Barracuda XT, Barracuda, Barracuda Green, DiamondMax and other desktop SATA and IDE drives

[fzabkar]

AIUI, clicking and freezing is often symptomatic of weak read heads or bad sectors. It happens because the hard drive is retrying several times to read a problem sector.

You can distinguish these freezes from normal wake-from-standy pauses by running a benchmarking tool such as HD Tune. You will see large dips in the performance graph whenever these freezes occur.

You can sometimes coax a drive into remapping marginal sectors by performing a zero fill operation using SeaTools or DiscWizard. Then run the long test in SeaTools.

If the clicking persists, you can use Makebad to transform weak sectors into uncorrectable (UNC) ones. A subsequent format operation will mark all these UNC sectors as bad clusters. Please read the warnings before use.

http://files.hddguru.com/index.php?&direction=0&order=&directory=Software/Makebad

The author states that "from this moment, performing a CHKDSK X: will tell you that bad sector does exists, but you no longer will be bored with slowdown while looking at a video.... ;-)"

WARNING: If your drive's problems are due to a weak read head, a sequential read test may accelerate its failure.

[AlanM · Accepted solution]

Thanks!  I've added this to the Common Issues thread.

[Cantbecanit]

To add to whats been said, if it's clicking and or freezing, get it backed up asap, and rma it if it's valid.

[likestheaction]

What if it clicks on cold boot up only?  Always 11 times, no more.  No other clicking was ever heard before this.

-lta

[fzabkar]

On power-up, a drive attempts to bootstrap itself by loading its firmware from a reserved section on the platters. It also attempts to read its bad block map, and LBA translator module, both of which are also stored on the platters. Clicking would suggest that it is having difficulty bootstrapping itself, ie your problem is inside the HDA (head/disc assembly), not on the PCB.

[zamar51]

Is this bootstrapping issue also causing locks on wakeup-from-standby? My 7200.11 500Gb HD (fw SD04, no new FW files) clicks 10-12 times on cold boot and also on wake-up-from-standby, then appears to work OK. At times it can't start spinning at all, so I need to reboot it several times or wait for a few hours, despite its temp is low 38deg, and I've 750W PSU far exceeding the PC needs. When the spin is delayed on wakeup, some files and dirs get damaged, so the system reboots on its own and runs chkdisk to repair. And it gets worse over time.

What causes the drives shaft to lock at start? Why it progresses over time, and not from the beginning of use? If its low quality hydrodynamic bearings materials or their design flaw, the assumption is shaft clearances in bearings should increase over time, not get smaller? So, why the shaft locks?

[zamar51]

So, what exactly causes Seagate drives click multiple times at start up under normal voltage conditions? What's the root cause? If the cause is in damaged servo zone records, can the drive read servo marks at bootstrapping at all, until it actually starts spinning? What comes first: reading servo zone or start spinning? Any way for a user to fix "cold clicking" issue? Can servo zone be verified by SeaTools or any other similar software - which one?

[fzabkar]

I can't answer your questions. You may have better luck at the HDD Guru forums.

That said, could the spin-up problem be due to one of the supply voltages being below spec??? This would result in the MCU being held in the reset state by the voltage monitor in the SMOOTH motor controller.

Alternatively, could it be an intermittent stiction fault as a consequence of the heads coming to rest on the platters outside their landing zone???

Can you see large numbers of SpinUp Retries in the SMART report?

IMHO, the clicking on startup still suggests a problem in the reserved System Area, assuming it occurs before the BIOS tries to identify the drive.

[zamar51]

Here is SMART stats of the problematic drive. In my view, the only thing it shows is that Seagate firmware performed SMART routing is designed to conceal the drive failures until warranty expires, provided it spins at all. Despite the drive now attempts up to 20 spin up retries at each boot or wakeup from standby, spin-up times always = 0. In fact, Seagate can't rely on SMART Error in denying warranty, because its impossible to read the drive until it actually starts spinning, so SMART is useless for diagnostics of cold boot non-spinning drive, especially if the drive eventually starts spinning before SeaTools or another tool is ready to record its performance, so the clicking period will be missed and entirely unnoticed.

That clear message is thoroughly missing on Seagate website.  Why - because its a typical failure, and denying warranty due to SMART giving no error unduly saves millions, while artificially improving drive failure stats then converted into false marketing and advertising tool to promote and charge more for mediocre at best HD products. Theoretically, the only way for SeaTools to register temp non-spinning drive is to hook it to a power rail after PC is operational and SeaTool is ready to measure. Is it safe to connect power and SATA data cables to a drive after the PC is powered up? Why this method is not described on Seagate website? More interesting, will it allow in practice to register a drive startup stalling with SeaTools?

And yes, clicking occurs at the BIOS trying to ID the drive. And it can't, until clicking is over. With 750W PSU with 4 19A rails nothing points to PSU deficiency. I wonder, why Seagate never publicly disclosed possible causes of startup stalling? Concealing the problems exist is equal to doing nothing to improve design, production, and materials selection, thus preventing it from happen again. In fact, Seagate never explained to customers that SMART feature is not capable to read reserved System Zone on drive platters at all, hence useless as a diagnostic tool or warranty services ground in particular, when service marks or firmware written in drive Service Zone are damaged.

Another interesting issue would be, why problems in Service Zone occur in the first place? And why firmware can't ID and log bad service marks, and next boot start straight from the mark it started on previous boot, thus properly registering and subsequently preventing cold boot clicks? And if firmware can't register cold boot clicks until the drive actually spins, why they demand from customers to demonstrate SMART error in such cases to receive warranty service?

[zamar51]

Here's my tip on what to do if your drive is clicking and freezing at startup or wakeup from standby. Make sure first, your Power Supply is adequate by using an accurate web PSU calculator, test voltages on your power cable at startup load (they shouldn't drop below 11.5V), and try connecting your problematic drive to a different power cable (hooked to a different PSU power rail - look at PSU schematics found on the web).

If that doesn't help, and you experience regular or worsening clicking and freezing at startup, don't waste your time on SeaTools tests. The Tools are not designed to register such drive failures, and will show Pass grade if the drive eventually starts spinning. You can not repair such mechanical & electrical failures of the dying drive yourself at all. Quickly backup all data from the drive, transfer OS partitions to another drive and make them bootable with a popular Boot Manager or Data Backup Tool, and RMA the drive to Seagate despite passing SeaTools mark. Keep in mind, the replacement drive quite likely will also fail at some point, since its probably going to be exact same model, which already failed before and get refurbished. You can't get your money back anyway. So, the earlier in your warranty period you attempt to get a normally working drive, the better chances you have in the long run, as you may need to go through several RMA cycles to find one that will normally serve 3-5+ years as you expected when you bought it.