PDA

View Full Version : Need some help with 8 inch floppy drive...



Adventurer
January 31st, 2017, 11:12 PM
I'm about to aquire an IBM Displaywriter, with double floppy drive and no software. I do need to setup my old 286 PC to be able to read 8 inch floppies, so I guess I need some help.

First - if I understand correctly, I can use (or create) and adapter from 34 pin floppy cable to 50 pin cable for 8 inch drive. I ordered this adapter:
35828

So far so good. Next, I do need 50 pin cable. Can these SCSI cables be used?
35829

Next - which drive can you reccomend? I need to be able to write both - DS/DD and SS/SD floppies, at least judging from IBM displaywriter specifications. I'm thinking about getting a Shugart SA 851.
35830

Last - as I understand, the drive motor is directly powered with AC power, and control board needs 24V and 5V DC?

Thanks in advance for any advice or help!

fritzeflink
February 1st, 2017, 12:32 AM
# 1st
The adapter is best. I have the older one. For SD format the floppy controller I prefer is

Adaptec AHA-1522A then
Adaptec AHA-1542B

Look at Registry of mainboards/floppy controller results
in http://classiccmp.org/dunfield/img/index.htm
In the CP/M FAQ are some answers, here is a link as I don't want to post it all.

Q14: Can I read my 8" disks with my PC?

http://oldcomputers.dyndns.org/public/pub/manuals/cpm-faq.txt

What drives you may use ?
I hope somebody other has an answer.

# 2nd

SCSI cable ?

I believe so .

modem7
February 2nd, 2017, 12:59 AM
First - if I understand correctly, I can use (or create) and adapter from 34 pin floppy cable to 50 pin cable for 8 inch drive. I ordered this adapter:
I use the D Bit FDADAP adapter as well. Note that it needs to be powered via the 4-pin connector.


So far so good. Next, I do need 50 pin cable.
To cater for the data connector on my TM848-02 drive, I use the FDADAP-to-drive cable shown at [here (http://minuszerodegrees.net/temp/3/temp_65_85891.jpg)].


Last - as I understand, the drive motor is directly powered with AC power, and control board needs 24V and 5V DC?
My TM848-02 drive only needs +24 Vdc and +5 Vdc. I power it via the setup shown at [here (http://minuszerodegrees.net/temp/3/temp_65_85892.jpg)].

If you do not buy a tested drive, then do not be surprised if the drive is faulty. I had faulty capacitors in both of the drives that I purchased (separately).

Having technical documentation for the drive is good as well (e.g. for jumper configuration, for repair).

Don't accidentally buy a single-sided drive when you are after a double-sided one. That is what I did.

Adventurer
February 2nd, 2017, 08:45 AM
If you do not buy a tested drive, then do not be surprised if the drive is faulty. I had faulty capacitors in both of the drives that I purchased (separately).

Having technical documentation for the drive is good as well (e.g. for jumper configuration, for repair).

Don't accidentally buy a single-sided drive when you are after a double-sided one. That is what I did.

Thanks, a very useful information it is! I have bought (ordered) Shugart SA-851, which I guess is a double sided drive after all. Still, getting it to work might be an interesting task, since usually no one lists them as "working", at least for the fair price not. I know how to replace the capacitors, and do have a soldering station, however, I hope there might not be some mechanical damage to drive heads or something, which can not be repaired unless having one more drive.

At least, so far:

FDADAP adapter - ordered, on the way
50 pin SCSI cable - ordered
8 inch drive - Shugart SA-851 - ordered
8 inch floppies - to be ordered
DC adapter for 8 inch drive - to be bought
286 PC with 5/14 floppy cables - in my possesion
All available software images for an IBM Displaywriter already downloaded, including ImageDisk software as well
Last, but not least - IBM Displaywriter, ordered, on the way, however, no documentation, no software provided.

Am I missing something in order to transfer the data?

Malvineous
February 18th, 2017, 04:53 PM
I would be interested to hear how this ends up working for you.

I have successfully used the KryoFlux (USB floppy interface) to read 8" disks with a Shugart SA-800 (http://www.vogonswiki.com/index.php/Shugart_SA800-2) single-sided drive and it worked fine. I haven't tried connecting the drive to a real PC though.

You will likely need to adjust jumper settings on the Shugart drive to put it in PC-compatible mode (the factory default) if it has been changed to work with another type of computer. Some drives were made with soldered wire instead of jumpers so hopefully if you have one of these then it is already configured for PC operation.

If you're using it with a standard PC floppy controller, make sure you get the right 50Hz/60Hz version for your local mains power. Where I live the mains is 50Hz and most drives for sale are 60Hz, likely because someone imported the wrong drive and is trying to sell it again when they discovered it didn't work. If you get the wrong drive then the motor will run at the wrong speed (assuming the voltage is compatible) but I think the KryoFlux might be able to compensate for this. A real PC controller can't, though.

The FDADAP with SCSI cables works fine, as long as you can get the 50-pin edge connector at one end. The ones in your photo look fine. As well as +5VDC, +24VDC, and 240VAC @ 50Hz, my SA-800 also needs -5VDC to operate. The manuals can be found online and helpfully give part numbers for the various power connectors, along with current requirements for each voltage.

Shadow Lord
February 20th, 2017, 09:40 AM
I went a bit over board with my setup but given that most of my drives are external it has proven to be quite versatile:

FDADAP (http://www.ibm51xx.net/)

Adventurer
February 21st, 2017, 11:11 AM
Thanks everyone for advices and tips, and here are the news - I finally got all the components I needed, so I started to do some soldering/connecting job. The only problem - I had to use two power adapters for the drive, since local electronics shop did not have one with +24 and +5 V.

At the beginning things looked pretty well...36356

Later when the main connections and adapters were connected, it was time to connect FDDAP adapter and time to test the drive.

36357

Brand new old stock DS/DD floppies for testing:
36358

Test results:

Motor working - OK
Drive set up as 5.25 1.2 MB in BIOS - OK
Access light lits when switching on the PC, drive solenoid makes a click - OK
No errors on starting PC - OK
Drive access, format: fail
36359

I did notice, that the rails are not moving - I can manually adjust them, and they just stay where they are. It seems there is a problem either with electronics, or the rail motor itself. Maybe I can not use it with an MS-DOS format? Even then I would expect the rails to move at least...

Adventurer
February 21st, 2017, 11:19 AM
Forgot to write, that FDADAP adapter shows no progress when accessing drive - instead of track number there is "- -"

archeocomp
February 21st, 2017, 11:22 AM
I had one 8" drive which did not move head assembly. One 74 TTL on board was dead. I would double check step pulses are there and are wide/slow enough for your particular drive. When that's OK it is time to study the schematics. At least make sure there is TTL high level on pin 36 - the STEP signal of the drive when it is powered on and idle.

EDIT: FDADAP staying in unknown state indicates there might be problem with Track0, STEP and DIRECTION signals. Check Track0 levels when you move head assembly by hand.

sciencedude100
February 21st, 2017, 11:39 AM
What would be a good 8" drive that's cheap.
Building one might be a possibility if I can get the circuits.

