All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 20 posts ] 
Author Message
 Post subject: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 19th, 2010, 10:12 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Ok, here's the continuing story of my attempts of recovery on a Western Digital WD1600 with a seized spindle motor.
I'm happy to report so far, so good with my rehearsals on a working WD1600 test drive.

My environment seems to be clean enough, as I have had the cover off the test drive on a number of occasions now and have not caused any slow blocks over and above what was initially already there.
I have also worked my way up to being able to fully unload and reload the hsa using the v-card technique and only cause 17 bad blocks.
I rehearsed and refined my procedure several times on a dead WD1600 before trying it on the test drive.

There are a few reasons I can think of why the bad blocks would have occurred.
One is the action of sliding the card between the platters and the hsa arms may have disturbed the platter suface on a tiny spot.
Another is that the small jiggling of the heads on one spot of the platters whilst inserting the cards, or the initial moving of the heads back to the park position after removing the cards.

In any event, the loss of only 17 blocks in a tight cluster near the start of the drive with the first of them at LBA 5795840 and the last at 5816576, is of no consequence to me and if this is the worst that happens I'll be ecstatic!!

But....my biggest hurdle still awaits! Can I do all that and have a method of clamping the platters together, remove them, refit them and still have a mountable drive.....stay tuned! :D


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 19th, 2010, 15:54 
Offline

Joined: January 8th, 2008, 5:21
Posts: 808
Location: uk
It's a good start!

Can you see any recoverable data on the 'test' drive?


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 19th, 2010, 20:25 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Hi Dick,
The test drive had been wiped before I got it. The first thing I did was to run a surface scan to have a baseline from which to compare subsequent scans and to drag n drop about 1gb of files and folders to it and to read them back to the desktop noting how fast it did it.
All those files and folders can still be read back to the desktop just as quickly as when I first dropped them on the drive.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 20th, 2010, 15:35 
Offline
User avatar

Joined: July 16th, 2008, 17:52
Posts: 486
Location: The Lab.
I am guessing that your bad blocks are most likely caused by contact with the platter. There is probably some sort of damage that is not perceivable by the human eye; you can have some fun with a good microscope, and test how much damage different materials actually do on the microscopic level.


Regards,


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 20th, 2010, 21:50 
Offline

Joined: January 20th, 2010, 21:49
Posts: 1
Location: location
would be nice if pictures are included... :)


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 21st, 2010, 11:38 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Russwinters wrote:
I am guessing that your bad blocks are most likely caused by contact with the platter.

I'm sure that is the case Russ, though some blocks have recovered since so it might just be a spot of contamination.

I've made up a device to clamp the platters, have tried it on a dead WD1600 [not the test drive] and it seems to work well. Even the spacer ring between the platters was held firm.
I have however, had my worst nightmare realized, eccentricity and concentricity of the platters.
I put a dial indicator on the edge of the test drive platters and neither condition exits. Both platters are off center to the spindle motor center and off center to each other.

The only small saving grace is that they seem to be off center to the spindle motor by the maximum amount possible and exactly opposite to each other - i.e. if the upper platter's high point is at 0 deg. , then the lower platter's high point is at 180 deg.
Replicating this is going to be tricky, but not impossible. I can only hope there is some tiny measure of guard band between cylinders and the drive will cope with a small amount of mistracking.

abubin wrote:
would be nice if pictures are included... :)

It shall be done :D


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: January 23rd, 2010, 15:23 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Ok, test drive No 1 is a bin job. Because of eccentricity problems, at least with these drives, imo there's no way you'll ever swap platters and still have a working drive no matter how you clamp them.

The only hope is to leave the platters locked to the rotor, press the rotor off the stator and change the bearing. So far, I have found that pressing the bearing out of these motors is a piece of cake.
Doing it with the platters still sitting on the rotor and not causing them any harm - that will be a mission!


