[SOLVED] PC worked for 4 years - Now in an infinite restart loop

Toggle sidebar Toggle sidebar
S

stjepan.herceg

Dec 9, 2018
32
0
30
My computer worked flawlessly for the last 4 years and suddenly it's in an infinite restart loop. Here are my system specs:

Motherboard: Gigabyte GA-Z87X-UD7-TH

CPU: Intel i7 4790K with Corsair H100i Cooling System

RAM: Corsair Dominator Platinum DDR3 4x8GB (2400 MHz)

GPU: EVGA GeForce GTX 780 Ti

PSU: Corsair AX1200i

The PSU has a constant green light flashing, as well as the GPU. The Fans are working and the startup usually lasts about 4 to 6 seconds. It doesn't matter if I go into the BIOS menu or continue with the regular booting process, it always restarts after about 4 to 6 seconds.

Is the CPU damaged? Is the cooling system damaged (I use two Noctua fans attached to the H100i which do spin in the booting process)?

It's really strange. I used to run this PC weeks on end without turning it off. It was on the last 2 days before yesterday, and now that I wanted to turn it on again, it simply restarts all the time.

The last two Motherboard codes I see before the computer restarts on its own are A2 followed by A0. A2 = Detect and install all currently connected IDE devices and A0 = IDE initialization is started.
 
Solution
justin.m.beauvais
Yes an optical drive can also cause this kind of problem. Usually an extremely long drive detection time points to a drive that is going bad though. So, if it is taking forever to do the drive detection, that can be a sign of a bad drive and any drive can cause that. If it is quick and doesn't boot that could be a bad boot drive or a motherboard controller.

If it reaches the bootloader then odds are much better that your boot drive is fine and something else is causing the issue. So... if I were you I'd disconnect your drives and try to boot to a bootable USB. Your USB drive can be anything really you are just seeing if it boots and ruling out most of the hardware. In fact, you can hook your drives up and boot to the USB again and see...
Sort by date Sort by votes
S

stjepan.herceg

Dec 9, 2018
32
0
30


Thanks for your answer, Justin!

Just to be sure: it's the last two codes i see before the reboot. Is that the correct error code? Or should I look at the code immediately after the reboot?

I have 2 HDDs and 2 SSDs. One of my Samsung 840 500GB SSDs is my primary drive (Mac OS X) and the other SSD contains Windows 10. The two 2TB Western Digital HDDs are only used for extra storage and have no OS data on it. Is the primary drive failing in my case or can any failing drive cause this error? Is the hardware damaged or can this problem persist due to a loose power connection to the drives etc.?

If the hardrive controller is failing, would I have to replace my current motherboard or is there anything i could try to do?
 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840


I clicked the wrong button there and said you solved your own problem. My bad.

The code that is the most likely problem is just before reboot. It could be a bad hard drive, in this case it would be the bad drive with the master boot record on it, most likely. One of the other drives might cause it, just disconnect them to be sure. A bad drive could possibly take the whole thing down, but it is much less likely than the boot drive having a problem.

If it is your motherboard the chances are that you will have to replace it. There is the option of buying a PCI-E drive controller, but depending on support on your motherboard it may not be bootable.
 
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


Apart from SSDs and HDDs, can optical DVD drives also cause this problem?

The boot process is sometimes long enough to reach the clover bootloader (hackintosh necessity), which I need to choose either between mac or windows OS. Does this mean that the primary drive is still intact, as it contains the clover bootloader which was reached in the boot process?
 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
Yes an optical drive can also cause this kind of problem. Usually an extremely long drive detection time points to a drive that is going bad though. So, if it is taking forever to do the drive detection, that can be a sign of a bad drive and any drive can cause that. If it is quick and doesn't boot that could be a bad boot drive or a motherboard controller.

If it reaches the bootloader then odds are much better that your boot drive is fine and something else is causing the issue. So... if I were you I'd disconnect your drives and try to boot to a bootable USB. Your USB drive can be anything really you are just seeing if it boots and ruling out most of the hardware. In fact, you can hook your drives up and boot to the USB again and see if you detect your drives. That will tell you if it is the controller. I'd just put a Ubuntu image on the USB. Or something that can read the Windows and Mac file systems so you can see all the drives and their data.

