I have a 2TB Hitachi 5K3000 that has an annoying issue of not always spinning up. No mechanical clicking or anything, it just doesn't spin up. Tried with both 3.3V and 5V supplies, this doesn't seem to make any difference. When the drive spins up, it works just fine for as long as it's not spun down. It's a rather fresh drive (<1000h on) but was cast aside due to this issue.
Anyone encountered this before, and is there a remedy?

AIUI, your drive requires +5V and +12V supplies.

Does the drive sound like it is trying to spin?

_________________
A backup a day keeps DR away.

Apologies for the confusion, I meant that I've tried it with a SATA power plug that incorporates the orange wire, as well as with a 'molex to sata power' which lacks this orange wire (so as to force it to use the 5V as opposed to 3.3V).


Yes. I'm not sure that what I'm hearing is supposed to spin at 5400rpm, but when the drive turns on correctly it takes about 3 seconds after plugging in the SATA power before it reaches peak velocity, so to speak. When it doesn't properly turn on, I would say it cuts out at ~2500rpm within 2 seconds. It does not attempt to spin up again, until I replug the power (or reboot).

@Spildit: It's a 2TB HDS5C3020ALA632, not in enclosure.

I'm not interested in the data. The power is not the issue, I've tried it in a multitude of different setups.
I don't have an identical drive handy, much less one that I'd be willing to use for spare parts. If this is likely a motor issue then I will probably discard the drive, but if this can be tracked down to a bad transistor or possibly an off-the-shelf IC I'd like to replace that.

If the +12V supply is sagging, then the fans should slow down audibly.

Alternatively, if there is a problem with the PCB, then there may be a way to fudge a "solution", provided that the motor controller uses external current sense resistors. Could we see a photo of the component side?

_________________
A backup a day keeps DR away.

I'll open her up.
The 12V is absolutely fine. The same issue exhibited on several top-brand psu's.

Here's a CCD scan: http://imgur.com/a/9tkvz

After some better listening, it seems the drive spins down as soon as it starts to seek.

It is not the same model drive (mine is 7K300 series, model HDS723020BLA642), and not quite the same issue (mine spins up, but will just click several times and spin down).

But it is a somewhat similar issue. The drive acts like it is not getting full power on startup, and the issue cannot be corrected without power cycle. And the power supply has been completely ruled out. It may or may not be related somehow...

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

I couldn't find an external current sense resistor for the spindle motor, so I expect that sensing is done internally. There is an 0.47 ohm sense resistor, but it appears to be for the VCM circuit. The other sense resistor (0.1 ohm) is in the Vcore circuit.

It may be worth checking Vboost. This voltage is used by the high-side MOSFETS in the VCM H-bridge. These MOSFETs are involved in head positioning.

Just FYI, the following device is used in a particular Seagate SCSI drive. It enables the MOSFET to monitor its own current, without the use of a resistor.

NIMD6302R2, ON Semiconductor, Dual N-Channel Self-protected FETs with 1:200 Current Mirror FET, 30V, 5A, SOIC-8, marking N6302:
http://www.mouser.com/catalog/specsheet ... 02r2-d.pdf



SSM3J314T, Toshiba, P-ch MOSFET, -30V, -3.5A, 54mohm RDSon, marking KDU:
https://toshiba.semicon-storage.com/inf ... ?did=10375

SSM3K14T, Toshiba, N-ch MOSFET, 30V, 4A, 39mohm RDSon, marking KDK:
http://www.maoye-smd.com/canshu/pdf/201 ... M3K14T.pdf
http://datasheet.octopart.com/SSM3K14T(TE85L,F)-Toshiba-datasheet-13716475.pdf

MX25U4035, Macronix, Serial Multi I/O (MXSMIO) Flash, 4Mbit, 1.8V:
http://www1.futureelectronics.com/doc/M ... MI-25G.pdf

K4T56163QN-HCE7, Samsung, DDR2-800 5-5-5 SDRAM, 1.8V, 256Mbit, 84FBGA:
http://www.mt-system.ru/sites/default/f ... rev103.pdf

_________________
A backup a day keeps DR away.

"5K3000" is not a model designator. Instead it refers to a "series".

Hard Disk Drive Specification, Hitachi Ultrastar 5K3000, Hitachi Deskstar 5K3000, 3.5" SATA HDD, Models: HUA5C3030ALA640 HUA5C3020ALA640 HDS5C3030ALA630:
https://www.hgst.com/sites/default/file ... Rev1.1.pdf

Ultrastar 5K3000 Datasheet:
https://www.hgst.com/sites/default/file ... 000_ds.pdf

_________________
A backup a day keeps DR away.

I have a couple quick questions about swapping the PCB. I do not need detailed answers, simple answers will do.

1) Is is possible to read/write the ROM/NVRam with software tools? Or does it require physical swap?

2) If #1 is possible with software, are there any free tools that can do this?

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone

That is good information to me for the future. In my case I actually have a near perfect match that has a bad head. According the the label they are only one month apart. The drive with the weird issue is JAN 2011, and the bad head is FEB 2011.

But unless there is a free tool that will read and write the NVRam, I am not willing to do any testing. I am actually still using my drive, and money is an issue. I can't afford to kill it. And I am not willing to do any testing with VSCs either, already burned myself out on WD VSC testing. Plus I do not want to hijack this thread. If it would be an easy thing to do then I would have been willing to do the swap just to see if it helped, and reported results. But I think I cannot help more with this thread.

_________________
http://www.hddsuperclone.com
Home of HDDSuperClone