• Please review our updated Terms and Rules here

Troubleshooting / Repairing floppy drive on a PS/2 Model P70

K

kobayashimaru

Member
Joined
Aug 21, 2015
Messages
23
Location
Jackson, New Jersey
Hey Everyone!
I've got an IBM PS/2 P70 that I got on eBay a few years ago and I just started playing with it again. It turns on and comes up with the normal errors for system configuration, wanting to boot to the reference diskette. However, with the diskette in the drive, it spins up, but nothing happens. I've seen references online that this can be caused by bad electrolytic capacitors on the circuit board on the drive, but also that they aren't always the problem. I could always replace the drive, but from what I understand, all of the PS/2 floppy drives had the same reliability. As I understand it, I could hack in a standard 34 pin PC floppy drive, but I don't want to lose the original look.

Thanks in advance.
Joe
 
Just a quick tip of advice: don't play or screw around with the motors and the alignment screws. Try to take them off one by one and you'll end up like me with a bigger issue that I can't even get out of now. The top head alignment screw doesn't want to secure anymore and now I feel embarrassed with the shivering disaster.

EDIT: I just realized you were talking about a PS/2 ModelP70, not a PS/2 Model 70! It's very to confuse the model numbers, and the former uses the IDE cable like design. My mistake!
 
Last edited:
kobayashimaru said:
IBM PS/2 P70 It turns on and comes up with the normal errors for system configuration, wanting to boot to the reference diskette. However, with the diskette in the drive, it spins up, but nothing happens. ...this can be caused by bad electrolytic capacitors on the circuit board on the drive, but also that they aren't always the problem. I could always replace the drive, but from what I understand, all of the PS/2 floppy drives had the same reliability.
Click to expand...

Ah, that new car smell... I have a few 8573 in P70 and P75 flavours. The original floppy is a dust magnet. Later 2.88MB floppies fit and configure as 1.44MB, no issues. The eject button is a bit long, and the floppy will not fold back into the case if a floppy is in the 2.88MB drive. Small sacrifice. It is my SWAG that the floppy cable is nylon tied to that small angle on the PSU just for assembly purposes, and snipping that cable tie has no affect on the floppy. Makes it easier to replace the original with a drive that has the 34 pin connector on the "wrong" side.

Here is my take on the floppy drive issue... YMMV.

First, there are bad caps. They can be replaced.
Next, the original drive with no shutter sucked dust into it.

Now to go even more gonzo... Earlier 8573 had issues with static electricity damaging gates on the floppy controller. Supposedly, that issue was fixed. Always make sure the ground cable is firmly fastened.

Now, this is an area of fact and supposition... Early 8573-P70 had issues with floppy read failures under Xenix or AIX, due to EMI inside the case. The source was tracked down to the original PSU design. The case halves had to be a tight fit, or the case halves would act as an antenna (credit to Don Hills from NZ). Two fixes were applied, depending on components and severity... Apply a 1/2" piece of copper tape on the PSU seams. Add a 1,000uF cap on the MCA slot riser.

Now this is where a big pot of coffee comes in. Let's posit that on the bright, sunny day that the PSU was born, the top and bottom halves of the PSU case were riveted together. The thin metal was coated and clean... Now add twenty years of indifferent use and storage. Corrosion (in vulgar terms, rust) occurs on the PSU case, forcing the halves just a teeny bit apart. Now we have an antenna.

So, there is more EMI than should be in the case. Whatever floppy is in there has to deal with it.

YMMV. But I have replaced a floppy, configured things, booted once or twice, then floppy errors...

http://ps-2.kev009.com/ohlandl/8573/8573-P70.html
http://ps-2.kev009.com/ohlandl/8573/8573_Power.html
 
1.2.12 DISKETTE DRIVE PROBLEMS, CONTAMINATION, ETC.

Record number: H09135

Device: D/T8550
Model: M
Tip key: 100
Date created: O91/11/22
Date last altered: A94/02/08

PERSONAL COMPUTER AND PERSONAL SYSTEMS/2 USERS MAY EXPERIENCE
DISKETTE DRIVE ERRORS, DAMAGED DISKETTES AND INTERCHANGEABILITY
PROBLEMS DUE TO CONTAMINATION AND ENVIRONMENTAL CONDITIONS.

