Corsair One i164 Bluescreens

[mistacool]

On 8/20/2021 at 8:39 PM, mistacool said:

I'm not fully convinced it's an NVME SSD specific issue. Having experienced the error on both 2.5" HDD as well as a replacement 2.5" SSD, and now getting errors with iCUE as well, I'm starting to think it is some conflict between Windows and iCUE firmware or drivers. Fresh Windows installs seem to work OK for only so long until it decides it wants to update. What's driving me nuts right now is that the latest version of iCUE can only pick up my keyboard, so I'm prevented from adjusting my system's fan speeds.

Just an update on my situation:

System has been pretty stable since my last post, no BSODs (yet), though current latest iCUE (v. 4.14.179) still can't detect the Corsair One. I'm no expert but this is leading me to believe that iCUE's fan control was conflicting with something, and perhaps causing the system to overheat / crash in combination with higher-than-avg ambient temperature. Part of me is tempted to install an older version and see if my system becomes detectable again and whether that triggers a BSOD, but I don't want to risk resetting Windows for the n-th time 😩

@wagagagaggag what version of iCUE are you running and what Corsair model do you have? Maybe try updating iCUE or uninstalling altogether and see if that improves anything? 

Best of luck 🙏

[frode.bjerkholt]

On 9/2/2021 at 4:30 AM, wagagagaggag said:

Do you have any updates as to confirming if replacing the boot SSD drive fix the issue?

It is now about 6 weeks since I replaced the SSD and I have not got any BSODs. I have used the computer approximately 8 hours 5 days a week - But I guess that I have to use the computer for more than 18 months to sure if the replacement has fixed the issue or not. 

[rick.sanchez]

Hey guys,

Having the same issue, after weeks of investigating the issue myself (following a "your machine is no longer under warranty" answer from the rep... thank you very much), here are some finding you may find useful:

Basically C1 has 3 weakness points, all of them may cause random boots/BSODs:
1. M.2 goes extremally hot (specifically it's controller, mine goes above 90c)
2. Incompatibility with some new Windows drivers
3. Corsair Link & iCUE

In my experience, #2+3 seems the most probable cause, for the last several days I reinstalled Windows several times, tried with/without installing new drivers / Corsair Link, and currently running stable (for couple of days only, but that's the most I had in a raw) with a fresh install, ignoring driver updates and NOT installing Corsair Link (Bare in mind turning CL off from startup service is not enough, as there is a service running once you installed it for the first time, you need to uninstall it completely).

Of course your mileage may vary, and burning hot SSD is something that needs to be addressed as well, but on my system the issue seems to be Corsair Link (had 2 fresh installs, 1 with and 1 without it, the one with failed immediately, the 1 without runs well so far).

PS: also turned off "smart fan" on the bios and put the fan on max, not sure if that did anything, just part of my experiments to mitigate the problem.

 

[hancui97]

Hey guys,

I am having the same issue with my i164 system. BSOD and reboots into BIOS frequently (In the worst case, BSOD every 5 minutes, making the system unusable).

Since the BSOD is because the NVMe SSD fails to be recognize, and someone mentioned it is a common problem for the Samsung PM981 SSD, I tried to install the latest (v3.3) Samsung NVMe SSD Driver (https://www.samsung.com/semiconductor/minisite/ssd/download/tools/) last night to see if it can mitigate the issue, and It works! I have left my system on for the past 24 hours, and no BSOD yet. My system has never this stable since the problem occurred earlier this year. I will try to leave my system on for another 48 hours to see if the drive really solved the issue.

 

 

Edited September 6, 2021 by hancui97

[wagagagaggag]

On 9/2/2021 at 5:43 PM, mistacool said:

Just an update on my situation:

System has been pretty stable since my last post, no BSODs (yet), though current latest iCUE (v. 4.14.179) still can't detect the Corsair One. I'm no expert but this is leading me to believe that iCUE's fan control was conflicting with something, and perhaps causing the system to overheat / crash in combination with higher-than-avg ambient temperature. Part of me is tempted to install an older version and see if my system becomes detectable again and whether that triggers a BSOD, but I don't want to risk resetting Windows for the n-th time 😩

@wagagagaggag what version of iCUE are you running and what Corsair model do you have? Maybe try updating iCUE or uninstalling altogether and see if that improves anything? 

