My new mobo won't post

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

I just put together a new system with an Asus P4SGX-MX motherboard and a
Pentium 4, 2.4A GHz 533 MHz System Bus.
I connected everything using a wrist strap, but I must have goofed up
somehow. I checked the voltage at the Power supply connectors on the board
(20 pin and 4 pin), as well as on the drives. The voltage all checks out,
but it won't post.
Are there any tests I can perform that will tell me whether it's the
motherboard, the CPU or (God forbid) both that have gone bad? I would like
to avoid replacing both simply because I can't tell which is bad.
Any advice?

Kyle

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Kyle Griffin wrote:

> I just put together a new system with an Asus P4SGX-MX motherboard and a
> Pentium 4, 2.4A GHz 533 MHz System Bus.
> I connected everything using a wrist strap, but I must have goofed up
> somehow. I checked the voltage at the Power supply connectors on the
> board
> (20 pin and 4 pin), as well as on the drives. The voltage all checks out,
> but it won't post.
> Are there any tests I can perform that will tell me whether it's the
> motherboard, the CPU or (God forbid) both that have gone bad? I would
> like to avoid replacing both simply because I can't tell which is bad.
> Any advice?
>
> Kyle

Forget about it being a static problem. I've never worn a wrist strap and
I've never had a problem.

Make sure the RAM and video card are seated properly. If the CPU was on
incorrectly, you'd be seeing the smoke by now. Disconnect all the drives
in case it's a problem with your power supply not supplying enough juice.

You do have a heatsink on there, right? If not the temp will shoot up and
the system won't turn on.

Are there any beeps? Try taking out the memory...there should be a series
of beeps now when you turn it back on.

--Mitchua

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Mitchua" <mitchuaNOSPAM@hush.com> wrote in message
news:Z0Rjc.1871$k%.26621@news20.bellglobal.com...
> Kyle Griffin wrote:
>
> > I just put together a new system with an Asus P4SGX-MX motherboard and
a
> > Pentium 4, 2.4A GHz 533 MHz System Bus.
> > I connected everything using a wrist strap, but I must have goofed up
> > somehow. I checked the voltage at the Power supply connectors on the
> > board
> > (20 pin and 4 pin), as well as on the drives. The voltage all checks
out,
> > but it won't post.
> > Are there any tests I can perform that will tell me whether it's the
> > motherboard, the CPU or (God forbid) both that have gone bad? I would
> > like to avoid replacing both simply because I can't tell which is bad.
> > Any advice?
> >
> > Kyle
>
> Forget about it being a static problem. I've never worn a wrist strap and
> I've never had a problem.
>
> Make sure the RAM and video card are seated properly. If the CPU was on
> incorrectly, you'd be seeing the smoke by now. Disconnect all the drives
> in case it's a problem with your power supply not supplying enough juice.
>
> You do have a heatsink on there, right? If not the temp will shoot up and
> the system won't turn on.
>
> Are there any beeps? Try taking out the memory...there should be a series
> of beeps now when you turn it back on.
>
> --Mitchua

Thanks for the reply.
Just to fill in the blanks, there is a heatsink with a fan on the CPU. The
fan spins when I power up, so I know that's working at least minimally. The
CPU is mounted in a 478 pin ZIF socket. I don't think I could get it in the
wrong way without hearing a resounding snap. I've also re-seated it several
times to be sure, inspecting the pins each time. They all look good.

I tried as you suggested and disconnected all but one drive and removed the
RAM, then I tried it with no drives connected. No beeps either way. I
cleared the CMOS previously, after the first few times it wouldn't post.

If it won't beep at all, does this sound like a motherboard problem, or is
there something else I should try?

Kyle

 

[jt]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Wed, 28 Apr 2004 14:12:57 -0700, "Kyle Griffin" <kcfg@comcast.net>
wrote:

>

>Thanks for the reply.
>Just to fill in the blanks, there is a heatsink with a fan on the CPU. The
>fan spins when I power up, so I know that's working at least minimally. The
>CPU is mounted in a 478 pin ZIF socket. I don't think I could get it in the
>wrong way without hearing a resounding snap. I've also re-seated it several
>times to be sure, inspecting the pins each time. They all look good.
>
>I tried as you suggested and disconnected all but one drive and removed the
>RAM, then I tried it with no drives connected. No beeps either way. I
>cleared the CMOS previously, after the first few times it wouldn't post.
>
>If it won't beep at all, does this sound like a motherboard problem, or is
>there something else I should try?
>
>Kyle
>
>

You have to have the ram to get beeps. Don't need a drive hooked up at all.
Check the clear CMOS jumper. Some used to ship it in the clear position,
which kept the board from working. In fact, if there are jumpers on the
board recheck them all. Restart with only cpu, heatsink & fan, ram, psu,
and speaker connected. Should get beeps (lots of them because of no video
.

