UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 16th Aug 2016, 6:46 pm   #1
itmustbesid
Retired Dormant Member
 
Join Date: Jul 2016
Location: Auburn, California, USA.
Posts: 8
Default 5.25 disks for BBC micro

Hi All

I am trying to get an old BBC to work with some external floppy drives, but getting nowhere, and the more I read, the more confusing things are.

First, I have two drives to play with, both are vintage Cumana drives designed for the BBC. They are both dual units, and are 40/80 switchable.

Second, the BBC has both an Opus DDOS ROM and a Disk Doctor ROM: both have their own "format" command and I have been trying both.

I have some new floppies that are "double sided, double density" and some new floppies which are totally unmarked but claimed to be "double sided, high density" by the seller.

I get the same error on both drives, both sets of disks, and both ROMS (I think I tried every combo, maybe I missed one) .... the format appears to work (which means, the disk lights come on, the disk spins, the head moves to the outermost position), but then the "verify" always fails on track 00.

At this point I'm not sure if I have a hardware problem or whether it's my disks. Does anyone have any advice?
itmustbesid is offline  
Old 16th Aug 2016, 9:32 pm   #2
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: 5.25 disks for BBC micro

What model BBC is it? I'm guessing B as later machines have the format comment built into the Acorn DFS.

I'd forget about the high density discs for now - double-density is what you need.

I have had this before, and it was the disc controller IC. If I recall correctly, it was the 1770 in a Master, rather than the 8271 used in the original B. The fault was that discs would read fine, but couldn't be written. I'm guessing that you don't have a known-good disc already? Anyone near to you that can write one for you?

I'd check the ribbon cables are OK - it's worth reseating the connectors at the drive end - and check the heads are clean. Drives are pretty reliable in my experience, but anything is possible at this age.
mhennessy is online now  
Old 16th Aug 2016, 10:23 pm   #3
itmustbesid
Retired Dormant Member
 
Join Date: Jul 2016
Location: Auburn, California, USA.
Posts: 8
Default Re: 5.25 disks for BBC micro

Thanks for the reply: It's an Issue 3 BBC B. I did try reseating all the ribbon cables, but since they're pretty old I don't rule out a broken connection perhaps. I also cleaned the heads but not sure I did that right (alcohol on a q-tip). You're right that I don't have any known-good discs, and unfortunately I'm an expat in the USA so finding someone else to swap disks with is pretty hard!
itmustbesid is offline  
Old 16th Aug 2016, 10:35 pm   #4
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: 5.25 disks for BBC micro

It sounds like you've done everything you reasonably can, short of getting the 'scope out and seeing what the interface is doing!

Without access to known-good discs, and/or another machine, it's really hard to know what eles to suggest. The BBC does have a small following over there, so it might be worth asking around - there are one or two BBC-specific forums like Star Dot - http://www.stardot.org.uk/forums/

Good luck,

Mark
mhennessy is online now  
Old 17th Aug 2016, 2:23 pm   #5
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: 5.25 disks for BBC micro

Can you post a picture of the drives and discs, please? Cumana were a computer distributor and built floppy drive units with lots of different manufacturers' mechanisms in. I've seen them with posh Teac FD55 units in, and I have two pairs myself which have Chinon and JVC mechanisms, both of which required attention to get them working.

When you say the format appears to work, does it format the whole disk and then fail when verifying, or just do track 00 and then fail?

The high-density discs definitely won't work. The way to spot them is that they don't have a reinforcing ring stuck round the centre of the disc. They are made with a different magnetic material which needs a stronger field from the heads to record data, so single- and double-density drives can't handle them. The symptoms you report (failing to format even one track) are exactly what I'd expect with the wrong type of disc.

If the discs are the right type, dirty heads are definitely the next suspect. Did you take the drives apart to clean them? I've found that drives of this age can have dirt really well stuck to the head, which a casual cleaning with a cleaning disc or q-tip won't shift. They can have been retired in working condition, but somehow the old oxide seems to 'set' and have a worse effect once it's been there for a couple of decades. The last set of drive heads I did only responded to a q-tip soaked with brake cleaner (cyclohexane) viewed under a magnifier so I could see that all the little lumps and bumps of deposited oxide had disappeared. It's hard to do this properly without removing the upper drive head, which is usually quite simple but means realigning it when refitting it. It's not too difficult to do this, though, even without any test equipment.

The disc controller electronics are pretty reliable in my experience so we should eliminate the simple things first.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 19th Aug 2016, 10:17 am   #6
mike_newcomb
Rest in Peace
 
Join Date: Apr 2012
Location: West London, UK.
Posts: 665
Default Re: 5.25 disks for BBC micro

Hi, can anyone advise exactly what format the discs should be for this BBC Computer.

e.g. hard or soft sectoring, single or double sided etc.

I wonder if it is having problems, because the disks are already (incorrectly for the BBC) formatted and this may stop it correctly formatting to the required format.