Odds are that if you can see the drives and the data that the bootloader might have gotten corrupted on your main drive. If one drive doesn't work right then that is your bad one. If none of the drives are working it is likely your motherboard's controller.
 
Upvote 0 Downvote
Solution
S

stjepan.herceg

Dec 9, 2018
32
0
30


Thanks! I'll try to disconnect the drives and use a bootable USB tomorrow. Will report back if there are any other problems or if the drives don't turn out to be the problem after all.

 
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30
I have disconnected all SATA cables, all external cables except for HDMI and I have also tried different RAM variations (1 RAM, 2 RAMs in different slots), and nothing seems to work. It doesn't matter if I go into the BIOS or if I let the PC find out that there's no bootable drive connected, it always reboots after a couple of seconds.

I now get the error code 00. I can't find anything in the manual that describes the 00 error code. From my google research it seems to be hardware failure.

It probably isn't the GPUs fault because I can clearly see the BIOS screen.

Is the CPU maybe overheating during the first few seconds so that the motherboard automatically shuts it down?

The light on my H100i cooler lights up and the fans are also spinning. The CPU power is probably well connected, otherwise I wouldn't be able to reach the BIOS screen. So what else can it be?

It probably isn't the GPUs fault, either, because I can clearly see the BIOS screen.
 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
Even if it is your coolers fault it wouldn't kick off so immediately. If your pump on the H100i had died you'd see it getting hot in the OS, then performance would get really bad as the CPU thermal throttles, then it would shut down, usually. You can check to see if your pump is the problem but my thought is that it is just fine.

So, we've ruled out everything but the CPU, cooler, video card, motherboard, and power supply at this point. I don't think it is the cooler because symptoms aren't tracking with that kind of failure. So, realistically we are looking at CPU, video card, motherboard, or PSU. If you have access to another computer you can test the PSU idea. Anything over a 350W will allow the computer to boot with CPU, mobo, RAM, Video, and a USB boot device. The CPU and motherboard are tougher. Usually people don't have spare CPUs lying around and even less likely for motherboards. The GPU is way easier. Just use the BIOS to enable the on chip video that your 4790K has, shut it down, pull the video card, hook up video to the motherboard, and start the system back up. You'll find out really quick if you have a problem there.

At this point I'm disappointed that it wasn't just a drive. Good luck, and let me know what you find.
 
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


I've seen a youtube video which shows that a broken cooler might also cause the pc to shut down after 10-20 seconds. It's way faster in my case. I can feel the water flow through the pipes when the computer is running, so that would be strong evidence that the pump is indeed still functioning.

The GPU's light is also constantly green and its fans are spinning, so I don't think that it's due to the GPU. The PSU also shows a constant green light, which would imply that the AX1200i functions as it should.

I've also disconnected the CPU power cable and then it immediately shuts down the pc.

I'd like to change some settings in the BIOS, only problem is that I can't view the BIOS screen for too long (less than 2 seconds) before it reboots again. I've also tried the backup BIOS settings with the hardware BIOS switch, but this hasn't helped, either.

Is dust maybe the problem? It's not too dusty, but I do see it on some components...
 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
If your pump is running, water is moving, and your fans spinning then you have a working cooling loop. That is all there is to a liquid cooling system. The only other thing that can degrade performance is dirt or buildup inside the loop, but the odds of that happening in an AIO are really low.

Dust isn't going to be the issue unless it is absolutely caked on there... possibly with metal shavings in it.

If you do a BIOS reset the on chip video SHOULD default to on (sometimes you just can't be sure), then you can maybe check that...

