high density disks

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
User avatar
Elk Towers
Posts: 489
Joined: Sun Apr 23, 2006 2:10 am
Location: Kettering, Northants
Contact:

high density disks

Postby Elk Towers » Thu Jun 12, 2008 6:02 pm

How long before problems arise using hd disks formatted in a dd drive.

are we talking hrs/days/weeks or months :lol:

As i Have just formatted a hd disk(yesterday) using my electron ap34 and the basic prog that i put on there is still there today :D

Nick

User avatar
sorvad
Posts: 2172
Joined: Wed Aug 24, 2005 12:13 pm
Location: Back of beyond

Postby sorvad » Thu Jun 12, 2008 7:58 pm

Not exactly sure, but I did some once, and a week later all I could do was read the catalogue. This was repeated with a couple of other HD disks, so I gave up using them. They were "brand new" sealed ones too.

OK for stuff you're only keeping for a very short time I suppose, but wouldn't trust them for much else.

guesser
Posts: 143
Joined: Mon Jun 26, 2006 9:21 pm

Postby guesser » Fri Jun 13, 2008 2:31 am

it varies from disk to disk

I've had disks where the data is unreadable later that day, but others that have been formatted for months and are still readable now.

as a general rule though, only use HD floppies for doing transfers if you can help it. I know finding DD disks is getting hard now :(

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Postby CMcDougall » Fri Jun 13, 2008 8:12 am

guesser wrote: I know finding DD disks is getting hard now :(


thats why ive got 4 big bags full! :D :D :D

i had no luck at all with HD discs, maybe cause i formated them in OmniFlop on PC, then into beeb /elk. Properly head allignment did
not help either.

i mind folk saying that it may work in one drive, but not in another.

see http://www.stairwaytohell.com/sthforums ... 78&start=0
half way down, post by Beardo.
ImageImageImage

User avatar
Elk Towers
Posts: 489
Joined: Sun Apr 23, 2006 2:10 am
Location: Kettering, Northants
Contact:

Postby Elk Towers » Fri Jun 13, 2008 12:21 pm

I am not using a hd drive

I am using a hd disk in an amiga dd drive connected to my elk.

Also can i connect a bbc 5.25in drive to my pc?

User avatar
alchresearch
Posts: 227
Joined: Mon Jan 03, 2005 9:14 pm
Location: Southport, UK
Contact:

Postby alchresearch » Fri Jun 13, 2008 12:39 pm

I'd say years. All of the HD ones I formatted on the +D disk interface on my Spectrum back in the early 90's still work.
Andy Davis.

SarahWalker
Posts: 1041
Joined: Fri Jan 14, 2005 3:56 pm
Contact:

Postby SarahWalker » Fri Jun 13, 2008 1:47 pm

The +D's a 3.5" drive isn't it? That's a different issue altogether, I've been using HD 3.5" discs in DD machines (Archimedes + Amiga) for years now without any problems. 3.5" DD and HD discs are largely the same, whereas 5.25" DD and HD discs are very different.

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Postby CMcDougall » Fri Jun 13, 2008 3:27 pm

corelki2005 wrote:Also can i connect a bbc 5.25in drive to my pc?


never tried that, as did not want to wreck my beeb setup at the time. But now i know the 31/2"PC drives run on beeb, so i could try!
....ow, hold it, i never found a PC that had the proper SD/DD disc controller to handle it :cry: , hence why i only use a P3 laptop to write /read to 31/2"

if u find a PC that has proper SD/DD disc controller, then use a beeb drive thats already set to drive 1/3 (need a double drive, or change 0/2 jumpers on top), put on a PC cable with original PC twist, and it should go :?
ImageImageImage

guesser
Posts: 143
Joined: Mon Jun 26, 2006 9:21 pm

Postby guesser » Fri Jun 13, 2008 6:22 pm

TomWalker wrote:The +D's a 3.5" drive isn't it? That's a different issue altogether, I've been using HD 3.5" discs in DD machines (Archimedes + Amiga) for years now without any problems. 3.5" DD and HD discs are largely the same, whereas 5.25" DD and HD discs are very different.


