Possible Seagate X22 ST22000NM001E Fake Drive

[unlucky]

Hello guys!

I need help and advice:

Right now i have 2 discs of same model Seagate X22 ST22000NM001E

On pictures it's the right one. Left one is original

My thoughts:

QR code not original (upper right one and bottom one)
SN of HDD not original
PN of HDD not original
Syntax errors on Seagate adress

Bold Font on sticker

Weight of HDD is 690 g. instead of 680

Different motherboard on back side


What type of tests should i do to check it properly? I read that scammers delete SMART data and FARM also... How i can prove it's fake by software?
Im ready to provide any additional photos you need. In 1-2 days i will have PC with SATA to test this drive

Attachments

3.jpg (31.95 KiB) Viewed 758 times

[fzabkar]

The bad spelling clearly points to a fake.

Check your serial numbers here:

https://www.seagate.com/au/en/support/warranty-and-replacements/

Code:

Your Product      Exos X22 22 TB
Model Number      ST22000NM001E
Serial Number     ZX299GC3

Your product is not under warranty. This product was originally sold as a part of a larger system. Please contact the system manufacturer or your place of purchase for warranty support.

Code:

Your Product    Exos X22 22 TB
Model Number    ST22000NM001E
Serial Number   ZX2881VZ

Warranty valid until September 13, 2029

The DOM on the label is is 22 Oct 2024, so that's a warranty period of 1 month less than 5 years. That's clearly not right.

The PCBs have date codes of 2439 (left) and 2403 (right). That's week 39 and week 03 of 2024.

Here is a 30TB HDD with the same PCB as your fake:

https://www.kitguru.net/components/hard-drives/simon-crisp/seagate-exos-m-30tb-hdd-review/all/1/

[fzabkar]

Same PCB here:

https://realhardwarereviews.com/seagate-exos-x24-review/6/

[unlucky]

Thanks for your reply!

Right now i checked SMART AND FARM on both drives.

I must say they delete almost all info even on FAKE one...

I add 2 files, SMART and FARM from original (left one from photo) and fake (right one) drives.

Can you help me with smart and FARM values to find critical moments ?

RN i find one : Assembly date YYWW on original is 2501 (DOM 30 dec 2024 = 1 week of 2025)
Assembly date on fake is 2531, but it's week started from 28.07.25 but DOM on disc is 22.10.24

My current plan for futherer investigation: Right now im checking fake drive for old files with file recovery tool R-Studio - if i find old files on disk it will be bingo! But it will complete scan in 24 hours...
Next - I want to check read/write speeds to compare with Seagate datasheet on this product. Which software shoud i use?
And final step - fake drive have FW SN02 - i want to update firmvare to 03 version and look what happens (if it's fake it wont update, or become to show real working hours/model name after update)

Attachments

fake_FARM.txt

(26.26 KiB) Downloaded 17 times

orig_FARM.txt

(26.95 KiB) Downloaded 14 times

[SWM]

If they (the Chinese?) rescan x24 (down) to x22 using WinFOF, you don't have to look for anything on the disk. I doubt there's anything left on the disk after the factory scan, except for the dates.

[unlucky]

My previous message lost...
Can you help me to analyze SMART and FARM data?

I add smart and FARM from both drives.

RN i find key differences:
On real one, assembly date from FARM = DOM Date from disc YYWW 2501 = 30 dec 2024
On Fake one, assembly date from FARM didn't match DOM date: YYWW 2531 = 28.07.25 and DOM is 22.10.24 - so obviously fake.

My plan:
RN im trying to find some recovery data on fake disk with R-studio Software
Next:

What type of Software to test read/write speeds you prefer? I want to compare them side by side with data sheet on x22 drive
Last step: I want to try to update firmware from SN 02 to SN03 and see what happens.

Attachments

orig_FARM.txt

(26.95 KiB) Downloaded 18 times

fake_FARM.txt

(26.26 KiB) Downloaded 18 times

[SWM]

If you want to return the drive, the label with errors in the legal address is sufficient. You won't find old files—I wrote about why.
If you update the firmware (no matter the result), the seller may refuse your return. I would return this HDD if I bought it new.
Unless you're refurbishing it yourself, of course. It's a convenient way to check for discrepancies.

[unlucky]

If you want to return the drive, the label with errors in the legal address is sufficient. You won't find old files—I wrote about why.
If you update the firmware (no matter the result), the seller may refuse your return. I would return this HDD if I bought it new.
Unless you're refurbishing it yourself, of course. It's a convenient way to check for discrepancies.

Thanks for the info.

The problem is that company bought 1000+ of this fake HDDS - all with fake labels, SN etc.

And seller tells to our stupid IT department, that it's just OEM drives, LOL.

Im working in Corporate Security/Risk Management department and try to find any proofs that it's fake - all what i can.

