• Please review our updated Terms and Rules here

Seagate ST-251 MFM HD problem

jwd

New Member
Joined
Jan 11, 2023
Messages
3
I have found my old ST251-1 HD from my parents loft and even though I should have all the data off it when i moved to another computer I would still like to check a do a proper copy of everything from it if i can.

Problem is when power is applied (nothing else connected) it spins up fine but doesn't do the head seek/calibration bit and so is stuck in its parked area.

I have moved the heads out manually and powered down, the stepper motor correctly auto parks the heads.

I have replaced the 2 power capacitors on the main control board as mentioned here :

https://www.lo-tech.co.uk/wiki/Seagate_MFM_Hard_Drive_Repair_Notes

to no avail, the logic would suggest get another ST251-1 off fleabay and swap the control board, as with anything retro now the prices are ridiculous, so would anyone have any ideas why the drive won't calibrate/seek after spinup and/or things to try?

This is exactly what it sounds like (link starts at 1:09) :


Spins then a small strain sound then nowt (except spin). Drive therefore doesn't initialise.

Any ideas what could be wrong?

many thanks
 
Once you've taken the drive apart, they are worthless. They likely have issues with the control boards. Beyond that, it is just a guess. You've trashed the drives by opening them up and exposed them to the environment.
Dwight
 
Personally, I've had bad luck with that generation of half height Seagate drive. Of a hand full of drives based on the same mechanism (a mix of MFM, RLL, and an SCSI), the best one works, but has lots of bad sectors. Several of them have at least one completely dead head, and at least one doesn't work at all. Swapping around controller boards confirmed that the mechanisms are at fault.

I've had much better luck with their full height drives that have the voice coil actuator.
 
I have one of the top comments on that video. Dead stepper or driver, most likely. Either that or the microcontroller is not booting - that said, in the video it was regulating the spindle speed so it must have been booting.

The Seagate ST-225 and ST-251 are some of THE most reliable stepper drives out there, the SCSI ones suffer from PCB issues unrelated to the mechanism. It sounds like you've got a bad batch. I hope you were LLFing each time you swapped boards, though.

For the record, drives of this age are not hugely affected by disassembly. My own ST-251-MLC0 had it's lid removed for replacement of the rotary band 5 or 6 years ago and still works as of December 2022. As long as you're not dumb and dump sand or let your cats play with them, they are fine. The tolerances are rather wide. The gaskets on these drives in my experience however are untrustworthy and you should replace them if you open the drive.

All this being said, you will almost certainly NEVER be able to recover any useful data from a drive like this. They are subject to rather extreme margin of error over the years from thermal expansion, any track that was present 20 years ago or even just 10 is probably in a physically different spot. Not to mention, the data might as well be inaccessible without owning the exact original controller associated with this drive. Even interchange between the exact same model controller is poor, in many cases there is analog adjustment involved which can be different for each unit.
You may see success with the MFM emulator from pdp8 com but you will be unlikely to read anything worthwhile. These drives were popular during a time when things like family photos and personal diaries were not exactly easy to do on home computers. If you find anything, it will probably be a DOS install with some home office software and associated documents.

Good luck!
 
For the record, drives of this age are not hugely affected by disassembly. My own ST-251-MLC0 had it's lid removed for replacement of the rotary band 5 or 6 years ago and still works as of December 2022. As long as you're not dumb and dump sand or let your cats play with them, they are fine.
I see old diagrams like [this]. The manufacturers had a business reason (i.e. monetary reason) to set up and use 'clean rooms'. So there is risk; how much I don't know. There is a lot of dust floating about in the air.

All this being said, you will almost certainly NEVER be able to recover any useful data from a drive like this. They are subject to rather extreme margin of error over the years from thermal expansion, any track that was present 20 years ago or even just 10 is probably in a physically different spot.
I am trying to work out how that occurs (I am curious and I like to learn). The platters are metal, and the heads are at the end of metal arms. I have a crude understanding of thermal expansion in metal; it expands as it heats up, but then it contracts when it cools down. Where is the long term thermal expansion occurring?
 
These drives are very tough, way more so than new drives.

There’s obviously *some* risk of dust, but for the most part if you are careful and don’t leave the top off for too long, it will be fine.

