Drive repair tutorial?

37 views
Skip to first unread message

B 9

unread,
Nov 7, 2025, 9:41:59 AM (13 days ago) Nov 7
to se...@googlegroups.com
Hi! Thanks to Glenn's help off-list, I am now booting H-DOS on my Z-80 and I am thrilled! 

As he warned me, my drive does appear to need some repair. I know Glenn will eventually have a full tutorial, but in the meantime, is there a reasonable howto on repair of the Siemens FDD-100.5?

The drive acts like it is working fine when I just use it for loading and saving files. However, TEST17 fails the 3-pass General Drive Test.

T: Drive Rotational Speed: 1.001
D: 3-pass General Drive Test: prints "ABCDEFG" and then goes *thunk* *thunk* *thunk* repeatedly until I hit ^C.
M: Media sector verification (No bad sectors found)
S: Minimum step time: 14 ms.

I believe the *thunk*  sound is the head load solenoid, but I could be mistaken.

—b9

Glenn Roberts

unread,
Nov 7, 2025, 10:05:19 AM (13 days ago) Nov 7
to se...@googlegroups.com
The thunk is probably the track 0 homing. Did you see any errors accessing the boot disk I sent? 

As I recall test “G” does do extensive back and forth on the disk. It may have been fine. It does run for a long time. I would say try INITing and SYSGENing a disk and if that succeeds just use it and keep an eye out for errors. The H17 driver normally retries up to 10 times when there’s an error, so after you’ve heard that sound a few times you will come to dread it

You’ve got a good drive speed, 14ms seek is about the best you’ll ever get out of this drive, and media check showed no errors. All good signs! Have fun!

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sebhc/CAG1tPesgG_StaLof6wqneDvf0FfEDYEs7x-WxbDHHno0z9v89w%40mail.gmail.com.

Glenn Roberts

unread,
Nov 7, 2025, 10:08:39 AM (13 days ago) Nov 7
to se...@googlegroups.com
On further thought test G probably does a lot of head loading and unloading. It’s noisy but not to worry…

B 9

unread,
Nov 7, 2025, 11:28:23 AM (13 days ago) Nov 7
to se...@googlegroups.com
Ah, thank you! There was no error printed, but I cancelled when the thunking noise continued; I feared irrecoverable damage. 😅

INIT worked without errors. I'll try SYSGEN next.

The manual seemed to say that INIT (at least in HDOS 2) doesn't check for bad sectors, implying one should always run TEST17's verification afterward. Is that what folks actually do in practice?  

—b9

Glenn Roberts

unread,
Nov 7, 2025, 11:49:45 AM (13 days ago) Nov 7
to se...@googlegroups.com
I think it would have completed test G, but then it does all of the tests 2 times again! The sound of constant head banging was always scary to me - now it’s music to my nostalgic ears 😀. Probably not worth the wear and tear on the drive and disk to run all three passes.

Try SYSGEN /MIN to copy over just what’s needed to boot. you can always use ONECOPY to install individual files as needed.

I sent you a bootable H89LDR disk so you can recreate any of the old disks we have archived on the SEBHC site. See my REMarks article.

You’re right, the Heath-recommended process is to INIT then test for bad sectors (the test will overwrite the formatting) then INIT again, blocking out any bad sectors. I think with one of the updates there was a new INIT that would also do the media check… (fuzzy memory)??


Sent from my iPad

On Nov 7, 2025, at 11:28 AM, B 9 <hack...@gmail.com> wrote:



B 9

unread,
Nov 7, 2025, 11:55:27 AM (13 days ago) Nov 7
to se...@googlegroups.com
I may have made a mistake. SYSGEN got to:

    Insert the Source Diskette in SY0:.  Hit Return when ready:

So, I inserted the HDOS disk and hit Return and, after some whirring, got this:

    ERROR - ?02 SYS ERR #018

HDOS 2 chapter 2 says #18 means "Read failure on the device." Curious. Right?
Curiouser and curiouser: Now the HDOS disk doesn't boot. 

Please tell me HDOS doesn't write to the Source Diskette...? Did I need to do something special to tell it I have only a single drive?

In the meantime, I'm going to try to recover using the other disk you sent me, H89LDR. Right after I write-protect it!

—b9


glenn.f...@gmail.com

unread,
Nov 7, 2025, 1:42:09 PM (13 days ago) Nov 7
to se...@googlegroups.com

Can’t remember if I write protected that system disk but that’s always a good idea (put opaque tape over the protect notch).

 

You can recreate any of these disks so no huge deal.