MikeS
February 21st, 2017, 12:02 PM
...
Test results:

Motor working - OK
Drive set up as 5.25 1.2 MB in BIOS - OK
Access light lits when switching on the PC, drive solenoid makes a click - OK
No errors on starting PC - OK
Drive access, format: fail
...
The drive light should not light, nor should the HL solenoid engage until the drive is actually selected. Check the cable.
What drive number is the drive jumpered as?

m

modem7
February 22nd, 2017, 01:04 AM
Drive set up as 5.25 1.2 MB in BIOS - OK
Note for later when you get the drive operational.

1.2M drives have 80 tracks. Like my TM848, the SA-851 has 77 tracks.

From memory:
* When I used an XT-FDC card, and DOS tried to format/access tracks beyond 77 (because DOS thought there were 80), I would hear a very nasty sound from the drive.
* When I used an FDADAP, and DOS tried to format/access tracks beyond 77, the FDADAP appeared to keep the drive at track 77.

So with the FDADAP, when I formatted a floppy via DOS, track 77 on the floppy ended up being formatted four times, first as track 77, then as 78, then as 79, then as 80.
The result was that the last six tracks ended up being marked as ..., 72, 73, 74, 75, 76, 80.
This meant that only 76 tracks on the floppy were usable, and I had to hope that DOS would not try to put anything beyond track 76.

archeocomp
February 22nd, 2017, 02:52 AM
What would be a good 8" drive that's cheap.
Building one might be a possibility if I can get the circuits.

Building only the electronics you mean?

Chuck(G)
February 22nd, 2017, 04:40 AM
I'm sure you know this already, but I'll say it anyway. On 8" floppies, unlike 5.25" floppies, you need a write enable tab over the write-protect slot. 5.25" are exactly the opposite.

Also, 3M produced a line of floppies packaged with transparent red write-enable tabs. Apparently, they're too transparent for some drives. That one took a bit of work to discover... :)

sciencedude100
February 22nd, 2017, 09:25 AM
Building only the electronics you mean?

I can build (most likely) the mechanics of a drive and (definitely) the electronics.
All I would need are the components and schematics. I could even make my own board if needed.\
So, is there anyone with schematics of a DS/DD 8" floppy drive?
Also, what's a good reference on it's standards? (Such as speed, timings, data flow, etc.)

MikeS
February 22nd, 2017, 09:27 AM
The result was that the last six tracks ended up being marked as ..., 72, 73, 74, 75, 76, 80.
This meant that only 76 tracks on the floppy were usable, and I had to hope that DOS would not try to put anything beyond track 76.
An interesting bit of info; thanks!

Chuck(G)
February 22nd, 2017, 09:41 AM
I can build (most likely) the mechanics of a drive and (definitely) the electronics.
All I would need are the components and schematics. I could even make my own board if needed.\
So, is there anyone with schematics of a DS/DD 8" floppy drive?
Also, what's a good reference on it's standards? (Such as speed, timings, data flow, etc.)

A good place to start are the Shugart OEM documents on bitsavers. Shugart drives did tend to be the standard for 8" in the 70s.

Good luck on making those heads...

Shadow Lord
February 22nd, 2017, 10:27 AM
So with the FDADAP, when I formatted a floppy via DOS, track 77 on the floppy ended up being formatted four times, first as track 77, then as 78, then as 79, then as 80.

The result was that the last six tracks ended up being marked as ..., 72, 73, 74, 75, 76, 80.
This meant that only 76 tracks on the floppy were usable, and I had to hope that DOS would not try to put anything beyond track 76.

Modem7,

Could this be bypassed using the DOS driver "driver.sys" to give formatting info to programs like format?

Also can't you can define format parameters on the command line for format.com:

https://technet.microsoft.com/en-us/...(v=ws.11).aspx

So IIRC this should work: format d: /a:1024 /t:77 /n:8 where d is your 8" drive. Since you specify /n: I don't BELIEVE you need the /8 switch. Please note this is for MS-DOS version of the format.com and not the windows version.

Adventurer
February 22nd, 2017, 10:56 AM
Glad to see more info coming in - I believe there should be something like a "using 8 inch floppy drive for dummies guide".

Anyway, I spent a rather sleepless night, working with the drive. Replaced some bad caps, had a long, but finally eventful fiddling with jumpers - finally I did get a seek operation when turning on the PC. However, it was not seeking when trying to acces/format the drive
36396

Later, more jumpers, two more added, and finally heads started to work as expected. What a relief!.

Next, some of you are already laughing, and well, I was not even reading the forum posts, so I got "Drive protect error". Instead of looking in forum (or google), I thought it must be the jumper configuration (indeed - it is possible to disable write protect on this drive, by soldering two connectors). Only after this operation I found out, that all I had to do is put a piece of tape on the floppy side. Oh well.

DOS 5.0 format - started to work, but the drive was making weird noises, heads jumping over sectors, so I simply aborted the operation. My main goal was to restore a floppy image to test with an IBM Displaywriter. After some trials and errors with ImageDisk software, I was finally making some progress:

36397

However, IBM Displaywriter acted as if I had inserted a blank floppy disk. I decided to test making an image of the disk, and here the picture was not that happy - basically the program was unable to read anything that was written.

36398

I strongly suspected this might had to do something with writing SD image on a DD drive/DD floppy (the only speed I could write was 250k, however, the image was written with 500K speed - maybe this also could cause some compatibility problems). Then, when I thought that very soon I will have everything under control, came the unexpected - I was transfering CP/M image for a Displaywriter, which was DS/DD format, and in the middle of the process, the 40MB hard drive of my system crashed, together with MS_DOS 5.0, and images to be tested. The fact that I had no floppy drives connected to it (except an 8 inch drive) did not help me at all.

While I work with fixing my system, may I ask some questions I really need answers for:

* Is DS/DD drive compatible with SS/SD floppies in terms of writing/reading the disk? I'm starting to think that there might be a problem.
* Has anyone succesfully restored and run any of the floppy images for IBM Displaywriter on bitsavers?

Time to rest now, to be continued tomorrow...

Chuck(G)
February 22nd, 2017, 10:59 AM
Or you could simply use ImageDisk for testing--there's a custom formatting dialog there.

The link produces a 404, BTW. But I believe that the article refers to the standard NEC PC98 formatting, which is, a double-sided format (i.e. won't work with a single-sided drive). This format is shared among all PC 98 media types; 8" 5.25" and 3.5" (eminently sensible). However, I'm not clear if later versions of non-PC 98 MS-DOS support the 8x1024 track format.

Al Kossow
February 22nd, 2017, 11:17 AM
* Has anyone succesfully restored and run any of the floppy images for IBM Displaywriter on bitsavers?


The single-sided disks have been tested, the recently added double-sided disks have not.

As Chuck says, you really should be using Imagedisk for testing.

and.. be EXTREMELY careful with the disk that was sent to you. Like I said, no double-sided disk images have been verified
to work and it would be really bad if the one you were given was destroyed.

also, the diagnostic disks require one track to be completely blank, which requires you to bulk-erase the diskette before copying to it.

