arga

Samsung 1TB HD103UJ: HDDScan problem (476 green blocks)

Recommended Posts

I have bought 2 Samsung hard disks, both are the same model HD103UJ, 1 Terabyte, SATA.

I bought the first one in July, and the second one in August, in different shops.

I have scanned both units using HDDScan.

The first one has an acceptable HDDScan (only 9 slow blocks-green)

But the second one has a lot of slow blocks (476 green blocks).

Both units pass the Samsung diagnostics OK.

I would like to return the second one, but I wonder if the shop will be willing to change it for a new one.

I fear that they will argue that the disk is not defective because it passes the Samsung test ok.

I bought this disk just last week.

I would like your opinion.

Here are the scans:

hddscancompqn8.th.jpg

Share this post


Link to post
Share on other sites

Run the test again. I have never received same numbers in any invocation of HDD Scan.

I doubt you can find bad sectors with new drives any more from windows. They get remapped by the firmware when they are encountered. Appropriate counters are incremented in SMART. HD Sentinel is a useful tool for interpreting SMART values.

If you think some parts are about to fail, run a full format from disk management in windows and it will have the drive remap failing sectors.

Share this post


Link to post
Share on other sites

Hello:

these are not bad sectors, just slow sectors.

I have run HDDScan several times, and always got a lot in this drive.

Of course, the number varies each time, but always a lot.

And inspecting the log, I suspect the slow sectors are concentrated in one platter.

I think one of the three platters is "sub-standard".

I got basically the same result before and after formatting.

I did a long formatting using Windows XP.

To make things worse, the shop has not answered my mail in the last 24 hours.

They told me that if the Samsung diagnostics fail, then my drive will be exchanged without problems. But I told them that the drive passes it because this utility doesn't detect slow sectors.

Of course, this is a bit absurd: just imagine you get a drive with heavy external damage, but it passes Samsung diagnostics... according to this logic, the drive can't be returned???

Run the test again. I have never received same numbers in any invocation of HDD Scan.

I doubt you can find bad sectors with new drives any more from windows. They get remapped by the firmware when they are encountered. Appropriate counters are incremented in SMART. HD Sentinel is a useful tool for interpreting SMART values.

If you think some parts are about to fail, run a full format from disk management in windows and it will have the drive remap failing sectors.

Share this post


Link to post
Share on other sites

The drive is not failing yet. Its health has diminished but those blocks have not yet been unreadable.

I posted you the graphs so that you can see the difference. There were no bad sectors on any of these drives either. But a full format reallocated the soon-to-be-bad sectors.

Short cirtuit the drive. I doubt they will understand what happenned. Dirty, but sometimes you gotta do what you gotta do. Better than waiting for it to die on you.

Share this post


Link to post
Share on other sites

Well, in my case the formatting didn't reallocate these sectors.

I kept the results:

-Before formatting: 476 slow blocks-green,

-After full formatting: 482 slow blocks (green) and 1 slow block (orange)

I am going to wait, I hope I won't have to resort to any dirty trick to enforce my warranty. sigh.

Share this post


Link to post
Share on other sites
Try running SMART > Long Test then test. If no improvement, try running a Secure Full Erase if your drive supports it.

ok, I am going to do that.

In the meantime take a look at HD Tune results. Indeed the slow sectors show up in the transfer rate curve:

hdtunecomple0.th.jpg

Share this post


Link to post
Share on other sites

I have not yet run the long smart test.

I am scanning again, this time with the HDDScan 3.0

Got 717 slow blocks this time.

I don't know if this is growing or just a fluctuation.

I will test the good one with 3.0

Share this post


Link to post
Share on other sites

@ 666:

what is your experience with HDDScan on brand new disks?

how many "greens" do you usually have?

In my experience, at most, I get a dozen. Normally zero, one, two...

Edited by arga

Share this post


Link to post
Share on other sites

arga, this drive is faulty. i would not trust my data on it.

I never had >5ms blocks 0.1% of any of my new drives. Yours is 7%!

How are you testing this drive? Is it the boot drive? If not, can you move it to another controller? For example, if you have Jmicron or some other controller on the board other than intel, can you connect it to that one and run the test there?

