1 - Possible motherboard failure - BIOS ROM checksum failure.
-----------------
Hmm, try
Replacing CMOS battery Default BIOS to get a good-clean- fresh reset on everything. Reseat and check video card(s) Reseat ram dimms and all PCIe cards Reseat all connectors
But I need to clarify that my situation has no 'repetitive' beeping. Each startup (whether a signal is successfully sent to the monitor and detected) has a single beep which, I believe, is the normal beep.
Please excuse that I forgot to say that I have already tried:
1. loading the default BIOS but the problem remains.
2. run the PSA test and it passed the regular test (but I did not run the 'extended memory' test)
3. updated to latest nVidia graphics driver but problem remains.
And it seems that the problem is getting worse in the sense that it happens more often, at least during today.
I checked some back posts in this forum and found a post below having similar problem, even though his is a X51 and mine is an Aurora R4.
So, I have now taken out the installed two GTX660 in SLI. To my surprise, a black small plastic 'rod' (10mm length) fell out when I removed the card from the motherboard slot. It does not look like anything related to the card PCB but looks like a part of the black plastic shroud. When I shake this card, something else is still rattling inside. This is definitely not normal.
Now, I have installed a single (a known working spare card) GTX570.
I have found out what that little plastic rod is. As there is some rattling noise inside the plastic shroud, I shake and a little spring fell out. I did an extensive online pic search and comparison, I am certain that little black plastic rod is one of the two hold-down pins responsible for holding down the heatsink for the VRM section (not for the GPU). With that pin broken off, its spring just has to come off as well, of course.
This means the VRM is not receiving enough contact pressure and the cooling is sub optimal. I think this is a high chance of being the culprit that makes the card not sending a signal sent to the monitor during startup. (But I do not remember whether this card is the primary or secondary card of the SLI pair. But I suppose, if one card has a problem, it will affect the SLI pair.)
Anyway, my spare card is now installed and has no problem for 1 day. If no problem comes in, say, 7 days, then I will put back the good GTX660 for the next testing period.
(This bad GTX660 needs to be replaced regardless if it is the trouble maker).
The problem is 'solved' and I like to document the rest of the story, which might be helpful to other people facing similar problem in the future.
Continuing from my post above:
I tested my spare GTX570 in both motherboard's slots and the problem did not come in 10 days.
Then, I installed 1st GTX660 back to the motherboard and tested both slots and the problem did not come in 10 days.
Then, I put the 1st GTX660 into the bottom slot (main slot of this motherboard) and put the 2nd GTX660 (with broken hold-down pin) to the upper slot. Reconnect them with the SLI bridge and enabled the SLI via Nvidia Control Panel. Under this testing configuration, I expected the problem to come back immediately. But strangely, it did not, after a few days. Perhaps I should test even longer.
But the 2nd GTX660 has a broken hold-down pin, I wanted the card to be replaced anyway. So, I called in and Dell dispatched a technician on the next day with a new card. He explained that the hold down pin was broken primarily due to the heat expansion of the thermal pad between the VRM units and the heatsink. He had seen a few of this cases.
Anyway, I have the 2nd card replaced and the system has been good for a couple days now. I hope the problem is solved or just has gone away on its own.
If nothing else happens, this thread can be changed to a 'solved' status.
Alienware-Naomi
4 Operator
•
1.8K Posts
0
April 25th, 2014 12:00
You can send me private message with your Service Tag and contact information once you've determined if one or both cards need to be replaced.
Tesla1856
8 Wizard
•
17.4K Posts
0
April 23rd, 2014 23:00
Repetitive Beep Codes
1 - Possible motherboard failure - BIOS ROM checksum failure.
-----------------
Hmm, try
Replacing CMOS battery
Default BIOS to get a good-clean- fresh reset on everything.
Reseat and check video card(s)
Reseat ram dimms and all PCIe cards
Reseat all connectors
painterhenry
46 Posts
0
April 23rd, 2014 23:00
Thank you for your quick response.
