• Please review our updated Terms and Rules here

Possible to recover from this kind of hard drive error?

Trixter

Trixter

Veteran Member
Joined
Aug 31, 2006
Messages
7,478
Location
Chicagoland, Illinois, USA
In attempting to resurrect my beloved 6300, the included Miniscribe 3650 42MB MFM drive doesn't want to function any more. (It's not stiction because the platter spins up and I can hear some seeking.) It spins up, I can hear some head seeks, and then there is a soft "clunk", after which the front LED flashes out an error code. According to http://stason.org/TULARC/pc/hard-drives-hdd/miniscribe/M3650-42MB-5-25-HH-MFM-ST506.html the error code is "Seek errror during burn-in or recal".

I have both hardcards and a single XT-IDE I can use in it, so it's not like I'm out of luck -- I just want to see what's on the drive before I throw it away. Is there any hope of recovering this data without a recovery service?
 
Jim, have you looked at the track 0 sensor? Is the positioner gunked up? On a drive like this (that can be low-level formatted), the procedure for self-test is to step to track 0, issue a seek to some cylinder, then step out the same number of pulses to see if the track 0 sensor registers again. If things get a bit fouled up (dry grease, oil, whatever) or the Track0 sensor gets obscured, then you'll get the error you describe.

Sometimes those error codes are unreliable and it's a completely different problem that's being seen.

Also, double-check your +5 and +12 supply voltages while the drive is operating. If the +12 sags too far while the drive is spinning up, you get problems.
 
Often this error occurs if the low level format is corrupted too... the answer being as simple as to try a low level format of the drive.

Though that would delete any data... You might want to try running spinrite on it.
 
deathshadow said:
Often this error occurs if the low level format is corrupted too... the answer being as simple as to try a low level format of the drive.

Though that would delete any data... You might want to try running spinrite on it.
Click to expand...

If the drive is spun down with a blinking LED, you're not going to be access it. It will be unresponsive until the power is cycled again.
 
The key to many errors involving magnetic media is to try again .. over and over again.

One thing you can try to do is to leave the machine turned on and let the interior get to a normal operating temperature. The drive generates a bit of heat when it is running; even when it is not running it will still get warm in there. After 30 or 40 minutes power off for a few seconds and then power back on, and try it again. If it really is a thermal recalibration issue that should help the drive by getting everything warmed up and realigned. And of course you want to make a few attempts under these conditions.


Mike
 
Chuck: I opened a hardcard to fix some stiction and it worked -- and then the drive never worked properly again. (I was ready and got my data off while it was working.) If I pop the cover on this drive to check the track 0 sensor, will that have the same behavior -- drive might work a few times but then be forever borked?

Mike: Good point, and I did not let the drive get to it's nominal operating temperature. The most it was spinning was 10 minutes. I will let it spin for an hour, then pulse it's power to get it to run the self-test. Hopefully I'll get lucky.
 
Chuck(G) said:
Jim, I think Mike has an excellent suggestion--all you need is for it to work once.
Click to expand...
Yup; just be ready for it in case it does! Nothing more annoying than to have it come to life unexpectedly with no place to copy it to; Murphy says that after you do set up a backup drive you'll never be able to read the original again.

As a very last-ditch effort you can try cooling it in the fridge/freezer; let it warm up a little bit to get rid of condensation before you try it. It's worked for me a couple of times.
 
Around 1986 I did some tech support for a company that sold a product which included a Miniscribe 20MB hard drive (3.5 inch?). There were a lot of flashing LED failures in the field that were due to the track zero interrupter getting out of place. Due to the way the drive was installed it might have gotten warmer than its design specifications allowed and that might have been related to the track zero interrupter working its way out of place.

If the failure was due to the track zero interrupter being out of position we could usually get a local tech to fix it by instructions over the phone. It involved loosening the lock screw on the track zero interrupter, then powering up the drive with a diagnostic mode jumper installed, then you would manually rotate the track zero interruper into the dectector until the LED just turned on, then manually rotate it back out until the LED just turned back off. Then tighten down the lock screw and Loctite it in place. At least that is how I remember it 26 years ago.

-Glen
 
mbbrutman said:
One thing you can try to do is to leave the machine turned on and let the interior get to a normal operating temperature. The drive generates a bit of heat when it is running; even when it is not running it will still get warm in there. After 30 or 40 minutes power off for a few seconds and then power back on, and try it again. If it really is a thermal recalibration issue that should help the drive by getting everything warmed up and realigned. And of course you want to make a few attempts under these conditions.
Click to expand...

GIVE THAT MAN A CUPIE DOLL!

I powered it on and left it spinning, blinking its error message, for 90 minutes. I then powered off, waited 1 second, and powered back up. It booted right up. (Granted, on the second boot it couldn't read command.com, so something is definitely off, but I plan to low-level format it using the printed defect list as soon as I'm done getting what I want off of it.)

Why did I care about getting this drive working? Some of the stuff I've copied off of it tonight:

  • The very first version of CopyIIPC
  • The very first version of SpinRite (NOT because I want to use it, but because it's historically significant as the birth of hard drive snake oil)
  • The very first version of XTree
  • Some AT&T 1980's-era internal software (editors, comm programs, nothing amazing, just interesting)
 
You must log in or register to reply here.
Back
Top