MSI X99A SLI Plus - Very odd mobo malfunction

[Alpha0mega]

About two months ago, my MSI X99S SLI Plus died. It stopped booting, both BIOS lights came on at the same time and the CPU fan connector stopped working too. Additionally, it killed my 5820k. Looking online, I am not the only one to have experienced this type of malfunction on this board, including it killing the processor (https://forum-en.msi.com/index.php?topic=252711.0).

Anyway, I RMA’ed processor to Intel and board to MSI, and the latter sent a refub board of the same make, which had two of its RAM slots non-functional. If they were populated, the board wouldn’t POST.

On sending it back, I was sent a X99A SLI Plus, a board identical to the X99S, apart from two USB ports being USB 3.1 Type A. Here is the where the odd malfunction happens.

This board POST’ed initially, I checked that all the ports and slots were working correctly, installed all my hardware and tested it for a few hours (RealBench, AIDA64). Everything seemed in order, so after doing some work, I turned it off for the night, which included turning off the PSU (as I normally do). Next morning I turned it on, to find that it didn’t POST. Both BIOS lights were on again. Tried the alternate BIOS, got POST screen, but the board wouldn’t go forward. The BIOS not currently selected would have its light flicker constantly.

Worried about having the processor killed again, I stopped experimenting at that point and sent the board back to MSI, after which they reported that the primary BIOS had been corrupted (for unknown reasons). They flashed it, and sent me a video to show that both BIOS were POST’ing. I also ordered a new PSU, the Seasonic SS-850KM3 and all new cables, just to be sure. On getting the board back, I installed only the bare minimum hardware, CPU/cooler, RAM, one GPU, one HDD. Didn’t even attach the front panel connectors or audio.

The board POST’ed on the primary BIOS (though I had to clear it first), and booted. I made no changes to the BIOS settings or run any tests this time. I just ran the system, browsed, worked etc. Then at night, I shut down, including the power to the (new) PSU. Next morning, exact same problem as last time. No POST, both BIOS lights on.

I have sent the board back to MSI, and since it has been two months since the initial RMA, with no end in sight, I have bought a new ASUS X99-A board.

MSI has been very unhelpful throughout the whole ordeal, currently saying that they will keep sending back the same board after re-flashing since BIOS corruption isn’t a “component problem”, and that leaving the board without power shouldn’t matter.

Can anyone here tell me what could possibly be happening with that board? Why leaving it completely powered down overnight should cause this? I have read about other people with faulty hardware who would have POST problems if the left the system off a significant time, but not why this happens. Could it be something do with temperature changes? Keep in mind that I did cut the power to the board numerous times during installation, for short periods (first boot was with only 1 RAM stick, then while checking each RAM slot and PCI-E connector etc.), but that didn’t cause a problem. Also, other mobo+processor combos I installed into the system between getting the board from MSI worked fine. Everything else was the same. My grounding is proper.

Other components:
i7 5820k
Noctua NH-D15 cooler
G.Skill 4x4GB DDR4 2400MHz RAM
Zotac GTX 970 x2
OCZ Vertex 2 120GB
WD Red 5TB
NZXT Phantom 530
Logitech G510+G402
CM Storm Sirus

TL;DR version: Replacement X99A SLI Plus primary BIOS gets corrupted when left overnight with no power (PSU off). Why?

 

[LTC8K6]

The obvious suggestion is the CMOS battery, which you probably already checked?

 

[nutternatter34]

I'm going for the dumbest solution ever, try a non-logitech keyboard. Seriously, preferably a keyboard that won't pull extra power to drive lighting and the lcd screen.

 

[Alpha0mega]

The CMOS battery was fine. Even if it wasn't, why would that lead to BIOS corruption, and only the primary BIOS? At most, it should mean that BIOS settings wouldn't be retained. Also, as I had mentioned, I had turned the power off multiple times, for over 30 minutes or more, before leaving it off for the night.

As for the keyboard, if the mobo is having a problem providing the paltry amount of power to a few LEDs, then it's not functional. External HDDs, USB audio, not to mention phones connected via USB, all draw far more power than the keyboard possible could. And why BIOS corruption of all things, after being left un-powered for a long period of time? Keep in mind, that I have been using the keyboard for quite sometime, including using it in the interim system, the first X99S replacement (the one with the bad RAM slots), as well as the new ASUS X99-A board. None of them had this problem.

I have read of people experiencing problems with a faulty system, after leaving it cold for half-a-day or so, but I have not read of anyone providing a possible reason for it. What I am really looking for is some sort of more-or-less technical explanation why this could possibly happen, so that I can tell MSI, who maintain that it shouldn't happen, despite being experienced by myself and others.

