I will buy a drive from eBay.
I will try to replace the diode. Where to get a board with dump?
eBay sellers cannot do that.
Will waiting over the years make it more likely to be repairable or having all the
common components replaced (resistors, coils, transistors, diodes,
capacitors, etc.)?
But of course I will ask the site I mentioned that they give me the ROM file.
Yes, eventually support for your drive should become available. Whatever you do, don’t damage the ROM. And don’t buy a donor drive just for a diode. Any unidirectional 5V TVS diode will do, if you really must.
Be aware that eBay is full of fakes.
I will but so that the repair service can move my ROM to the
good PCB.
I will see first if the diode fixes it.
By the way, if I cover the SATA data connection of the 8TB Seagate enclosure the failed drive doesn’t spin down (those enclosures that give 1 year of cloud storage).
I know some command is sent to spin it down as no other adapter does that.
It could be that the bridge controls a dual MOSFET which switches the 5V and 12V power. That’s how some WD My Books do it. To confirm whether this is the case, measure the SATA power pins before and after spin-down.
If the 5V and 12V rails are not switched, then the bridge must be communicating with the drive. That would mean that the drive must be working.
I will test the voltages.
The enclosure where I put it normally, that emulates 4096-byte sectors
from 512-byte sectors, says that the drive is half the size, 4657.53TB,
does that indicate some specific symptom, or a strange need for the diode?
Says the device is Not initialized, the device is not ready.
It has 5V and 12V always in the test enclosure, 3V seems disconnected.
Windows Server 2008 R2 spins it down before detecting it,
Windows 10 doesn’t. After a while, it blinks blue faster than 1 second,
as when it reads (it is a slow-blinging enclosure) then it blinks faster and stronger
in error.
It is not the OS, but the circuitry of the enclosure, that spins it off.
When I want to put it offline or do anything in Windows 10, it says:
“The request could not be performed because of an I/O device
error.”
This enclosure, that uses 512-byte sectors, gives the right capacity,
7452.04GB, Unknown.
The drive is formatted for the 4KB enclosure.
CrystalDiskInfo doesn’t detect it. It took some time for
it to detect the inaccessible drive, I had to move it to
a specific USB port.
The format is OK, I just can’t access sectors not even with iBored.
It spins much much longer with Windows 10. In Windows Server 2008 R2
spins fully maybe for 1-5 seconds.
4657.53TiB is 5TB. That doesn’t appear to have anything to do with 4KB sector sizes.
If the drive is spinning down, and if the 5V and 12V supplies remain switched on, then the bridge must be sending a spin-down command. That means that the drive is communicating with the bridge.
If the USB-SATA bridge is getting the correct capacity from the drive, then the drive must be responding to an Identify Device command.
In short, the bridge is just an annoying obfuscation. A direct SATA connection is the way to go.
Are you saying that it is possible to read the disk sectors once it spins?
If so, what is the procedure I need to follow step by step to read the drive
even if it doesn’t get properly detected at IDENTIFY or partition level?
I just want to copy the data to another drive if it’s readily possible now
even if I need the lowest level and strangest Linux utilities.
All I’m saying is that the bridge is communicating with the HDD PCB firmware. That says nothing about the discware, i.e. the firmware on the platters. The SMART data would be one component of the discware, so a failure to fetch a SMART report would point to internal problems. That’s assuming that the bridge isn’t getting in the way.
I have found that SAS disks, like 1TB 2.5 Constellation ones,
stop spinning fully when there is a sudden power loss with an
HBA, like Microsemi 2100 ones, but you can easily put them online in Windows
with a simpler HBA like LSI-3041 (B), but never back with the Microsemi one.
Is there a similar procedure for enterprise SATA?
SAS drives are preferable, they are much friendlier to re-enable
after hardware or power faults, but I find no information about
how to re-enable SATA drives at all, but it is surely as easy as with a SAS.
What tools are there to re-enable SATA drives in that way?