[SWM]

Your files show that the fake drive has two heads disabled before testing. The board tests and initial head tests are skipped.
Factory testing of such a drive takes many days, and testing the board requires specialized factory equipment.
But this is my subjective opinion. I do not represent the manufacturer.

[Hardcore Games]

It seems that scammers are still at it with 30TB class disks with 10 plates at 3TB in the drive. If there are 2 plates faulty then then the disk would be 24TB which sells for a lot lower price than the 30TB disk.

Sure seems like some scam is based in south east Asia somewhere. I suggest reporting the scam to authorities so that the miscreants can be discovered.

[SWM]

Not two heads are disabled, but three. I made a mistake when I skimmed through it. But it doesn't matter.

Code:

Orig                                                              Fake
Device Interface: SATA                                 Device Interface: SATA
Device Capacity in Sectors: 42970644480         Device Capacity in Sectors: 42970644480
Physical Sector Size: 4096                             Physical Sector Size: 4096
Logical Sector Size: 512                                Logical Sector Size: 512
Device Buffer Size: 536870912                        Device Buffer Size: 536870912
Number of Heads: 20                                     Number of Heads: 17

[unlucky]

Thanks for the info!

Sooo- what type of drive is that monster with 17 heads for 22 TB drive?

P.s. - i continue tests, attached more tests

Attachments

[SWM]

I can't understand what you want. The label is fake and there are 17 heads.
https://www.seagate.com/content/dam/sea ... 12000a.pdf
Table on sheet 7:
ST22000NM001E,
ST22000NM002E
Heads 20
Disks 10
What do your tests tell you? After the WinFof, the HDD is just like factory, except for dust in the hermetic block with damaged surfaces. How long will it take for it to get under the remaining working heads?

[unlucky]

I can't understand what you want. The label is fake and there are 17 heads.
https://www.seagate.com/content/dam/sea ... 12000a.pdf
Table on sheet 7:
ST22000NM001E,
ST22000NM002E
Heads 20
Disks 10
What do your tests tell you? After the WinFof, the HDD is just like factory, except for dust in the hermetic block with damaged surfaces. How long will it take for it to get under the remaining working heads?

This fake HDD is Around 10% slower in seq. read/write tests. Another question - what happens with fake HDD if i try to update firmware on it ?

[SWM]

You'll tell us that after the update. Especially if the HDD grows three heads after the update, the label errors disappear, and the dust from the hermetic block disappears! Please don't mock us...

[fzabkar]

The FARM data must be fake. The manufacture dates on the PCB and in the FARM log are 18 months apart.

Assembly Date (YYWW): 2531
Date on PCB : 2403

Both drives have the same capacity, but the fake has a transfer rate which is 10% slower. This means it must have a lower areal density, which in turn means that it must have more heads.

(No. of heads for fake) = (No. of heads for original) x (1.10)^2 = 20 x 1.21 = 24.2

To me, it looks like the actual number of heads is 24. If you can dump the ROM(s), I should be able to determine the real number of heads from the adaptive segments.

Another way to see the real number of heads would be via a short-stroke read benchmark in HD Tune or Victoria.

How to determine number of heads using HD Tune:

https://www.hddoracle.com/viewtopic.php?f=59&t=650

You will need Firefox to access the site (because of ongoing DDOS attack mitigation).

[unlucky]

You'll tell us that after the update. Especially if the HDD grows three heads after the update, the label errors disappear, and the dust from the hermetic block disappears! Please don't mock us...

Update failed... I tried to update to SN03 via seatools and it fails instantly.
Update to SN02 from fake "SN02" also failed.

[unlucky]

The FARM data must be fake. The manufacture dates on the PCB and in the FARM log are 18 months apart.

Assembly Date (YYWW): 2531
Date on PCB : 2403

Both drives have the same capacity, but the fake has a transfer rate which is 10% slower. This means it must have a lower areal density, which in turn means that it must have more heads.

(No. of heads for fake) = (No. of heads for original) x (1.10)^2 = 20 x 1.21 = 24.2

To me, it looks like the actual number of heads is 24. If you can dump the ROM(s), I should be able to determine the real number of heads from the adaptive segments.

Another way to see the real number of heads would be via a short-stroke read benchmark in HD Tune or Victoria.

How to determine number of heads using HD Tune:

https://www.hddoracle.com/viewtopic.php?f=59&t=650

You will need Firefox to access the site (because of ongoing DDOS attack mitigation).

Thanks i will try to do dump of ROMs to send for you- but with which software i can do it?

[unlucky]

" If you can dump the ROM(s), I should be able to determine the real number of heads from the adaptive segments." - With which software i can dump the Rom(s) from this fake drive?

[pepe]

need to use an spi memory programmer, coz all interfaces are locked...