New motherboard/CPU/RAM not completing POST
Alex Fraser
get a bit more processing power but it is not getting as far as POST and
behaving ... strangely. Having reinstated the old components, the
system, which includes a Seasonic S12 430W PSU, seems to be working fine.
The new bits are an Asus P5Q SE Plus motherboard, C2D E8400 and 2x
Crucial 2GB DDR2-1066.
Regardless of what additional components are connected, the system
behaves the same as with just PSU/motherboard/CPU/RAM. On pressing the
power button, the fans spin etc but there are no other signs of life.
From this state, if I press the reset button, the system appears to
reset but after a second or so goes completely dead - as though the PSU
switches the main supply off. A couple of seconds later, it springs back
to life.
I can turn the power off by holding the power button down for a few seconds.
Any thoughts on what's going on?
I can swap the RAM, and I have another PSU I can try, but not until
Sunday. Given the behaviour described above - which feels like some kind
of PSU protection kicking in - I'm not sure whether trying the other PSU
(an Antec Earthwatts 380W) is a good idea. I don't have another S775 CPU
or motherboard to test with.
What would you recommend I do?
Alex
><(((°>
Some things to try:
Remove the CPU and give the socket & pins a good visual check.
Are all pins same height and aligned?
The slightest knock to a pin can cause it to not connect to the CPU
Try clearing the BIOS
Check the CMOS battery is 3V (or try another one)
Is clear CMOS jumper in correct place
Try singularly both your memory sticks
Try alternate memory
Try another graphics card
Check CPU compatibility with current BIOS and board rev levels (label on
BIOS chip?)
BIOS flash might be needed
Give the board a good visual inspection - any scratches over tracks etc
Try an alternate CPU
All else fails RMA the board for replacement.
Tippunmon
"Chronos" <m...@privacy.net> wrote in message
news:hfda0t$5cn$1...@news.eternal-september.org...
> Alex Fraser wrote:
>
>> What would you recommend I do?
>
> on
> the board. In fact, remove the battery (between the SATA ports and the
> southbridge heat sink) for ten minutes. Some Asus boards do power
> themselves
> down and back up again when re-reading the NVRAM (my M3A78-Pro does the
> same
> thing, even when simply changing a setting in setup - quite unnerving when
> you've never seen a board do this before).
>
> I've seen boards arriving out of the box with a bad CMOS battery. If the
> voltage has dropped below 2V, I have found boards refusing to POST. Worth
> trying the lithium battery from the old motherboard.
>
> ref: http://www.newegg.com/Product/Product.aspx?Item=N82E16813131347
> comment
> by "tommy" on 29th Nov 2009
>
> Try the memory modules one at a time. Try it with no memory fitted at all
> to
> ensure you get a warning from the speaker.
>
> My next job would be voltmeter on the PSU and a POST card, but I'm
> guessing
> you have neither of these.
>
> Other than that, just ensure you have a speaker connected to the front
> panel
> connector and that, if it's one of those little piezo-electric
> transducers,
> it's the right way around as polarity matters with these. If you still get
> no response and no beep codes on either PSU (I'd say the 380W will be fine
> as long as you're not using a monster graphics card), I'd be tempted to
> look
> closely at the LGA775 carrier for the word "Foxconn."
> --
> Chronos
Also check the board is correctly identifying the memory speed in the BIOS
as this board takes several different memory speeds.
Flushing the BIOS to default settings as has already been suggested is the
first thing to do, then run the board with only the "bare necessities" to
allow it to boot.
Regards
Tippunmon
Alex Fraser
All looks fine as far as I can see.
> Try clearing the BIOS
No change.
> Check the CMOS battery is 3V (or try another one)
I will be able to do that tomorrow.
> Is clear CMOS jumper in correct place
Checked as part of clearing CMOS :).
> Try singularly both your memory sticks
> Try alternate memory
Will be doing that tomorrow.
> Try another graphics card
Presence or absence of the graphics card doesn't change the behaviour,
and the one I used works fine in my old board. But I have another I can try.
> Check CPU compatibility with current BIOS and board rev levels (label on
> BIOS chip?)
> BIOS flash might be needed
I couldn't see anything on the board indicating the BIOS version.
According to the Asus site, the CPU is supported in the oldest BIOS and
all PCB revisions.
> Give the board a good visual inspection - any scratches over tracks etc
Looks fine as far as I can see.
> Try an alternate CPU
Not an option unfortunately.