Will it be possible to do that if I buy a server grade 8TB+ PCIe
SATA controller or HBA to directly command the drive?
There are probably purely SATA LSI HBAs.
Maybe they are trying to reverse-engineer a problem that for now
can be fixed with proper drive/HBA compatibility matrix documents.
Probably these business critical drives are meant to behave like that,
you cannot really fix it, it is a required firmware feature that you cannot
reverse-engineer or bypass, you can only equip yourself with all the
different HBAs, server machines that specifically list those drives in
their compatibility reports, compatibility matrix, etc., for the exact
drive model and HBA/server, so what is required is always be up
to date with the servers/HBAs you have for maybe extracting their
hardware reenabling sequences for faulted datacenter enterprise critical drives.
I said Kroll Ontrack Recovery:
It is just 2 datacenter drives, 8TB HUH721008ALE600 DC HC510, and WUH721414ALE6L4 (14TB) helium drives.The 14TB one just lost access along a failed e=
nclosure, and the 8TB one had a short circuit in several pointsin the PCB a=
nd had the 5V TVS diode shorted and hot. I removed the TVS and now it spins=
. Both drives have no media accessand no SMART data. No matter where I conn=
ect them, they have no access to the media. Some enclosures spin them down,=
others can try to access them more easily, but they just return I/O, CRC or=
drive not ready errors.They appear with the right capacity in R-Studio wit=
h some enclosures, the firmware and model.
They are the SATA versions.They contain basic Windows GPT NTFS partitions, =
just like normal end-user 8TB+ enclosures.
I need to know if there is a way or a tool I can use to bring them to an us=
able state again.I know that my SAS drives can be reactivated after power f=
ailures with Windows and LSI HBAs,but I need to know if you can offer me a =
data recovery option that allows me to reenable these datacenterdrives as i=
f they were SAS drives, with a simple command or tool I can download and pa=
y for. They seem tohave no physical damage, they spin OK, they just don’t e=
nd their initialization procedure, just spin up,do slow seeks, but their ch=
aracteristic fast seek is not done and media cannot be read.
This is the first option I want to try, a tool I can download and pay for t=
oreenable datacenter WD drives that are locked due to power failure or encl=
osure failure.
I need to know if the firmware chip can be transferred to a new PCB of the =
same model(it seems to have 2 firmware chips), or if I can update the drive=
firmware, and how, sincethe manuals for these drives say that updating the=
firmware clears all error logs in the driveas if it was new, which could r=
eactivate the drive for media access.
I know a bit of assembly and DOS programming, if you have tools that can se=
ndrecovery commands I could run them, I expect them to be paid.
Would sending a new drive of the same model for spare parts with the PCB of=
the failed drivefor these models?
If not I need to know if I can recover this kind of drives with just USB Ra=
pidSpar or PC-3000,or other options with certain models of servers, HBAs, o=
r SATA PCI adapters that can just reactivatethe drives just by plugging the=
m into such servers/boxes. I want to be ableto reenable the drives with sim=
ple commands like I do with my SAS drives, if possible, beforeconsidering s=
ending the physical parts as a last resort. It seems to need just a server =
boxdatacenter reactivation sequence, not really damaged, so it would be ide=
al to try knownprofessional software tools I can download, or suggestions o=
f which SATA-capable devices/servers/HBAsfor handling this kind of datacent=
er drive problems.
It seems to be a simple datacenter SATA drive problem that would need me to=
send thedrives every time the power fails or an enclosure stops working in=
stead of justrunning a command and probably getting some specific model of =
SATA adapter/boxproven to reactivate these drives.
