Is a scratched HD still usable? Is it going to get worse?


Recommended Posts

So I have this brand new Seagate 1TB external USB hd.  About a week after it was purchased, someone dropped it (not me ;) ), and it became unreadable.  Windows said the partition was "raw", and partition table recovery software was not able to restore the partition.  After recovering as much data as possible with R-Studio (only got about 70% of the data back), I formatted it as NTFS (64kb allocation unit size, as it is a file storage drive and not a system drive) and ran a chkdsk /r on it, with the drive plugged directly into my mobo via SATA, instead of the USB enclosure.

 

 Chkdsk found:

  •  6144 KB in bad sectors.
  •  Adding 96 bad clusters to the Bad Clusters File.

 

Now I know chkdsk is looking for bad clusters, and not bad bits, is this going to make a difference?  Or will it just mean that slightly more of the drive will be unreadable (96 bits unreadable vs 6144kB in clusters unreadable)?

Should I do a hard format to write all bits to 0 to check for bad bits, or is that what chkdsk /R did?

 

My second question, because these are the result of an injury and not age, does that mean the drive should be relatively fine from now on?  Or can this scratch propagate and make things worse?  Because Windows 8 kept warning me that the drive was about to fail, although that was before I did the chkdsk, haven't had any warnings in the 5 mins since chkdsk finished.  Anyone here ever had experience using a physically damaged drive afterwards?  (yes I should just buy a new one, no I'm not going to)

 

I should also post the SMART warnings:

  • Raw Read Error Rate: Current 1, Worst 1, Threshold 51, Data 646544 [WARNING! # of hardware read errors: 646544]
  • Calibration Retry Count: Current 100, Worst 100, Threshold 0, Data 12 [# of calibration attempts after calibration failure: 12]
  • Current Pending Sector: Current 96, Worst 96, Threshold 0, Data 777 [# of unreadable sectors: 777]

 

Thanks for all your help!

Link to comment
Share on other sites

The simple answer: try it and find out. More than likely if it's already damaged a solid head-crash is in your near future, but as long as you're not relying on it, why not see what happens.

Link to comment
Share on other sites

Problem with a HDD with surface damage, is that the scratches will leave tiny little fragments inside. When these get between the heade and disk, they will cause more damage, remember the distance between the head and surface is microscopic. Even the scratch on the surface could cause he head to lose height and crash again causing more and more damage.

I once sent in a important disk for a client to one of those expensive disk recovery places. The answer we got back was there was no recoverable data due to damage cause by fragments and repeatedly trying to run the disk after the crash causing further damage until most of the surface was now unreadable.

Link to comment
Share on other sites

Problem with a HDD with surface damage, is that the scratches will leave tiny little fragments inside. When these get between the heade and disk, they will cause more damage, remember the distance between the head and surface is microscopic. Even the scratch on the surface could cause he head to lose height and crash again causing more and more damage.

I once sent in a important disk for a client to one of those expensive disk recovery places. The answer we got back was there was no recoverable data due to damage cause by fragments and repeatedly trying to run the disk after the crash causing further damage until most of the surface was now unreadable.

 

Good to know, I didn't think of that!  Maybe I shouldn't use it for any data that I want to keep... All I really want it for is a temporary scratch disk to copy to and from, so that I can combine two partitions on a different drive.

 

How do you know it is scratched? Did you open it up?  :s

I guess I don't.  But what else would it be, if someone dropped it and parts of it suddenly became unreadable?  I imagine if any other component was damaged (the head, the motor, the axle...) it would be 100% unreadable.

Link to comment
Share on other sites

Try spinrite which might better mark the bad areas. I personally dd if=/dev/zero or /dev/(u)random disks over and over and over, with reads (dd=disk of=/dev/null) in between to make sure the bad sectors are marked so, and then I label the disk in windows "USEWITHCAUTION" and use it only when I need scratch space that isn't critical.

Link to comment
Share on other sites

There is one thing you can say about HDD's: the rate of read errors always increases with age. This is also why they are doomed to get slower with age. So if things are screwed up now, they aren't going to get any better.

 

Edit: cybertimber2008's suggestion about spinrite is a pretty good one, you can at least get rid of more of the bad sectors with a better program

Link to comment
Share on other sites

You should immediately look to getting the drive replaced either under warranty (don't say anything and I won't!), or via another purchase.

 

Continuing to use the drive will cause it to degrade, or fail completely.

Link to comment
Share on other sites

Do a full format, not a quick format. Use Seatools and do a full test. If device still functions then use it as a non-critical storage device.

Link to comment
Share on other sites

You should immediately look to getting the drive replaced either under warranty (don't say anything and I won't!), or via another purchase.

 

Continuing to use the drive will cause it to degrade, or fail completely.

 

I second trying to get an RMA. Since it's still a basically brand new drive, try to submit and RMA request. Check this link: http://www.seagate.com/support/warranty-and-replacements/

Link to comment
Share on other sites

I don't see why everyone is telling you to bin it. I have a ton of old drives kicking around that are in various states of decay that can still be used. I have one hooked up to my Wii... it still runs, it's not critical data and it's all backed up, so I don't really care how many crunching/wheezing noises it makes.

Link to comment
Share on other sites

Drive upside down and a large knock during an intensive test via either SeaTools, Samsung disk tools or IBM HDD tools generally results in a SMART failure, drive head lock and a RMA/error code. I tend to use this for intermittent drives that should be replaced under warranty rather than something that was dropped. Do as the cricket once said with this information if it helps.

Link to comment
Share on other sites

i would only use the drive as a backup drive, but this is what i would do.

 

1 run check disk and scan for bad sectors (which usually needs to be selected)

 

there is a section of the drive that is reserved for bad sectors, once that is filled you will get 'SMART error, Replace HDD' on startups or HDD checkers. it also fragments the hdd and you dont know what files are fragmented because they wont be listed as fagmented due to the clusters being reallocated to the reserved section of the drive.

 

2, once that is complete, i would run something like this on the empty drive:

 

H2Testw

http://www.softpedia.com/get/System/System-Miscellaneous/H2testw.shtml

 

select english when you run it. this will fill the whole drive and then verify the data is correct, if it is the drive should be fine to use. 

 

Note, this will make recovery of any data impossible.

 

i've always used this for data recovery

http://www.piriform.com/recuva

 

seems to do better than most, do a deep scan.

Link to comment
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.