ah, didn't realise you were referring to five and a quarters, my experience is just with three point fives (on the spectrum too funilly enough, but the +3 not a +D)

User avatar
sorvad
Posts: 2172
Joined: Wed Aug 24, 2005 12:13 pm
Location: Back of beyond

Postby sorvad » Fri Jun 13, 2008 7:32 pm

Yes the coercivity on 3.5" HD disks is much close to the DD ones. Whereas on 5.25" disks it's significantly different (about half the amount) which is where your problems arise. These figures were recently posted on the mailing list;


5.25" DD (48 tpi) and QD (96 tpi) soft sectored: 300 Oe
5.25" HD: 600-675 Oe (different sources)

3.5" DD: 600 Oe
Oe3.5" HD: 750 Oe3.5"


So you cn see there's a very big difference for 5.25" but not 3.5".

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Sat Oct 04, 2008 12:23 am

I've done some reading of the plethora of technical, scientific and anecdotal evidence available on this HD/DD thing and, as sorvad says above, the coercivity difference is a definite along with the different head current (and hence magnetism or flux) used by floppy drives to read and write the two types of media. It is also variously suggested that in the case of the 3.5" discs, the difference in media coercivity is small enough to be virtually ignored as a variable in the confusion. However, the different flux levels used to write a HD or DD disc are significant and cannot be ignored.

Onto a practical test - About three months ago, I took two identical 3.5" HD floppies which had both been previously used on a PC and hence in HD mode. On both discs I blanked the HD 'hole' and formatted the first on a Beeb. For the second disc however, I first erased it using a magnetic media bulk eraser (we all have one don't we :roll: ) before formatting it on the Beeb. I then copied several random files onto both sides of both discs in the same order so that both discs were identical. As I said, I did all this about three months ago and, cutting to the chase, the bulk-erased disc still seems to be fine whereas the format-on-format disc appears to be significantly corrupted. Could be coincidence but it does lend some credence to stuff I have read which suggests that whilst the surface coercivity difference is small, the magnetic 'marks' written in HD or Quad-Density by a PC are not properly overwritten or replaced by a DD or Double-Density format and write. It is suggested that this mix of residual and new magnetic 'marks' leads to the magnetic equivalent of smudging and hence the eventual corruption. So, if you need to use a HD floppy in a DD drive by covering the ID hole, make sure the disc is totally blanked on a bulk eraser first. Easy! :wink:

Now, since we don't all have a Bulk Eraser and because the above may be complete codswallop, onto a possible hardware solution if someone knows whether it's feasible. This is to a large extent a question to anyone who knows 3.5" drives intimately and/or has a circuit diagram for a typical still available type.

The two relevant function changes that a drive makes when it detects a HD or DD disc are 1 - to change the data interface between Quad and Double Density and 2 - to change the head current used to write the disc. The media type sensor identifies the ID hole (or not) and must signal to the drive which processes apply. I was wondering, if the two functions are seperately selected, would it be possible to override the QD/DD 'switch' and tie it in the DD mode such that HD floppies could then be freely used and would be written with the correct flux. I've drawn a simple schematic below to illustrate the idea.

disc.JPG
(17.18 KiB) Downloaded 3828 times

If the mod were achievable, the drive would always work in DD mode as we need on Beebs and the like but the head current would still correctly change to match the type of disc being used. Any thoughts?

Martin

User avatar
sorvad
Posts: 2172
Joined: Wed Aug 24, 2005 12:13 pm
Location: Back of beyond

Re: high density disks

Postby sorvad » Sat Oct 04, 2008 6:17 am

Seems reasonable to me, if what you say is correct, I've never looked so closely, your research here is excellent. top job. explains a lot.

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Sat Oct 04, 2008 9:08 am

Thanks Steve - I was only trying to make sense of lots of thinking actually done by other people. There's nothing new really but I thought there might be a trick amongst all the scientific evidence to resolve the issue for us Beeb users.