Take a look at the spindle replacement I recently performed on a Miniscribe 3650 for example.


Drive worked fine afterwards!
 
I've had this era of Seagate open when the environment wasn't a clean room and it wasn't immediate death.

However depending on how clean your workspace was, what you handled inside the drive and how well you attempted to keep the drive clean when closing it back up plays a major factor. After that point if a full head seek and verify (head turbulence will blow anything off the surfaces) doesn't crash it, whatever contamination you missed will be caught in the filter and it will be fine.
Otherwise there is not a lot of reasons you ever need to open these drives. The heads, stepper motor and platters are non-serviceable. Later products from Quantum and the likes however you do have to go into to deal with sticky bumpers.
 
I see old diagrams like [this]. The manufacturers had a business reason (i.e. monetary reason) to set up and use 'clean rooms'. So there is risk; how much I don't know. There is a lot of dust floating about in the air.
Of course, there is some risk. If you are reasonably conscious about yourself, it will not affect drives of this age. Don't go smudging greasy fingers on platters or coughing into it or comb your hair or whatever right over the drive, but nine times out of ten not letting your cat play with it is good enough to keep these drives sane. For the most part even then, they will often send particles through their own filter before causing damage.

It is worth mentioning that the ST-251 is a tub chassis design, do not run the drive with the cover off, as airflow will not be directed under the heads and this will cause damage in rather short order.
I am trying to work out how that occurs (I am curious and I like to learn). The platters are metal, and the heads are at the end of metal arms. I have a crude understanding of thermal expansion in metal; it expands as it heats up, but then it contracts when it cools down. Where is the long term thermal expansion occurring?
Most of this comes from two components - The steel rotary band, which changes shape and loses it's rigidity over time, and the stepper itself. As good as these steppers are, the repeatability of ANY mechanical motion over the course of 30 years will eventually change. You're relying on that stepper to fall in the exact same position every step as it did 30+ years ago to reliably read that data - Without immediate on-disk feedback, this simply won't happen. You'd be pretty lucky but I suppose it's vaguely possible.

Take a look at the spindle replacement I recently performed on a Miniscribe 3650 for example.
Nice! Spindle replacement is just about one of the most invasive surgeries you can successfully do at home.

The heads, stepper motor and platters are non-serviceable. Later products from Quantum and the likes however you do have to go into to deal with sticky bumpers.
In a select few drives the head tension is adjustable and in some cases it is advisable to do so. A few drives have mechanical end stops that are adjustable, and of course like compaq above, it is possible to replace spindle and stepper motors yourself if you really care about the drive for some reason. Personally, I only undergo this type of surgery if the drive is particularly rare, interesting, or both. I've been working on perfecting a method of re-coating and in some cases outright re-making platters for a while now, my success rate is still sadly rather low. That said, if you have a drive with a head crash and a drive with a dead stepper, nothing's saying you can't swap the stepper between the two, or the remaining good platters. There's very little that these stepper drives won't tolerate in terms of service.
 
Just an update on this if anyone is interested.

I managed to find a donor ST251-1 drive in good condition and most important cheapish. Swapped the logic board and powered up, this time with the correct sounds and seek test after spinup as it should, yay!

Now when connecting to my controller board its not seen but at least i can here 'trying' from the drive when the controller is communicating.

Progress albeit slowly . . .
 
The Seagate ST-225 and ST-251 are some of THE most reliable stepper drives out there, the SCSI ones suffer from PCB issues unrelated to the mechanism. It sounds like you've got a bad batch. I hope you were LLFing each time you swapped boards, though.
I guess I got a bad batch. I did LL format each time. From memory, I have at least two that have one or more dead heads. The SCSI mechanism wouldn't format when connected to an MFM board. I forget exactly what it was doing. Ironically, the best one (has about 300k of bad sectors) appears to have been opened.
 
It is true that the large particles will end up in the filter. The smaller ones don't It may work this week but start to fail several months later.
The worst particles, according to what I read would be tobacco smoke. Not only a particle but because of incomplete burning, oily as well.
Anyway, there are only two things one might do by opening the drive up. You might replace the bearings. If you has donor drive, you might swap heads or even platters.
Dwight
 
Now when connecting to my controller board its not seen but at least i can here 'trying' from the drive when the controller is communicating..
Does the access lamp come on or flash? Check your write fault latch is disabled and your drive select is set correctly. Make sure the terminator resistor is installed correctly as well.