A work around for this is to 'break' the floppy, by formatting on a (preferably DOS) PC that can read its current floppy format. Switch off the PC immediately the formatting starts. This corrupts the floppy, such that the 'original' format does not cause problems when formatting anew.

Good Luck - Mike
mike_newcomb is offline  
Old 19th Aug 2016, 11:10 am   #7
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: 5.25 disks for BBC micro

Soft sectored, double density. Double sided if the drives are (haven't seen a single-sided disc in a long time).

We only had single-sided drives at school, so we crudely cut out openings for the index hole!

Assuming the hardware is working correctly, the format routine will successfully overwrite whatever is on a DD disc - no worries there. Though it'll struggle with high density discs, as noted earlier.
mhennessy is online now  
Old 19th Aug 2016, 11:14 am   #8
cmjones01
Nonode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
Default Re: 5.25 disks for BBC micro

BBC discs need to be soft sectored, so with only one index hole, but they may be single or double density depending on the filing system in use. The standard Acorn DFS and its clones were single density (10, 256-byte sectors per track) whereas the Acorn ADFS and some special double-density DFSs (like the Opus DDOS) could use double density (usually 16 or 18, 256-byte sectors per track). The density is specified at the time of formatting the disc. The physical media are exactly the same in both cases, though.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline  
Old 19th Aug 2016, 11:34 am   #9
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
Default Re: 5.25 disks for BBC micro

I don't think I've ever seen a single-density disc. Were they ever sold?

As you say, there's no physical difference: https://en.wikipedia.org/wiki/List_o...al_composition

Unless it's been modified, this machine has an 8271 controller which can't do MFM.
mhennessy is online now  
Old 19th Aug 2016, 3:52 pm   #10
Dave Moll
Dekatron
 
Dave Moll's Avatar
 
Join Date: Feb 2005
Location: West Cumbria (CA13), UK
Posts: 6,118
Default Re: 5.25 disks for BBC micro

I seem to remember my first Beeb discs were sold as 40-track single-sided single-density (which was all my first disc drive would handle) - probably, these were the ones that failed quality control for higher densities. What I do remember is what a wonderful new world these seemed after loading from and storing to audio cassettes!

As stated, the original 8271 disc controller only supported single density, but (unlike a non-HD PC) it did support 80 tracks.
__________________
Mending is better than Ending (cf Brave New World by Aldous Huxley)
Dave Moll is offline  
Old 19th Aug 2016, 6:53 pm   #11
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
Default Re: 5.25 disks for BBC micro

Quote:
Originally Posted by Dave Moll View Post
As stated, the original 8271 disc controller only supported single density, but (unlike a non-HD PC) it did support 80 tracks.
Actually an IBM 360K disk has 80 tracks. 40 on each side of the disk. The 40 positions of the head assembly are 'cylinders'. Few people seem to get this right any more, but it causes confusion on multi-head disk drives, etc.

The original, DD-only IBM PC disk controller supports 80 cylnders. You can use 720K drives (either 5.25" or 3.5") on the original IBM 5150 PC with the original controller board. I do it.

I seem to remember an issue with one of the Intel floppy disk controller chips (can't remember if it was the 8271 or the 8272 or both) that it would only step out 77 cylinders on a 'restore' command. Meaning that if the head was on one of the innermost 3 cylinders you'd get an error on restore as it would not trip the track 0 sensor. The cure, of course, was to repeat the restore command if this happened and only give an error if the second one also failed.

As stated, the original BBC micro disk controller was the 8271, also used in the Acorn 'System' machines on Eurocards. This is a single-density (FM) controller only. Later BBC versions (B+, Master, etc) used the Western Digital 1770, which was also available on a plug-in daugherboard for the BBC B. It needed a different DFS ROM (obviously) and I think there was a routine that accepted 8271 commands and turned them into the appropriate 1770 commands. So that copy protected software that used the OS routines in odd ways to read the disk would still run.

Debugging floppy disk problems is helped if you have a drive exerciser (but few people do). Basically you should check :

The disk is rotating (not just the motor is rotating in a belt-driven unit).

It's rotating at the right speed (some drives have strobe markings on the spindle pulley)

You are getting an index pulse once per revolution (the frequency of this is a guide to the disk speed, of course).

The head can be stepped in and out

The track 0 sensor operates when the head gets to the outer endstop.

The write-protect sensor is not stuck on with a non-protected disk. Note that on all decent drives this sensor not only tells the controller that the disk is write-protected, it also disables the write current in the drive. You cannot write to (or format) a disk under those conditions.

You can get write current when you try to write to the disk.

You can get something from the read chain when you read what you have just written.

Note that most drives have a pair of testpoints at the output of the differential amplifier before the first filter in the read chain. These are often TP1 and TP2. A differential-input 'scope connected there will tell you if the drive is reading anything.

