Random Shutdowns possibly dry solder joint

[hbake17]

So I brought a new CPU a Xeon E3-1231-V3 and borrowed a motherboard from a mate, an asrock z87 extreme9/ac, it worked fine for a few days then began to crash randomly, as of now I have tested the ram, GPU, HDDS, PSU, I have flashed the bios, made sure the bios's were sitting properly. and just about everything else, I am unable to test the CPU in another board or put another CPU in this board at the moment as I don't have an 1150 CPU or board, I might be able to test one when another mate gets back to me. However I think it might be a dry solder joint as the system runs fine sometimes and will crash on the bios others or a moment after powering on or on one occasion 12 hours after it turned on. So I am wondering if anyone knows an easy way to identify where to look for the dry joint, but any advice would be great thanks.

 

[LinuxDevice]

Although you tested the PSU it doesn't mean the PSU has enough power to do the job. You should list the PSU details.

Also, you could be running into software issues, it doesn't have to be hardware at fault. Cold joints often fail right away and not after heating from being used for 12 hours.

Crashing in the BIOS can simply occur if the button battery is low and CMOS becomes unreliable/corrupt. A new battery and a reset of BIOS would probably answer that. How old is the CMOS battery?

 

[hbake17]

Although you tested the PSU it doesn't mean the PSU has enough power to do the job. You should list the PSU details.

Also, you could be running into software issues, it doesn't have to be hardware at fault. Cold joints often fail right away and not after heating from being used for 12 hours.

Crashing in the BIOS can simply occur if the button battery is low and CMOS becomes unreliable/corrupt. A new battery and a reset of BIOS would probably answer that. How old is the CMOS battery?

The PSU is a corsair HX850w I have tested each line with a multi-meter, but when I said I had tested it I mean I put another PSU in there, a 700w so power supply is not the issue, especially given the 700w runs all day every day with no issues.

I have reinstalled the OS a couple of time, made sure drivers were up to date, disabled win 10s shitty auto drivers installation and installed all the drivers my self all from the manufactures website all up to date for the board, I even tested slightly out of date drivers, and drivers slight more out of date that that.

The bios battery is fine I have tested it as well with a multi meter I have updated bios B to P2.50 and re-flashed bios A with P2.30 so the bios it self is not the issues.

I know that it is unlikely that heat is causing expansion in the solder causing a cavity to grow but I have gone though all the basics and a dry joint fits the intermediate nature of the fault hence why I am here asking. Oh quick edit, the board sat in my mates cupboard for about a year before I began using it.

 

[LinuxDevice]

A multimeter is not capable of measuring stability. All it can measure is if it was outright failed. Switching supplies are complicated.

If you want to rule out software, then run memtest86+ overnight. It is essentially its own mini operating system, and it doesn't need to actually install (it can run from a bootable thumb drive or DVD). Reinstalling Windows in no way adds validity to whether the software is good...you're still running Windows.

Once again, you need to measure the output of a CMOS battery under load. A multimeter does not measure that. If the battery is more than about 4 years old, then it would probably need replacement anyway. Admittedly a CMOS battery is not normally under much of a load...but it is under that load 24/7 for years.

Motherboards are typically 8-layer. There is very little chance you could spot a cold joint. If it does go bad, then you probably need to replace it anyway. There is so much which could go wrong though it may end up that all you can do is replace parts until you find out what is wrong. A button battery and memtest86+ are about the best bang for the buck you are going to get.

 

[hbake17]

A multimeter is not capable of measuring stability. All it can measure is if it was outright failed. Switching supplies are complicated.

If you want to rule out software, then run memtest86+ overnight. It is essentially its own mini operating system, and it doesn't need to actually install (it can run from a bootable thumb drive or DVD). Reinstalling Windows in no way adds validity to whether the software is good...you're still running Windows.

Once again, you need to measure the output of a CMOS battery under load. A multimeter does not measure that. If the battery is more than about 4 years old, then it would probably need replacement anyway. Admittedly a CMOS battery is not normally under much of a load...but it is under that load 24/7 for years.

Motherboards are typically 8-layer. There is very little chance you could spot a cold joint. If it does go bad, then you probably need to replace it anyway. There is so much which could go wrong though it may end up that all you can do is replace parts until you find out what is wrong. A button battery and memtest86+ are about the best bang for the buck you are going to get.

I hate to be rude, but I have been working on computers now for 14 years could we avoid advice given to kids at school? I have run memtest, twice. To prove a point I even took the bios battery out and ran the system, the same problem occurred and yes I re-flashed the bios before running that test.

I know a multi-meter won't test a PSU or a battery under load but replacing the PSU proved that the problem was not related to the load of the device as both PSUs had the same problem and as I said the 700w runs all day every day in another computer that acts as a server to house my data.

As I said I know that it is unlikely but on the off chance that I could fix it the board is still worth $500 so I would like to continue using it, given this is the first time I had this problem I thought I would ask a community of people who might have had it before, maybe even fixed it, or given I did ask for other advice you know like perhaps someone had this problem before and it turned out to be a problem in the southbridge or that the VRM's were dead or dying, not run a test that I have already run because I have been working on this for two weeks already.

Thanks for your attempt but clearly you are not the one to help me fix this problem.