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: February 15th, 2010, 5:40 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Hi all, just an update for those watching this thread. I'm still working on/rehearsing procedures and machining up anvils and dies to do the bearing swap.
So far it's at least looking promising and I should be finalizing the procedure this weekend before trying it out on test drive No 2.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 17th, 2010, 13:42 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Success!! After months of tool making and rehearsals, on my first attempt I've succeeded in pressing the bearing out of a working test drive, press another bearing back in and still have a working drive :D
I did loose 5 blocks, 2 at 28% and 3 at 38%, but this level of success is waaay in excess of my expectations.

There was a touch of irony along the way. I initially intended to just press the bearing down in the base a bit, then press it back into position then test it, only to discover that the bearing in this drive was on it's last legs anyway!

The bearing felt ok initially after pressing it back home, but when I reassembled the drive and powered it up, it spun up ok, the heads did their initialization routine, then the motor suddenly stopped.
I took the cover off and discovered that the spindle was rather stiff to turn by hand. The load on the dodgy bearing from pressing it back home was the final straw for it!
So I was then forced to pull it apart again and press another bearing into it.

Also, because I couldn't easily turn the motor by hand anymore, I was forced to move the heads to the outer edge by rotating the hsa only.
This may have been what caused the lost blocks, and is something I'm going to have to do on the data drive as well, so was handy to know that whilst not desirable, it's not total death moving the heads in this way.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 17th, 2010, 13:47 
Offline
User avatar

Joined: August 15th, 2006, 3:01
Posts: 2376
Location: CDRLabs @ Chandigarh [ India ]
Pixels wrote:
Success!! After months of tool making and rehearsals, on my first attempt I've succeeded in pressing the bearing out of a working test drive, press another bearing back in and still have a working drive :D
I did loose 5 blocks, 2 at 28% and 3 at 38%, but this level of success is waaay in excess of my expectations.

There was a touch of irony along the way. I initially intended to just press the bearing down in the base a bit, then press it back into position then test it, only to discover that the bearing in this drive was on it's last legs anyway!

The bearing felt ok initially after pressing it back home, but when I reassembled the drive and powered it up, it spun up ok, the heads did their initialization routine, then the motor suddenly stopped.
I took the cover off and discovered that the spindle was rather stiff to turn by hand. The load on the dodgy bearing from pressing it back home was the final straw for it!
So I was then forced to pull it apart again and press another bearing into it.

Also, because I couldn't easily turn the motor by hand anymore, I was forced to move the heads to the outer edge by rotating the hsa only.
This may have been what caused the lost blocks, and is something I'm going to have to do on the data drive as well, so was handy to know that whilst not desirable, it's not total death moving the heads in this way.



Wooo ,
A Picture Is all What We Poor Mortals Are Demanding ...

_________________
Regards
Amarbir S Dhillon
CDRLabs Data Recovery - [ Chandigarh ,India ]
My Data Recovery Experiments - > https://www.facebook.com/chandigarhdatarecovery


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 18th, 2010, 0:46 
Offline

Joined: March 22nd, 2009, 0:19
Posts: 272
Location: behind the platter
Good Job!!! :D


Rare Case of Successful DIY!!!


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 18th, 2010, 4:07 
Offline

Joined: October 3rd, 2005, 0:40
Posts: 2579
Location: Hungary
yeah, hats down...

pepe

_________________
Adatmentés - Data recovery
No bitcoin donations :)


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 18th, 2010, 10:02 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Thanks crew! It's been a labor of love so far. Just wanted to see if I was smarter than a western digital hard drive :lol:
So far I seem to have their measure!!
I'm starting to gather some pics of the tools, will post up soon.
But first I have to translate this success to the drive with the data on it. If I succeed with that drive, well....I'm off to the moon for a while!! :D


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 18th, 2010, 10:03 
Offline

Joined: October 28th, 2009, 14:35
Posts: 778
Location: Toronto
Congrats m8. Good job =) U lucky ur heads didnt die when u took em off while being unable to turn the platter =) They have a tendency to fold into themselves when smbd tries to do that =)


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 18th, 2010, 10:15 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Alexii wrote:
Congrats m8. Good job =) U lucky ur heads didnt die when u took em off while being unable to turn the platter =) They have a tendency to fold into themselves when smbd tries to do that =)

Thats for sure!! They let out a soft squeal as I rotated the stack. [I grimaced!]
I'm figuring that for these situations where the platters can't be rotated, it's best to rotate the stack with reasonably quick velocity to avoid sticking and to allow some float on the lubricant coating?