Best of luck 🙏

Hi @mistacool I have the i165, i tried uninstalling icue before but it did not help. The same BSOD occured and is just currently waiting for my RMA to proceed. 

[rick.sanchez]

Updating following my prev post, unfortunately random boots have returned (clean win install, no new drivers, no Corsair Link) 😟

[wagagagaggag]

15 hours ago, hancui97 said:

Hey guys,

I am having the same issue with my i164 system. BSOD and reboots into BIOS frequently (In the worst case, BSOD every 5 minutes, making the system unusable).

Since the BSOD is because the NVMe SSD fails to be recognize, and someone mentioned it is a common problem for the Samsung PM981 SSD, I tried to install the latest (v3.3) Samsung NVMe SSD Driver (https://www.samsung.com/semiconductor/minisite/ssd/download/tools/) last night to see if it can mitigate the issue, and It works! I have left my system on for the past 24 hours, and no BSOD yet. My system has never this stable since the problem occurred earlier this year. I will try to leave my system on for another 48 hours to see if the drive really solved the issue.

 

 

I am trying your method now, and will let you know if this fix my BSOD. 

[frode.bjerkholt]

I installed the Samsung driver after hancui97's recommendations. I also installed Samsung Magician to monitor the temperature of the SSD. This is what I found when testing different iCUE (v4.14.179) cooling presets :

iCUE Cooling presets	SSD Temperature (°C)
default	72
Fixed 58%	63
Fixed 75%	55

This is under relatively low disk activity and with very low GPU activity. 

According to this article the Samsung SSD is designed to operate at temperatures ranging from 0°C to 70°C. 

[Corsair Caliente]

For the time being, we suggest you guys request an RMA or repair if you're experiencing BSOD issues related to WHEA_UNCORRECTABLE_ERROR.  Seems like swapping out the SSD has resolved the issue for a lot of people which is something we can do for you.

[rick.sanchez]

On 9/8/2021 at 2:13 AM, Corsair Caliente said:

For the time being, we suggest you guys request an RMA or repair if you're experiencing BSOD issues related to WHEA_UNCORRECTABLE_ERROR.  Seems like swapping out the SSD has resolved the issue for a lot of people which is something we can do for you.

Thanks Caliente,

Any chance to have a tutorial how to replace the M.2 ourselves? for the folks with expired warranty.
My One Pro's M.2 controller sits around 80-90c idle and occasionally tops 100 - then the whole machine freezes / shuts off, which 100% is the cause for my shut-downs.

I really love my C1 and want to keep on using it as it has all the power I need, just need to sort this heat issue out.

 

Thanks!

Edited September 12, 2021 by rick.sanchez

[frode.bjerkholt]

15 hours ago, rick.sanchez said:

Thanks Caliente,

Any chance to have a tutorial how to replace the M.2 ourselves? for the folks with expired warranty.
My One Pro's M.2 controller sits around 80-90c idle and occasionally tops 100 - then the whole machine freezes / shuts off, which 100% is the cause for my shut-downs.

I really love my C1 and want to keep on using it as it has all the power I need, just need to sort this heat issue out.

 

Thanks!

Such high drive temperatures indicates that something is wrong. Have you by any change disabled the iCUE application? As far as I know the iCUE application needs to run to get the main fan running.

[rick.sanchez]

6 hours ago, frode.bjerkholt said:

Such high drive temperatures indicates that something is wrong. Have you by any change disabled the iCUE application? As far as I know the iCUE application needs to run to get the main fan running.

Yes it's on, also forced the fans to run at full speed.

I tried to change the M.2 myself but once opened the machine I couldn't find the right screws / order of operations to dis-attach the graphic card and get to it.