I do like the hardware idea though because there would be no compromises, it would use discs exactly as designed. Indeed, if it were possible, we could all get a brand new 3.5" drive and use brand new HD discs which, for the floppy side of the hobby, would reset the clock and perhaps give another 25 years or so of playing :D

I wouldn't take the bulk erasing idea as gospel, I'll try some more samples but I'm not sure we could ever be certain unless someone could confirm that this method fits the underlying science.

I'll wait to see if anyone has some better disc drive knowledge for the hardware idea but if not, I'll maybe have a poke around with a scope (and a magnifying glass :shock: ) to see what I can learn.

Martin

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: high density disks

Postby CMcDougall » Mon Oct 06, 2008 2:33 pm

MartinB wrote: a magnetic media bulk eraser (we all have one don't we :roll: ) before formatting it on the Beeb


ehh, this u mean
Image
first ive knew of them :shock:

of do you just mean : -
Image
ie leaving HD/QD 31/2" discs on top of my car box for 2 hours??

wonder if this is the truth, or pure luck :?
i never had any luck with them, just the DD ones that are Blue and more of a brown disc, than 'black'
ImageImageImage

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Mon Oct 06, 2008 3:26 pm

Actually, mine was originally sold as a Compact Cassette eraser by Weircliffe and it's classed as a 'Flat Bed' type. It's just a wooden box with a flat black plastic top, a mains lead and a push button switch - for interest, I'll post a piccie later.

I think it's just some sort of monstrous transformer and switching it on with nothing on or near it you would wonder if it actually does anything. However, sprinkle a few nuts and washers on it and press the button and you get something akin to a nail bomb :shock: I guess that explains why there's a warning label telling you to remove your wristwatch :lol:

For discs, 3.5" and 5.25", you just hold the button in and slide the discs accross a couple of times and they are well and truly blank. There's certainly something to this because sometimes I've had floppies that a PC said were unformattable (is that a word?) but after an erase have worked fine. I wouldn't be in the least bit surprised if doing this allows HD discs to be used on DD drives, despite all the discussions about coercivity and the coating particle size etc.

I still do fancy that drive mod though but I can't for the life of me find any schematics or circuit diagrams for any 3.5" drive on the net :evil: Of course, it might be simply impossible and I'm only guessing at how the QD/DD stuff might be implemented - it's probably much more complicated and likely all just in one inaccesible surface mount array [-X We shall see......

Martin
Last edited by MartinB on Mon Oct 06, 2008 4:12 pm, edited 1 time in total.

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: high density disks

Postby CMcDougall » Mon Oct 06, 2008 4:10 pm

MartinB wrote:I've had floppies that a PC said were unformattable (is that a word?) but after an erase have worked fine.


yip, and yip, and cheers for the info Martin :wink:

i have the blue ones, and if i format them to DFS 80T DS with OmniFlop, then try to format them to (bad word coming!) ADFS :evil: , its also says [-X
wonder why that is :? maybe because DFS is =P~
ImageImageImage

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Mon Oct 06, 2008 8:28 pm

Here's a pic of my bulk eraser - nowhere near as fancy as either of Col's examples :wink: I think the one Col has shown (not the Beat Box!) feeds the disc through at a constant rate which is fairly important and you have to pull the disc away from the machine before switching the power off to prevent the collapsing field causing 'streaks' on the disc. It's actually dead simple to do manually though and I've never had any problems.
bulk eraser.JPG
Oersted Hell...
(56.06 KiB) Downloaded 3653 times

Incidentally, I know several people have mentioned using brand new HD discs yet still had problems. I had a quick look at some new boxes I've got and they're actually all pre-formatted for DOS/Windows, even one box which say they aren't but according to a PC they are! If there's any truth to this bulk erasing lark then this would mean that these 'new' HD discs are, to a DD Beeb, actually used and hence 'pre-corrupted' if you get my drift.

And, off at a tangent......

