Question How to prove a NVMe drive is defective for warranty service?

[zhwu]

I have a ~3 year old Western Digital Black NVMe M.2 disk started showing signs of failure. It comes with 5 year manufacture warranty but I am not sure how to prove the issue to the manufacturer in order to claim the warranty services.

Here is what's happening: this NVME drive is my Windows OS drive: for the last 4 ~ 5 days, after booting up the PC to Windows 10 for the first time every day, within 5 minutes I will get a Blue Screen of Death error in Windows for a split second and then PC will immediately reboot itself (not enough time for me to take a photo of the BSOD error). The PC will boot into BIOS because no OS disk can be found, when I tried to look for NVME device in BIOS, no NVME device can be detected there. The only way to solve the problem is to power off PC and turn it back on after a minute. The drive seems to be crash less after it has warmed up a bit. I use this PC for ~6 hours daily and will get random crash 2~3 times a day. The first crash always happens within 5 mins after initial boot, but the other crash happens more randomly, all due to the same problem: lost the NVME device.

The challenge is when this NVME drive is working, nothing seems to be wrong with it. I have tried CrystalDiskInfo and few other disk diagnose tools: everything appears to be perfectly normal when this drive is working. Then all of the sudden the PC will lose the NVME device. I have configure Windows to write Crash information into Event logs but nothing is being captured. I haven't made any hardware change to this PC for 2 years and all the Windows drivers are up-to-date ( haven't install any new software for a few months)

I believe this NVME SSD is the problem and it's slowly dying ( although my motherboard may develop problem with the NVMe port but that's more unlikely?)
Any suggestions on how I can prove/validate (to Western Digital) this drive is at fault and claim warranty service from them? The drive does not crash immediately and I don't have any evidences from disk diagnose tools or Windows Event logs to show there any error with the drive. (just random NVMe disconnect/crash daily)

 

[UsandThem]

WD would run their diagnostic software on it to see if it was showing any errors.

It really could be your M.2 slot developing errors if other SSD utilities aren't reporting any issues. You could move the SSD to different M.2 slot to likely know for sure which one could be the issue (although you didn't list any hardware specs, so just a guess that your motherboard has more than one M.2 slot).

 

[zhwu]

Thanks for the feedback. If you think it's common for M.2 slot on the motherboard to develop error over time, I will try to move the NVMe drive to another M.2 slot on my motherboard (Gigabyte H370M D3H) to give that a try.

 

[mikeford]

i would replace the M.2, this isn't something to let go for weeks, eventually things will get corrupted you don't want corrupted. If you have another M.2 slot, sure try that first.

Maybe also make a few searches using keywords of the problem with first the M.2 model info then MB to see if issue is common to either.

 

[zhwu]

I just want to provide an update here: I moved the NVMe SSD to another M.2. slot on the motherboard and for the last 3 ~ 4 days I haven't experienced any crash at all. (It's used to crash within 5 minutes from a cold boot every time) So it looks like I have a defective M.2. slot on my Gigabyte motherboard. Since none of the SSD diagnose tools has detected any problems with the SSD drive itself. ( all the SMART checks are good, it still has 96% life left), I will continue use this NVMe SSD and just be extra caution with the data I put on the Drive. (only use it for OS & software I can easily re-install )

 

[solidsnake1298]

Out of curiosity, which slot was the SSD in and which is it in now?

 

[Jimminy]

Just taking it out and reinserting it could have helped. Sometimes contacts get crusty.

 

[SKORPI0]

Is the Gigabyte Mobo still under warranty for an RMA replacement?

 

[aigomorla]

WD would run their diagnostic software on it to see if it was showing any errors.

correct they have there own diagnostic which talks directly to the controller.
They also can view smart logs to see how many TB was written.

OP if you think you can squeeze a RMA though after ripping the drive from CHAI plotting.... no... drives are a lot easier to find out abuse on them vs GPU's because they log how many hours they been on, how many times they been power cycled, and most definitely how many TB's were written.

RMA will cover cover until you exceeded the Warranty Time / MTBF on the drive advertised. Either one will invalidate your RMA.