[Corsair Caliente]

On 9/12/2021 at 12:53 PM, rick.sanchez said:

Thanks Caliente,

Any chance to have a tutorial how to replace the M.2 ourselves? for the folks with expired warranty.
My One Pro's M.2 controller sits around 80-90c idle and occasionally tops 100 - then the whole machine freezes / shuts off, which 100% is the cause for my shut-downs.

I really love my C1 and want to keep on using it as it has all the power I need, just need to sort this heat issue out.

 

Thanks!

If you'd like to change out the SSD yourself you can use the following instructions:

1. Open Up CORSAIR ONE by pressing release button on rear of system and lifting the top fan assembly.
2. Disconnect top fan.
3. Remove screws securing side panel to video card side of system. (Side panel has radiator for video card cooling system connected to it, take care to just move this out of the way while you're working on it).
4. Remove video card (this'll involve a number of screws, take note of size and location as you work around the graphics card). Be careful of any cables connected to the GPU cooler.
5. Gently remove graphics card (unplug power/PCIe riser cables).
6. Remove stock M.2 SSD (make sure you backed up your data), keep it safe in case you need to replace the stock SSD for whatever reason.
7. Install new M.2 SSD (note that if it's brand new, you'll need to install Windows 10 and drivers)
8. Reinstall graphics card.
9. Close side panel and secure to chassis.
10. Reconnect and install fan assembly to top of chassis.

[mistacool]

Error has occurred for me again. I should reiterate that I am currently using a windows instance booted from a SATA connected SSD rather than NVMe. Happened while I left my computer alone for a bit to download a game. Logs around the time of the crash don't show any alarmingly high temperatures.

[LeDoyen]

On 9/12/2021 at 9:53 PM, rick.sanchez said:

Thanks Caliente,

Any chance to have a tutorial how to replace the M.2 ourselves? for the folks with expired warranty.
My One Pro's M.2 controller sits around 80-90c idle and occasionally tops 100 - then the whole machine freezes / shuts off, which 100% is the cause for my shut-downs.

I really love my C1 and want to keep on using it as it has all the power I need, just need to sort this heat issue out.

 

Thanks!

 

If you open Windows' resource monitor (from the task manager) and look on the disk tab, do you have some process generating a lot of disk read and/or writes ?

It's very unusual to see the controller going so hot, even without heatsink, without massive amounts of disk access

The CPU may be   somewhat idle, but it's possible something in the background keeps your disk busy

[Corsair Caliente]

19 hours ago, mistacool said:

Error has occurred for me again. I should reiterate that I am currently using a windows instance booted from a SATA connected SSD rather than NVMe. Happened while I left my computer alone for a bit to download a game. Logs around the time of the crash don't show any alarmingly high temperatures.

That's alarming.  Were you still using the NVMe drive in the system?

Did you have the same BSOD?  After booting into BIOs were you able to detect both drives?

[xSh4d0W]

I've been having the same exact issue for the past few weeks. Installed samsung's NVME drivers as a previous user suggested which stopped the bluescreens but replaced them with freezes instead (PC goes into limbo mode, you can move windows around but can't interact with them until force restart, audio/video stops playing etc.).

Symptoms

- BSOD booting into BIOS. Can't boot into windows without force power off first. No error in BSOD and no dump file created "due to error during dump file creation" according to event viewer. 

- Restarting from BIOS will go into BIOS again. Can't exit BIOS without force shutdown by holding power button.

- The issue slowly increased until it reached a point where every day I start the PC i'll keep getting BSOD for 5-10 times, until finally it reboots normally. The last boot after the last BSOD will usually come back to windows normally instead of  BIOS (without force shutdown).

 

- After the episode of BSODs stops, the PC will (majority of the time) remain normal without BSOD until I use it the next day. (stays on for all of the day).