JT

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"JT" <spam@dcplus.dyndns.info> wrote in message
news:68476201f8128adc747440185dd57bb1@news.teranews.com...
> On Wed, 28 Apr 2004 14:12:57 -0700, "Kyle Griffin" <kcfg@comcast.net>
> wrote:
>
> >
>
> >Thanks for the reply.
> >Just to fill in the blanks, there is a heatsink with a fan on the CPU.
The
> >fan spins when I power up, so I know that's working at least minimally.
The
> >CPU is mounted in a 478 pin ZIF socket. I don't think I could get it in
the
> >wrong way without hearing a resounding snap. I've also re-seated it
several
> >times to be sure, inspecting the pins each time. They all look good.
> >
> >I tried as you suggested and disconnected all but one drive and removed
the
> >RAM, then I tried it with no drives connected. No beeps either way. I
> >cleared the CMOS previously, after the first few times it wouldn't post.
> >
> >If it won't beep at all, does this sound like a motherboard problem, or
is
> >there something else I should try?
> >
> >Kyle
> >
> >
>
> You have to have the ram to get beeps. Don't need a drive hooked up at
all.
> Check the clear CMOS jumper. Some used to ship it in the clear position,
> which kept the board from working. In fact, if there are jumpers on the
> board recheck them all. Restart with only cpu, heatsink & fan, ram, psu,
> and speaker connected. Should get beeps (lots of them because of no video
> .
>
> JT
>

Actually, I've already tried it with only the CPU, heatsink & fan, RAM and
speaker, and since the first time I tried to start it up, I've been doing it
with no monitor (the video is on-board) because the monitor remained black
and in stand-by mode.

I've compared the CMOS jumper to the position indicated in the manual and
it's correct. The only other jumpers are for wake from sleep by USB devices
and jumpers to bypass the connector for front panel audio.

What am I forgetting?

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Kyle Griffin wrote:

>
> "JT" <spam@dcplus.dyndns.info> wrote in message
> news:68476201f8128adc747440185dd57bb1@news.teranews.com...
>> On Wed, 28 Apr 2004 14:12:57 -0700, "Kyle Griffin" <kcfg@comcast.net>
>> wrote:
>>
>> >
>>
>> >Thanks for the reply.
>> >Just to fill in the blanks, there is a heatsink with a fan on the CPU.
> The
>> >fan spins when I power up, so I know that's working at least minimally.
> The
>> >CPU is mounted in a 478 pin ZIF socket. I don't think I could get it in
> the
>> >wrong way without hearing a resounding snap. I've also re-seated it
> several
>> >times to be sure, inspecting the pins each time. They all look good.
>> >
>> >I tried as you suggested and disconnected all but one drive and removed
> the
>> >RAM, then I tried it with no drives connected. No beeps either way. I
>> >cleared the CMOS previously, after the first few times it wouldn't post.
>> >
>> >If it won't beep at all, does this sound like a motherboard problem, or
> is
>> >there something else I should try?
>> >
>> >Kyle
>> >
>> >
>>
>> You have to have the ram to get beeps. Don't need a drive hooked up at
> all.
>> Check the clear CMOS jumper. Some used to ship it in the clear position,
>> which kept the board from working. In fact, if there are jumpers on the
>> board recheck them all. Restart with only cpu, heatsink & fan, ram, psu,
>> and speaker connected. Should get beeps (lots of them because of no video
>> .
>>
>> JT
>>
>
> Actually, I've already tried it with only the CPU, heatsink & fan, RAM and
> speaker, and since the first time I tried to start it up, I've been doing
> it with no monitor (the video is on-board) because the monitor remained
> black and in stand-by mode.
>
> I've compared the CMOS jumper to the position indicated in the manual and
> it's correct. The only other jumpers are for wake from sleep by USB
> devices and jumpers to bypass the connector for front panel audio.
>
> What am I forgetting?

Do you have your old video card around to test it out?

