Pegatron M2N78-LA (rev. 3.02) freezes on POST with SATA drives attached
Like the title says, I have a Pegatron M2N78-LA (HP P/N: 513430-002) motherboard out of an HP Pavilion PC. The motherboard hard-locks during POST when it gets to the part of trying to detect the SATA devices *only* if there is a SATA device (be it HDD or optical drive) connected to any of the SATA ports. I cannot enter BIOS or do anything when this occurs. Pressing Ctrl+Alt+Del cannot restart the system either. In fact, the keyboard locks up too (Num Lock status LED cannot be toggled anymore.). However, if I disconnect everything from the motherboard's SATA ports, the motherboard can POST successfully and then goes on with the usual message that there is “No Boot Device Found” (which is true.) Also, when I have everything disconnected from the motherboard's SATA ports, I can go into BIOS without problems and change settings as I like. Actually, the first thing I tried was resetting CMOS settings (several times, too) and making sure everything is set correctly there.
On that note, I've noticed that if I set the SATA device type to either IDE or RAID, the motherboard won't hard-lock during POST, but instead will just sit on the SATA device detection screen and never goes past it. Pressing Ctrl+Alt+Del can restart the system fine and keyboard doesn't lock up. Only when I have the SATA device type set to ACHI (which is the default value, BTW), the board hard-locks
Unfortunately, the M2N78-LA doesn't have onboard IDE controller, so I can't test with an IDE HDD or optical drive. However, the board does have an FDD / floppy controller. Interestingly enough, that doesn't appear to work either - the motherboard never sees the floppy drive/device as connected. As such, inserting a bootable floppy does nothing (not even an attempt is made to spin it up.) Just to be sure, I did test with two known working FDDs and cables that I used recently, too. Now if I am not mistaken, FDD goes through the Super IO / LPC chip, which traditionally also provides PS/2, Serial, and Parallel port functions (and on this board, at least my test PS/2 keyboard was working fine.)
With that said, probably first thing I should mention is that the M2N78-LA does use a GeForce chipset / MCP (and I suspect that may be the root of the problem.)
In fact, there appears to be noticeable heat discoloration on the PCB around and under the chipset, along with a few other hot areas - though that is more or less something I regularly see on these GeForce chipsets, since manufacturers rarely bother to put adequate heatsinks on them (and these suckers run hot too.)
https://www.badcaps.net/forum/attach...1&d=1606626406
In terms of what I've tested with / so far…
- I've tried with different HDDs and optical drives - one device or multiple devices at a time, and tested on all ports. Interesting note here: when an optical drive is connected to the SATA ports, the drive becomes locked and trying to eject the tray does nothing.
- I've tried different and known-good sticks of RAM in different slots, one stick at a time.
- I've tested with two known good PSUs – one of them being a fully recapped HiPro and the other a partially-recapped Delta.
- I've tried resetting CMOS (in fact, I keep the motherboard without a CMOS battery, as I can't really use it for a PC build, obviously.)
- I've tried different CPUs too - both the original Athlon II X4 that came with the HP Pavilion PC and also a known-good X2 5200+.
- And finally, I did try a quick reflow/reheat of the nVidia chipset. But I only managed to heat it to ~180C, so perhaps it wasn't hot enough. Nothing changed in terms of function (or lack thereof) after this.
- Oh, and regarding caps - I checked (out of circuit) all of the single ones that filter important rails and even replaced a few, just in case… but still no change in operation.
So any ideas what else I could try?
I think I pretty much exhausted most options. Maybe reflow GF MCP again? Otherwise, I was also thinking maybe I can get a PCI-E 1x SATA expansion card (though I feel like that may not be worth the investment, considering this is a GeForce chipset motherboard.) What's your take on that?
As far as history of the board goes… well, I picked up this PC as scrap, along with 10-12 others, from a PC repair shop that was cleaning up storage. So unfortunately, I don't have any history of when the motherboard failed or how. All I can say is that about 2/3 of the other PCs from that place ended up working. Many just needed minor work, like new caps, working PSU, and/or good cleaning, along with parts (most lacked RAM, HDD, and etc.)
Anyways, here is a picture of the motherboard, so you can see what I'm dealing with:
https://www.badcaps.net/forum/attach...1&d=1606626406
Also, here are a few pictures of the chipset, from before the reflow (with gunk thermal compound still on there) and after the reflow (cleaned).
https://www.badcaps.net/forum/attach...1&d=1606626406
https://www.badcaps.net/forum/attach...1&d=1606626406
The chipset is a GeForce GF8100P-A-A2. Are these affected by the bumpgate? And if so, would you say I'd be wasting time trying to reflow it again? I have a similar Pegatron motherboard from another HP PC from that repair shop. Not sure if it also is an M2N78-LA or not, but from what I could tell when I had it opened, it was the same board, but only with 2 RAM slots, and IIRC also a few other ports missing. It came out of a lower-end HP Pavilion, and thus only packed an Athlon II X2. That PC ended up working fine, though. I added a small 40 mm fan on the chipset / MCP heatsink, and it seems to run adequately cool. I don't remember if that board had the same kind of PCB burn / darkening as this one, though.
Anyways, thanks for reading and cheers!
On that note, I've noticed that if I set the SATA device type to either IDE or RAID, the motherboard won't hard-lock during POST, but instead will just sit on the SATA device detection screen and never goes past it. Pressing Ctrl+Alt+Del can restart the system fine and keyboard doesn't lock up. Only when I have the SATA device type set to ACHI (which is the default value, BTW), the board hard-locks
Unfortunately, the M2N78-LA doesn't have onboard IDE controller, so I can't test with an IDE HDD or optical drive. However, the board does have an FDD / floppy controller. Interestingly enough, that doesn't appear to work either - the motherboard never sees the floppy drive/device as connected. As such, inserting a bootable floppy does nothing (not even an attempt is made to spin it up.) Just to be sure, I did test with two known working FDDs and cables that I used recently, too. Now if I am not mistaken, FDD goes through the Super IO / LPC chip, which traditionally also provides PS/2, Serial, and Parallel port functions (and on this board, at least my test PS/2 keyboard was working fine.)