 

[XLer8or]

The BIOS settings are being corrupted which is part of it. UEFI settings to be specific.

run mem test
dif mem combo & slots.
Try dif kb & mouse as the kb and mouse interact with UEFI bios.
try brand new CMOS battery, abrupt power losses will corrupt such as PSU shutoff.

Basicly memory issues, abrupt power losses and interactions can cause it.

 

[bradly1101]

The CMOS battery was fine. Even if it wasn't, why would that lead to BIOS corruption, and only the primary BIOS? At most, it should mean that BIOS settings wouldn't be retained. Also, as I had mentioned, I had turned the power off multiple times, for over 30 minutes or more, before leaving it off for the night.

As for the keyboard, if the mobo is having a problem providing the paltry amount of power to a few LEDs, then it's not functional. External HDDs, USB audio, not to mention phones connected via USB, all draw far more power than the keyboard possible could. And why BIOS corruption of all things, after being left un-powered for a long period of time? Keep in mind, that I have been using the keyboard for quite sometime, including using it in the interim system, the first X99S replacement (the one with the bad RAM slots), as well as the new ASUS X99-A board. None of them had this problem.

I have read of people experiencing problems with a faulty system, after leaving it cold for half-a-day or so, but I have not read of anyone providing a possible reason for it. What I am really looking for is some sort of more-or-less technical explanation why this could possibly happen, so that I can tell MSI, who maintain that it shouldn't happen, despite being experienced by myself and others.

That is a puzzler. The only thing I can think of is that there might be a problem with these early revisions of the boards. When I went X79 I didn't buy right away, and when I did I received rev. 1.1 of the board. I'm not sure what the early problems were, but I'm happy they were worked out before I bought.

 

[eton975]

What does the board look like? This guy's X99S looked like it had no lacquer on it.

 

[PhIlLy ChEeSe]

Why are you shutting off the power? You also did not mention what power supply you have? Also define corrupt in your eyes?
I have seen certain motherboards not work well with certain power supplies.

 

[Alpha0mega]

run mem test dif mem combo & slots. Try dif kb & mouse as the kb and mouse interact with UEFI bios. try brand new CMOS battery, abrupt power losses will corrupt such as PSU shutoff.

The CMOS battery was new. The keyboard and mouse work fine with all other boards (including the two previous X99S) and even the faulty board except when it has been left off for a long period. If they were to blame, would it matter how long the system was left off?

What does the board look like? This guy's X99S looked like it had no lacquer on it.

I had seen that video, but the board in question was visually ok. Nothing stood out.

Why are you shutting off the power? You also did not mention what power supply you have? Also define corrupt in your eyes?

I have always shut off mains power during long periods of non-usage, like at night. In case there is a power surge etc. the hardware will be protected. I do have a surge protector+UPS, but still. I even used to literally pull the plug (which did help once) . I am also a very light sleeper, and the clicking of the UPS switching to backup in case of voltage fluctuation is enough to break my sleep. So I have to turn it off.

I mentioned the new PSU model, the SeaSonic SS-850KM3, but I see I forgot the previous one. It was SeaSonic S12-D 850w. The latter ran the system fine for around 8 months, and another system before that, before the board died (taking the CPU with it). The first time I got the X99A, I ran it off that PSU, since it was fine after test. I had gotten the new PSU before I got the board back the second time, just in case, but same behavior.

As for what corrupt is, that's MSI's terminology not mine. Power comes on, but the BIOS does not POST (no display), the secondary BIOS LED flickers constantly (it's not supposed to be on at all).

Another thing, perhaps my focus on no-power is a mistake, or maybe tunnel-vision on a single aspect. It's possible that the problem's source is temperature change related (hours of usage, then leaving it for long enough to cool), and the power off is just coincidental.

 

[bradly1101]

Have you put a meter on the power-out of the UPS when the PC's power supply is off? I'm grabbing at straws.

I don't know enough about the internal switching of PSU's and UPS's but if ground is maintained while the PC is off and a spike comes through in the middle of the night when the PSU is off and can't deal with it, who knows? Perhaps there's a dying compressor motor in the house causing a spike...

A friend of mine used to live next door to a ham radio operator with a huge antenna on his roof. When my friend's stereo was off the transmissions would energize it as if it was on. Anything like that in the area?

The problem with MSI finding out if it's a board problem is that so few people turn off their power supplies. I'm also leery of the fact that you have to clear the BIOS before it will boot - it should come in a bootable (BIOS comes up) state.

I would also meter the battery just to make sure it's correct (3V?).