MikeS
February 22nd, 2017, 12:33 PM
Pictures on the web seem to suggest that the Displaywriter uses single-sided disks (Type 1); if the drives are in fact single-side drives then they will not normally be able to read double-sided disks (Type 2 or 2D) such as the SA-851 might create, because the index holes are in different locations.

Chuck(G)
February 22nd, 2017, 01:19 PM
I gotten them (DW data disks) in both single- and double-sided varieties.

Adventurer
February 22nd, 2017, 08:47 PM
The single-sided disks have been tested, the recently added double-sided disks have not.

As Chuck says, you really should be using Imagedisk for testing.

and.. be EXTREMELY careful with the disk that was sent to you. Like I said, no double-sided disk images have been verified
to work and it would be really bad if the one you were given was destroyed.

I wil, first thing I will do - make a backup of the contents on the Displaywriter itself. However, the floppy has not arrived yet, and I'm afraid it might be lost in transit...

Adventurer
February 22nd, 2017, 08:52 PM
Pictures on the web seem to suggest that the Displaywriter uses single-sided disks (Type 1); if the drives are in fact single-side drives then they will not normally be able to read double-sided disks (Type 2 or 2D) such as the SA-851 might create, because the index holes are in different locations.

I think there might be a problem - it is possible, that Displaywriter will not take DS/DD floppies, only SS/DD. At least drive doors have marks 1/2D. I should have been more careful with purchasing 8 inch floppies, and should not have bought DS/DD format at all, whicn I think will have no use for me now.

Besides, it was probably not the best idea to write single side images on a double sided floppy - I'm not sure if it works that way at all. Whatever, now it is time to order SS/DD floppies for further testing and troubleshooting...

Chuck(G)
February 22nd, 2017, 09:28 PM
Or get yourself a hand punch and punch the proper aperture holes for single-sided media.

modem7
February 23rd, 2017, 01:01 AM
Modem7,
Could this be bypassed using the DOS driver "driver.sys" to give formatting info to programs like format?
Also can't you can define format parameters on the command line for format.com:
https://technet.microsoft.com/en-us/...(v=ws.11).aspx
So IIRC this should work: format d: /a:1024 /t:77 /n:8 where d is your 8" drive. Since you specify /n: I don't BELIEVE you need the /8 switch. Please note this is for MS-DOS version of the format.com and not the windows version.
My use of an 8" drive in DOS was short-lived; the novelty wearing off quickly. Maybe others here (for whom the novelty has yet to wear off) will experiment and report results.

If it does not work, the impact may be low. I seem to recall that DOS keeps data together as much as possible, for performance reasons. If that is the case, then post format, a problem may only be experienced if someone tries to fill the floppy to near capacity.


However, the floppy has not arrived yet, and I'm afraid it might be lost in transit...
I posted it almost three weeks ago. It does seem hard to believe that a large envelope would take that long to get from Australia to Europe. Maybe customs intercepted it and stuck the floppy on their wall as a reminder of the good old days when they used CP/M systems. Maybe the kangaroo got lost.

Adventurer
February 23rd, 2017, 06:21 AM
My use of an 8" drive in DOS was short-lived; the novelty wearing off quickly. Maybe others here (for whom the novelty has yet to wear off) will experiment and report results.

If it does not work, the impact may be low. I seem to recall that DOS keeps data together as much as possible, for performance reasons. If that is the case, then post format, a problem may only be experienced if someone tries to fill the floppy to near capacity.


I posted it almost three weeks ago. It does seem hard to believe that a large envelope would take that long to get from Australia to Europe. Maybe customs intercepted it and stuck the floppy on their wall as a reminder of the good old days when they used CP/M systems. Maybe the kangaroo got lost.

Too bad about the floppy, I still hope it might arrive, since I'm not making great progress so far with Displaywriter images on Bitsavers...

As for format with line "/a:1024 /t:77 /n:8", the problem is that switch "a" has been implemented starting with Windows 2000, it can not be run on MS-DOS 5.0 I have now. Still, I'm trying third party utilities, the results are coming in.

Adventurer
February 23rd, 2017, 06:55 AM
I'm still getting nowhere.

ImageDisk - I can successfully format/verify/read in DS/DD format, that is, until I have to write an image.
36468

As a matter of fact I can successfully write only SS/SD format, SS/DD is not working - do not really understand now.
It seems that ImageDisk starts to write SS/DD image first as SS/SD, then, after track 6 it tries to switch to SS/DD format with no success - see attached picture:

36469

As for writing SS/SD image - at first, everything seems to be good, no errors, disk seems to be written correctly:
36470

However, when I try to copy disk image to hard drive, nothing works:
36471

I started to think that there might be a drive head alignment problem, however, freshly formated disk shows no errors in Imagedisk, when testing read/write/seek, I manually tested all tracks:
36472

Adventurer
February 23rd, 2017, 07:32 AM
Here are the results of testing format/writing in MS-DOS mode. So far the only uility, which seems to work on formatting this disk, is Fdformat. It can format it as 620K disk, or 1.2 MB:

36474

Both formats allow read/write/copy with no problems. Even sys.com works on 8 inch drive.

36473

However, this is where the fun ends - if I do one of the following:

* Take out the floppy
* Restart/Switch off the PC

Then it is no longer readable, and I get either "sector not found" error or generic drive failure error. Actually it seems that as soon as the drive stops spinning, my data is gone. Only once I could get the contents of the floppy with one retry, and there were two small files on it. In all other occasions I had to format the floppy again to get it to work. Could there be a problem with a drive seek unable to return to exactly the same position when they are moved away?

MikeS
February 23rd, 2017, 07:43 AM
I think there might be a problem - it is possible, that Displaywriter will not take DS/DD floppies, only SS/DD. At least drive doors have marks 1/2D.
That sounds like it can handle all three types; 1 or 2 is no. of sides, D=double density:
https://en.wikipedia.org/wiki/List_of_floppy_disk_formats

Chuck(G)
February 23rd, 2017, 08:41 AM
Open up the drive box, look at the top of the drive. Single-sided drives have only a pad instead of a head on top.

Adventurer
February 23rd, 2017, 11:01 AM
Open up the drive box, look at the top of the drive. Single-sided drives have only a pad instead of a head on top.

I will do that, but first I need to be able to write and read the Displaywriter images, so far - one big FAIL.

So far I can only read/write DS/DD format:

Heads: 77
Sectors per track: 8
Sector size: 1024
Sides: 2
Data rate: 500K MFM

Low level format - PASS
Write - PASS
Read - PASS

Other formats tried, but failed:

DS/SD

Cylinders: 77
Sectors per track: 26
Sector size: 128
Sides: 2
Data rate: 250K FM

Low level format - PASS
Write - PASS
Read - FAIL, can read only first sector of each track

SS/SD

Cylinders: 77
Sectors per track: 15
Sector size: 256
Sides: 1
Data rate: 250K FM

Low level format: PASS
Write : PASS
Read: FAIL, can read only first sector of each track

What even more puzzles me, is that SS/SD images of Displaywriter software seem to be written with 500K instead of 250 - I have to use translation to 250 in ImageDisk settings to be able to write the image, but still - it fails with the same problem - can read only first sector of each track.

