Used SSD - high power on count, low total hours?

[chubbyfatazn]

Got myself a used 128GB Samsung PM830 off of ebay. When I plugged it into my laptop and ran CrystalDiskInfo on it, the power on count was 3290 with 10 total power on hours.

Would this have been some sort of review or test drive? It's an mSATA SSD, and isn't available via the retail channel. Health is reported at 94%. Reallocated sector count is still far from the limit threshold, so I'm not sure what happened to this drive or if I should keep it. So far so good. What say you guys?

 

[mrpiggy]

Probably lifted from wherever they were doing automation testing. That power-on count comes to 5.5 power cycles per minute or a power cycle every 11ish seconds. As that's faster than even PC's will go through a typical BIOS POST, it was probably done on a special drive testing unit like a Flexstar drive tester.

Or someone had it hooked to a very flaky power supply for exactly 10 hours...

 

[chubbyfatazn]

Probably lifted from wherever they were doing automation testing. That power-on count comes to 5.5 power cycles per minute or a power cycle every 11ish seconds. As that's faster than even PC's will go through a typical BIOS POST, it was probably done on a special drive testing unit like a Flexstar drive tester.

Or someone had it hooked to a very flaky power supply for exactly 10 hours...

I'm not too knowledgeable about SSDs, so forgive me for this. What exactly are the downsides, if any, of having a high power cycle count? I've had regular mechanical HDDs that have well exceeded that number, but I don't recall reading anything about power cycles (only power on hours) for SSDs.

 

[hhhd1]

That is strange stats, also 94% remaining, which means it has used up 6% of its write cycles ... there is no way it could have used that amount of cycles while being on for 10 hours only, and while staying most of the time booting up and shutting down ....


Try this:

1. take a note of the power on count and power on hours smart values.
2. Power it on for 30 minutes, then shut down.
3. Repeat step 1 and 2 few times.

see if power on count increases by not keeping it on for a complete hour at a time.

i am thinking that it 'might' be a bug in the samsung firmware, that it doesn't count the hour until a whole hour has passed

 

[chubbyfatazn]

That is strange stats, also 94% remaining, which means it has used up 6% of its write cycles ... there is no way it could have used that amount of cycles while being on for 10 hours only, and while staying most of the time booting up and shutting down ....


Try this:

1. take a note of the power on count and power on hours smart values.
2. Power it on for 30 minutes, then shut down.
3. Repeat step 1 and 2 few times.

see if power on count increases by not keeping it on for a complete hour at a time.

i am thinking that it 'might' be a bug in the samsung firmware, that it doesn't count the hour until a whole hour has passed

Nope, definitely not a firmware bug... it's been logging the cycles and hours properly since I installed it. It's gained 6 cycles and 4 hours since I've installed it, which is about right for me (installed Windows and drivers). Thanks for the suggestion, though.

 

[tweakboy]

Well its on its way, dying a slow agonizing death. NEVER buy used SSD you dont know how the powner handled it. I dont want you to lose any important stuff. Honestly back up your stuff my friend cuz you dont know when the drive will go BOOOOM! then your in a bigger hole... Just back up your stuff to external drive and keep a eyes on that 94 percent health.

 

[Vinwiesel]

Does it have to be a hard shutoff before it counts as a power cycle? For example, will a reboot add 1 to the counter? I once had a CPU fan die in the middle of the night and the computer was rebooting every 20 seconds for god knows how long.

 

[tweakboy]

Each reset counts.

 

[mrpiggy]

Wouldn't worry about the power cycles. If it was going to blow up from power cycles/thermal expansion tests, chances are it would have already (that's why SSD's get those kinds of tests in the first place). 94% useage is fine too, it's got plenty of life left. Just use it and enjoy it.

 

[tweakboy]

Yes piggy is right. However I would still consider a external backup solution.

 

[chubbyfatazn]

Does it have to be a hard shutoff before it counts as a power cycle? For example, will a reboot add 1 to the counter? I once had a CPU fan die in the middle of the night and the computer was rebooting every 20 seconds for god knows how long.

It incremented after I rebooted after installing some crap. It also counted going to sleep/waking up from sleep as a cycle.

Wouldn't worry about the power cycles. If it was going to blow up from power cycles/thermal expansion tests, chances are it would have already (that's why SSD's get those kinds of tests in the first place). 94% useage is fine too, it's got plenty of life left. Just use it and enjoy it.

Yeah, I guess so. For the price I'm pretty happy with it. Always have the regular mechanical running alongside (thank you Lenovo) it if I ever need to revert.

 

[hhhd1]

Overall, i wouldn't worry about it .. however the combination of numbers seems really strange.

Can you post a screenshot of crystal disk info, with all smart data visible ?
make sure to be using a recent version of crystal disk info.

 

[chubbyfatazn]

Overall, i wouldn't worry about it .. however the combination of numbers seems really strange.

Can you post a screenshot of crystal disk info, with all smart data visible ?
make sure to be using a recent version of crystal disk info.

Yeah. The seller keeps telling me that it was a working pull from a laptop with cosmetic damage... obviously not. Total host writes was probably around 330GB before I plugged it in and installed everything.