ANALYSIS OF FIELD RETURNED DISKETTE DRIVES AND DISKETTES HAVE
REVEALED THE FOLLOWING PRIMARY CAUSES OF DISKETTE DRIVE
PROBLEMS:

1. HEAD CONTAMINATION DUE TO "BURNISHED-ON OXIDE*."
2. HEAD CONTAMINATION DUE TO DUST AND/OR PARTICULATE MATTER.
3. POOR QUALITY MEDIA, WITH POOR OXIDE BONDING.
4. POOR QUALITY MEDIA, WITH IMBEDDED PARTICULATE CONTAMINANTS.

* BURNISHED-ON OXIDE, IS THE OXIDE AND BINDER RESIDUE WHICH HAS
MIGRATED FROM THE DISKETTE MEDIA AND FIRMLY BONDED TO THE
READ/WRITE HEADS.

INTERCHANGEABILITY PROBLEMS:

CONTAMINATION, WHICH INTERFERES WITH NORMAL HEAD/MEDIA CONTACT,
MAY CAUSE INTERCHANGEABILITY PROBLEMS (DISKETTES MAY READ ON ONE
DRIVE, BUT NOT ON ANOTHER). THIS CAN CAUSE LOWER THAN NORMAL
AMPLITUDE SIGNALS AND RESULTING READ ERRORS. POOR QUALITY MEDIA
MAY BE A CONTRIBUTING FACTOR TO THIS SITUATION.

FORMAT FAILURES AND DIAGNOSTIC ERRORS:

UNNECESSARY FRU REPLACEMENT HAS BEEN ATTRIBUTED TO POOR QUALITY
OR DAMAGED MEDIA BEING USED DURING TROUBLESHOOTING. SUBSEQUENT
USE OF IBM BRAND MEDIA HAS RESULTED IN ERROR FREE DIAGNOSTICS
AND SUCCESSFUL DISKETTE FORMATS ON DRIVES WHICH HAD BEEN
"FAILING" PREVIOUSLY.

AS DISKETTE QUALITY IS NOT OBVIOUS BY VISUAL INSPECTION, ONLY
IBM BRAND DISKETTES SHOULD BE USED TO TROUBLESHOOT. THE
FOLLOWING BLANK DISKETTES ARE AVAILABLE FOR DIAGNOSTIC USE BY
SERVICERS THROUGH THE MECHANICSBURG PARTS DISTRIBUTION SYSTEM:

FRU P/N6023450 FOR 5.25" DSDD (360KB)
FRU P/N6109660 FOR 5.25" DSHD (1.2MB)

DUSTY ENVIRONMENTS AND INFREQUENT USAGE:

MANY FAILURES ARE CAUSED BY DUST AND FIBER PARTICLES WHICH PICK
UP A MINUTE ELECTRICAL (STATIC) CHARGE AND ADHERE TO THE
RECORDING HEAD SURFACES. THIS IS PARTICULARLY TRUE IF THERE IS
NO DISKETTE DRIVE USAGE FOR LONG PERIODS OF TIME. THEN, WHEN A
DISKETTE IS INSERTED, THE HEADS ABOVE AND BELOW THE DISKETTE
COME TOGETHER AND TRAP THE CONTAMINATION BETWEEN THE DISKETTE
AND THE HEADS. DAMAGE RESULTS FROM ONE OF THE FOLLOWING:

A. TINY CIRCULAR DAMAGE MARKS MAY OCCUR AT THE HEAD CONTACT
POINT DUE TO CONTAMINATION PARTICLES BEING COMPRESSED INTO
THE DISKETTE SURFACE.
B. THE CONTAMINANT STICKS TO THE HEAD LONG ENOUGH TO DEVELOP
A SCRATCH INTO THE MAGNETIC RECORDING COATING.
C. THE CONTAMINANT MAY BOND (STICK) TO THE MEDIA RECORDING
SURFACE.
D. A PARTICLE MAY BOND TO THE RECORDING SURFACE OF THE HEAD AND
STRIP THE COATING FROM THE MEDIA, IF THE COATING BONDING
(ADHESION) IS WEAK.