Could anyone help me with this issue? It seems I am stuck right there...

Al Kossow
February 23rd, 2017, 11:24 AM
I will do that, but first I need to be able to write and read the Displaywriter images, so far - one big FAIL.

So far I can only read/write DS/DD format:

Heads: 77
Sectors per track: 8
Sector size: 1024
Sides: 2
Data rate: 500K MFM

Low level format - PASS
Write - PASS
Read - PASS

Other formats tried, but failed:

DS/SD

Cylinders: 77
Sectors per track: 26
Sector size: 128
Sides: 2
Data rate: 250K FM

Low level format - PASS
Write - PASS
Read - FAIL, can read only first sector of each track

SS/SD

Cylinders: 77
Sectors per track: 15
Sector size: 256
Sides: 1
Data rate: 250K FM

Low level format: PASS
Write : PASS
Read: FAIL, can read only first sector of each track

What even more puzzles me, is that SS/SD images of Displaywriter software seem to be written with 500K instead of 250 - I have to use translation to 250 in ImageDisk settings to be able to write the image, but still - it fails with the same problem - can read only first sector of each track.

Could anyone help me with this issue? It seems I am stuck right there...

many PC floppy controllers cannot reliably write single-density format.
there is a test program on Dave's Imagedisk site which can be used to verify if your controller can write SD

8" disks always write at 500K

MikeS
February 23rd, 2017, 11:38 AM
...
8" disks always write at 500K
Is that correct? I was under the impression that 8" FM was 250K @360RPM...

Chuck(G)
February 23rd, 2017, 11:39 AM
What Al says is correct--don't confuse the clock rate with the bitrate. All 8" diskettes are written at 500KHz clock (high density), whether they're FM or MFM (or MMFM, but that's a different subject). The bitrate of FM is half that of MFM, but the clock is the same.

What are you using for a floppy controller? Has it passed the FM test of [url=http://www.classiccmp.org/dunfield/img54306/testfdc.zip]TestFDC{/url]?

MikeS
February 23rd, 2017, 11:41 AM
...
Could anyone help me with this issue? It seems I am stuck right there...
Do they only fail on the DW, i.e. can you reliably read back the disks on the PC that created them?

MikeS
February 23rd, 2017, 11:45 AM
What Al says is correct--don't confuse the clock rate with the bitrate. All 8" diskettes are written at 500KHz clock (high density), whether they're FM or MFM (or MMFM, but that's a different subject). The bitrate of FM is half that of MFM, but the clock is the same.
So, you're saying that the "data rate" in the OP's settings (which is what I think we're talking about) means "clock rate" and not "bit rate" and he should select 500K regardless of density?

Chuck(G)
February 23rd, 2017, 11:49 AM
Very likely yes, but I don't have his program, so it's hard to say for sure. But that's the usual convention.

MikeS
February 23rd, 2017, 11:58 AM
Very likely yes, but I don't have his program, so it's hard to say for sure. But that's the usual convention.
Kind of counter-intuitive; I'd think "data" would refer to bits and not the clock, sort of like "baud" and "bps" in the modem world...

Clock rate and/or transfer rate would certainly make it a little less ambiguous.

Learn something new every day.

Chuck(G)
February 23rd, 2017, 12:09 PM
Nope--bitrate depends on the encoding (modulation). FM reserves a clock bit for every data bit. MFM is more complex, but is self-clocking.and doesn't use separate clock bits. The raw maximum signal frequency is the same for both.

http://www.datarecoverytools.co.uk/wp-content/uploads/2009/08/fm-and-mfm-encoding-write-waveform-for-the-byte-10001111.jpg

MikeS
February 23rd, 2017, 12:15 PM
Nope--bitrate depends on the encoding (modulation). FM reserves a clock bit for every data bit. MFM is more complex, but is self-clocking. The raw maximum frequency is the same for both.
Sorry, what's the "nope" refer to? I thought I understood FM/MFM and bit rate vs. clock rate, but now my head's starting to hurt.

But to get back OT, the question was should the OP indeed always enter 500K for the Data rate in his various experiments?

Chuck(G)
February 23rd, 2017, 12:19 PM
My opinion is "yes". Always 500KHz for 8".

MikeS
February 23rd, 2017, 12:31 PM
Nope--bitrate depends on the encoding (modulation). FM reserves a clock bit for every data bit.

Not to drag this on longer than necessary, but doesn't FM require two clock bits per data bit (which is why the bit rate is 250K vs. clock rate of 500), whereas with MFM they're more or less the same? Or am I now having a problem with clock "bit" vs. clock "rate"?

As I said, my head's starting to hurt...

Chuck(G)
February 23rd, 2017, 01:18 PM
I find that most folks find the diagram posted above confusing because it traces the magnetization direction and not what's seem on reading. If you look at the write stage of most floppy drives, you'll see that it starts with a "T" type flip-flop.

Remember that magnetic recording, with very few exceptions, is a "pulse" technology on the read side. So, in the above diagram, every time the magnetization "flips", the read circuitry will see a pulse. Let's reproduce the diagram again:

http://www.datarecoverytools.co.uk/wp-content/uploads/2009/08/fm-and-mfm-encoding-write-waveform-for-the-byte-10001111.jpg

Armed the information that we see a pulse occurs every time the magnetization changes direction, so let's write a "1" for every clock interval with a pulse and a "0" if there is none.

So for the FM trace we have : 11 10 10 10 11 11 11 11 - note that every data bit is prefixed by a "1" clock bit.

MFM encoding is a bit more involved, as it depends on history of what was recorded--you can't have more than 2 successive cells with no transition/pulses. Further pulses can occur on half-clockimes, but no pulse can ever be closer from the previous one than a whole clock period. A "1" bit is signified by a transition on a half-clock; a transition on a full-clockl boundary and signifies a data clock bit.

You can see that FM has a spectrum with two peaks: t and 2t (for 1 and 0 cells) and MFM has a spectrum of 1, 1.5 and 2t components. The bandpass frequencies are still the same and you can see where MFM might be a bit more susceptible to timing "jitter", which is why most MFM data separators are PLL (or the digital equivalent) stabilized.

I've tried to be as brief as possible and hope I haven't over-simplified things to the point of error.

MikeS
February 23rd, 2017, 01:48 PM
I get all that; As I said I'm just confused by your "FM reserves a clock bit for every data bit" when it seems to me that every FM data bit requires two clock bits, and the ambiguity that while "data transfer" rate generally refers to bit rate apparently "data" rate usually refers to clock rate.

But confusion is more and more my normal state these days... ;-)

Anyway, interesting as it is, other than whether the OP should enter 500 or 250 at the data rate prompt this is getting a little off topic...

m

Chuck(G)
February 23rd, 2017, 02:37 PM
He should use 500Khz as the clock rate.

Don't confuse FM with the NRZ used in, say, async comms. Very different animals. FM always has (with certain "flagging" exceptions) a 1 clock bit followed by a 1 or 0 data bit. Got it? 2 bits, one of which is always 1; the other may be 1 or 0 depending on the data being sent. Clock bits have no intrinsic information content. MFM introduces clock bits only as needed to maintain synchronization.