> All else fails RMA the board for replacement.
I get the feeling I'll end up doing that :).
Thanks for the suggestions,
Alex
Smurf
Some suggestions, some are obvious, so dont be insulted.
i) Are you sure you have plugged in the 12v 4pin atx power cable.
ii) Have you tried resetting bios/ checking the bios jumpers are in the
correct position (some boards, mainly pcchips, are known to come with the
pins jumpered wrongly)
iii) Try removing the bios battery, a battery giving out a lower then normal
voltage can cause some boards to not start.
Alex Fraser
> Alex Fraser wrote:
>
>> What would you recommend I do?
>
> the board. In fact, remove the battery (between the SATA ports and the
> southbridge heat sink) for ten minutes. Some Asus boards do power themselves
> down and back up again when re-reading the NVRAM (my M3A78-Pro does the same
> thing, even when simply changing a setting in setup - quite unnerving when
> you've never seen a board do this before).
>
> I've seen boards arriving out of the box with a bad CMOS battery. If the
> voltage has dropped below 2V, I have found boards refusing to POST. Worth
> trying the lithium battery from the old motherboard.
>
> ref: http://www.newegg.com/Product/Product.aspx?Item=N82E16813131347 comment
> by "tommy" on 29th Nov 2009
>
> Try the memory modules one at a time. Try it with no memory fitted at all to
> ensure you get a warning from the speaker.
>
> My next job would be voltmeter on the PSU and a POST card, but I'm guessing
> you have neither of these.
>
> Other than that, just ensure you have a speaker connected to the front panel
> connector and that, if it's one of those little piezo-electric transducers,
> it's the right way around as polarity matters with these. If you still get
> no response and no beep codes on either PSU (I'd say the 380W will be fine
> as long as you're not using a monster graphics card), I'd be tempted to look
> closely at the LGA775 carrier for the word "Foxconn."
Thanks Chronos - some of the above I've covered in my previous reply.
I've arranged to borrow a multimeter. I'll experiment with memory and
PSU tomorrow. The speaker is definitely the right way round, but no I
don't have a POST card.
Alex
Mike Tomlinson
<m...@privacy.net> writes
>What would you recommend I do?
Take the board out of the case, and test it 'bare' (sit it on the mobo
box to test with the bare minimum of hardware - cpu, memory and video
card). If it then works, make sure you have the motherboard stand-offs
lined up with the holes in the board and there isn't one 'spare'
shorting the bottom of the board.
--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png
BigH2K
"Alex Fraser" <m...@privacy.net> wrote in message
news:hfd6ti$nbg$1...@news.eternal-september.org...
Hi Alex, are you saying that when you plug the base unit it, switch on the
PSU and then hit the on switch that you have power for a split second and
then it goes dead and then no matter how long you leave it afterwards that
it stays dead?
Or... when you hit the power button you have power for a split second and
then it goes dead but if you leave it a few seconds it then boots?
If it's the second then it's quite normal on modern systems but doesn't
happen if the machine is left plugged in with power as opposed to switched
off completely at the mains.
Alex Fraser
> In article <hfd6ti$nbg$1...@news.eternal-september.org>, Alex Fraser
> <m...@privacy.net> writes
>
>> What would you recommend I do?
>
> Take the board out of the case, and test it 'bare' (sit it on the mobo
> box to test with the bare minimum of hardware - cpu, memory and video
> card). If it then works, make sure you have the motherboard stand-offs
> lined up with the holes in the board and there isn't one 'spare'
> shorting the bottom of the board.
Thanks Mike, that's a good idea. I'm fairly sure it isn't a short - when
I have seen a short before (though with a different PSU) it resulted in
the PSU shutting down until it was power cycled, which is not what I am
seeing. But it's a simple thing to do that puts the matter beyond any doubt.
Alex
Alex Fraser
Neither :). Although closer to the latter.
The behaviour I see is that pressing the power button - whether the
first time since switching on at the mains or after holding the power
button until the system switches off - fans etc power up immediately and
seemingly not much else happens. I probably didn't leave it like this
for more than a few tens of seconds but suspect nothing would have
changed if I did.
From this condition, pressing and releasing the reset button causes it
to behave as if starting to reset, then the PSU switches off, then
switches on a few seconds later. And then it seems to be in the same
condition again.
Alex
Mike Tomlinson
<m...@privacy.net> writes
>It's always the simple
>things that trip you up, isn't it?
