BestBuy - WD - Easystore 14TB External USB 3.0 Hard Drive - $190

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

madbrain

Active Member
Jan 5, 2019
212
44
28
Mine were all WD140EDFZ . All 8 tested fine with h2testw while in their USB enclosures.

Once shucked, I ran into a problem with one of the 8 drives. It wouldn't spin up.
It doesn't spin up with the SATA cables from the PSU - a Raidmax RX-AE1200 .
It doesn't spin up with the so-called SATA "EZ-dock" of the Cooler master HAF-932 case, even though this EZ-dock is powered through a molex cable.
It does spin up in the following internal SATA dock, which is powered by SATA power cable from the PSU.
It also spins up in the following external USB SATA dock :

All 7 other drives spin up on matter where they are hooked up ... Very strange !

I was hoping to have 6 drives internally in my case, and 2 in the EZ-dock.
Instead, I ended up with 6 internally in the case, 1 in the EZ-dock, and 1 in the Icy-dock 971.

I really wonder what's different about this particular drive, especially given that it looks identical - same model number, same firmware. I haven't checked for external physical differences yet. I'm considering unshucking it and returning it to BB for another less capricious drive. I may do that and resilver my array since I'm running raidz2 .

I'm transferring the data from my old array now. I have all 8 new 14 TB drives hooked up, plus all 6 old 10TB drives. Both arrays are RAIDZ2.
The old 6 drives are hooked as
1 on the EZ-dock
3 in the following icy dock I just purchased
(warning: fan is extremely loud, not suitable for 24/7 use. But at least it only turns on if a drive is plugged in).
and 2 in the aforementioned Inateck USB 3.0 dock

I have about 32TB to transfer. Old array was 90% full. Hopefully this will take less than a day to transfer.
 
  • Like
Reactions: foureight84

Stephan

Well-Known Member
Apr 21, 2017
944
712
93
Germany
@madbrain Check SATA plug specs and your cable if this one drive gets 3.3V on PWDIS pin. Test with a Molex to SATA power converter temporarily, drive should then spin up. Because Molex has no 3.3V just 5+12V. If that is the case you need a "3.3v pin mod" for this drive.

What is happening is that some shucked drives need this mod and some don't. The HDD's label will not show this in any way, have to test it. I ended up buying a spare PSU to 4x SATA cable and removed the 3.3v rail in a safe way (no chance for short circuit).
 
  • Like
Reactions: BoredSysadmin

madbrain

Active Member
Jan 5, 2019
212
44
28
@madbrain Check SATA plug specs and your cable if this one drive gets 3.3V on PWDIS pin. Test with a Molex to SATA power converter temporarily, drive should then spin up. Because Molex has no 3.3V just 5+12V. If that is the case you need a "3.3v pin mod" for this drive.

What is happening is that some shucked drives need this mod and some don't. The HDD's label will not show this in any way, have to test it. I ended up buying a spare PSU to 4x SATA cable and removed the 3.3v rail in a safe way (no chance for short circuit).
Thanks. As far as I know my PSU is up to the latest ATX specs.
I will try with a Molex to SATA adapter when my data transfer is complete. I'm not sure if that will be today. It might be tomorrow.

However, as i mentioned in the previous post, one of my docks, the dual drive EZ-dock from the Cooler Master case, is powered by Molex (straight Molex, not Molex to SATA). So it's pretty strange to me that the drive won't spin up on that dock.

The MB971 Icydock dock that does work is powered by Molex to SATA.
 

josh

Active Member
Oct 21, 2013
615
190
43
Waiting for Amazon to match with the elements. Free global shipping and EMFZs instead of EDFZs :)
 

madbrain

Active Member
Jan 5, 2019
212
44
28
What's the difference between the two other than EMFZ doesn't suffer from the 3.3v issue?
I don't think the 3.3V issue has anything to do with the model. According to a poster above, it is random on certain drives. Out of 14 drives (6 x 10TB, 8 x 14 TB), I have one drive that presents something that might be this issue, but not confirmed yet.

I believe the difference between the EDFZ and EMFZ has to do with support for hardware encryption.
 

josh

Active Member
Oct 21, 2013
615
190
43
What's the difference between the two other than EMFZ doesn't suffer from the 3.3v issue?
The lack of TLER is a deal breaker on the EDFZ. People say you can write a script to turn it on every boot but I don't believe that a value that doesn't persist on reboot is actually being set at all.
 

madbrain

Active Member
Jan 5, 2019
212
44
28
The lack of TLER is a deal breaker on the EDFZ. People say you can write a script to turn it on every boot but I don't believe that a value that doesn't persist on reboot is actually being set at all.
Not sure why you would think that. It's pretty standard for firmware to keep state in RAM.
Many others have been using various WD drives and setting various features on via smartctl for years, not just TLER.

 

josh

Active Member
Oct 21, 2013
615
190
43
Not sure why you would think that. It's pretty standard for firmware to keep state in RAM.
Many others have been using various WD drives and setting various features on via smartctl for years, not just TLER.

Well you only know if the drive actually has TLER on failure and by then it's too late for your data
 

madbrain

Active Member
Jan 5, 2019
212
44
28
Well you only know if the drive actually has TLER on failure and by then it's too late for your data
Well, the other values that are changed dynamically like sleep timeouts can be observed to take effect after being set via smartctl. I don't see any reason why the TLER wouldn't be the same. Certainly if you want to theorize that's the case, you should come up with some evidence to back it up. I don't have access to the WD firmware source to tell you. But as a firmware developer (though not HD firmware developer), I think it would be pretty shitty for the drive to accept a value and then just throw it away. The proper response would be to reject the command. Anything else would be a bug.
 

josh

Active Member
Oct 21, 2013
615
190
43
Well, the other values that are changed dynamically like sleep timeouts can be observed to take effect after being set via smartctl. I don't see any reason why the TLER wouldn't be the same. Certainly if you want to theorize that's the case, you should come up with some evidence to back it up. I don't have access to the WD firmware source to tell you. But as a firmware developer (though not HD firmware developer), I think it would be pretty shitty for the drive to accept a value and then just throw it away. The proper response would be to reject the command. Anything else would be a bug.
Like I said, there's no way to know if it works until a drive actually fails. I'm not very willing to risk my data or drives to test this so I'll just wait for the reports. I also have enough of 14TB EMFZs so no reason for me to gamble with my data. Hope it works out for you anyway.
 

thedman07

New Member
Sep 14, 2020
24
23
3
I was finally able to pick my last two drives up at Best Buy. Now all 8 are cooking away on a bad blocks test before I shuck them.