Using v4.6.2:

 

[Phynaz]

Well its on its way, dying a slow agonizing death.

Ignore Tweeky. He's off his meds.

 

[Phynaz]

Got myself a used 128GB Samsung PM830 off of ebay. When I plugged it into my laptop and ran CrystalDiskInfo on it, the power on count was 3290 with 10 total power on hours.

Don't worry about it. My laptop does the same thing.

Edit:
I just checked mine. 1400 power cycle count, 268 hours. 3.28TB written.

 

[chubbyfatazn]

Ignore Tweeky. He's off his meds.

Lol, we're all well aware of that.

I bet my problems stem from the fact that the PCB isn't black.

Don't worry about it. My laptop does the same thing.

Edit:
I just checked mine. 1400 power cycle count, 268 hours. 3.28TB written.

None of my other five SSDs have any issue with CDI or being logged.

I guess I wouldn't be too worried if the drive had a similar number of power-on hours as total cycle count. I'm just interested in what testing procedure the thing went through beforehand at this point.

 

[Phynaz]

Lol, we're all well aware of that.

I bet my problems stem from the fact that the PCB isn't black.



None of my other five SSDs have any issue with CDI or being logged.

I guess I wouldn't be too worried if the drive had a similar number of power-on hours as total cycle count. I'm just interested in what testing procedure the thing went through beforehand at this point.

It didn't. It's most likely Windows power management. You know, every time you close the lid of a laptop to go to a meeting power to the drive gets killed. Stuff like that. My laptop gets at least 9 hours a day of use, and it's only saying the drive has been powered on for 268 hours - that's a month.

 

[hhhd1]

It didn't. It's most likely Windows power management. You know, every time you close the lid of a laptop to go to a meeting power to the drive gets killed. Stuff like that. My laptop gets at least 9 hours a day of use, and it's only saying the drive has been powered on for 268 hours - that's a month.

That 'kinda' proves my theory.

@chubbyfatazn

a reboot to the computer might not record as a power cycle.

try to make sure its powered off .. maybe like repeat it every 15 minutes for a while.


EDIT:

although i might not be totally correct, but still, the point is, the power cycle counts does not get counted properly.

 

[mrpiggy]

It didn't. It's most likely Windows power management. You know, every time you close the lid of a laptop to go to a meeting power to the drive gets killed. Stuff like that. My laptop gets at least 9 hours a day of use, and it's only saying the drive has been powered on for 268 hours - that's a month.

While it could be a Windows thing, it's doubtful for this particular drive.
The Samsung 830 mSATA was announced in Nov/Dec and didn't get any reviews until the Feb/Mar time frame. It was probably released to OEM's and Retail channels a couple of months after that. So on a extremely tightened estimate, actual early retail and/or OEM availability was May/June. So let's say the orignal owner had a good three months of ownership because he was 1st in line. That 3290 drive power cycles in 90 days is too high a number.

If we assume that it was simply caused by Win power management and open/shut laptop, that's around 36-37 open and closes per day with only 10 hours of use in 90 days... That's a bit of a stretch. Even pretending the seller got it 6 months ago ~180 days (which is pretty doubtful considering the mSATA 830's were only available to a select few review sites 6 months ago), that's still 18+ power cycles or lid openings/closings per day and only 10 hours of use; which is still stretching the numbers.

I think it's great that the OP got a cheap 830 mSATA, but the power cycles, power on-time, and release dates of the Samsung mSATA 830 SSD's says it's likely a q/a type OEM test drive that probably got "liberated" along the way as opposed to the seller's claim it just happened to be in some damaged laptop (and SSD equipped mSATA latops are still rare) he happened to have.

 

[chubbyfatazn]

It didn't. It's most likely Windows power management. You know, every time you close the lid of a laptop to go to a meeting power to the drive gets killed. Stuff like that. My laptop gets at least 9 hours a day of use, and it's only saying the drive has been powered on for 268 hours - that's a month.

Every time I shut the lid my X220 goes to sleep - and the power-on cycle counter increments by 1 each time. All the SSDs I've ever put in a Thinkpad (two M4s, one Kingston V+100) have always exhibited this behavior.

I think it's great that the OP got a cheap 830 mSATA, but the power cycles, power on-time, and release dates of the Samsung mSATA 830 SSD's says it's likely a q/a type OEM test drive that probably got "liberated" along the way as opposed to the seller's claim it just happened to be in some damaged laptop (and SSD equipped mSATA latops are still rare) he happened to have.

That's what I've been thinking. This drive, along with the mSATA M4s, are some of the newest mSATA SSDs out there. The numbers would make somewhat more sense if this were a drive with an older controller (SF2200?).

The only things this SSD has been definitely found in are ultrabooks, notably Samsung's own line, Dell, and Apple afaik. He's sold at least three others under the same cosmetic damage guise, so I'm guessing they're system pulls. He's also been dodging my questions asking about what procedures the drive went through. I'm banking on the established reliability of the controller in the 2.5" Samsung SSDs in keeping this drive. Would've nabbed the M4 instead, but Newegg was OOS, so the next best alternative was the Sammy.

 

[Compman55]

94% is still good. With SSD you need a mindset to keep your data backed up anyway.