- I noticed that when the bsod happens and it boots into BIOS, the ssd drive is labeled as "UEFI disk" only. After force power off then restart again, BIOS displays the disk as "UEFI disk - windows boot manager - samsung...". This furthers the idea something is up with the ssd.

 

Notes:

- Reinstalled fresh clean windows, didn't help.

- Ran disk checks and sfc, nothing comes up.

- HWInfo/magician show the high ssd temps described by other users, but according to quick online search a lot of other samsung nvme SSD users report similar temps so it might be a red herring.

- lastly installed samsung nvme controller driver. This stopped the BSODs but then replaced them with computer freezes and pc becoming unresponsive. After installing this driver, I'm also noticing abnormalities e.g the mouse cursor will occasionally stutter or stop for a few seconds. Spontaneous short pc freezes, etc.

Since several corsair one users started simultaneously mentioning this problem over the past year, my suspicion is that there is a defect with this specific SSD model. If installing samsungs nvme driver stopped the BSODs and replaced them with freezes, this leads me to conclude that either people in this thread all have faulty ssd drives which are dying early, or this specific ssd drive is not properly supported by the windows/samsung nvme controller drivers. 


Sadly I'm 2 months out of warranty; very frustrating. I don't know if Corsair would be willing to own up to the problem if it turns out to be a common hardware defect in this particular model.

Hope this helps.

[xSh4d0W]

Update: The frequency of freezes (which were BSODs before nvme controller driver) has increased. 

 

The freezes almost always happen while doing regular web browsing / idle. Never happened while gaming so far.

[wagagagaggag]

Hello everyone, I wanted to share how I fixed my WHEA_UNCORRECTABLE_ERROR blue screen:

this issue is most likely caused by the NVME SSD drive, and replacing it does fix the blue screen. 

I tried installing new firmware for the SSD, while that fixed the blue screen, the computer would freeze instead of blue screen, so that was not a fix. The freeze was gotten so bad that even after window reinstallation, the screen would freeze and I wouldn't be able to click on anything to even complete the window setup. 

I contacted corsair support about a month ago, and they were able to send me an replacement SSD in 2 weeks, I have just finished replacing it and reinstalling the window. So far everything has worked perfectly as designed. They sent me an identical drive but is produced this year in April rather than my old drive which was produced in 2019 November. 

The replacement process could be a bit frustration for first timers, but it is overall pretty easy to do. You can follow the instructions stated by the corsair employee above, and let me know if you have questions. 

Here are a couple tips in replacing your NVME:

1. After you open up the panel where your GPU is (the right side when the computer is facing you), the heatsink could be a bit frustration to hold on to while you unscrew the GPU (4 screws for the GPU), so it would not be a bad idea to buy yourself some new thermal paste so you could unscrew the heatsink before taking the GPU out. 

2. watch out for the cables connected to the GPU fan when you are taking your GPU out. also remember to unplug the 3 displayport and 1 HDMI cable, 2 power cables, before taking your GPU out. The power cables were difficult to unplug, so use a flathead screwdriver to pry it loose first. 

3. After taking your GPU out, the NVME is right there, and you just have to unscrew it and replace with a new one. I would recommend getting magnetic screwdrivers because some screws are in places where your fingers won't fit. 

4. Have a 16gb usb stick for windows installation media, after getting installation media on your USB drive, plug that into your computer, and your computer should boot right into bios after you installed your NVME, and you just need to go to "boot option", and select UEFI USB Hard drive as your #1 boot drive and restart your computer.  The rest of windows installation should be self-explanatory, but remember to go back to bios to switch back your samsung drive as your #1 boot drive after windows installation. 

Hopefully you guys were able to fix your computer like I do, and hopefully my computer does not die on me again in the near future. 

 

[xSh4d0W]

9 hours ago, wagagagaggag said:

I tried installing new firmware for the SSD, while that fixed the blue screen, the computer would freeze instead of blue screen, so that was not a fix.

 

That was exactly my experience as well. I think the driver just masks up the BSOD by 'handling' some hardware exception, which hides the exception instead of passing it to windows which prevents the BSOD, but does not actually fix the underlying issue and hence the computer freezes. 