I guess I got a bad batch. I did LL format each time. From memory, I have at least two that have one or more dead heads. The SCSI mechanism wouldn't format when connected to an MFM board. I forget exactly what it was doing. Ironically, the best one (has about 300k of bad sectors) appears to have been opened.
Interesting, even my opened ST-251 has less bad sectors than that.

It is true that the large particles will end up in the filter. The smaller ones don't It may work this week but start to fail several months later.
The worst particles, according to what I read would be tobacco smoke. Not only a particle but because of incomplete burning, oily as well.
Anyway, there are only two things one might do by opening the drive up. You might replace the bearings. If you has donor drive, you might swap heads or even platters.
Dwight
The ST-251 I opened up has been working perfectly fine for at least 5 or 6 years now, as I already said earlier. If you are reasonably careful with it, things will be fine. I've repaired at least several dozen drives from this era in my own home without issue, even in the case of some drives pretty invasive surgeries. My most recent project drives were a Seagate ST-212 whose sled bearings came dislodged during transit and a Shugart SA-712 whose spindle motor had died. Both drives are currently working with no further ailments than they had to begin with - The SA-712 shows just one bad DOS cluster in SpinRite 5.
 
I have ST251 20 MB drive which was lying on the shelf for 30 years. No computer and no controller i have (unless the controller was on the same card as the videocard? I have videocard from the same computer with some suspicious connector on top of it) . Is the data on it recoverable?
 
I have ST251 20 MB drive which was lying on the shelf for 30 years. No computer and no controller i have (unless the controller was on the same card as the videocard? I have videocard from the same computer with some suspicious connector on top of it) . Is the data on it recoverable?
Connectors would be 34 and 20 pin to connect the drive. The drive format is controlled by the controller card and multiple formats used so not interchangeable like modern drives. Without knowing the original controller finding a compatible controller would be hit or miss.

If you really want the data my board should be able to read it if the drive is still good.
http://www.pdp8online.com/mfm/
 
I have ST251 20 MB drive which was lying on the shelf for 30 years. No computer and no controller i have (unless the controller was on the same card as the videocard? I have videocard from the same computer with some suspicious connector on top of it) . Is the data on it recoverable?
Welcome to these forums. Whereabouts on the planet are you? For example, I have lots of different make-model of 'MFM' hard drive controllers for just this kind of situation.

Consider adding your location to your profile:
Click on your profile name at the top-right of the screen, then select {Account details}, and there you will find a field named {Location}.
 
Once you've taken the drive apart, they are worthless. They likely have issues with the control boards. Beyond that, it is just a guess. You've trashed the drives by opening them up and exposed them to the environment.
Dwight
Oh, they are not as fragile as all that. Back in the 1990s I had a ST-235 running without a cover for nearly 2 months on a dare, what finally killed it was me sneezing on it by accident. While I don't "recommend" doing that, merely taking the lid off in an intentional way to inspect things is unlikely to cause an issue.

From a confirmation bias perspective, people generally don't remove the lids without the drive not working to begin with.

-- Bob
 
Oh, they are not as fragile as all that. Back in the 1990s I had a ST-235 running without a cover for nearly 2 months on a dare, what finally killed it was me sneezing on it by accident. While I don't "recommend" doing that, merely taking the lid off in an intentional way to inspect things is unlikely to cause an issue.

From a confirmation bias perspective, people generally don't remove the lids without the drive not working to begin with.

-- Bob
Haha, that's awesome. I've had plenty of drives apart before but usually not for more than a few hours at a time. The longest I had one apart was my Epson HD-860, which suffered from stiction and needed a bump to get started. In a vain attempt to try and cure this, I had spun it up and set it to seek test one night before bed. I wound up going to sleep, getting up in the morning, and going to work. When I got home from work I realized it was still going and the lid was still off. Sadly, despite being boiling hot and despite seeking for something like 14 hours straight, none of the surface lube seemed to move around enough to make it spin up easily. Oh well.
 
The original stiction problem was that the surfaces were too smooth. Van Der Waal attraction but your drive could have been lube.
Dwight
 
Back
Top