Corsair One i164 Bluescreens

[eric418]

On 6/18/2022 at 3:46 AM, Z1970 said:

Sorry don’t have before data.

Currently running bios default fan control, and temp is 53….. which is slightly hotter than i would like.

What ought to be usable airflow over the SSD is I think compromised by the GPU flat cables, which a) largely occupy the airspace over the SSD and b) as you can see in the photo serve to stop heat coming off the SSD; note the fabric and foil.

I suspect the nature and position of the GPU cables are material to the overheating issue.

Yes, this is what I was thinking too.

I didn't add any heatsink or cooling solution to the M.2, as I believe there is virtually no airflow at all there between the M.2 and GPU, so the temp improvement of adding an M.2 heatsink should be minimal, unless using a M.2 heatsink with mini fan which is 7mm thick. 

Do you think a 7mm thick heatsink will fit?

[Z1970]

The MP600 pro lpx I have is 11mm including the heatsink, so 7mm should be good.

Braided cables sitting over the fan will be the issue I guess.

[eric418]

I suddenly think that actually we should use an M.2 NVMe extension riser cable, and this should solve all the problem.

1. By routing the M.2 from the back of the motherboard to the CPU side, there will be plenty of space and airflow, which will allow us to run the fan at a much lower speed while maintaining the M.2 in good temp. Probably it will be below 50c even with top fan at zero RPM.

2. All future maintenance will be much easier, just open the CPU side to replace or upgrade the M.2 (just like the 2.5" SSD)

3. With the plenty of space at the CPU side, we can use any type of M.2 without the space constraint between the MB and GPU.

4. We can use any M.2 cooling solutions, like heatsink, or heatsink with fan.

So my question now is if we will be able to route the M.2 riser cable from the GPU side to the CPU side? Is there a channel or hole somewhere we can pass through this cable?

https://www.moddiy.com/products/NGFF-M.2-Key-M-Extender-Cable-Adapter-Support-NVMe-Ultra-SSD-R44SF.html

 

[flexflex]

Yepp, had the exact same issues. And no warrenty left. Bought a new nvme Kingston 500gb chip and replaced the current one. A bit tough to get access to it under the GPU... but kept track of all the screws and the order to reassemble everything. So far so good and no crashes.

[lucarav8]

I give my two cents for an alternative trick.

I had exactly the same issue everyone is describing here (bluescreens, hwinfo showing high temperatures of the NVME...).

I got to the point in which I was getting a crash every few minutes, and the PC had basically become unusable. I followed this discussion for long time and I really appreciate the effort everyone has put in trying to solve the issue, but I am not fully convinced it is an issue with the M.2, since my crashes were happening not only when the temperatures were at their max (about 85-90 C), but also much lower (55-60 C).

With a bit of luck, I found this YouTube video in which a computer guy was trying to understand and fix the issue:

Despite the title, he thinks the problem might be actually caused by the RAM (he guesses about some oxidation in the pins), since for him it disappeared once he removed and reattached the RAM blocks.

Well, I decided to give it a try and just removed and reinserted the RAM blocks a couple of times (without reinstalling windows or anything else). After that, the issue disappeared (I left the PC on for 5 days 24/7 with external temperatures reaching 35 C and not a single crash). To date, it did not crash a single time anymore.

I am not a computer expert, so I cannot say if this solution would work for everyone, or if at a certain point the crashes will restart, but I thought it was a good idea to share it in case it can help others.

 

Edited July 30, 2022 by lucarav8

[PappaDragons]

16 hours ago, lucarav8 said:

I give my two cents for an alternative trick.

I had exactly the same issue everyone is describing here (bluescreens, hwinfo showing high temperatures of the NVME...).

I got to the point in which I was getting a crash every few minutes, and the PC had basically become unusable. I followed this discussion for long time and I really appreciate the effort everyone has put in trying to solve the issue, but I am not fully convinced it is an issue with the M.2, since my crashes were happening not only when the temperatures were at their max (about 85-90 C), but also much lower (55-60 C).