IN AN AREA OF HIGH DUST OR PARTICULATE CONTAMINATION AND
INFREQUENT DISKETTE DRIVE USAGE (DAILY OR LESS), SUCH AS A
BANKING, STORE SYSTEM ENVIRONMENT OR SOME LAN ENVIRONMENTS,
THE FOLLOWING RECOMMENDATION SHOULD BE VERY HELPFUL:

USE AN "EXPENDABLE" DISKETTE AND DO A DIRECTORY LISTING
("DIR A: ",ETC), PRIOR TO INSERTING A "CRITICAL" DATA DISKETTE.
THIS WILL "WIPE" THE HEADS AND MINIMIZE THE EXPOSURE OF
CONTAMINANTS DAMAGING DISKETTES AND CAUSING ERRORS.

ADDITIONAL INFORMATION:
IBM BRAND MEDIA WAS USED IN A TEST OF FIELD RETURNED DISKETTE
DRIVES WHICH HAD INITIALLY FAILED DIAGNOSTIC TESTING. THE HEADS
ON THESE DRIVES WERE CONTAMINATED WITH "BURNISHED-ON" OXIDE.
BY PERFORMING REPETITIVE FORMAT ROUTINES UNTIL ALL ERRORS
WERE ELIMINATED, THESE DRIVES WERE RETURNED TO SERVICEABLE
CONDITION.
SUMMARY:
THE USE OF "HEAD CLEANING KITS" ON PC AND PS/2 5.25 AND 3.5 INCH
DISKETTE DRIVES IS NOT SUPPORTED. ENGINEERING INVESTIGATIONS
INTO "CLEANING" OF DISKETTE DRIVES HAVE SHOWN NO CURRENTLY
AVAILABLE METHOD TO BE ACCEPTABLE. SHOULD THIS SITUATION
CHANGE DUE TO CONTINUING REVIEW, THIS RETAIN TIP WILL BE
UPDATED WITH THE APPROPRIATE INFORMATION.

DURING THE SECOND QUARTER OF 1993, AN ENGINEERING REVIEW OF THE
USE OF "HEAD CLEANING KITS" ON PS/2 DISKETTE DRIVES REVEALED
THE FOLLOWING FACTS:

Due to recent environmental issues, FREON based solvents, used
in some diskette drive cleaning kits, have been eliminated and
replaced with Isopropyl Alchohol. This exposes the user to
the following situations:
1. Freon evaporates faster then Isopropyl Alchohol., therefore,
a user can put too much fluid on the cleaning disk which will
leave a liquid on the head surface. If a data diskette is
then immediately inserted and the heads are loaded, the
magnetic recording surface under the head can be dissolved.

2. Unlike FREON, Isopropyl Alcohol, as it dries, can leave a
film on the heads. Any film on the head surface can cause
read problems.

THE USE OF QUALITY MEDIA, SUCH AS THAT SUPPLIED BY IBM,
AWARENESS OF THE "DUST" FACTOR AND USE OF THE "EXPENDABLE
DISKETTE" CIRCUMVENTION TECHNIQUE, AS DESCRIBED ABOVE, IS
THE BEST LONG TERM SOLUTION TO GAINING THE BEST RELIABILITY
FROM DISKETTE DRIVES.
 
I take it that is a TSB from IBM? Neato! I never knew that Isopropanol left residue. Unfortunately, Freon is quite pricey!

The machine I have belonged to an IBM employee and came with his business card, the original Reference disk, an IBM PS/2 mouse and a custom fit carrying case. I may try and see if the caps on the drive are the issue. If they don't fix it, I may do as ardent-blue suggested and use a 2.88 drive and just eject my floppies when I close the system up
Thank you!
 