MikeS
February 23rd, 2017, 03:01 PM
He should use 500Khz as the clock rate.

Don't confuse FM with the NRZ used in, say, async comms. Very different animals. FM always has (with certain "flagging" exceptions) a 1 clock bit followed by a 1 or 0 data bit. Got it? 2 bits, one of which is always 1; the other may be 1 or 0 depending on the data being sent. Clock bits have no intrinsic information content. MFM introduces clock bits only as needed to maintain synchronization.
Let's hope the OP read that among all the other rambling... ;-)

Still sounds like 2 clock bits per FM data bit to me, so I still don't get what you meant by "FM reserves a clock bit for every data bit;" but that's OK; I'm sure folks are getting bored by now...

And of course baud & bps are different animals from FM clock &. bit rates, I just meant that it's another example where the bit rate is not necessarily the same as the "clock" rate.

m

Chuck(G)
February 23rd, 2017, 04:01 PM
Let me try again by saying this: Clock period is not the same as bit cell. At a 500KHz clock a bit cell (4 usec) in FM is 2 clock periods; it however encodes only one explicit clock bit per bit cell. It results in a data rate of 1 bit per bit cell, or 250Kbits/second. In MFM, by dint of clocking either on the edge or the middle of a clock period, a bit cell is 2 usec and results in a data rate of 500K bits/second. In MFM, clock bits (i.e. nondata transitions) are supplied only as needed to maintain good synchronization. They always occur on the edge of a clock period.

I think what you're trying to say is that you see FM as requiring two clock periods/cycles. That's true--but you have to differentiate the notion of "data" (i.e. information bits) and "clocks" (non-information bits intended for synchronization).

Communications protocols such as asynchronous rely on the fact that the implied clock domain is the same on both the talker and listener side--and, aside from small errors, is essentially the same and need only be accurate over the period of a single character, so including explicit clocking information would be superfluous. Note that synchronous communication requires a separate clock signal.

Adventurer
February 23rd, 2017, 09:02 PM
What Al says is correct--don't confuse the clock rate with the bitrate. All 8" diskettes are written at 500KHz clock (high density), whether they're FM or MFM (or MMFM, but that's a different subject). The bitrate of FM is half that of MFM, but the clock is the same.

What are you using for a floppy controller? Has it passed the FM test of [url=http://www.classiccmp.org/dunfield/img54306/testfdc.zip]TestFDC{/url]?

I use an old ISA controller, which works with a 360K 5.25 inch floppy. There should be no problems with writing a single density media. By the way, testfdc is not usable with an 8 inch drive anyway, since it tries to format 80 cylinders instead of 77, and, of course, fails for both - single and double density formats...

Chuck(G)
February 23rd, 2017, 09:22 PM
AFAIK, there are no non-ISA legacy PC controllers, even if they're embedded into a non-ISA board chipset; the DMA and interrupt structure remains a child of the ISA bus scheme (IRQ 6, DMA 2, Port 3F2-3f7).

However, you'll need to use a controller that's capable of driving 1.44MB 3.5" floppies--an old one-data-rate 360K controller won't pass muster. You need to use a controller that's capable of 500KHz "high-density" operation. (500KHz applies to both high density 1.2M (5.25') and 1.44M (3.5") drives). I'm not suggesting that you drive your 8" drive with testfdc--but rather hook up a 3.5" or 5.25" high-density drive and run the test. If it passes for both FM and MFM high-density data rates, then you're good.

Adventurer
February 24th, 2017, 07:54 AM
AFAIK, there are no non-ISA legacy PC controllers, even if they're embedded into a non-ISA board chipset; the DMA and interrupt structure remains a child of the ISA bus scheme (IRQ 6, DMA 2, Port 3F2-3f7).

However, you'll need to use a controller that's capable of driving 1.44MB 3.5" floppies--an old one-data-rate 360K controller won't pass muster. You need to use a controller that's capable of 500KHz "high-density" operation. (500KHz applies to both high density 1.2M (5.25') and 1.44M (3.5") drives). I'm not suggesting that you drive your 8" drive with testfdc--but rather hook up a 3.5" or 5.25" high-density drive and run the test. If it passes for both FM and MFM high-density data rates, then you're good.

Did a test - is it really that bad as it looks? Basically, it looks that my floppy controller is not capable of writing any of the formats I need:

36499

I had another PC with a floppy controller - this time Celeron 2.8 - the results does not seem to look better anyway...

36500

Meanwhile my 5V adapter for feeding 8 inch drive gave up ghost - IC circuit exploded (I strongly suspect this was manufacturing fault), so I need a new one.

Now I need a compatible floppy controller for writing 8 inch floppies. I will make a research, but are there any known models, which work? That would be better rather than getting all possible floppy drive controllers/computers with controller.

MikeS
February 24th, 2017, 08:43 AM
...
Now I need a compatible floppy controller for writing 8 inch floppies. I will make a research, but are there any known models, which work? That would be better rather than getting all possible floppy drive controllers/computers with controller.
Check out the registry at Dave Dunfield's site referenced in post #2
http://classiccmp.org/dunfield/img/index.htm

Is your controller 8-bit or 16-bit ISA?

MikeS
February 24th, 2017, 08:57 AM
...I think what you're trying to say is that you see FM as requiring two clock periods/cycles. That's true--but you have to differentiate the notion of "data" (i.e. information bits) and "clocks" (non-information bits intended for synchronization).
Yup; that's what I was trying to say; no need to reiterate that bit and clock rates are not the same thing, that's obvious.

We've gone off on a discussion of diskette and comm protocols (mostly talking past each other somehow); I only wanted to clarify Al's ambiguous remark that "8" disks always write at 500K"; most references to "data rate" out there refer to bps (250Kbps for 8"FM) whereas Al is obviously talking about clock rate (500KHz).

The relevance to this thread was what the OP should enter in response to IMD's Data rate question; apparently it should actually be the clock rate.

e.g. (From Patterson's blog)
Those 5” disks would spin at 300 RPM (5 revolutions per second), a little slower than the 360 RPM (6 revolutions per second) of 8” disks. And the 5” disks had exactly half the data rate, 125,000 bits per second, vs. 250 kbps for 8”.

Also Wikipedia:
https://en.wikipedia.org/wiki/List_of_floppy_disk_formats

Chuck(G)
February 24th, 2017, 09:23 AM
Mike, it's worse than that--when I talk about clock bits, I'm talking about an actual recovered clock from the data stream. This gets to be important when talking about address marks with "missing" clock bits.

Yes, there's data rate and clock rate. Perhaps the best way to talk about them is "bit cell time". It can get to be really confusing if you talk about the high-end spectrum. Both MFM and FM have the same high frequency limit in their spectrum; but the encoding efficiency differs. Given the "dumbness" of diskette drives, it's the bandpass that's probably of most interest to an engineer; whether FM, MFM, MMFM, GCR or RLL encodings are used doesn't really matter for him.
----------------------------
@Adventurer, your FDCTEST results show why you're not having luck.