With a bit of luck, I found this YouTube video in which a computer guy was trying to understand and fix the issue:

Despite the title, he thinks the problem might be actually caused by the RAM (he guesses about some oxidation in the pins), since for him it disappeared once he removed and reattached the RAM blocks.

Well, I decided to give it a try and just removed and reinserted the RAM blocks a couple of times (without reinstalling windows or anything else). After that, the issue disappeared (I left the PC on for 5 days 24/7 with external temperatures reaching 35 C and not a single crash). To date, it did not crash a single time anymore.

I am not a computer expert, so I cannot say if this solution would work for everyone, or if at a certain point the crashes will restart, but I thought it was a good idea to share it in case it can help others.

 

That’s a great find! Having the same issues just after my warranty expired. Just refitted the ram modules and will try it out the next couple of days. 

[LeDoyen]

On 6/21/2022 at 4:42 PM, eric418 said:

I suddenly think that actually we should use an M.2 NVMe extension riser cable, and this should solve all the problem.

1. By routing the M.2 from the back of the motherboard to the CPU side, there will be plenty of space and airflow, which will allow us to run the fan at a much lower speed while maintaining the M.2 in good temp. Probably it will be below 50c even with top fan at zero RPM.

2. All future maintenance will be much easier, just open the CPU side to replace or upgrade the M.2 (just like the 2.5" SSD)

3. With the plenty of space at the CPU side, we can use any type of M.2 without the space constraint between the MB and GPU.

4. We can use any M.2 cooling solutions, like heatsink, or heatsink with fan.

So my question now is if we will be able to route the M.2 riser cable from the GPU side to the CPU side? Is there a channel or hole somewhere we can pass through this cable?

https://www.moddiy.com/products/NGFF-M.2-Key-M-Extender-Cable-Adapter-Support-NVMe-Ultra-SSD-R44SF.html

 

That looks like a better solution and well worth trying.

My fear when attaching heatsinks in place is that, being efficient thermal transfer devices, they may end up transferring even more heat from the GPU to the SSD 😛

 

[FrustratedCSOne]

Getting the same BSOD on a Corsair One I purchased on January 2021. BSOD is occurring more and more frequent at various times. 

- CMO reset didn't work

- Reinstalled Windows didn't work. Made it worse actually.

- Submitted support.info zip file and fail log screen to Corsair, waiting on response. 

- $3k+ USD computer is UNUSABLE.

[Queeg]

4 hours ago, FrustratedCSOne said:

Getting the same BSOD on a Corsair One I purchased on January 2021. BSOD is occurring more and more frequent at various times. 

- CMO reset didn't work

- Reinstalled Windows didn't work. Made it worse actually.

- Submitted support.info zip file and fail log screen to Corsair, waiting on response. 

- $3k+ USD computer is UNUSABLE.

You might try removing and reinserting the memory sticks.  There was a post here recently about that fixing BS problems.  Worth a try. 

[FrustratedCSOne]

On 4/24/2022 at 5:57 AM, AlanLiang1999 said:

after I uninstalled icue. Every problem solved. NO longer exists BSOD. I have no idea why. just accidently watched on the youtube which said that icue would crash the computer. This BSOD problem has been struggled me for 1 year and now I completely solve it.  

I uninstalled iCue and the BSOD continued multiple times after. Computer is unuseable now. I'm waiting on Corsair response for this. 

[FrustratedCSOne]

22 hours ago, Queeg said:

You might try removing and reinserting the memory sticks.  There was a post here recently about that fixing BS problems.  Worth a try. 

Tried that and it didn't work either. 

[lucarav8]

On 7/30/2022 at 6:45 PM, lucarav8 said:

I give my two cents for an alternative trick.

I had exactly the same issue everyone is describing here (bluescreens, hwinfo showing high temperatures of the NVME...).

I got to the point in which I was getting a crash every few minutes, and the PC had basically become unusable. I followed this discussion for long time and I really appreciate the effort everyone has put in trying to solve the issue, but I am not fully convinced it is an issue with the M.2, since my crashes were happening not only when the temperatures were at their max (about 85-90 C), but also much lower (55-60 C).

With a bit of luck, I found this YouTube video in which a computer guy was trying to understand and fix the issue:

Despite the title, he thinks the problem might be actually caused by the RAM (he guesses about some oxidation in the pins), since for him it disappeared once he removed and reattached the RAM blocks.

Well, I decided to give it a try and just removed and reinserted the RAM blocks a couple of times (without reinstalling windows or anything else). After that, the issue disappeared (I left the PC on for 5 days 24/7 with external temperatures reaching 35 C and not a single crash). To date, it did not crash a single time anymore.

I am not a computer expert, so I cannot say if this solution would work for everyone, or if at a certain point the crashes will restart, but I thought it was a good idea to share it in case it can help others.

 

Unfortunately, also for me the solution does not seem definitive.

After about 2 weeks, I started having blue screens again, although much less frequently than before (like it can happen or not during an entire day, while before I got to a point in which it was happening every 10 minutes).

So, maybe the RAM trick can alleviate the issue, but it is not the final fix.

I will probably try to repeat the procedure every now and then, and hope it keeps the PC more usable. I guess at this point we are into voodoo territory 🙂

 

[Mutts2]

I165 from 2019. Was solid for like over a year at first. Then bouts of blue screens which really shocked me after it being so rock solid. Ram trick helped but never truly solved it permanent. However I have had good luck with reseating that really long connector just below the ram running horizontal front to back.   It has no locking mechanism, just press fit in. Reseating it, not just once but do it a few times to really scrape the contacts, has left me with long periods without BSOD, like I think a year,  then it started again a month ago, reseated that long connector again and so for no problems.  I think reseating this, keeping things cool has helped me greatly.  Oxidation sounds like as good a explanation as any, need the scraping of the connections back and forth a few times to restore good contact again. Probably temperature variations on these machines with going way up and way down all the time contributes too, expansion and contraction of the contacts?   I did upgrade the top fan to a newer one that goes 2000 RPM so when it's hot weather I'll bump it up to max.

[frode.bjerkholt]

I changed the SSD to a Samsung 970 EVO a year ago and my i164 has worked without any blue screens for a long time. However late in July this year, the BIOS suddenly stopped recognizing the SSD from time to time after reboots. I have now installed a new Samsung 980 Pro with a Heatsink and I see that the temperature reported by Samsung Magician is much lower than before. For the 970 EVO the temperature almost always was above 60°C. Now for the 980 Pro with a heatsink it is mostly below 50°C.  Hopefully this will make the SSD last longer. 

[MJ Alexander]

For what it's worth, I've owned my ONE for two years... has been a quiet, reliable beast, no issues whatsoever. About two weeks ago, I started seeing the BSOD... often multiple times per day. This year, we've experienced heavier than average rainfall here (Central Arizona mountains) during our monsoon season, with very high relative humidity.  At 6200' elevation, we've never needed A/C, so the resultant ambient conditions in my workspace (home studio) have become less than optimal for a PC, even a liquid-cooled ONE. After reading this entire thread (and others) searching for a fix, I read a BSOD comment somewhere (possibly here) that suggested SDRAM contact oxidation as a suspect. Applying Occam's Razor, I pulled the sticks, blew out the ports, swabbed the contacts with isopropyl alcohol, re-inserted the sticks into their opposite ports, and haven't have a BSOD since... knocking on wood here. Four and one-half days non-stop up-and-running so far...   

[Ryan LeBlanc]

Yet another problem to add to the growing pile of unhappy Corsair One i164 customers.  I purchased mine in June 2020 and 1-year to the date I started having random BSOD -> BIOS boot loops.  I was right at the end of my warranty and Corsair accepted it as an RMA return in early July 2021 - unfortunately I had already paid several hundred in software fees trying to get it repaired locally.

My new RMA device worked great for about 9-10 months and then had the exact same problem with increasing BSOD -> BIOS boot loops.  Resetting Windows worked for a short period of time however the problem has returned.  Essentially the computer is unusable now.

Like many here I've had almost identical problems - when booting to BIOS the SSD is not recognized (only the back-up non-SSD is found).  I've run a RAM diagnostic tool (Memtest86) at Corsair's recommendation which fully passed with no errors over 4 passes.

As RMA returns only have a 3-month warranty I am doubtful they will be able to help me much here.  I'm debating replacing the SSD like others have tried (which will cost several hundred) or just giving up altogether and having a new machine built.  I spent $4,200 CAD on this device new.

Given the number of people just in this forum alone all with IDENTICAL problems including different RMA machines this appears to be a manufacturing error/problem and I would surmise that Corsair is liable from either a Class Action or Consumer Proposal standpoint.  If the issue can't be resolved satisfactorily I'll be pursuing this option.

[MostlyICareaLot]

Hi all,

I can confirm many of the problems forum users in this thread have experienced. My rig is just over 2 years old and has had an exponentially increasing amount of BSOD. The BIOS failed to show the SSD  (also the boot drive in my setup) in the end rendering the unit useless.

After following frode.bjerkholt comment plus a £100 poorer I received the replacemtn EVO 980 with heatsink. I used to know my way around the i286 generation pcs which in comparision the ultra tight setup was a doddle. However, after 15 minutes of careful removing and refitting the procedure of replacement was complete. Better still it runs fine (for now) with an SSD temp of 55degC at modest use and no chassis fan running.

Below I posted some photos of the journey (apologies for the dust, I did a complete clean after!). Make sure your screwdriver tip is properly magnetised ;D

[Superval]

Same issue, a $6k system now an expensive paper weight and unusable. M2 drive seems to overheat and shuts down just from booting. According to diagnostics it hit 82 degrees c. This has to be a design flaw and Corsair should be looking to fix

[MostlyICareaLot]

On 9/19/2022 at 4:17 AM, Superval said:

Same issue, a $6k system now an expensive paper weight and unusable. M2 drive seems to overheat and shuts down just from booting. According to diagnostics it hit 82 degrees c. This has to be a design flaw and Corsair should be looking to fix

Quite right. Being wedged between CPU/MCB and GPU inevitably leads to heat problems. Ironically enough, the cooling parts take up most of the width of the unit and disallow the option for direct cooling of the SSD slot. Some may say form over function

[eric418]

I have done it finally and it works perfectly! Hope this info can help other Corsair One users here.

In my previous post in this thread, I was thinking of adding a NVMe extension cable to route the NVMe to the CPU side for better cooling and ease of management.

Now, I finally did it with https://www.moddiy.com/products/NGFF-M.2-Key-M-Extender-Cable-Adapter-Support-NVMe-Ultra-SSD-R44SF.html (30cm)

Now the NVMe is placed right below the top fan with perfect airflow, and I can also change/replace it anytime in future easily without touching the GPU side at all. 

There is no need to add any heatsink (which I had actually prepared as well) because the temperature is so low already without any heatsink.

Originally, the SDD temp is around 50c at idle, and now it is 34~37c only (ambient temp is 31c).

On top of that, the temp will not go up even during full load because the airflow is so good now.

The NVMe controller temp was like 60~70c before, but now it is always at 42~47c only.

A surprising bonus which I wasn't expect is that the Sequential Write speed is MUCH faster now! I didn't know the SSD was throttling for the whole time with original configuration, probably due to the NVMe controller high temp without airflow.

It was 1580MB/s before, and now it is 3330MB/s!

The other performances are basically the same, but Sequential Write speed is doubled now.

 

[MostlyICareaLot]

Glad to hear it has improved the temps.

Wondered though what you meant by"Now the NVMe is placed right below the top fan ". Did you mount the SSD to anything or is the SSD literally hanging on a cliff xd?

[RD3N]

Like @eric418 I confirm that using a NVMe extension cable to move the SSD below the top fan solves the problem.
I did it 10 days ago and didn't get any bsod since then whereas, before that, my Corsair One had become unusable exactly with the symptoms described all along this thread.

Before using an extension cable, I had previously:

• reinstalled Windows,
• updated the bios to its latest version,
• installed Samsung NVMe controller driver,
• removed, cleaned and put back the RAM sticks,
• set the fan speed to 60% instead of 25% by default in iCUE,
• added a heatsink on the SSD.

But these actions hadn't solve the bsod, the boot loop to bios etc.

After monitoring the SSD temp, I opened the machine and put an external fan 😅 which drastically reduced the SSD temp and vanished the bsod.

I then bought a 20cm extension cable and moved the SSD up, just below the top fan. I didn't moved it to the CPU side like @eric418 but keeping it to the GPU side also did the trick.

[eric418]

10 hours ago, MostlyICareaLot said:

Glad to hear it has improved the temps.

Wondered though what you meant by"Now the NVMe is placed right below the top fan ". Did you mount the SSD to anything or is the SSD literally hanging on a cliff xd?

Somehow, I can just place it next to the chipset heatsink and it can be placed it firmly and securely there without any mounting. I don't know how to explain but it is not moving at all. 

But with the extension cable, you can actually place the NVMe anywhere you want within the case.

[eric418]

5 hours ago, RD3N said:

Like @eric418 I confirm that using a NVMe extension cable to move the SSD below the top fan solves the problem.
I did it 10 days ago and didn't get any bsod since then whereas, before that, my Corsair One had become unusable exactly with the symptoms described all along this thread.

Before using an extension cable, I had previously:

• reinstalled Windows,
• updated the bios to its latest version,
• installed Samsung NVMe controller driver,
• removed, cleaned and put back the RAM sticks,
• set the fan speed to 60% instead of 25% by default in iCUE,
• added a heatsink on the SSD.

But these actions hadn't solve the bsod, the boot loop to bios etc.

After monitoring the SSD temp, I opened the machine and put an external fan 😅 which drastically reduced the SSD temp and vanished the bsod.

I then bought a 20cm extension cable and moved the SSD up, just below the top fan. I didn't moved it to the CPU side like @eric418 but keeping it to the GPU side also did the trick.

Thanks for sharing!

It seems your choice of placing the NVMe is better than mine. There is more space on the GPU side.

But I'm too lazy to take everything apart again, will just keep it like this for now.

[Andy036]

On 9/2/2022 at 1:56 PM, Ryan LeBlanc said:

Yet another problem to add to the growing pile of unhappy Corsair One i164 customers.  I purchased mine in June 2020 and 1-year to the date I started having random BSOD -> BIOS boot loops.  I was right at the end of my warranty and Corsair accepted it as an RMA return in early July 2021 - unfortunately I had already paid several hundred in software fees trying to get it repaired locally.

My new RMA device worked great for about 9-10 months and then had the exact same problem with increasing BSOD -> BIOS boot loops.  Resetting Windows worked for a short period of time however the problem has returned.  Essentially the computer is unusable now.

Like many here I've had almost identical problems - when booting to BIOS the SSD is not recognized (only the back-up non-SSD is found).  I've run a RAM diagnostic tool (Memtest86) at Corsair's recommendation which fully passed with no errors over 4 passes.

As RMA returns only have a 3-month warranty I am doubtful they will be able to help me much here.  I'm debating replacing the SSD like others have tried (which will cost several hundred) or just giving up altogether and having a new machine built.  I spent $4,200 CAD on this device new.

Given the number of people just in this forum alone all with IDENTICAL problems including different RMA machines this appears to be a manufacturing error/problem and I would surmise that Corsair is liable from either a Class Action or Consumer Proposal standpoint.  If the issue can't be resolved satisfactorily I'll be pursuing this option.

Hi, i am heaving same issue after reading all these posts i have now realized that Corsair already know this issue reason why they are very cold in response on the BSOD tickets- and its still going on. now i know i am just another victim of this fraud  company but willing to know did you get any final resolution?