As for the status lights being on and all, those are usually just checking one specific thing about the hardware. Do I have voltage? Am I overheating? Those kinds of things. They are quick and dirty idiot lights most of the time. They don't tell you if you're suffering current drop offs, dirty power, low voltage, circuit failures, or any number of other things. However, if they are to be trusted then your problem is CPU and/or motherboard... which gets expensive. If you are planning an upgrade, now would be a good time... but only if that PSU is working right, otherwise you risk killing new hardware. So, that is why I'm trying to rule that out.
 
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


I removed the GPU and connected the HDMI cable directly to the motherboard, so that it uses the intel onboard gpu. It still didn't help. I did reach the BIOS screen once again, but soon afterwards it fell into the same reboot pattern. It seems that the GPU can definitely be ruled out. I took the CMOS battery out and pressed the power button for about 60 seconds to drain all the current and then I restarted the PC once again, and the same problem still persists.

As far as the CPU is concerned: If I can reach the BIOS screen with the GPU removed, does that mean that the CPU is not damaged? If it is indeed so, only the motherboard and the PSU will be left.

Which pins do I need to check with a multimeter to rule out a PSU failure? Since the AX1200i is a modular PSU, can I check the pins directly on the PSU without using the power cables or would you advise against it?

Thank you once again, Justin! Your help is much appreciated.

 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
The easiest way to check a PSU is with a PSU tester. Using a multimeter you are going to be checking voltages primarily because without a good PSU tester you aren't going to be able to put anything under load. You want to make sure that you are getting your 3.3V, 5V, and your 12V. Everything in the PC runs on these voltages. Any other voltages in the computer are derived from those, so if they aren't spot on things like the 1.3V the CPU uses (converted by the motherboard) won't be the right level.

So, I'm going to give you a link with instructions on how to check the voltages.
Here: https://www.lifewire.com/how-to-manually-test-a-power-supply-with-a-multimeter-2626158

If your voltages look good it could still be a load issue... but the odds of that are pretty low. Usually a PSU that has issues under load causes crashes, reboots, and blue screens after launching a taxing program.

Unfortunately we can't rule out the CPU being the problem without another system to test in, but the likelihood of it being the problem is much lower than the likelihood of the motherboard being the issue.
 
  • Like
Reactions: chronometric
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


Ok, I will buy the PSU tester and see what happens.

What I still don't understand is how the CPU can't be ruled out? I mean, the BIOS screen can be reached, even when using the onboard GPU HDMI output, so it must be working correctly? Or is the CPU damaged in a way that it still functions but handles voltages etc. wrongly?

I sure hope that it doesn't turn out to be the motherboard... it's no longer under warranty and I can't find a replacement.

I've read some blog posts that claim to fix motherboards by swapping the bios chip. I'm not sure if my motherboard only uses one bios chip. There is the option of switching between 2 bios settings directly on the motherboard, so maybe the bios chip swap wouldn't even be an option for me...

 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840


A PSU tester is a good thing to have on hand. I used one when I was working computer repair but just haven't gotten around to getting my own yet.

Ruling out the CPU or even seeing if it is the motherboard or CPU is pretty tricky. A while back they began integrating motherboard functions into the CPU. So not only is there just the processing cores, but you have video, RAM controler, basic IO functions, all on top of the cache and everything that interconnects all that. They are pretty darn complicated and if you lose a little function in one part it can throw the whole system off. So without testing it in a known good motherboard there really isn't a definitive way of ruling it out.

As for swapping the BIOS chips, sometimes that does work. The BIOS has a lot more function these days than just being the Basic Input Output System. It controls how the CPU interacts with the motherboard. From what you are saying you have the dual bios setup, which has pretty much ruled out the BIOS as the problem for you. It keeps one chip offline and runs the PC on the other. So you've had one chip that shouldn't have been affected by anything else, and you tried it already, so that pretty much ruled out the BIOS.
 
  • Like
Reactions: chronometric
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


I have finally received the PSU tester, and here are the results:

6XAkkdm.jpg


This picture was taken with solely the Motherboard 24pin power connector plugged into the tester. As you can see, the +12V2 is showing LL (Low Level) insufficiency. Then I added the CPU 8 pin power connector to the tester, and it showed me this on the display:

WNDaHoP.jpg


With the Motherboard and the CPU power connected, everything seems fine. I do wonder: if there is an LL signal with just the Motherboard power connector, does it mean that it is definitely the power supply that's causing these constant reboots? The +12V2 seems to be responsible for CPU power, so this might very well be the case...