I've asked before but does anyone have a circuit diagram for a Watford Electronics 32K Shadow Ram board? I'm still trying to pimp this Elk with one but still having some issues. I wanna get this working before corelki gets his hard drive up and running :lol:
elk we32.JPG
Plug 'n' Play Hell...
(59.62 KiB) Downloaded 3653 times

Martin

frankoid
Posts: 53
Joined: Wed Jan 10, 2007 9:18 am
Location: London

Re: high density disks

Postby frankoid » Mon Oct 06, 2008 9:05 pm

I just found my copy Archimedes Elite, which I bought in January 1993, and the disc in there is actually an HD disc formatted to DD 800K. The disc is black and has the "HD" logo on it and 2 holes, one of which I covered up with a paper label. It can still be read - I made a disc image of it.

I'll defer to the opinions of those more knowledgeable than me about whether it's safe to use HD discs formatted as DD, but just thought people might find it interesting that some commercial software came on the wrong type of disc. Perhaps someone put the wrong type of disc in the duplicator?
Attachments
ArcElite_Disc.png
(58.42 KiB) Downloaded 3645 times

SarahWalker
Posts: 1041
Joined: Fri Jan 14, 2005 3:56 pm
Contact:

Re: high density disks

Postby SarahWalker » Mon Oct 06, 2008 10:44 pm

I've got a few commercial Archimedes discs that are HD formatted to DD. The Arc doesn't look at the density hole, so it will quite happily over/underformat discs if you aren't paying attention.

I would have thought this would work okay for commercial stuff - ie read but never written. I suspect Hybrid simply found HD discs for cheaper than DD.

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Mon Oct 06, 2008 10:56 pm

Hmm.. interesting stuff - certainly not the 'just can't be done' as some suggest and it looks like you guys can claim the record then at an impressive 15+ years =D>

The Arc 800k 'D' format was indeed still DD, but with the sectors/bytes juggled to give that little bit extra storage space.

I suppose I could turn (twist?) this in favour of the 'blank before use' theory because early HD floppies weren't pre-formatted, that luxury(?) started to appear and become the default sometime after the appearance of HD discs. It's the other way round now in that you can still buy unformatted ones but they're not as common as pre-formatted.

Anyway frankoid, to complete your contribution, could you give HYBRID a ring please and ask them whether they used unformatted discs or, if they were pre-formatted to FAT, did they bulk erase them first? Now please..... :D

Martin

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: high density disks

Postby CMcDougall » Tue Oct 07, 2008 12:33 pm

was the Arc not :evil: ADFS?? :?

if so, then thats why it will work, as :evil: ADFS is MFM, and DFS is FM (what ever that is :roll: ). So hence it still works!

the 12+ PCs i tried to get OmniFlop to work with, that did not have DDensity disc controllers, DFS was a no go, but they did work for ADFS S/M/F (elk discs).

also, when covering the 2nd hole (HD/QD) on floppy, i used blue tack, but some formatted half way then said Err, this was due to not a light sensor, but drive tries to stick a 'pin' through the hole, so i had to stick cornflakes cardboard onto the blue tack, then it was 'ok'.
ImageImageImage

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Tue Oct 07, 2008 2:02 pm

Not sure - my head hurts :(

I don't think that's necessarily relevant though - what seems to matter is that the the HD discs have a finer coating and were therefore designed to be written with a reduced head current which then in turn allows more data to be packed into each track & sector. ADFS (1770) uses Double Density and DFS (8271) uses single density but both need the 3.5" drive to have a DD disc (so the interface doesn't switch to Quad/HD) and for DD the drive then uses the higher head current for the DD media type. Therefore, I think these Arc discs should still be on DD media, not HD.

It would appear that a truly blank HD disc will work as a DD disc when the hole is covered but I guess the question is whether that is ok in the long term. Certainly I have found that in the short term a pre-used HD (on a PC) disc is likely to fail and that a bulk-erased one seems fine. However, unless there's a way of proving this is good practice without waiting another 15 years then it seems we're back to square one again :evil:

Definitely need some way of forcing a 3.5" drive to permanently function in DD mode for the interface but to write to the disc using the appropriate DD/Hi or HD/Lo head current. I'm sure it would be technically achievable but without any info for a given drive it will be difficult to achieve in practice. Usual story I suppose #-o

Martin

User avatar
CMcDougall
Posts: 5620
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: high density disks