With that said, probably first thing I should mention is that the M2N78-LA does use a GeForce chipset / MCP (and I suspect that may be the root of the problem.)
In fact, there appears to be noticeable heat discoloration on the PCB around and under the chipset, along with a few other hot areas - though that is more or less something I regularly see on these GeForce chipsets, since manufacturers rarely bother to put adequate heatsinks on them (and these suckers run hot too.)
https://www.badcaps.net/forum/attach...1&d=1606626406
In terms of what I've tested with / so far…
- I've tried with different HDDs and optical drives - one device or multiple devices at a time, and tested on all ports. Interesting note here: when an optical drive is connected to the SATA ports, the drive becomes locked and trying to eject the tray does nothing.
- I've tried different and known-good sticks of RAM in different slots, one stick at a time.
- I've tested with two known good PSUs – one of them being a fully recapped HiPro and the other a partially-recapped Delta.
- I've tried resetting CMOS (in fact, I keep the motherboard without a CMOS battery, as I can't really use it for a PC build, obviously.)
- I've tried different CPUs too - both the original Athlon II X4 that came with the HP Pavilion PC and also a known-good X2 5200+.
- And finally, I did try a quick reflow/reheat of the nVidia chipset. But I only managed to heat it to ~180C, so perhaps it wasn't hot enough. Nothing changed in terms of function (or lack thereof) after this.
- Oh, and regarding caps - I checked (out of circuit) all of the single ones that filter important rails and even replaced a few, just in case… but still no change in operation.
So any ideas what else I could try?
I think I pretty much exhausted most options. Maybe reflow GF MCP again? Otherwise, I was also thinking maybe I can get a PCI-E 1x SATA expansion card (though I feel like that may not be worth the investment, considering this is a GeForce chipset motherboard.) What's your take on that?
As far as history of the board goes… well, I picked up this PC as scrap, along with 10-12 others, from a PC repair shop that was cleaning up storage. So unfortunately, I don't have any history of when the motherboard failed or how. All I can say is that about 2/3 of the other PCs from that place ended up working. Many just needed minor work, like new caps, working PSU, and/or good cleaning, along with parts (most lacked RAM, HDD, and etc.)
Anyways, here is a picture of the motherboard, so you can see what I'm dealing with:
https://www.badcaps.net/forum/attach...1&d=1606626406
Also, here are a few pictures of the chipset, from before the reflow (with gunk thermal compound still on there) and after the reflow (cleaned).
https://www.badcaps.net/forum/attach...1&d=1606626406
https://www.badcaps.net/forum/attach...1&d=1606626406
The chipset is a GeForce GF8100P-A-A2. Are these affected by the bumpgate? And if so, would you say I'd be wasting time trying to reflow it again? I have a similar Pegatron motherboard from another HP PC from that repair shop. Not sure if it also is an M2N78-LA or not, but from what I could tell when I had it opened, it was the same board, but only with 2 RAM slots, and IIRC also a few other ports missing. It came out of a lower-end HP Pavilion, and thus only packed an Athlon II X2. That PC ended up working fine, though. I added a small 40 mm fan on the chipset / MCP heatsink, and it seems to run adequately cool. I don't remember if that board had the same kind of PCB burn / darkening as this one, though.
Anyways, thanks for reading and cheers!
Attached Files
just give the shitset an mma punch! 
if it fixes it, 99% its the shitset! no need to waste energy on a reflow if a punch doesnt fix it then its not the shitset. 
)
), followed by a prompt to restart the PC… so I did. The motherboard went through its usual boot-loop cycle that it does when the CMOS has been reset/erased. Thus, I had to go in BIOS and re-enter some of the settings that got erased. After this, I let Windows 7 start normally again through the PCI-E 1x to SATA card, just to verify that the system is stable with the new BIOS.
which ran the fan slower, but still provided more than adequate cooling performance. Not only that, but now when I try to run SpeedFan (a software utility used for hardware monitoring), Windows crashes to a brown/gray/black (usually one of the three) -colored screen, whereas before, I had no problems running it. That’s not too big of a deal, though. Lastly, I also noticed the CPU seems to switch between different performance states (and V_core voltages) way too often - almost constantly, in fact. Perhaps the new v5.22 BIOS is optimized for the newer Athon II CPUs better and not so much for the older Athlon X2 5200+ (Windsor) CPU I have in the PC right now. In any case, the new BIOS did not seem to add any positive improvements in system performance. 
… if I couldn’t get Windows 7 to install from a DVD, then why not clone an already installed Windows 7 HDD? After all, I have a nearly identical HP Pavilion PC, which I originally thought had the same motherboard. 
So I cloned the HDD from that PC anyways. But upon installing it in the M2N78-LA mobo and trying to boot, Windows 7 crashed like it would when a boot HDD from a completely different system is used.
I get my fancy optical drive eject tool (really just a paper clip with a ring formed in one end 
And of course, at least half of the “driver” entries were for HP Update software and various wireless keyboard drivers - as if these are soo damn important. 
Needless to say, I had to go to 3rd party websites to get my drivers… which can be a shady endeavor. But I found one place that seemed to have legit drivers, so I stuck with that. And more importantly, the website worked… unlike HP’s giant turd posing as a website.
Comment