If you're looking for 486-P3 integrated FDC motherboards, support of FM (single density) is hit or miss. For example, I have an FIC KC19 (Intel 820) P3 motherboard, that has great support for FM floppies, but it stands as an island in a sea of similar (including 440BX) motherboards with no support at all for FM. Unfortunately, support for ISA slot (it has 2) DMA is completely absent.

The ISA FDC controller done here by James Pierce and others does support FM.

Adventurer
February 24th, 2017, 11:47 AM
Check out the registry at Dave Dunfield's site referenced in post #2
http://classiccmp.org/dunfield/img/index.htm

Is your controller 8-bit or 16-bit ISA?

Thanks, I did not pay attention at the beginning. It seems that FD controller is no less important than FDAP adapter or the 8 inch drive itself.

I do think it is 16 bit adapter, but I am not sure. One thing I know - I need another one :)

Chuck(G)
February 24th, 2017, 11:51 AM
Find yourself a 16-bit ISA SCSI controller with FDC. Look at the FDC chip--if it's by NSC, you're in luck. Other possibilities exist, however.

Don Maslin and I started compiling a list years ago--I suspect that Dave's list is derived from that.

Adventurer
February 24th, 2017, 11:55 AM
Thanks to Modem7, today I finally received Textpack 4 floppy for IBM Displaywriter. It works! However, it is really frustrating to wait 6 minutes for it to boot, with long gaps of complete silence, with floppy drive being idle.
36506

Unfortunately, I can not make a backup copy of it, because Displaywriter notices a read error and aborts the operation. Is there a setting somewhere to ignore read errors? I did not find one...

With the Displaywriter floppy being in DS/DD format, it makes it compatible even with my current floppy controller, however, because of read errors on track 5 and 6, all I'm getting is unusable image, because ImageDisks no longer "understands" the format of the floppy, and writes nothing after track 6

Tried Teledisk with similar results - the only difference is that Teledisk stops writing after track 7. Is there a remedy to this problem, or the floppy has to be with 0 read errors?

Slowly starting to think about ordering Kryoflux USB controller...

Chuck(G)
February 24th, 2017, 12:23 PM
It sounds as if Track 6 is completely blank, though I'd check that with an FM-capable FDC.

Adventurer
February 24th, 2017, 10:43 PM
It sounds as if Track 6 is completely blank, though I'd check that with an FM-capable FDC.

Found another PC - here are the results - will this be enough?

36516

archeocomp
February 25th, 2017, 01:35 AM
Found another PC - here are the results - will this be enough?

36516

Looks promising, only third line from bottom makes me unsure :-) Give it a try

Adventurer
February 25th, 2017, 07:35 AM
Looks promising, only third line from bottom makes me unsure :-) Give it a try

Did it. Tried. Mess on my desk becomes bigger and bigger.
36537

The results:

I can write images to floppy and back, both in SD and DAD formats, and now with 500K rate even for SD images, no errors. However, Displaywriter fails to recognise any of them.

Now I can copy Textpack floppy to 18-th sector, then it appears there is no data. However, Displaywriter is not unable to read my image, and I believe the floppy is not fully copied...

36538

I did an experiment - formatted a floppy on a Displaywriter, then wrote the DS/DD image of Textpack. It did not work, but Displaywriter was still able to read old label "DOC", so it is clear the data are missing.

Maybe interleave, Format, Read/Write gaps should be set manually. Would be good to hear something from someone, who had been able to restore Displaywriter images back to floppy disk...

Al Kossow
February 25th, 2017, 08:25 AM
try setting imagedisk to do an analysis on every track

a double-sided double-density Imagewriter disk should look like this

Cyl:0 Hd:0 500K FM 26 sectors size 128
Cyl 00 Hd 0 128 .CCC.C..dddddddddddddddddd 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Cyl:0 Hd:1 500K MFM 26 sectors size 256
Cyl 00 Hd 1 256 dddddddddddddddddddddddddd 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Cyl:1 Hd:0 500K MFM 26 sectors size 256
Cyl 01 Hd 0 256 ..........C............... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Cyl:1 Hd:1 500K MFM 26 sectors size 256
Cyl 01 Hd 1 256 .......................... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Cyl:2 Hd:0 500K MFM 26 sectors size 256
Cyl 02 Hd 0 256 .......................... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Cyl:2 Hd:1 500K MFM 26 sectors size 256
Cyl 02 Hd 1 256 .......................... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Chuck(G)
February 25th, 2017, 08:43 AM
I believe that Al might have something--the DW, IIRC, does use Deleted Data Marks on the early cylinders. IMD should be able to copy them.

Adventurer
February 25th, 2017, 09:46 AM
Already did that - without full analysis ImageDisk is unable to make an image to hard drive.

Now there are some good news after all - all day trying, but finally I managed to make a copy of the damaged Textpack 4 floppy on a Displaywriter itself. Now it boots in about 20 seconds instead of 6-7 minutes before. I now have 3 master floppies + the original damaged one, which can be read only in a Displaywriter.

36548

Now I was able to make an image of the copied floppy with no read errors, however, some problems still remain.

First - it seems that floppy for the Textpack should be completely blank - if not, Displaywriter shows previous floppy label, and treats it as damaged document disk.

Second - even with blank floppy it is not a complete success - for the first time Displaywriter starts to read it, but few seconds later shows *02*.

I have a feeling that now I am very close to success, but something seems to be missing - could there be data on a floppy, which are not read by ImageDisk? Besides, judging from the file size (170K), it appears to be an image of SS/SD floppy, duplicated to DS/DD, rather than native DS/DD image.

Next task - will try the same with Teledisk...

Chuck(G)
February 25th, 2017, 09:58 AM
IMD compresses its images, so size can be misleading. Use the "IMDU /B" utilitiy to extract an uncompressed image.

Adventurer
February 25th, 2017, 02:31 PM
No luck so far...

ImageDisk settings:

Interleave 1:1 (tried best guess)
Double stepping: off
Sides: As read (tried double as well)
Tracks: 77
Full analysis: on
Increased step rate from 8 to 15 - the same, just writes slower

Run align/diagnostics - high pitched noise on all tracks, data read/write fine.

One more thing puzzles me - I took Displaywriter system floppy (it's named NEGKKE), opened in ImageDisk, and choosed "Erase".

Expected: When put in a Displaywriter, it would display it as a blank floppy: "*******"

Real result: Displaywriter still was able to read the disk name "NEGKKE". The question is - from where?

Chuck(G)
February 25th, 2017, 04:58 PM
If you create an image file, how many sectors does it say it copied? It should say 2002.

If you scan an "erased" floppy with IMD, do you get any reports of data?

Adventurer
February 26th, 2017, 10:03 AM
If you create an image file, how many sectors does it say it copied? It should say 2002.

If you scan an "erased" floppy with IMD, do you get any reports of data?

According to ImageDisk, it has 2548 sectors:

36607

Meanwhile I think I have found the problem - instead of real writing/formatting/erasing, it was only pretending to write. Textpack 4 floppy, erased and formatted, was still readable on IBM Displaywriter, although it had some trouble reading data from it.