If it is the boot drive, can you make sure the system is idle (no indexing, a/v, etc running at the back ground). Even if it is on a seperate controller or if it is a seperate drive, make sure nothing is doing any disk intensive operations on your boot drive. I have awful results whenever there is disk operations on my boot drive (even though the drive being tested is on a seperate port or controller).

Share this post


Link to post
Share on other sites

I will post you some screenshots later so that you can see what i mean.

Did you run HD Sentinel on this drive and examine smart values?

If nothing works and they do not accept the drive back, just short circuit it and let it die. Dirty, unethical, yes... but sometimes the only solution to get your right back since this drive is faulty and you paid for it.

Share this post


Link to post
Share on other sites

For example, if i am running HDDScan and i start Everest from my boot drive, i immediately get few orange sectors. Taht is probably Everest is querying the drive for SMART info. It is important those tests are done on a clean system with no software running.

I hate that i cannot edit anything after few seconds in this forum. At least make it 5 mins or something.

Share this post


Link to post
Share on other sites

Hi:

I am almost finishing the scans using HDDScan 3.0. Seems it counts the fast (gray) blocks better.

I will post the results.

The tested disks are absolutely clean and empty. And the OS is in a separate disk.

Also, I leave the computer idling. Just HDDScan running.

The controller is the integrated controller (Intel ICH7). In combined mode.

I have no other controller. Gigabyte 945GC mobo.

I have tested both Samsungs in the same conditions. Unit 1, and then, I replace Unit 1 with Unit 2.

Unit 1 results are ok, Unit 2 are horrible.

So there is no doubt the bad result is because of the disk itself.

I will download HD Sentinel and post the SMART results.

How do you short circuit it?

Edited by arga

Share this post


Link to post
Share on other sites

Where did they go? It was half a million, now they are 1700? Is the HDDScan or HDTune graph still erratic?

Run a seperate PSU, not the one connected to your mobo, reverse pins inside molex connector, and just start the PSU hd connected.

Wait a sec, i dont know how it is with SATA if your drive is SATA. You might need to cut the PSU cable for that.

Get the windows version of Sentinel. I think it is shareware.

here they are the results with HDDScan 3.0.

13 vs 707 green blocks

hddscan3compqd2.th.jpg

Edited by 6_6_6

Share this post


Link to post
Share on other sites

>>Where did they go? It was half a million, now they are 1700? Is the HDDScan or HDTune graph still erratic?

<<

I think the old version reports the "gray blocks" incorrectly (<50ms).

As you see, the (<15ms) was zero in the old version.

But the green blocks (<150ms) is quite consistant.

The shop has finally answered. They say that Samsung won't exchange a disk just because of slow sectors. They say that under 500ms is not a problem!!! Just imagine the case of a brand new disk full of 500ms sectors and I can't return it !!

I am going to insist diplomatically. If this doesn't work, I will use other methods.

Share this post


Link to post
Share on other sites

yes, the transfer rate curve is still dented-jagged in the bad disk.

This is something totally consistant. It happens 100% of the times.

Edited by arga

Share this post


Link to post
Share on other sites

I have run SMART selftest-extended (using HDDScan 3.0). More than 3 hours.

I am scanning now. So far, I see no improvement.

Edited by arga

Share this post


Link to post
Share on other sites

I am using CMRR secure eraser.

I didn't choose the enhanced mode, just the normal one (1st option).

And chose not to erase the special diagnostic areas of the disk.

Now it will take 3 hours, probably.

Still, I don't know... if full format didn't work, I guess this won't work either.

Share this post


Link to post
Share on other sites

arga, there is nothing more you can do for this drive. Have a rest.

Did you consider using it for offline storage/backups? I mean, fill it up with stuff, do md5 checksums of the files, and if no errors, just take the drive offline with your data. Its health would not detoriate if it will not be used.

Share this post


Link to post
Share on other sites

hey dude, I think I have destroyed the disk.

After CMRR secure eraser, all the sectors are Bad in HDDScan, and Samsung utility reports a size close to zero.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now