First off, sorry to bump up this old thread. My P70 has a very similar issue. When the computer is powered on, the floppy drive would make normal noise. But when it attempts to boot from the reference diskette, the head moves back and forth for 4 times and then nothing. The motor spins fine. I can press F1 key to continue and it will boot into ROM basic. Here is the link to the video that I shot earlier today. Any thoughts?
https://youtu.be/OpPcDdt4M_g
 
Sounds like you did a manual clean.

I always use a cleaning disk. It takes less than a minute, does a better job with no required effort or disassembly, works every time, and doesn't carry the possibility of causing head damage of any type which includes, among other issues, misalignment.
 
I agree.

A cleaning disk is always my first step when dealing with a floppy problem.

I encounter lots of bad disks (that foul the drive's heads) but virtually no bad drives. :)
 
So I don’t have a cleaning disk on my hands and that’s why I used the q-tip and alcohol method. Actually it didn’t really make a difference before and after the cleaning. It makes feel that the heads have a hard time finding track 0?
 
Since you opened the drive to clean it, I think it would be possible to rule out dirt around the track sensor. You could try replacing the capacitors which is a common failure for PS/2 floppy drives. Of course, if the read/write heads were bent, replacing the capacitors won't help.

https://abitoutofplace.wordpress.com/2015/07/02/ps2-model-70-floppy-drive-repair/ has a description of how to replace capacitors.
 
krebizfan said:
Since you opened the drive to clean it, I think it would be possible to rule out dirt around the track sensor. You could try replacing the capacitors which is a common failure for PS/2 floppy drives. Of course, if the read/write heads were bent, replacing the capacitors won't help.

https://abitoutofplace.wordpress.com/2015/07/02/ps2-model-70-floppy-drive-repair/ has a description of how to replace capacitors.
Click to expand...

The capacitor issues were also my concern. Would they cause the symptom that was shown in the video? If so, I will go ahead and do the recap stuff. I am not sure if the heads were bent during the cleaning but since it did not act differently before and after, it is possible that it was bent before the machine came to me - I can definitely see some repairs on the drive (some flywires and resistors were soldered on the pcb).
 
I think it is possible that a capacitor failing could cause the symptoms. Other causes may exist but I can't think of other easily fixable components. If you can find the matching maintenance manual, it may have circuit diagrams and testing methods to find exactly what had failed.
 
krebizfan said:
I think it is possible that a capacitor failing could cause the symptoms. Other causes may exist but I can't think of other easily fixable components. If you can find the matching maintenance manual, it may have circuit diagrams and testing methods to find exactly what had failed.
Click to expand...
Thanks! I’m going to get some capacitors in the next couple of days and see if recapping can fix it. Will keep you guys posted.
 
Okay, so I completely recapped the drive and it said "Non-system disk or disk error, Replace and press any key when ready." - Much much better than before. Since the heads were already cleaned, I believe there were some alignment issues. I then tried to adjust the upper head and was fortunate enough to get it read the reference disk and successfully boot to DOS. However, now I am facing another issue. The floppy drive has no problem reading the first side of the disk but would very likely fail/produce errors when reading the second side. I am saying this because if I format a disk with this drive, everything was fast and smooth before 60%. After that, it starts to make abnormal noise until the end of the process. If I copy almost a full disk of files from another machine and then try to copy them to my P70, some of the files are okay but some would cause "Data error reading drive A" error. One interesting observation is that if I format the disk with the parameter /F:720, to format it as a 720KB disk, everything works like a charm.

I believe something is still not spot on in terms of the head alignment. I'd like to try more sophisticated method to align it. Certainly I don't have a oscilloscope so I'm looking for something like ImageDisk. You might be curious why I don't use it - that's because it always throws an "Internal Stack Overflow, System Halted" error message and the system then just hangs there. It works fine on all my other computers and only has issues on this P70. Don't know if it has to do with the PS/2 platform but if anybody can share another utility that can align the heads, I would greatly appreciate it.
 
You must log in or register to reply here.
Back
Top