I think it has to do something with unterminated input lines, strange, but with previous PC the drive was also the last one, but was able to read/write, although not in the formats I needed.

I tried to terminate them with 150 Ohm resistors, but floppy controller showed failure on boot - will try tomorrow with 1K resistors again. The drive write head is OK, since it was writing on another machine.

Still, the written images look OK in HEX editor - there is data, and the format is very similar to the ones uploaded in Bitsavers, so at least there is hope, that at least read function is working correctly.

Al Kossow
February 26th, 2017, 10:12 AM
have you been able to read any of the copies you've made on the imagewriter that can boot?

it is very strange that track 0 side 0 is double-density. In the images I've made t0 s0 is always single density.

Adventurer
February 26th, 2017, 10:27 AM
have you been able to read any of the copies you've made on the imagewriter that can boot?

it is very strange that track 0 side 0 is double-density. In the images I've made t0 s0 is always single density.

I do not know if the images will work - I need to fix the drive first, as it is currently in a "read only" mode. I will report the results of testing as soon as the drive problems have been sorted out. If you are interested, I can mail you a copy of this particular Textpack 4 floppy (tested).

Adventurer
February 27th, 2017, 08:44 AM
This looks like a dead end for experimenting - even with input lines terminated (both - 150 Ohms and 1000 Ohms), both computers still manage to recognise two drives instead of one. I strongly suspect the write signal goes to the wrong lines because of it.

286 PC with ISA floppy controller:

36616

Intel Celeron 2,6 Ghz with onboard floppy controller:

36617

Input lines have been terminated:

36618

Still, the PC with Celeron and best testfdc results (only unable to write SD in 128K - but completely fails when 8 inch drive is connected) seems suitable for my tasks, but ImageDisk fails to write/erase anything, but it looks like it is writing normally - heads move, FDADAP adapter shows correct track numbers etc. The only program which can format the disk is FDFORMAT, I can read/write in DOS after that, but it does not helps with Displaywriter data in any way. Whats even worse - it appears the reading is worse after terminating input lines, since it shows completely wrong data:

36619


286 PC is in fact able to erase the disk with ImageDisk, but it fails in reading/writing, and BIOS recognises two 360K floppy drives.

Can anyone suggest what might be wrong? Tried different floppy cables, etc - still the same - two drives instead of one in the BIOS. I'm not sure if getting Adaptec SCSI card will solve the problem with data signals. I have no more ideas left...

Chuck(G)
February 27th, 2017, 09:09 AM
I'm going to be lazy, but (a) what 8" drive are you using and (b) how is it jumpered (all jumpers, not just the ones that you think are important). I'm willing to go through the OEM specs and figure out how things should be.

8" drives on a PC are no big deal, really.

Adventurer
February 27th, 2017, 09:17 AM
I'm going to be lazy, but (a) what 8" drive are you using and (b) how is it jumpered (all jumpers, not just the ones that you think are important). I'm willing to go through the OEM specs and figure out how things should be.

8" drives on a PC are no big deal, really.

The drive is Shugart SA-851. Seek was not working until I added jumpers to DS and C. Drive select set to 1, but tried all other positions. By mistake I removed write protect feature - cut WP and soldered NP joints. What else - some pins are bent/broken due to shipping, but there were no jumpers in the package. I can solder the joints if needed. I used Shugart SA 850 OEM manual for troubleshooting, but it appears I'm out of ideas for now.

I hope the photo is clear enough to see? I can upload a larger version somewhere, if needed.

36620

Chuck(G)
February 27th, 2017, 09:57 AM
Good enough--I'll get back later today on what I see.

Adventurer
February 27th, 2017, 10:04 AM
Good enough--I'll get back later today on what I see.

Just in case I added labels to the jumpers connected for an easier reading:

36622

Adventurer
February 27th, 2017, 11:22 AM
Made adjustments according to this picture:

36626

PC-286
Now there is finally a single floppy drive in BIOS, stepper engages without DS and C connected. However, when i try to access the disk, stepper moves, but then a message appears "No interrupt from FDC controller". A jumper missing somewhere?
Strange, but drive is displayed as B: on Drive select1, and A: on drive select 2

PC-Celeron - shows error 601 on switching on - floppy controller failure, with the same configuration. Drive select change has no effect.

Chuck(G)
February 27th, 2017, 02:58 PM
A PC doesn't use the "polled" feature of the FDC--and most controllers have it disabled. In general, a PC assumes the DS is active when a drive is being stepped.

No "no interrupt" is because the FDC doesn't see an index pulse, so something's still wrong.

Note that Shugart numbers their drive select from 1, not 0 (as do other makers), so you want DS2 in this case.

You don't want the drive select to function as a side select, so 1B..4B should be open.

In fact, but for DS2, the factory (in black) defaults should be the ones to use, if I'm reading them right.

Adventurer
February 28th, 2017, 12:35 AM
Well, the driveToday I connected the drive back to 286 PC - and suddenly it appears no longer working there - the BIOS recognised a single drive, but no seek and no access light anymore.

Even worse, after a while I have no access light /solenoid working without controller connected. I suspect 5V electronics have failed. Time to get a new(old drive)? I can remove all jumpers again and start debugging, but it appears something has gone completely wrong, since drive access light should light and solenoid should make a click when applying power without controller connected.

Chuck(G)
February 28th, 2017, 06:28 AM
If you apply power and ground the appropriate drive select, you should get the LED. Otherwise, nothing should happen, assuming that you have the correct pullups installed.

Adventurer
February 28th, 2017, 06:39 AM
If you apply power and ground the appropriate drive select, you should get the LED. Otherwise, nothing should happen, assuming that you have the correct pullups installed.

Thanks, got it sorted out - 5V power supply ceased to exist, so installed a new one. Drive setup is currently like this:

36640

The differences from standard are:

Could not find trace 5E, not from the picture, it is either not connected or permanent trace without jumpers
Drive select set to DS2
Sector output set to 850 (851 was waiting for hard sectored floppy)
Allow write when write protected
Could not find trace M anywhere, it is either not connected or a permanent trace without jumpers
Drive input is terminated with 7 1K resistors
Every other jumper is set as standard, according to the picture - either connected or free

Now:

286 PC

The 286 is able to read/write in limited formats due to FDC incompatibility, but it works, there is a single drive A: in BIOS now

Intel Celeron 2.6

It does recognise drive as A: on switching on, drive access light goes on, solenoid makes a click (no seek yet), but, after saving changes to BIOS, it reboots, and shows "Error 601 floppy controller", and drive A: is removed from BIOS. Can repeat as many times I want:

First time:

BIOS recognises the drive, access light and solenoid engages, BIOS reboots - and drive A: is lost

What am I missing, and why it works properly on a 286?

Chuck(G)
February 28th, 2017, 07:21 AM
What are you using for pullups? The original spec called for 150 ohms, which means that drivers need to sink about 30 ma. A lot of later motherboards are oriented to 3.5" drives and can't sink the required current. (the old controllers used something like 7438 buffers to drive the lines). 150 ohms on a short line is a bit excessive; try 1K for pullup and see if that helps.