This seems like it could potentially be a driver issue, or hardware conflict, or just a faulty hardware(nvme controller). It's impossible to tell without trying the SSD on a different machine.

 

Quote

this issue is most likely caused by the NVME SSD drive, and replacing it does fix the blue screen. 

I'm glad to hear confirmation that the SSD drive is the faulty hardware. Thanks for following up.

Does anyone have experience with Corsair warranty in common defects? Anybody knows whether they'd be willing to make an exception for faulty hardware shortly after end of warranty if it is a common issue? 

 

Thanks!

 

 

[ElliottBear]

Just chiming in to mention that this is happening with my One (i64) as well…

I’m currently working up the courage to try swapping the SSD (my unit is 3 months out of warranty) - could someone confirm the correct model of SSD? I’ve seen one going for £600, which will sting but if that’s what it is then that’s what it is…

The fact that Corsair can get away with such lacklustre QA is staggering, but the truly damning thing is that this community seems to have single-handedly troubleshooted, patched and then fixed the issue themselves - all CS seem to do is ask you to follow a series of rudimentary fixes before telling if it’s out of warranty you have a 4 figure paper weight 

[zero_scripts]

Not sure if you peeps are still having this issue, but thought I'd give an update on mine.

For the past few months my Corsair One i164 (bought in April 2020) started experiencing exactly the same issues posted here. I notice that the NVMe was not showing in the BIOS, and also had constant BSOD.  After updating the NVMe drivers, I stopped getting BSODs; however, it was replaced with constant freezes - the only way to get back to a working desktop was to do a hard reset, to then be greeted with another desktop freeze.

Reading up a bit further I installed version 3 of the iCUE software and set the fan speed to Extreme; since doing this I have not had one freeze or BSOD. Currently 3 days of constant use (I use it for work and gaming) and counting.  I found that my NVMe was idle and sitting at a temperature of 66c. After setting the fans on the extreme setting, they sit idle at around 44c.

This leaves me to believe that it's not a problem with the NVMe drive, but a heating/air flow issue for the Corsair one desktops.

Edited September 21, 2021 by zero_scripts

[frode.bjerkholt]

I have replaced the SSD in my i164 with this one: 

I have had no problems for 2 months now. However I will recommend setting the iCUE cooling presets to 65% to keep the drive temperature below 60°C for most of the time. I agree with @zero_scriptsthat this probably is a heating/air flow issue.

[xSh4d0W]

I have also set my iCue performance setting to "extreme" two days ago and have not gotten a BSOD/freeze so far.

Before changing the fan settings, the SSD controller temp was reaching 90-100C temps rather frequently. The freezes/BSOD tend to happen when the controller temp is near 100C. 

 

After changing the profile to extreme the controller temp doesn't exceed 60-70. Sadly this also means the PC is no longer as quiet.

 

Note that after I set the fan setting to extreme, I actually exit iCue software because it keeps making "hardware disconnected" sounds extremely frequently while turning on/off the case RGB. Rather annoying.

 

Although the high temps are what's causing the freezes/BSODs, I disagree that this is a cooling/airflow issue since the PC was fine for a couple years for me and several other posters. If it was constricted airflow it would've been there since day 1. Something changed that is causing the SSD drive to reach much higher temps than normal. 

 

[ElliottBear]

On 9/22/2021 at 7:44 AM, frode.bjerkholt said:

I have replaced the SSD in my i164 with this one: 

I have had no problems for 2 months now. However I will recommend setting the iCUE cooling presets to 65% to keep the drive temperature below 60°C for most of the time. I agree with @zero_scriptsthat this probably is a heating/air flow issue.

That's brilliant thank you - I did try using a Custom 65% fan preset, and that solved the issue for 3 days, however no the BSoDs have returned; I'll try 80% moving forward and will likely look to upgrade as soon as I can work out a good way of moving everything from one SSD to the other