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 19th, 2010, 8:10 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Hmmm....I've been presented with another hurdle to jump :(
I'd noticed some time ago that there's two different bearing - housing axial depth positions.
With some wd1600jb's, the bottom face of the bearing is near flush with the housing face and others are recessed by about 1mm.
Those that are recessed, press out fairly easily. Those that are flush seem to have been pressed in with some bonding agent and are very tough to get to move, though once cracked, it's then easy to move.

Some time ago, I'd pressed apart a dead drive with a flush bearing and noted the heavy effort needed to initially move it.
After the above-mentioned successful bearing transplant which was a recessed type, I then tackled another flush bearing type.
However, such was the strength of the bonding on this drive, that all I succeeded in doing was stripping the 3 motor mounting screws clean out of the frame!! :shock:
Needless to say, that was an instant bin job, as your then left with no way of getting the bearing out without damaging the platters.
And you guessed it, the drive with the data is a flush bearing type Image

So, it's back to more r 'n' d for me.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: March 19th, 2010, 14:31 
Offline

Joined: October 28th, 2009, 14:35
Posts: 778
Location: Toronto
the hits just keep on coming m8 =) I wish u best of luck.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: April 12th, 2010, 21:29 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Woo-Hoo!! Break out the champagne. After months of blood, sweat and tears, I've got nearly all my files back off the drive. Those files that crashed are totally unimportant, so I'm one VERY happy camper!! :D
Will write up some more about it later, have to go to work now.


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: April 13th, 2010, 12:04 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Ok, here's some pics of the tools I made + purchased mandrel press.
I cut the V cards from light card then bent them around a wood ruler which formed a square spine of just the right height to suit the thickness of the platters.
Attachment:
File comment: Mandrel press, machined tools and V cards
Tools and V cards.jpg
Tools and V cards.jpg [ 136.57 KiB | Viewed 6380 times ]


Here are some close-up's of the tools.
The clear round cylinder supports the drive base and provides a space for the rotor assembly to move into as the bearing is pressed out of the base.
The black plastic parts are a firm sliding fit on each other. The outer one slides down the inner one, supporting the rotor as it's pressed down.
The small aluminium bearing press pin slides through a hole in the larger part, which acts to center and guide the pin as the bearing and rotor are pressed out of the base.
The larger part also has a notch in it's rim for clearance to the motor connector.[see pic further down]
Attachment:
File comment: Tools to press bearing and rotor from base
Bearing from base and rotor capture press.jpg
Bearing from base and rotor capture press.jpg [ 103.56 KiB | Viewed 6380 times ]


These tools are to press the bearing out of the rotor.
The outer black plastic part from the previous step is used again to support the rotor while another pin and guide is used to press the bearing out.
Attachment:
File comment: Tools to press bearing from rotor
bearing from rotor press.jpg
bearing from rotor press.jpg [ 115.21 KiB | Viewed 6380 times ]


These tools are used to press the replacement bearing back into the rotor.
The bearing is first sat into the rotor by hand. Then the assembly sits in the larger support which has a 0.1mm raised center [see pic further down] to limit the bearing movement to the right height in the rotor.
The spring loaded press tool has a center hole which is a close fit around the bearing body. The center pressing pin has several mm clearance to the bearing, so initial movement will only compress the spring which holds the rotor and bearing in close vertical alignment.
Vertical run-out of the platters does not have to be perfect, but it must be very close to perfect.
Attachment:
File comment: Tools to press replacement bearing to rotor
bearing to rotor press.jpg
bearing to rotor press.jpg [ 118.27 KiB | Viewed 6380 times ]