Adventurer
February 28th, 2017, 07:28 AM
What are you using for pullups? The original spec called for 150 ohms, which means that drivers need to sink about 30 ma. A lot of later motherboards are oriented to 3.5" drives and can't sink the required current. (the old controllers used something like 7438 buffers to drive the lines). 150 ohms on a short line is a bit excessive; try 1K for pullup and see if that helps.

I'm already using 1K - was no difference :(

MikeS
February 28th, 2017, 09:25 AM
Intel Celeron 2.6

It does recognise drive as A: on switching on, drive access light goes on, solenoid makes a click (no seek yet), but, after saving changes to BIOS, it reboots, and shows "Error 601 floppy controller", and drive A: is removed from BIOS. Can repeat as many times I want:

I still think that the activity light and head load should not be active "on switching on," but only after actually selecting the drive. Could the cabling be different on the Celeron?

Adventurer
February 28th, 2017, 10:25 AM
I still think that the activity light and head load should not be active "on switching on," but only after actually selecting the drive. Could the cabling be different on the Celeron?

It does not switch on immediately - only when PC checks its hardware few seconds later - switching head load and moving to track 0, the same what happens with a 3.5 inch drive. It works as it should.

Meanwhile I found another PC with Celeron 2.8 - BIOS recognises it fine, I can access it, format, but FDC controller is completely incompatible, so I can not really do much.

For a moment I was able to access the drive even on Celeron 2.6 with the most compatible FDC controller, but after a moment, in the middle of reading it stopped responding with ImageDisk error - no interrupt from FDC. Maybe I need 2K resistors instead of 1K for termination? It appears that floppy controller engages, but shuts down after a while, but it can read/write data and access the drive at this short moment.

Chuck(G)
February 28th, 2017, 10:51 AM
That "no interrupt from FDC" refers to the fact that the FDC isn't seeing an index pulse.

The 765 as used in a PC doesn't have any "Drive Ready" signaling (that signal is NC in the PC design). So a timer is used to determine the drive's ready status and the FDC itself is programmed to abort if two index pulses go by without a sector being found. So, if IMD assumes that the drive is ready and there's no index pulse timeout, it puts up that message.

A floppy drive is pretty brain-dead--it mostly tries to do what it's instructed by the FDC and, in the case of the PC implementation, only gets back READ DATA, INDEX, WRITE PROTECT, and optionally, DISK CHANGE (on 360K and some 720K drives, that status line isn't used).

Adventurer
March 1st, 2017, 03:27 AM
Finally a progress - I was finally able to make an Image of IBM Displaywriters document floppy, and was able to confirm, that my written text is really there. In fact, IBM used an odd code table, so it has to to converted first to be read on anything:

Original text as seen on IBM Displaywriter:

36651

The same portion of the text, as seen in DOS mode, in an uncompressed image file:

36652

And finally, after making conversions for the text to be readable:

36653

Was also able to make an image of IBM Displaywriter DS/DD with a Textpack 4 - image size is 534KB, and in fact, Displaywriter needs a part of it to be in SD/128 format, which my controller can not write, so I'm not sure if the image is usable before I get a compatible FDC adapter. Still, things are looking much better than before:

36654

Meanwhile thanks everyone for helping out, it could have taken much longer to solve everything. Still, for me this was not an easy/peace of cake experience...

KC9UDX
March 1st, 2017, 03:39 AM
Odd code table? EBCDIC?

Chuck(G)
March 1st, 2017, 08:55 AM
Yes, the DW is an EBCDIC machine. (Well, it's IBM, isn't it?) That includes the printer.

KC9UDX
March 1st, 2017, 12:57 PM
As much as I cringe thinking about coding for EBCDIC, (how bad would it be anyway, in those rare cases where I'd need it, using a look-up table), it would be fun to own a machine that uses it. But it would have to be something that I'd write software for on a regular basis, as opposed to something single-purpose.

Chuck(G)
March 1st, 2017, 02:48 PM
There are a fair number of those (https://en.wikipedia.org/wiki/PC-based_IBM-compatible_mainframes)

But realistically, it doesn't matter most of the time--you do have to switch mental gears when reading memory dumps, however.

KC9UDX
March 1st, 2017, 06:22 PM
[url=https://en.wikipedia.org/wiki/PC-based_IBM-compatible_mainframes]There are a fair number of those

But realistically, it doesn't matter most of the time--you do have to switch mental gears when reading memory dumps, however.

I had an XT/370, or was it AT/370, for a short while. It was part of a large cache that I obtained and couldn't keep all of, unfortunately. I powered it up just enough to not be able to use it, just once. It just didn't seem to me like something I could really play with without very serious effort. But this was before the days of finding everything you needed to know on the WWW.

Chuck(G)
March 1st, 2017, 06:36 PM
Part of my "baptism by fire" when I was starting out was the assignment of finding the cause of failure as reported by users. The guy I was assigned to looked like he'd been through hell and he had a pile of these things, each stapled to about a 400 page dump about 5 feet high and 15 feet long. It was probably a safety hazard, should an earthquake have hit. Each dump had 65Kwords of central memory (60 bit words), ten 4K (12 bit words) dumps of PPU and 2M words of ECS--and a yellow sheet stapled to the top with the problem.

I don't recall how many of the things I actually solved, but I learned to read octal something fierce, reading text as well as instructions and tables. It was frustrating, daunting and educational.

Adventurer
March 1st, 2017, 09:17 PM
I'm wandering if anyone has ever made a program to get text from a Displaywriter document floppy? So far the only option seems search/replace utilities/macros, but still - if text is in different segments, I can not automatically patch it together...

Adventurer
March 2nd, 2017, 05:26 AM
Just when I thought I got everything under control, a problem again.

First- the Celeron 2.8 refused to turn on (5 seconds, and it switched off), if the connected 8 inch drive was switched on. Then, after multiple retries, it did switch on, and it seemed that everything works, fine, however, it is no longer reading one side of the disk, in fact, ImageDisk thinks it is SSDD instead of DS/DD.

36682

If I manually select two sides, I get this picture - it tries to read, but can not:

36683

The data are read correct from one side.

Tried - replace SCSI cable, remove floppy cable, reconnect, restart - the same. Besides, it does not matter which disk I put in the drive - it reads them all as single sided.

Overloaded floppy controller?
8 inch drive electronics failure?
Something else?

This is one of the moments, when I really want to take a heavy hammer, and crush all this into pieces. It feels like there are small gremlins living inside my equipment, just waiting for a moment to strike...

Needless to say, that everything worked yesterday, and I have not moved/connected/reconnected anything.

Adventurer
March 9th, 2017, 11:01 PM
Finally got a fully compatible system - an old Compaq Deskpro:

36844

According to FDADAP it can read/write in all available formats:

36845

However, when I connect my 8 inch drive, it can really do nothing - It can write in none of the formats, and has very poor media reading results with ImageDisk. BIOS recognised the drive correctly, I can format it with ageDisk, but that is all.

Two other systems with less compatible floppy adapters at least are able to get data from the floppies, this can not.

Will try to terminate with 2K and then 3K resistors, I am really out of ideas...