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"Mitchua" <mitchuaNOSPAM@hush.com> wrote in message
news:M30kc.8341$k%.284251@news20.bellglobal.com...
> Kyle Griffin wrote:
>
> >
> > "JT" <spam@dcplus.dyndns.info> wrote in message
> > news:68476201f8128adc747440185dd57bb1@news.teranews.com...
> >> On Wed, 28 Apr 2004 14:12:57 -0700, "Kyle Griffin" <kcfg@comcast.net>
> >> wrote:
> >>
> >> >
> >>
> >> >Thanks for the reply.
> >> >Just to fill in the blanks, there is a heatsink with a fan on the
CPU.
> > The
> >> >fan spins when I power up, so I know that's working at least
minimally.
> > The
> >> >CPU is mounted in a 478 pin ZIF socket. I don't think I could get it
in
> > the
> >> >wrong way without hearing a resounding snap. I've also re-seated it
> > several
> >> >times to be sure, inspecting the pins each time. They all look good.
> >> >
> >> >I tried as you suggested and disconnected all but one drive and
removed
> > the
> >> >RAM, then I tried it with no drives connected. No beeps either way.
I
> >> >cleared the CMOS previously, after the first few times it wouldn't
post.
> >> >
> >> >If it won't beep at all, does this sound like a motherboard problem,
or
> > is
> >> >there something else I should try?
> >> >
> >> >Kyle
> >> >
> >> >
> >>
> >> You have to have the ram to get beeps. Don't need a drive hooked up at
> > all.
> >> Check the clear CMOS jumper. Some used to ship it in the clear
position,
> >> which kept the board from working. In fact, if there are jumpers on the
> >> board recheck them all. Restart with only cpu, heatsink & fan, ram,
psu,
> >> and speaker connected. Should get beeps (lots of them because of no
video
> >> .
> >>
> >> JT
> >>
> >
> > Actually, I've already tried it with only the CPU, heatsink & fan, RAM
and
> > speaker, and since the first time I tried to start it up, I've been
doing
> > it with no monitor (the video is on-board) because the monitor remained
> > black and in stand-by mode.
> >
> > I've compared the CMOS jumper to the position indicated in the manual
and
> > it's correct. The only other jumpers are for wake from sleep by USB
> > devices and jumpers to bypass the connector for front panel audio.
> >
> > What am I forgetting?
>
> Do you have your old video card around to test it out?

I already did that too. No good. I guess I should have mentioned that
before. I apologize for leaving all of this out.
I've tested it with all drives connected, some drives connected (in several
combinations), no drives connected, with memory, without memory, with a
spare video card and without. I even tested the speaker to make sure it
wasn't dead. I believe it's twice now that I've cleared the CMOS, I've
checked the jumpers and all of the power supply cables for voltage. I don't
have an ammeter, so I can't check that, but it's a 350W psu, so it should
handle it. Should I have reached a conclusion based on this or do I need to
check anything else?

-Kyle

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Wed, 28 Apr 2004 22:37:10 -0700, "Kyle Griffin" <kcfg@comcast.net>
wrote:


>I already did that too. No good. I guess I should have mentioned that
>before. I apologize for leaving all of this out.
>I've tested it with all drives connected, some drives connected (in several
>combinations), no drives connected, with memory, without memory, with a
>spare video card and without. I even tested the speaker to make sure it
>wasn't dead. I believe it's twice now that I've cleared the CMOS, I've
>checked the jumpers and all of the power supply cables for voltage. I don't
>have an ammeter, so I can't check that, but it's a 350W psu, so it should
>handle it. Should I have reached a conclusion based on this or do I need to
>check anything else?

This 350W PSU, it's a name brand so you can trust the wattage on the
label? It might be a stupid question, but do you have the 4-pin 12V
connector, connected? Perhaps it's not the PSU, but "assuming" it isn't
could make your efforts a waste if that's the problem.