I am thinking on buying:Broadcom 9300-8e, -4i4e, -16e=C2=A0
According to this official HBA/drive compatibility list, which list my 2 drive models:https://documents.westerndigital.com/content/dam/doc-library/en_us/assets/public/western-digital/product/platforms/ultrastar-data60-hybrid-platform/compatability-matrix-ultrastar-data-60-storage-platform.pdf
Compatible SATA Drives =E2=80=93 WD, WDC
|=20
|=20
| |=20
Compatible SATA Drives =E2=80=93 WD, WDC
|
|
|
Do you think that these HBAs will be capable to list my drives as FAULTED o=
r DEGRADEDand allow me to reenable them?
Firmware 16.00.17.00 16.00.17.00Driver39.100.00.00CentOS/RHEL 8.0 and above=
:41.00.00.00-12.51.27.01
Note: To increase the responsiveness of the topology during the event of a =
cable pull, IOM/ESM reset, or drive pull, use ScrutinyCLI (version 32 or la=
ter) to change the Task ManagementReset Type on the HBA(s) from Target Rese=
t to I_T Nexus Reset, and change the DMDsettings on the HBA from 0 to 21 se=
conds. See Error
058 for more details.
I have looked over sick disks given to me and most logic damage is due to bad power sources. The board on modern disks is rather spare but probing suggests transient power damage can trash the VLSI.
I use multiple disks redundant backups in case of a disk failure. I use seval USB cases and get new models frequently. Mostly so that testing fixes is easier.
1. Replace the TVS Diode
The TVS diode (Transient Voltage Suppression diode) is designed for overvoltage protection. A shorted TVS can disrupt power delivery.
Steps:
Purchase a 5V TVS diode with identical specifications (verify voltage rating, package size, and polarity).
Carefully solder the new diode using a hot air gun or soldering iron (ensure anti-static precautions).
Expected Outcome: Restoring the TVS may stabilize power delivery, but further diagnostics are required if issues persist.
2. Inspect Other Circuit Issues
Check 0Ω Resistors: Confirm they maintain proper continuity and resistance (rarely, they may degrade under overcurrent).
Measure Critical Voltages:
Use a multimeter to test stability of 5V and 12V lines (e.g., motor driver IC, main controller power).
Inspect the PCB for broken traces or burn marks (especially near power lines).
Check Main Controller IC: If the TVS failure caused voltage spikes, the main controller may be damaged (look for overheating or communication failure).
3. Address Firmware/Recognition Issues
Use Professional Tools:
Test with tools like PC-3000 or MRT to diagnose firmware corruption and repair damaged modules.
If tools are unavailable, attempt a “safe mode” entry via PCB shorting (refer to model-specific guides).
Swap PCB:
Source an identical PCB (ensure firmware version matches) and transfer the original ROM chip (requires soldering expertise).
4. Test for Enclosure Compatibility
Some enclosures may spin down the drive due to insufficient power or protocol mismatches.
Testing Steps:
Connect the drive directly via SATA to a motherboard (bypassing the enclosure).
Use a powered SATA dock with independent 12V/5V supply.
Critical Notes：
If replacing the TVS diode fails to resolve the issue, the main controller or motor driver IC may be damaged (requires professional repair).
If data is critical, consult data recovery professionals like ease us data recovery immediately to avoid irreversible damage.
“ease us” ???
No software tool can recover data from the OP’s drive. Moreover, few, if any, professional services can handle helium drives.
I was expecting at least 1 of the drives to be just faulted/offline/bad
at the datacenter firmware level, as has happened to me a few times
with SAS drives.
If I try several HBA models, LSI, MegaRAID, and even proper servers
in the future, I will have to say that drives are much safer when they
are SAS. I have seen documentation in many places that says that you
can mark online or offline drives (SATA supposedly too, not only SAS)
in hardware whenever they are datacenter drives.
I will be really surprised to have seen around 2 faulted SAS drives, while
I have never seen a faulted SATA drive that doesn’t even spin, but starts
spinning as soon as you use the proper HBA, like LSI 3041 (B), and put
it Online in the disk manager.