 

Though I don’t necessarily recommend a full disassembly and cleaning of your drive at this time (since it is reporting good numbers) it is always a good idea to clean the disk head. It is quite possible that during that extended disk test you ran some oxide came off on the head. I do this by removing the drive and taking the PC board off the top, then you can reach the head directly (gently lift the pad arm to hold it out of the way).  Use a cotton swab and alcohol to clean it.  That could easily explain sudden issues reading disks.  It could also be the disk is damaged.  The oxide and oxide bonding on the disk are 40 years old   

 

If you have one of those old disk head cleaning floppies that could work, or if you have a cotton swab on a long stick you might be able to access it from the front (be careful though not to damage the pressure pad).  But still the best way is to open up the drive as shown.

 

If you need to recreate an HDOS 2 boot disk the originals are here in H8D format:

https://sebhc.github.io/sebhc/software/HDOS/HDOS_2-0.zip

 

just follow the direction in my REMarks #3 edition.

 

  • Glenn
image001.png

B 9

unread,
Nov 7, 2025, 5:11:05 PM (12 days ago) Nov 7
to se...@googlegroups.com
The HDOS disk is probably fine. I know that because the H89LDR disk isn't booting either.

I suspect you're right about the oxide on the head. (I refrained from running the "Head Cleaning" option, whatever that does, in TEST17 for fears of exactly that.)

The HDOS disk was write-protected at one point in its life, given the gummy residue, but it's not now. If you did write-protect it, then the solution may be as simple as finding the piece of tape that's gone missing inside the drive. While I'm in there, I'll use some 99% isopropyl alcohol and give the head a gentle clean.  Thanks for the helpful picture.

And a big thank you for the timely tip on opaque tape. I had just used some strong transparent label tape (leftover from printing out "[10 HARD SECTORS]"). I guess I'm too used to old drives with mechanical write-protect detection. Were optical detectors the norm on Heathkit computers?

—B9 



Glenn Roberts

unread,
Nov 7, 2025, 5:30:11 PM (12 days ago) Nov 7
to se...@googlegroups.com
I think all the drives of that era, including Tandons, used an optical sensor on the write protect notch

Sent from my iPad

On Nov 7, 2025, at 5:11 PM, B 9 <hack...@gmail.com> wrote:


The HDOS disk is probably fine. I know that because the H89LDR disk isn't booting either.

I suspect you're right about the oxide on the head. (I refrained from running the "Head Cleaning" option, whatever that does, in TEST17 for fears of exactly that.)

The HDOS disk was write-protected at one point in its life, given the gummy residue, but it's not now. If you did write-protect it, then the solution may be as simple as finding the piece of tape that's gone missing inside the drive. While I'm in there, I'll use some 99% isopropyl alcohol and give the head a gentle clean.  Thanks for the helpful picture.

And a big thank you for the timely tip on opaque tape. I had just used some strong transparent label tape (leftover from printing out "[10 HARD SECTORS]"). I guess I'm too used to old drives with mechanical write-protect detection. Were optical detectors the norm on Heathkit computers?

—B9 



On Fri, Nov 7, 2025 at 10:42 AM <glenn.f...@gmail.com> wrote:

Can’t remember if I write protected that system disk but that’s always a good idea (put opaque tape over the protect notch).

 

You can recreate any of these disks so no huge deal.

 

Though I don’t necessarily recommend a full disassembly and cleaning of your drive at this time (since it is reporting good numbers) it is always a good idea to clean the disk head. It is quite possible that during that extended disk test you ran some oxide came off on the head. I do this by removing the drive and taking the PC board off the top, then you can reach the head directly (gently lift the pad arm to hold it out of the way).  Use a cotton swab and alcohol to clean it.  That could easily explain sudden issues reading disks.  It could also be the disk is damaged.  The oxide and oxide bonding on the disk are 40 years old   

<image001.png>

retrogear

unread,
Nov 9, 2025, 11:22:33 AM (11 days ago) Nov 9
to SEBHC
Just recently I was using cut up white labels on my 8" Qumetrak 842's then realized it was causing false write protect's until I switched to black colored labels so the sensor must also detect reflections ?
Never a dull moment ...

Larry G

glenn.f...@gmail.com

unread,
Nov 9, 2025, 11:27:44 AM (11 days ago) Nov 9
to se...@googlegroups.com

Mainly needs to be opaque.  I used to like the silver ones which were essentially some kind of metal with adhesive on the back.  Those stayed in place. Others had a tendency to come loose.

Mark Garlanger

unread,
Nov 9, 2025, 1:57:41 PM (11 days ago) Nov 9
to se...@googlegroups.com
On Fri, Nov 7, 2025 at 4:11 PM B 9 <hack...@gmail.com> wrote:
The HDOS disk is probably fine. I know that because the H89LDR disk isn't booting either.

I suspect you're right about the oxide on the head. (I refrained from running the "Head Cleaning" option, whatever that does, in TEST17 for fears of exactly that.)

For the clean head option, you'd need a cleaning disk like this - https://www.ebay.com/itm/276049263070

I've found it's easier to just use a q-top and alcohol to clean the heads, just be sure to do it gently.

Mark
Reply all
Reply to author
Forward
0 new messages