Of the parts you mentioned, motherboard and CPU, it's much more likely the
motherboard. However if the PSU is generic you ought to try a name-brand,
300W or more.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" <spam@spam.com> wrote in message
news:ggd190l02qlan9kufhqbl11ohgbr5ga55r@4ax.com...
> On Wed, 28 Apr 2004 22:37:10 -0700, "Kyle Griffin" <kcfg@comcast.net>
> wrote:
>
>
> >I already did that too. No good. I guess I should have mentioned that
> >before. I apologize for leaving all of this out.
> >I've tested it with all drives connected, some drives connected (in
several
> >combinations), no drives connected, with memory, without memory, with a
> >spare video card and without. I even tested the speaker to make sure it
> >wasn't dead. I believe it's twice now that I've cleared the CMOS, I've
> >checked the jumpers and all of the power supply cables for voltage. I
don't
> >have an ammeter, so I can't check that, but it's a 350W psu, so it should
> >handle it. Should I have reached a conclusion based on this or do I need
to
> >check anything else?
>
> This 350W PSU, it's a name brand so you can trust the wattage on the
> label? It might be a stupid question, but do you have the 4-pin 12V
> connector, connected? Perhaps it's not the PSU, but "assuming" it isn't
> could make your efforts a waste if that's the problem.
>
> Of the parts you mentioned, motherboard and CPU, it's much more likely the
> motherboard. However if the PSU is generic you ought to try a name-brand,
> 300W or more.

It's a Raidmax psu and case, and yes, I did have the 4-pin connected and the
voltage checked out, but again; no ammeter, so I can't tell if the wattage
adds up.
It's just a little bit frustrating having to choose from 3 possibilities.
The power supply has 2 quality control stickers on it, but that's just their
say so. The motherboard, on the other hand, is a model number that isn't
even listed on the Asus website. There are a few similar ones with similar
model numbers, but not this one. The store said I could exchange it for a
different board and pay the balance, if any, so I can try that and see if it
solves my problem.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

Load the factory default CMOS settings.

"kony" <spam@spam.com> wrote in message
news:ggd190l02qlan9kufhqbl11ohgbr5ga55r@4ax.com...
> On Wed, 28 Apr 2004 22:37:10 -0700, "Kyle Griffin" <kcfg@comcast.net>
> wrote:
>
>
> >I already did that too. No good. I guess I should have mentioned that
> >before. I apologize for leaving all of this out.
> >I've tested it with all drives connected, some drives connected (in
several
> >combinations), no drives connected, with memory, without memory, with a
> >spare video card and without. I even tested the speaker to make sure it
> >wasn't dead. I believe it's twice now that I've cleared the CMOS, I've
> >checked the jumpers and all of the power supply cables for voltage. I
don't
> >have an ammeter, so I can't check that, but it's a 350W psu, so it should
> >handle it. Should I have reached a conclusion based on this or do I need
to
> >check anything else?
>
> This 350W PSU, it's a name brand so you can trust the wattage on the
> label? It might be a stupid question, but do you have the 4-pin 12V
> connector, connected? Perhaps it's not the PSU, but "assuming" it isn't
> could make your efforts a waste if that's the problem.
>
> Of the parts you mentioned, motherboard and CPU, it's much more likely the
> motherboard. However if the PSU is generic you ought to try a name-brand,
> 300W or more.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Thu, 29 Apr 2004 09:39:05 -0700, "Kyle Griffin" <kcfg@comcast.net>
wrote:


>> This 350W PSU, it's a name brand so you can trust the wattage on the
>> label? It might be a stupid question, but do you have the 4-pin 12V
>> connector, connected? Perhaps it's not the PSU, but "assuming" it isn't
>> could make your efforts a waste if that's the problem.
>>
>> Of the parts you mentioned, motherboard and CPU, it's much more likely the
>> motherboard. However if the PSU is generic you ought to try a name-brand,
>> 300W or more.
>
>It's a Raidmax psu and case, and yes, I did have the 4-pin connected and the
>voltage checked out, but again; no ammeter, so I can't tell if the wattage
>adds up.

Oh... yeah I have a Raidmax 350W here, brand new/unused, and never will be
used due to it being such a piece of junk. I'd speculate it's worth about
230W watts of dirty power, is quite crude compared to something like a
240W Delta. Replace the power supply. Even if it's not the only problem,
it's still a problem-waiting-to-happen.

>It's just a little bit frustrating having to choose from 3 possibilities.
>The power supply has 2 quality control stickers on it, but that's just their
>say so.

Quality control stickers could mean two things:

1) A person or machine with a roll of stickers was at their post when the
unit rolled down the conveyor line.

2) THEY are happy that it's what they intended to make. It is quality
relative to THEIR standard. For example, if they spec'd a rubber band to
power a Fighter Jet engine, a properly produced rubber band could have a
quality control sticker.