*nods vigorously*
Mike Tomlinson
<m...@privacy.net> writes
>Thanks Mike, that's a good idea.
It'll only take a moment to check.
> I'm fairly sure it isn't a short - when
>I have seen a short before (though with a different PSU) it resulted in
>the PSU shutting down until it was power cycled, which is not what I am
>seeing.
It seems to have different results with different boards (which makes
sense as the layout differs between boards). In the 286 and 386 era it
more often than not caused a 'CMOS checksum failure', for instance.
Good luck.
PeeGee
are not shutting down. That, in turn, suggests the CPU is not running
(correctly) as nothing else happens.
I have only had that situation once (when looking after 300+ computers
at a local school) and replacing the CPU fixed it.
>
> From this condition, pressing and releasing the reset button causes it
> to behave as if starting to reset, then the PSU switches off, then
> switches on a few seconds later. And then it seems to be in the same
> condition again.
>
Does the board go into "sleep" mode for a short press, according to the
manual?
A slightly different situation I had at home, which probably doesn't
apply here, was a system that shutdown within a second of being started.
It had a three speed CPU fan and the BIOS shutdown the system if the CPU
fan failed (ie ran slower than 2000rpm) - it obviously didn't recognise
1500rpm as valid (then I RTFM) :-(
--
PeeGee
"Nothing should be able to load itself onto a computer without the
knowledge or consent of the computer user. Software should also be able
to be removed from a computer easily."
Peter Cullen, Microsoft Chief Privacy Strategist (Computing 18 Aug 05)
Alex Fraser
[snip]
Thanks again to everyone who replied.
Set it up outside the case, as per Mike's suggestion (why I didn't think
to try that I don't know...), and it got as far as attempting to boot
(but nothing was connected). The behavior on pressing reset also
disappeared.
So, it seems like the PSU was partially shut down due to a short to the
case (although the reset behaviour still seems odd), and the next step
is to work out where. Suggestions for how to do that would be gratefully
received, as it'll be a couple of days before I have time to do anything
more (sigh!).
Alex
Mike Tomlinson
<m...@privacy.net> writes
>(but nothing was connected). The behavior on pressing reset also
>disappeared.
Oh good. So you're now getting something on screen? This means it's
99% working.
> the next step
>is to work out where. Suggestions for how to do that would be gratefully
>received
Just eyeball it. Much of the time it's an extra standoff on the back
centre of the motherboard plate which doesn't match up with a hole in
the board. (hint: count the number of standoffs fitted and compare with
the number of mounting holes in the board)
I guarantee you'll feel such a numpty when you spot it :o)
Rob Morley
Mike Tomlinson <mi...@jasper.org.uk> wrote:
> In article <hfgvp7$hb2$1...@news.eternal-september.org>, Alex Fraser
> <m...@privacy.net> writes
>
> >(but nothing was connected). The behavior on pressing reset also
> >disappeared.
>
> Oh good. So you're now getting something on screen? This means it's
> 99% working.
>
> > the next step
> >is to work out where. Suggestions for how to do that would be
> >gratefully received
>
> Just eyeball it. Much of the time it's an extra standoff on the back
> centre of the motherboard plate which doesn't match up with a hole in
> the board. (hint: count the number of standoffs fitted and compare
> with the number of mounting holes in the board)
>
> I guarantee you'll feel such a numpty when you spot it :o)
>
Or it can be a loose screw that was lodged between mobo and case, which
fell out when you removed the mobo so you never find it and it remains
a total mystery.
Alex Fraser
Well, I am certain it wasn't a screw because I am always meticulous
about counting them out/in... But the parts now seem to be fine in the
case and I'm prepared to live with the mystery :).
Alex
Nick Le Lievre
>
> Well, I am certain it wasn't a screw because I am always meticulous about
> counting them out/in... But the parts now seem to be fine in the case and
> I'm prepared to live with the mystery :).
You must of angered the GOD of PC building and he sent you that little
hiccup to throw you off course... little things like that always seem to
happen when dealing with PCs and I am sure that there is some higher power
involved.
Johnny B Good
from "Nick Le Lievre" <nickle...@jerseymail.co.uk.invalid> contains
these words:
Most of us know _this_ particular god's name... it's Murphy! A very
'Hands On' mischeivious minor god with a special interest in all things
technical (not just PC building).
HTH & HAND ;-)
--
Regards, John.
Please remove the "ohggcyht" before replying.
The address has been munged to reject Spam-bots.