 

[Alpha0mega]

That RF radio incident was interesting, but nothing to cause any inductive effects anywhere near where I am. Not to mention that the problem is only with that one mobo, not any others.

I am leaning towards it either being a temperature variation caused problem, or maybe some thing the boards do when powered on after a significant period, like checking/retaining RAM or something. I don't have the requisite technical knowledge to do more than speculate, which I why I was hoping someone here might shed light on such instances experienced by myself and others.

 

[PhIlLy ChEeSe]

Temps wouldn't cause this, unless your not cooling it well or reasonable temps.
Not sure but your shutting off power, but that is the whole issue. You wouldn't have an issue if you did not. I know you have your reasons but you have to leave back up power on bottom line.............
Or else keep sending back new stuff from different companies.

 

[Alpha0mega]

Wait, you are saying that a board that becomes non-functional when left without power for a few hours is fine? That I should just work around it? What if I were to go on vacation and turn off utilities? What if there is a thunderstorm and power is out for a long time? What if I move? If the board dies the first time it loses power for a few hours, then it's defective, and I will not accept a defective product, and live in trepidation. I doubt most others will either.

I have been turning off power, ever day, on every system I have had, with not problems whatsoever, except with this one board, so it's not normal behavior.

 

[LTC8K6]

Instead of turning off the switch on the power supply, try unplugging the plug instead. See if that makes a difference.

I'm going for noise from the PS switch, and the board can't handle it.

Leave the PS switch on, shut the computer down, then pull out the plug.

Wait overnight. Plug it back in, and try to boot it.

Change some other variables and see what happens. Might try turning off power via a power strip, too.

 

[Alpha0mega]

I don't have the board now, since it's gone back to MSI (again), so I can't try your suggestion. Even if I did, it would be currently dead.

I have tried my normal nightly procedure with two different power supplies, the first one being S12-D 850, and the new one being SS-850KM3. Also, prior to the longer shutting off for the night, I have turned off the PSU switch multiple times, for a period of up to 30 mins or so, while checking all the different expansion ports/slots of the board and changing around the hardware. You know, boot with 1 RAM stick and GPU in 1st PCI-E, then turn off, do it with two RAM sticks and/or second PCI-E slot etc. This is both times on that board. It worked until it was left off overnight.

I still think it's possible that the problem is temperature change related, where heating while testing (or even normal usage) followed by a long period of cooling to ambient could be disrupting some connection point that's loose. The power off might just be something I focused on, but wasn't really related i.e. the board might have had the same problem if it was left with mains power, but left to cool overnight after usage.

 

[eton975]

-snip-

Is it possible to get a refund from MSI and get some AsRock board instead if it still doesn't work?

 

[vailr]

OP: what UPS are you using?
May be unrelated, but: in case you maybe weren't aware: many older UPSes don't provide proper sine wave AC power for modern computers. Using an older square wave UPS with an modern "Active PFC" computer power supply can cause early battery failure in the UPS, as well as other problems.

 

[Alpha0mega]

I took possible problems with UPS into account, even though it has worked with all other systems before and since that single board. Second time I had removed it completely and connected directly to mains.

 

[bradly1101]

I did some searching around (as I'm sure you did) and I couldn't find anyone with a similar problem, and it's happened to you with multiple boards, which leads me to think that there's something going on with the power. Does anyone know if ground is maintained when PSU's and UPS's are switched off?

Yeah, when you get the new board see if unplugging it causes the same problem. If not I think it's your power.

 

[Alpha0mega]

No, no. I didn't happen with multiple boards. It's happened with one board, the X99A, twice and in the exact same way. The same board was sent to me twice.

The first replacement I got, a X99S, had two bad RAM slots, out of the box, not this problem. I had turned off PSU and left it over night multiple times (for around 5 days, during which time I was using it), before it was picked up by courier. It never had this problem.

I have also used other CPU/MoBo combos, with everything else being the exact same (including the case), while the MSI boards were in RMA. An i7 950+X58 Sabertooth, i5 3960k+ASUS Z97-A. Currently I am running the i7 5820k with an ASUS X99-A, for the last 10 days. All were/are fine. I bought the ASUS since it has been over 2 months now, and MSI RMA center was/is being quite unhelpful, with no end in sight.

 

[Ketchup]

I think the latest MSI board has a bad rom. Just not holding the information. And it's going to keep happening if they just keep reflashing it.

 

[bradly1101]

No, no. I didn't happen with multiple boards. It's happened with one board, the X99A, twice and in the exact same way. The same board was sent to me twice.