I would say that SATA drives have no real way of unfaulting where SAS
drives fault/lock/stop spinning due to failed format or surprise powering them off
by accident. But from what I have read from documentation like Intel, Broadcom
and the like, you can make an LBA-inaccessible drive readable again whenever
you use a full-featured RAID controller/HBA, both for SAS and SATA datacenter
drives like mine.
Maybe that’s what happens with many external WD enclosure drives,
maybe they happen to be datacenter drives for some reason, then they
power-fault or have some minor problem overnight, then they lock themselves,
and then the enclosure says they are password protected when in fact they
are faulted waiting for a compatible full-featured SAS HBA to put them back
fully online in hardware/firmware.
It could be worth trying to revive external WD disks in such HBAs, in several
Intel, Western Digital, MegaRAID, Dell, and other servers until finding one capable
of putting those SATA drives back online.
I think I will NEVER buy a SATA drive for my critical massive data EVER again, only
SAS disks, be them SSD, DAS, and maybe some widely compatible enterprise
models as soon as I prove to myself that I can unfault them in case of problems.
I gave up on NAS when my old 4 disk unit no longer met my needs. I copied the sata to a HGST HUH721212ALE601. This disk has racked up some 31,000 power on hours since I procured it. I also have the more recent WDC WUH721818ALE6L4. I use this to handle more data.
Now I have an Orico 5-disk JBOD enclosure that handles my disks along with some USB individual enclosures. The old NAS has 4TB Seagate disks that still work.
What exact machine/server hardware/firmware model am I supposed to use with WUH721414ALE6L4 firmware version 240? It could be a virtually firmware-locked drive.
Can you PLEASE tell me what server/HBA do you use for your WUH721818ALE6L4 drive?
It has to be a controller capable of putting SATA drives back online if they become inaccessible.
I bought a MegaRAID 9440-8e HBA, but I will see if it can unfault my drives within 1 week.
For my drive, WUH721414ALE6L4, the 94xx card says “This specific drive model passed Broadcom interoperability testing. Any related drive models are qualified by association.”.
It seems that these drives may easily fault in less than 1 year if anything goes wrong with
cutting the power or losing the SATA connection, then they seemingly become unreadable
as if they were damaged.
I have the WUH721414ALE6L4, I have tried an LSI 3041 (B) and Microsemi/Microchip 2100 4i4e
but they just cannot detect it. SATA ports just say that it is locked, not frozen, so I have no way
to right click it and Mark Online, or Unconfigured Good, or maybe “JBOD Good”.
I think to even buy a WD Data60 server, Intel, Cisco, etc., over the years, I have not so much money.
New 16 bay NAS servers are usually not super expensive.
The firmware on my old disk is LEGL0002
If the disk is locked out it may be due to the logic board failing
I tried LSI 9300-8e, but could only disable read-only in diskpart,
online it in Disk Manager, but cannot remove the locked status,
with no password set and no security enabled.
I can see the model, 14TB capacity, but since it is locked from reading
sectors it shows as unallocated, unknown, not initialized. I will try with
MegaRAID 9440-8i with SuperMicro X11 for trying to mark as unconfigured good or JBOD.
I will try with Data102 enclosure much later.
Are these drives only functional with hardware RAID controllers?
Could someone just tell me how to make this datacenter-oriented drives accessible again
when they lock access to reads/writes? What tools, hardware, HBAs, servers, OSes?
Have you never found WD DC HC datacenter (or other brand) of SATA server drive
that disables access to media and you can reenable? Are these drives only manageable
fully with a hardware RAID tri-mode HBA or the like to mark them Good and readable/JBOD?
See this video to see what I’ve tried so far with LSI 9300-8e, which seems to have
NO SUPPORT for RAID commands:
A less important attempt from a normal EliteDesk 800 G1 SATA port:
Next i will try to regain access with MegaRAID 9440-8i tri-mode and Supermicro X11.
Do you have some tips on how to proceed or the right server motherboard/software tools?
Do NAS servers have tools to mark Unconfigured Good or otherwise regain access to
suddenly locked SATA media?
