a problem, before there is a mountain of duff circuit boards
out in back of the factory.

I cannot imagine the logistics of the RMA department in a
company like that. There must be a staggering head count
in that department.

Paul