Alignment problems tend to prevent disks from being read on other drives. A drive that will not format a blank disk then write to it and read from it has problems that are not due to head alignment, etc. However, if you are going to dismantle a drive to repair it, do not remove anthing that might upset the alignment (head carriage, stepper motor, index sensor, track0 sensor) unless you know what you are doing and have a CE disk, etc.
TonyDuell is offline  
Old 19th Aug 2016, 7:22 pm   #12
itmustbesid
Retired Dormant Member
 
Join Date: Jul 2016
Location: Auburn, California, USA.
Posts: 8
Default Re: 5.25 disks for BBC micro

Well, some good news, and some pictures. Yesterday I pushed and poked at all the IC's mentioned in the install guide for the disc controller one more time: nothing seemed loose and I didn't feel like I really did anything. However, this morning, my attempts to format were more successful, so something did change. Maybe elves came overnight and fixed something.

If I use the *FORMAT command from the Opus DDOS 3.35 ROM, drive 0 formats OK (I formatted two disks that way). I tried formatting with both single and double density options, and with 40 and 80 tracks. Drive 1 makes unhealthy sounds and the head does not move, so not so lucky there. I did not try the reverse sides yet.

If I use the *FORM command from the "DiscDoctor" ROM, formatting does not work. Sometimes the head moves a little before formatting fails, and sometimes it doesn't move at all. Not very concerned about that, since I should use the Opus DDOS ROM for an Opus controller, but it would be nice to know whether it should have worked anyway?

The disks I am using are Sony double sided, double density, 48 TPI. They do have an index hole, and they do appear to have a reinforcing ring on the inside.

One of the pics is a close up of a drive head (drive 1, but drive 0 looks the same). Looks like a white block with a fine wire over it. I really don't know what it is supposed to look like though.

My other Cumana drive has the 40/80 track selector switches and the power switch on the front: I will try that one next.
Attached Thumbnails
Click image for larger version

Name:	20160818_212723.jpg
Views:	407
Size:	34.8 KB
ID:	128918   Click image for larger version

Name:	20160818_212749.jpg
Views:	265
Size:	85.9 KB
ID:	128919   Click image for larger version

Name:	20160818_213827.jpg
Views:	236
Size:	48.5 KB
ID:	128920   Click image for larger version

Name:	20160818_214706.jpg
Views:	286
Size:	51.1 KB
ID:	128921  
itmustbesid is offline  
Old 19th Aug 2016, 7:58 pm   #13
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,191
Default Re: 5.25 disks for BBC micro

The head looks fine. The dark line is the ferrite core of the head, there is a tiny gap across it (just like a tape recorder head).

The disks you are using should be fine too. Strictly 48tpi is for 40 cylinder operation. I have had problems using poor-qualitiy 48tpi disks for 80 cylnders (96tpi) but it's not a common problem. If they work, use them.

On drive 1, can you (carefully) slide the head carrage back and forth by hand with the power off? Is the fact that the head doesn't move a mechanical issue or a problem with the drive to the stepper motor? Those drives look very repairable, the electronics is all small ICs (not one big ASIC as on later drives). Probably mostly TTL. The drives will not be made by Cumana (or whoever), is there a maker's name on the drive chassis itself? There may be a service manual available for it.

What is the 40 pin IC on the Opus daughterboard (4th photo)? I am going to guess it's a Western Digital controller chip, either a 179x series or a 279x series.
TonyDuell is offline  
Old 19th Aug 2016, 11:31 pm   #14
itmustbesid
Retired Dormant Member
 
Join Date: Jul 2016
Location: Auburn, California, USA.
Posts: 8
Default Re: 5.25 disks for BBC micro

Quote:
Originally Posted by TonyDuell View Post
On drive 1, can you (carefully) slide the head carrage back and forth by hand with the power off? Is the fact that the head doesn't move a mechanical issue or a problem with the drive to the stepper motor? Those drives look very repairable, the electronics is all small ICs (not one big ASIC as on later drives). Probably mostly TTL. The drives will not be made by Cumana (or whoever), is there a maker's name on the drive chassis itself? There may be a service manual available for it.
The stepper seems to have serious issues. It moves well enough with no power (the same range as the other drive), but powering on and off a few times gives random results: sometimes it does nothing, other times it will move all the way to the outer edge (track 00 I guess) but then keep on trying, making nasty noises as the motor tries to move a head that is up against some mechanical stop. When trying things like *FORMAT and *. from the Beeb, it doesn't move at all. Oh and it also seems to get quite a bit hotter than the good one.

The drives themselves say TEC FB-504 on their chassis; haven't found much online as yet but will keep digging.

Quote:
Originally Posted by TonyDuell View Post
What is the 40 pin IC on the Opus daughterboard (4th photo)? I am going to guess it's a Western Digital controller chip, either a 179x series or a 279x series.
Good guess, WD2793A-PL02
itmustbesid is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:22 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.