But I need to clarify that my situation has no 'repetitive' beeping. Each startup (whether a signal is successfully sent to the monitor and detected) has a single beep which, I believe, is the normal beep.
Please excuse that I forgot to say that I have already tried:
1. loading the default BIOS but the problem remains.
2. run the PSA test and it passed the regular test (but I did not run the 'extended memory' test)
3. updated to latest nVidia graphics driver but problem remains.
And it seems that the problem is getting worse in the sense that it happens more often, at least during today.
I checked some back posts in this forum and found a post below having similar problem, even though his is a X51 and mine is an Aurora R4.
http://en.community.dell.com/owners-club/alienware/f/3746/t/19579222.aspx
So, I have now taken out the installed two GTX660 in SLI. To my surprise, a black small plastic 'rod' (10mm length) fell out when I removed the card from the motherboard slot. It does not look like anything related to the card PCB but looks like a part of the black plastic shroud. When I shake this card, something else is still rattling inside. This is definitely not normal.
Now, I have installed a single (a known working spare card) GTX570.
Anyone's input is welcome. Thanks in advance.
painterhenry
46 Posts
0
April 24th, 2014 00:00
If I understand correctly, 'repetitive beep codes' means a continuous beeping when POST fails.
1-beep means: beep..............beep.............beep...........beep.........
2-beep means: beep beep.........beep beep..........beep beep.........beep beep.........
But I only have one single beeping sound, which I think is the most common sound for a successful POST.
Alienware-Naomi
4 Operator
•
1.8K Posts
0
April 24th, 2014 16:00
You're right about about beeps, one single beep is not the same as the No post Beep Codes.
I think the plastic part could one of the video covers. Test the computer with the single card and keep us updated with the results.
painterhenry
46 Posts
0
April 25th, 2014 01:00
Thank you for responding.
I have found out what that little plastic rod is. As there is some rattling noise inside the plastic shroud, I shake and a little spring fell out. I did an extensive online pic search and comparison, I am certain that little black plastic rod is one of the two hold-down pins responsible for holding down the heatsink for the VRM section (not for the GPU). With that pin broken off, its spring just has to come off as well, of course.
This means the VRM is not receiving enough contact pressure and the cooling is sub optimal. I think this is a high chance of being the culprit that makes the card not sending a signal sent to the monitor during startup. (But I do not remember whether this card is the primary or secondary card of the SLI pair. But I suppose, if one card has a problem, it will affect the SLI pair.)
Anyway, my spare card is now installed and has no problem for 1 day. If no problem comes in, say, 7 days, then I will put back the good GTX660 for the next testing period.
(This bad GTX660 needs to be replaced regardless if it is the trouble maker).
I will keep this updated.
In the meantime, any input/comment is welcome.
painterhenry
46 Posts
0
June 26th, 2014 17:00
The problem is 'solved' and I like to document the rest of the story, which might be helpful to other people facing similar problem in the future.
Continuing from my post above:
I tested my spare GTX570 in both motherboard's slots and the problem did not come in 10 days.
Then, I installed 1st GTX660 back to the motherboard and tested both slots and the problem did not come in 10 days.
Then, I put the 1st GTX660 into the bottom slot (main slot of this motherboard) and put the 2nd GTX660 (with broken hold-down pin) to the upper slot. Reconnect them with the SLI bridge and enabled the SLI via Nvidia Control Panel. Under this testing configuration, I expected the problem to come back immediately. But strangely, it did not, after a few days. Perhaps I should test even longer.
But the 2nd GTX660 has a broken hold-down pin, I wanted the card to be replaced anyway. So, I called in and Dell dispatched a technician on the next day with a new card. He explained that the hold down pin was broken primarily due to the heat expansion of the thermal pad between the VRM units and the heatsink. He had seen a few of this cases.
Anyway, I have the 2nd card replaced and the system has been good for a couple days now. I hope the problem is solved or just has gone away on its own.
If nothing else happens, this thread can be changed to a 'solved' status.