Or am I missing something?
 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
Everything there looks great. The +12V2 doesn't come from the 24 pin connection, it comes from the CPU header. So, the power supply has good power to the motherboard. It actually looks great. That expensive power supply really puts off some good voltages.

So here we are, back at CPU or Motherboard. My suspicion is motherboard, but without at least a known good CPU or a known good motherboard I can't be sure. I just had the idea of getting a super cheap Celeron to test the motherboard, but they still aren't cheap enough to justify picking one up as a testing chip.

At this point you have a decision to make. The decision is to buy replacement parts or new parts. A working motherboard can probably be had at a reasonable price (not your motherboard, that thing is crazy expensive), but if you have to replace that 4790K it is going to be like buying a new CPU in terms of price. I was looking at upgrading my CPU from my 4590 to a 4770K or 4790K and those CPUs are still selling for like $230 or more. IF you guess the broken part correctly and replace just that it will save you money over the new PC approach, but if you have to buy the other part as well you're spending so much that an upgrade would make more sense. For the same price you are looking at paying for a replacement 4790K and a replacement Z87 motherboard you're in the ballpark of an i5 8600K and a motherboard (Ryzen 2600X if you don't mind AMD and want to pay a little less). The only thing about a new CPU is new RAM as well. You need to upgrade to DDR4.

Sorry we couldn't get your system all back together. The rest of your hardware should be fine, so you don't have to buy a full new system. I'm afraid that replacing your motherboard might make the Hackintosh harder to make work, but at this point you can take the Hackintosh requirements into consideration and select parts that will work for it.
 
  • Like
Reactions: chronometric
Upvote 0 Downvote
S

stjepan.herceg

Dec 9, 2018
32
0
30


I've tested the 4790K CPU on a hackintosh that was using a 4770K CPU. The test computer was not stuck in a reboot loop. The BIOS has recognized the 4790K and has shown the proper performance metrics. It did, however, fail to boot completely. Maybe because the hackintosh was installed with the 4770K in mind, so it might not work with a 4790K.

Anyway, now that there is no reboot loop with the 4790K in the test computer, is it safe to say that it has to be the motherboard's fault?

I could use a paid service by Gigabyte, but I'm not sure how much this will cost or if it is even worth it. A motherboard that is failing might have accumulated a heap of problems by now, and fixing one thing just so that it boots up, might leave other problems unsolved and looming.

Is it better to switch to a completely new system?

 
Upvote 0 Downvote
justin.m.beauvais

justin.m.beauvais

Estimable
Dec 15, 2017
1,527
199
3,840
If you just want it working it is fine to replace the motherboard. The question comes down to price. If the motherboard is expensive, then you are better off getting brand new hardware. If it is cheap, or at least a price you consider reasonable, then just go for the motherboard. I mean, you've got a 4790K, there isn't anything under $200, as far as CPUs go, that can be considered a substantial upgrade from it. If you are looking at a Ryzen (which is the only upgrade I'd take in that price bracket) the best you can say is that it gets you on a new platform so that upgrading later costs less. With something like the 2600X, you are gaining 10% or so on performance. You also open yourself up for drop in upgrades to the new Ryzen 3000 series when they launch. If you go Intel at that price you are pretty much just looking at the i5 8400, which is a 6% performance increase. However, it offers an upgrade path to the mighty 9900K, which as the fastest consumer CPU right now is only 44% faster than what you have, and costs over $500.

The problem here is that I have no idea how any of the new hardware will play with a hackintosh setup. Apple doesn't use AMD CPUs, and they JUST started using more modern Intel ones.

So, if it were me, the new system depends solely on price. "Can I replace the motherboard for a price that feels reasonable?" is the question you should ask yourself. If the price doesn't seem reasonable, upgrade.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom