Failing A PCIe 5.0 NVMe SSD In Less Than 3 Minutes Without Extra Cooling

[DrLecter]

unacceptable.

[bug77]

unacceptable.

It's perfectly acceptable if your motherboard has its own heatspreader. Or if you stick it into a PCIe 4 slot, it's still a pretty speedy drive. Don't know if you'd want to pay the PCIe 5 premium in the latter case, though.

[Michael]

Phison has now commented on the article, update at the end.

[rob-tech]

It's perfectly acceptable if your motherboard has its own heatspreader. Or if you stick it into a PCIe 4 slot, it's still a pretty speedy drive. Don't know if you'd want to pay the PCIe 5 premium in the latter case, though.

No, you're wrong, it's not acceptable under any circumstances to generate file system errors as the drives thermal throttling behaviour should account for the worst case scenario of no heatsink and the specs and box do not clearly mention heatsink usage as necessary.

This problem, among others such as burning Ryzen CPUs is everything that is wrong with new PC parts and proves that QC seems to be minimal or non existent with the new stuff. This is also a big black mark on Phison and Corsair, further reinforcing the notion that now you should never be an early adopter of any new PC hardware.

It is nice that this is being fixed with firmware, however, the issue should have never been allowed to surface.

[AndyChow]

I have a bunch of Samsung 980 Pro. It's pcie4, but the point is it throttles if it gets hot. There's no universe where something that's not overclocked should start corrupting your data.

Hopefully Phison's new comment is real, and they are fixing the throttle curve. But not sure how that's related to cooling. Even with cooling, the drive should throttle, but here it will still clearly corrupts before it throttles. So I'm not sure I believe their final explanation. (to convince yourself, imagine it had a cooler but the ambient temperature was higher. Why would that change the drives behavior? Clearly it won't.)

[ChrisRamseyer]

https://www.youtube.com/watch?v=zRvJ...lectronicsCorp.

That should give everyone a better understanding of the kind of thermal testing that we do at Phison. It's not just a cold basement with a bunch of fans blowing on components.

As I mentioned in the statement to Michael that is now posted with the article, this is an issue we will address. Gen4 and Gen5 SSDs have the same 11.55w ceiling dictated by JEDEC for the M.2 slot. With Gen5 it just ramps up faster since you move 10-14GB/s instead of 7-7.5GB/s. The polling rate has to be faster to catch the drive as it scales up in load generating heat. You can actually use the drive without a heatsink as long as you have .2 LFM air moving across it. That's not a lot of air but we found that many reviewers use open table test benches without a lot of air moving over the SSD and that is creating issues in testing. The TechPowerUp system is a closed system like many end user systems, but it only has exhaust fans and no intake fans.

Regardless, we are glad reviewers reported the issue to us and we have already started the process of fixing the issue as it pertains to the E26's base FW code.

[Paradigm Shifter]

Phison's strict validation process

I presume they mean the strict validation process which allowed this critical issue to actually reach consumer release?

[rob-tech]

https://www.youtube.com/watch?v=zRvJ...lectronicsCorp.

That should give everyone a better understanding of the kind of thermal testing that we do at Phison. It's not just a cold basement with a bunch of fans blowing on components.

As I mentioned in the statement to Michael that is now posted with the article, this is an issue we will address. Gen4 and Gen5 SSDs have the same 11.55w ceiling dictated by JEDEC for the M.2 slot. With Gen5 it just ramps up faster since you move 10-14GB/s instead of 7-7.5GB/s. The polling rate has to be faster to catch the drive as it scales up in load generating heat. You can actually use the drive without a heatsink as long as you have .2 LFM air moving across it. That's not a lot of air but we found that many reviewers use open table test benches without a lot of air moving over the SSD and that is creating issues in testing. The TechPowerUp system is a closed system like many end user systems, but it only has exhaust fans and no intake fans.

Regardless, we are glad reviewers reported the issue to us and we have already started the process of fixing the issue as it pertains to the E26's base FW code.

Thanks for the info, you guys seem to have robust testing, however, most of us believe that your strict validation process should have caught something like this before release to consumers especially since you mention that it pertains to the E26 base firmware because corrupting a customer's data due to such a bug is a big deal.

That having been said, I own two high end Gen 4 Phison SSD's (one with E16 and another E18) and I really like both of them, this situation likely will not turn me away from Phison powered SSD's in the future, however, you guys should ensure a catastrophic bug like this does not happen again.

[theriddick]

Makes me think about those solid state coolers that were showed off and could remove like 5-10W heat, seems like these devices would be perfect candidate for them. The solid state coolers I can't remember the company but they are basically flat little squares that pump out heat pretty well. No moving parts.

[oiaohm]

Well, it makes sense to require cooling to reach top performance. But just like CPUs, when cooling doesn't cope, they're supposed to throttle back, not crap out. There is definitely something to fix in the firmware.

SSD have a hard limit. At 80C the NAND flash in them is at risk of being destroyed. So a SSD without enough cooling/fails to throttle soon enough there is no other option bar absolutely stop to get the NAND flash chips out of the danger temperature. . Repeated cycling into 80C also is reducing the life span of the Nand.

Sorry, but - nope. This is per definition not a proper testing procedure as it obviously lacks an important aspect. Complete testing involves testing under thermal stress and not in an ideal air-conditioned environment.

ChrisRamseyer response next is what I suspected the case to be. They had done the testing got a airflow requirement then that fact did not get to us instead they presumed M.2 slot would have that amount of airflow so gave out no warning that active cooling could be required with the device.