I've got a 3.5 drive that has one groove on the inside edge of the top platter. Is there any way to get the data? I've preformed a successful head stack swap with what I believe to be a matching donor. The heads quickly found their way to the original groove. I was hoping that their would be some way to seal, coat or treat that groove to keep the head above it. I'm new to data recovery. I'm sure many of you have already faced this situation and know what to do. I greatly appreciate any and all time and information you supply.

Although I never tried it, and I doubt it would work, I've been thinking of ways to fill in grooves. Perhaps some kind of varnish? Maybe a hard wax? Has anyone tried?

I like the wax idea. I haven't tried it myself. At least not yet. I've got plenty of drive to work with...practice makes perfect. One thing is for sure, if it can be done I will find a way. Most likely, someone has already succeeded here. I will be sure to post my results!!! Thanks for you wax idea. It's got merit. I'm not sure if that would gunk up the heads or not but it sure got the ideas rollin'.

The reason I thought of Wax was that you could heat it up, and HOPEFULLY it would settle in level. Heat also makes it easier to remove it. From doing research for homemade machineable wax for my new CNC machine, you can add Stearic acid to regular candle wax to harden it. Most craft stores also have wax hardener, as well as wax. I'm guessing a big failure point is centrifugal force pulling the wax out. And, you'd HAVE to keep the drive cool!

Forget it, guys - in theory, possible.

But in practice - you'd need machinery costing millions.

The principle of "filling in" gaps / holes / etc in a platter, working to measurements of microns (or less) throughout the scratch - not a home-brew job.

Wax wouldn't survive the centrifugal force - it would distort almost immediately, and foul up the heads.


Duncan

What about some type of machined disk? Maybe something that could just keep the head from moving to that inner region? My brother has 1/2 a million screw machines so I could get just about anything done. Thanks by the way for all the info.

Hi ,
Wax its impossible ..............

_________________
Regards
Amarbir S Dhillon , Chandigarh Data Recovery Labs [India]
Logical,Semi Physical And Physical Data Recovery
Website-> http://www.chandigarhdatarecovery.com

Odiferous mentioned machines costing millions - highly doubtful. However, there must be a method. I can imagine a machine that would be programmed to scan an undamaged section of platters. Using the original working PCB and incorporating any other necessary hardware should do the trick. I'm definitely up for experimenting. If there is someone out there with useful information then we would love to hear from you.
For all those who no nothing about the subject - which means that you have no first hand knowledge on the subject - do not waste your time or ours. I'm not interested in what you have read or what you think unless it is constructive. 10,000 tries to get the light bulb and I bet he had to deal with a whole ton of knuckleheads who knew he would never succeed! So thanks anyway! If you know - have experience - then let us know. If your ideas support this concept then great - otherwise keep it to yourself!
Someone has to have tried to recover data from damaged platters - and someone has to have succeeed! I want to hear from them!!!

There's a difference between things to say and job to be paid for...

Wrong - on both counts.

If there's a method - something along those lines would have been at least partially developed by now.

Signal Trace Technology. Google. Go for it.


Duncan

Hi,

Signal Trace Technology.

Thats what all the good recovery engineers dreaming from.
I think all guru thinking about this before, but too hard to made it.
The general question is: How to "trace the signal" without a standard MR sensor in the standard head?
The next question is: if the signal map is made done, how can make a new clone from the original plate.
For timing plates together, i have a (half) solution.

Regards,
Janos

Hi,

Now i have a little time, and found this:
http://www.actionfront.com/ts_whitepaper.aspx

No pictures, no end, but anyway there are a lot of useful info.

Janos

(Maybe this will be my last post in this forum in my life)

1) SIGNAL TRACE = decoding ?
2) If you know the structure of the data patterns and servo for the specific brand, you will be able to decode sector data.
3) After that, you can rebuild the SA/MC... (to get the error list / skip bad sectors mainly)
4) ... and then rebuild sector by sector.
5) and then perform a logical recovery on the image.

You only need to know how data is recorded.
Heads are all equal in structure, drives do differ in hardware and mechanism i.e. head switching and preamplifiers, and data patterns on the platters.
Head switching and preamp are controlled by SW.

A solution for reading data is not complicated to implement , we have a lot of powerful and cheap (for me, maybe not for you) solutions that can do 90% of the work, the rest can be accomplished with pure software. Of course it will need a lot of work.
About head positioning, it is NOT a problem. I have discovered modes to successfully drive the VCM, to drive the spindle motor and reach rotational speed and to READ the channel (was an experiment on low density drive, but the principle is the same). With some software overhead and the use of simple solutions could be possible to implement a closed loop to lock the servo and make rotational speed stable (anyway my 3 phase driver was sufficiently stable - I used a professional power supply as source and the 3 pahse driver circuit used a quartz crystal as reference just beacuse there was no closed loop between the reading channel/index and the motor driver as in real HDD).
I didn't decode the data but this was not the purpose, I only wanted to see the patterns on the scope / logic analyzer.

For some models I have pointed out the head switching mechanism, for others not.

I know the majority on this forum only want (and is an actual user of) commercial solutions for work with drives, but maybe some of us interested in SERIOUS R&D (that has serious side effects like frustration, headaches, painstaking experiments and so on )could join together to develop a non-standard solution that maybe can do what the other solutions can not. Anyway buying some equipment, click on buttons on a screen is easier than study, experiment and make your own system work so why worry ?

Try reading this rather technical book:

SPIN STAND MICROSCOPY OF HARD DISK DATA by Isaak Mayergoyz and Chun Tse

Spin stands at $150,000 (as a starting point) but then it gets real complicated...


Good luck!

sceggy

Just dropping by to say hello. Found the forums in my old favorites list.

For those who don't know me - I'm Nick Majors - the founder and former president of ActionFront Data Recovery, before I sold the company to Seagate - 2 1/2 years ago and retired.

I was the one who pushed for and invested heavily in Signal Trace Technology and published the white paper, linked to above.

These days, I am living the good life and pretty much out of the loop. Not even sure what Seagate has or has not done with our research.


All the best

Nick

hi Mr.Nick Majors，

your Actionfront data recovery lab is so great.. I never imagine that i could see your post on the froum!

Best Regards

Laura

_________________
SalvationDATA--Profesionnal Hard drive solution Provider. MSN: sales.laura@salvationdata.com

Wasn't he the "Fall Guy" with that sexy blonde?

_________________
All went well until I plugged the drive in.

On a serious note. Signal trace worked on an old WD Enterprise where (maybe!) the company had some good info. After that it's gone quiet because, in reality it's expensive and time consuming. Only government can afford to spend money and have the technical ability to do this...If they want...Which I guess they don't

Lets not worry about what could happen but work on what can happen

_________________
All went well until I plugged the drive in.

the only people who will use this signal trace technology

is the fbi and the usa goverment and allso the uk goverment
no one else.

goverment owned

Maybe not only fbi and u.s. gov. and developing a solution doesn't cost the world. Anyone want to try with me ?