> The motherboard, on the other hand, is a model number that isn't
>even listed on the Asus website.

???
http://www.asus.com/products/mb/socket478/p4sgx-mx/overview.htm


>There are a few similar ones with similar
>model numbers, but not this one. The store said I could exchange it for a
>different board and pay the balance, if any, so I can try that and see if it
>solves my problem.

I'd try a different power supply first, and even if the board turns out to
be bad, replace the power supply anyway.

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

"kony" <spam@spam.com> wrote in message
news:t5s290h7p3f3oq0oeh6gf1pdad1gbqm3ku@4ax.com...
> On Thu, 29 Apr 2004 09:39:05 -0700, "Kyle Griffin" <kcfg@comcast.net>
> wrote:
>
>
> >> This 350W PSU, it's a name brand so you can trust the wattage on the
> >> label? It might be a stupid question, but do you have the 4-pin 12V
> >> connector, connected? Perhaps it's not the PSU, but "assuming" it
isn't
> >> could make your efforts a waste if that's the problem.
> >>
> >> Of the parts you mentioned, motherboard and CPU, it's much more likely
the
> >> motherboard. However if the PSU is generic you ought to try a
name-brand,
> >> 300W or more.
> >
> >It's a Raidmax psu and case, and yes, I did have the 4-pin connected and
the
> >voltage checked out, but again; no ammeter, so I can't tell if the
wattage
> >adds up.
>
> Oh... yeah I have a Raidmax 350W here, brand new/unused, and never will be
> used due to it being such a piece of junk. I'd speculate it's worth about
> 230W watts of dirty power, is quite crude compared to something like a
> 240W Delta. Replace the power supply. Even if it's not the only problem,
> it's still a problem-waiting-to-happen.
>
> >It's just a little bit frustrating having to choose from 3 possibilities.
> >The power supply has 2 quality control stickers on it, but that's just
their
> >say so.
>
> Quality control stickers could mean two things:
>
> 1) A person or machine with a roll of stickers was at their post when the
> unit rolled down the conveyor line.
>
> 2) THEY are happy that it's what they intended to make. It is quality
> relative to THEIR standard. For example, if they spec'd a rubber band to
> power a Fighter Jet engine, a properly produced rubber band could have a
> quality control sticker.
>
>
> > The motherboard, on the other hand, is a model number that isn't
> >even listed on the Asus website.
>
> ???
> http://www.asus.com/products/mb/socket478/p4sgx-mx/overview.htm
>
>
> >There are a few similar ones with similar
> >model numbers, but not this one. The store said I could exchange it for
a
> >different board and pay the balance, if any, so I can try that and see if
it
> >solves my problem.
>
> I'd try a different power supply first, and even if the board turns out to
> be bad, replace the power supply anyway.
>

Well, it seems it was the board after all. I exchanged it for a different
brand and it posted just fine. I may take your advice on the power supply
and switch it out real soon. Now that the whole thing actually works, I'd
like to avoid frying it. It was kind of my intention to swap things in and
out as I went. The only real problem was getting it going in the first
place.

Thank you for your time and advice, all who replied. It was helpful in
speeding up the diagnosis. Now, does anyone have any recommendations on a
good motherboard/chipset and CPU to use to build a small Linux server?

-Kyle

 

[Guest]

Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Fri, 30 Apr 2004 08:21:56 -0700, "Kyle Griffin" <kcfg@comcast.net>
wrote:


>Well, it seems it was the board after all. I exchanged it for a different
>brand and it posted just fine. I may take your advice on the power supply
>and switch it out real soon. Now that the whole thing actually works, I'd
>like to avoid frying it. It was kind of my intention to swap things in and
>out as I went. The only real problem was getting it going in the first
>place.
>
>Thank you for your time and advice, all who replied. It was helpful in
>speeding up the diagnosis. Now, does anyone have any recommendations on a
>good motherboard/chipset and CPU to use to build a small Linux server?

Good to see you sorted it out.
That looks like a good end to this thread and a topic for a new thread.

Briefly the parts for the "small Linux server" might depend on what it's
doing exactly, type of serving, number of clients, number/size of files,
network speed or bottleneck potential.

Supposedly it's harder to set up an nForce platform though I've not tried
one. If "small" means little need for performance then consider that the
LAN may be the primary bottleneck and any-old box might be suitable so
long as it's stable and fit for long->24/7 uptimes.