The first replacement I got, a X99S, had two bad RAM slots, out of the box, not this problem. I had turned off PSU and left it over night multiple times (for around 5 days, during which time I was using it), before it was picked up by courier. It never had this problem.

I have also used other CPU/MoBo combos, with everything else being the exact same (including the case), while the MSI boards were in RMA. An i7 950+X58 Sabertooth, i5 3960k+ASUS Z97-A. Currently I am running the i7 5820k with an ASUS X99-A, for the last 10 days. All were/are fine. I bought the ASUS since it has been over 2 months now, and MSI RMA center was/is being quite unhelpful, with no end in sight.

Sorry, my bad. Until the board is eliminated as a problem... They should have sent a new board right away.

Although I haven't had any problems with MSI, it seems more and more that hardware vendors are using their customers as QC. I had to send two Dell products back immediately. The replacements were fine thank goodness.

 

[Alpha0mega]

RMA has become a case of hardware musical chairs, where the same items seem to be circulated until some poor fool accepts a defective item. Have heard of people being given bad replacement after bad replacement, particularly with graphics cards. Replacements so bad that they often don't boot or graphical corruption is evident instantly. With monitors, Dell, I have had to go through 5-7 monitors until an acceptable replacement arrives. The rejected monitors are really bad too, splotches of colors on screen, bunches of lit pixels etc.

With this motherboard case I pretty much had them admit they did this. I called MSI and complained about the poor quality of the boards, and the tech told me that they never had this problem before. In fact, they had given the first X99S replacement (the one that had 2 bad RAM slots) to another RMA case, and hadn't gotten any complaints about it. I was dumbfounded. I asked him to confirm that he had given a known defective board, for which I had them on record accepting that the slots were bad, to another customer. He was silent for a few seconds, and then said "we... ah... we managed to fix it". Total BS! They had told me that since the slot problem was unfixable at their end, they would send me another, which was the even worse X99A.

Most likely, the poor schmuck that was handed the X99S, was either running two sticks and didn't check all the slots, or figured that two bad out of 8 was acceptable for him (it isn't for me).

 

[PhIlLy ChEeSe]

OK, And sorry if I ask you dumb questions(as I don't know what you know about computer)..............
The X99S Asus board should have a second bios chip, have you tried switching the bios button to boot of the other bios?

also if your removing power,(no matter what board) if your booting from a none default(say raid 0 for example)setting in SATA, then the board is not defaulting back to the way you had it(again stupid question but)prior to shut down?
Do you have none default SATA settings? What are you using for the operating system, and how many? Do you own a 32GB or larger USB you could put an OS on to try it? If so install an OS on the jump drive then put the SATA settings in the bios to ACHI(should be default).
Ignore all that^ if you can't enter the bios, if yer desperate try this. Remove the mother board battery, leave it out over night(clear CMOS) over night or more then 12 hours. See if that works?
I have never seen a bios get corrupted simply by removing power, except it will lose my Raid settings in SATA. In which case I just point the bios to the raid n boom back in business. Then again the UPS I can't give experience with except to remove it till you get this figured out. Most if not all computer problems are very simple, we just make them rocket science. Try booting with the board out of the case? Try different memory, or can you swap out other parts?
Try the video card in the lower PIC-e Slot, try bare minimal one stick of memory the OS on a drive no mouse no keyboard, no CD or DVD drive nothing else except the PSU.

 

[mcmcmcmcmc]

Going to double the count of people experiencing OPs issue here.

I bought a new MSI X99A Sli Plus less than 6 months ago. No issues at first but started to notice that when the missus unplugged the extension lead from the wall, PC failed to POST when next turned on.

Had electrician in the other day so power was off same issue. You can reproduce the issue every time. If you take the plug out but get it back into a socket in less that a second you can get away with it. Otherwise its pull the case out and clear the CMOS via shorting the jumper. Would go with the corrupt bios explanation as the system is unplugged before and after shorting the jumper but seems to recover when next plugged in. Im using default bios settings anyway at the mo so its not like restoring defaults by clearing CMOS should help.

Its a pain in the backside as everythings rigged up and normally impossible to sort until the inquisitive 12 month old has gone to bed!

Will try swapping the CMOS battery next time i get a chance and post more details, but fearing its dodgy ROM on primary bios.

Annoyingly i already shelled £100 on a decent Corsair PSU assuming my old one had packed up, before realizing it was something more complicated. I suppose i have a decent spare now which is always handy.

Will also try keyboard idea, and possibly flash bios to latest. Dont want to RMA as Windows 10 registered to the board and cant be dealing with the hassle. Its powered most of the time so only an issue if someone unplugs or the electrics off. Probably wont buy MSI again though - its not like its a budget board.