And finally, these tools are used to press the bearing and rotor back into the base.
The drive base sits on the guide with either a pin in the hole or no pin, depending on bearing type.
For those wd1600 drives with a bearing that is shorter and recessed back in the base a bit, I used the guide with a pin sitting in it of just the right length, to limit the bearing to the right height in the base. For longer flush bearings, I used just the guide only. The hole in the guide is a bit smaller than the bearing body.
The pressing tool is the same one with the 0.1mm raised center from the previous step.
It sits on top of the rotor and presses the rotor with bearing into the base.
It's raised center also presses onto the bearing pin, ensuring the rotor does not move any further down the bearing pin.
Attachment:
File comment: Tools to press rotor and bearing assembly into the base.
bearing to base press.jpg
bearing to base press.jpg [ 123.04 KiB | Viewed 6380 times ]


Whilst the system worked, it is a bit "agricultural" and clumsy in some areas. There's a lot of refinement that could be done.
For any 'novices' reading this, bear in mind that while mechanical concepts are carried across most, if not all drive make/models, the exact requirements of individual drives will vary.
The WD1600JB's are probably one of the easiest drives to work on. If you try this yourself, be prepared to shed a lot of blood, sweat and tears, late nights and trashed test drives - and that will be just to get the head stack out and back in without screwing it up!!
My body count is 3 trashed drives and another 3 which whilst they still work, are virtually bin jobs due to their compromised performance + countless brain cells :lol:

There's more I'll write up on, but I need sleep right now.....


Top
 Profile  
 
 Post subject: Re: Recovery of jammed WD1600 Caviar SE - 2 platters, 4 head hsa
PostPosted: April 17th, 2010, 14:04 
Offline

Joined: January 6th, 2010, 7:38
Posts: 72
Location: Australia
Here's the last rant to finish this thread! :D
For those WD1600JB's with the flush and tough to move bearings, I decided that heating the base around the bearing was going to be my only chance.
The theory being that because aluminium expands faster than steel, this will reduce the tension on the bearing.

After finding a suitable candidate, I first removed the hsa, then screwed the cover back on, then used a hairdryer to heat the base around the bearing to about 80c.
Then quickly into the press [with the cover still on at this stage] and noted that the bearing moved with considerably less pressure.

This in itself wasn't absolute proof, as maybe the bearing in this test drive was one that was going to move more easily anyway, but it looked promising. Also, I was looking to see if there was going to be any adverse side effects from the heating.
After fully removing the bearing and reassembling the whole drive, it still worked fine with only a few lost blocks.

This success was somewhat unexpected, as I did get into a complete nightmare with this drive!
It was a WD1600JB, but with a shiny bare metal cover rather than the black cover and had a slightly different bearing in it.
It was slightly smaller in diameter, a bit longer and had a brass/bronze appearance.

The smaller diameter meant I had to improvise with the tools a bit, but my nightmare began when I discovered that after pressing the bearing out, it wouldn't rotate any more.
This was because the disc that's swaged into to bottom of the bearing case had moved inwards, jamming the bearing.
I overcame this by putting the bearing in the press and modest pressure on the bearing shaft moved the disc back again - but worse was to come!

Pressing the bearing back into the rotor was ok, but when it came to pressing the assembly back into the base, I neglected to consider the consequences of reverse pressure on the disc, which ejected it completely out of the bearing case along with some of the lubricant!! Image Tapped the disc back into the bearing, pressed the whole thing apart again, discovered that I had the disc in upside down...'stuff it, no time to worry about that', use a small socket to press the bearing into the base first, press the rotor onto the bearing - too much vertical runout!! press it all apart again.......exasperation is now setting in!!!
It took another two attempts to get the vertical runout acceptable.
Hence my amazement that after all this, the drive still worked at all!

It's because of vertical runout issues that I prefer to press the bearing into the rotor first, if it's not good enough it's one easy step backwards to press it out and try again.
Fortunately, the discs in the steel bearing black tops are swaged in a lot tighter than the brassy silver tops and recovery of my drive was a piece of cake - relatively speaking!! :D


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 20 posts ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: Exabot [Bot], Majestic-12 [Bot], maximus, Xeox and 39 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group