Postby CMcDougall » Tue Oct 07, 2008 2:59 pm

your head hurts, think about mine 3yrs ago with i done my 51/4" & 31/2" DFS to PC images :lol: & the PC drive fix =P~

Can't say I've seen a system be particular about the drive. But there is a definite issue with the discs themselves. My experience on the computers I've used is:-
DD data + pre-HD drive + HD disc => problems on one machine, apparently the drive erased one side as it wrote the other. OK on another computer.
DD data + HD drive + HD disc with tape over the hole => almost works, but full of errors a few minutes later.
DD data + HD drive + HD disc without tape => as above, or no response whatsoever.
DD data + HD drive + DD disc => works OK
If you're using HD discs you'd have gone from case (1) to case (3). Have you any 'blue' floppies?

^^ from other thread, by Beardo.
3½" Double Density disks:
are often marked "720kB", "DD", "QD","1MB capacity (unformatted)"
have one notch in the corner, for write protect.

3½" High Density disks:
are often marked "1.44MB", "2HD", "HD","2MB capacity (unformatted)"
have two notches in the corners, one for write protect, the other to indicate High Density (you can tape over this hole with black tape as a last resort to use the disk as Double Density, but the results aren't guaranteed).

^^ from OmniFlop page.

cant seem to find anything about 'that' FM /MFM thingy :cry: i think thats the key answer, but if your magnet sorts them Martin, then i will go to the corner in 15yrs time :lol: :wink:
ImageImageImage

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Wed Oct 08, 2008 8:56 pm

Anyway, whilst I wait 15 years so I can send Colin into the corner.... :)

I picked a random 3.5" drive, a Panasonic as it happens, and had a prod around. Col was right about the disc density sensor - it's a plastic pin switch which pokes through the HD hole, or doesn't for a DD disc. Actually, on this drive, the write protect and disc present sensors are all the same mechanical switches - I think I was expecting optical sensors but obvioulsy not Panasonic's preference.

In the absence of a circuit diagram, I randomly lurked around with a scope probe whilst alternately inserting a HD and a DD disc. I soon found a signal whose level reflected the type of disc loaded and followed it round the drive desperately hoping it would split into two before I got to the sneering silicon centipede which was hugging the PCB just centimetres away. Alas not to be, the signal vanished into the chip thingy along with the 34 way interface lines and it looks as if this same smart ass-chip directly controls the read/write heads and presumably sets both the DD/HD interface and the matching head current from within :evil:
So, on a sample of one, the little idea I proposed in my drawing isn't going to be possible. However, I have plenty more of the rascals to try before I give up :)

A question - does anyone know how the coils in a disc drive head are implemented? This drive has 5 connections to each head (upper and lower) and I can't quite figure what they might be. Am I right in thinking there's a read coil, a write coil and an erase coil? If so, five seems an odd number of connections unless the read/write coils maybe share a common line and the erase coil is independent? Any knowledge would be useful.

...and a spin-off thought - had the Beeb come along just a few years later, it would probably have been squeezed onto a few of these inpenetrable all knowing chips and a whole generation of young people would have been denied the opportuinity to learn about electronics and computers by hands-on tinkering.

Martin

SarahWalker
Posts: 1041
Joined: Fri Jan 14, 2005 3:56 pm
Contact:

Re: high density disks

Postby SarahWalker » Wed Oct 08, 2008 9:23 pm

Just as a note, I believe all Arcs/RiscPCs treat DD/HD discs exactly the same - no change in any of the drive lines - other than the data rate.

I'm not aware of any changes on PC drives either. The HD hold detection is only used by the BIOS/OS to determine the format.

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Wed Oct 08, 2008 10:49 pm

Thanks Tom, and I understand your point but the issue I'm trying to overcome is the way that a PC 3.5" HD drive (with DD backwards compatiblity) makes the executive decision to switch itself to HD in the presence of a HD floppy.

For example, if you connect such a drive to a Beeb, Master or Electron (assuming 1770 DFS), insert a HD drive with hole uncovered and try to do anything with the drive, you will get a drive error and will get no further. This is because the drive has, in response to the disc hole, switched itself to HD with the higher rate interface such that the drive and the 1770, which doesn't support HD, are now unable to communicate. So, we cover the disc hole, the drive drops back to a DD interface and the two can communicate again.

That's all fine but the 'hidden' problem, if we believe all the published guidelines, is that the two types of disc require different levels of flux during writing and, when we fool the drive with our sticky tape, we are causing the drive to write to the HD disc with the incorrect flux. This flux level control is a private drive function and I agree is of no interest to the host computer.

Therefore, what I would like to achieve is to have a PC HD drive which is always in DD mode for the interface but still correctly sets the write flux level appropriate to the inserted HD or DD media.

I think the real unknown in all this is how safe is data written to a HD floppy by a DD (or HD in DD mode) drive. My other findings suggest that a fullly degaussed HD disc works ok but I have also read that repetetive writing to a HD disc at DD flux levels will slowly degrade the disc. Trouble is, there is so much conflicting information, anecdotal evidence, urban myths and just pure nonsense out there that I think a hardware fix such as the one I am (so far fruitlessly :( ) persuing would be the only guaranteed route to 100% safe use of HD floppies on a Beeb and it's derivatives.

Martin

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Fri Oct 10, 2008 11:41 pm

If nothing else I'm determined to at least fully understand this stuff and sort the truth from the myth. I've read some fascinating things about floppy disc data recording and some of you may be interested as I go along. (Stop tutting, the forum's a bit quiet at the moment so just humour me...... :roll: )

I asked what the head configuration is like and it's surprisingly novel :shock: To get 'neat' tracks of data, a method called Tunnel Erasure is used where there is a single Read/Write head and this head is slightly trailed by two erase heads, one on either side of the Read/Write head. During a read, just the Read/Write head is used but during a write, the data is written by the Read/Write head and then, immediately afterwards, the two erase heads come along and effectively trim the ragged edges (in magnetic terms) of the written data to form a nice tight line of recording. It's like spray painting a stripe using two strips of masking tape slightly apart - when the spray has passed along the line there is a broad strip with fuzzy edges but when you pull off the masking tape you have a nice straight line.
From my point of view, knowing this means I can probably figure out what the head connections are and can then look at head currents etc. for the different types of drive and disc.

This coercivity thing is interesting - sorvad mentioned the figures so here's a light definition of oersted :

'Coercivity is the measurement of the level of difficulty to magnetise the material. For storage media, a high coercivity is desired in order that information stored will be preserved in the presence of stray magnetic fields that may be present. High coercivity also implies that a strong magnetic field is needed to record information onto it. Magnets with high coercivity are called hard magnets.'

Steve's figures show that HD discs have a higher coercivity than DD discs so, using the definition, it would seem fair to assume that for HD discs we will need to use a stonger magnetic field (flux) to store the data. This in turn means that the head current will have to be higher to generate this higher field. Now, reading various sources on the net, the most popular citation is that the flux used to write HD discs is lower than that used for DD discs which is the complete opposite of that which we have just deduced :?

However, if our deduction is true, this would fit the practical tests I have carried out with the bulk eraser since it is likely that in DD mode, the lower flux used during write could be insufficient to fully over-write all existing PC data which will have been laid down in the 'stronger' HD mode. It would also suggest that writing to a HD disc with a drive in DD mode may not fully saturate the higher coercivity coating of the HD disc which could explain why some people have reported that data subsequently vanishes.

Stay tuned...

Martin

User avatar
MartinB
Posts: 4555
Joined: Mon Mar 31, 2008 9:04 pm
Location: Obscurity

Re: high density disks

Postby MartinB » Thu Oct 16, 2008 12:14 am

Hello, me again :)

My research is finished and the truth was indeed out there – it just needed finding. For those following this I’ll cut the story back and try to keep it simple and relevant. My final discovery was, I think, pretty surprising and as a result I’ve tested the basis for a smart little hardware mod which will allow the humble Beeb (and possibly it’s relatives) to use a HD drive with HD discs exactly as they were designed to be used.

It’s important to say that I have collected information from dozens and dozens of sources and it’s impossible to attribute anything to anyone one person but I’m certainly not pretending that this is all my own work. I’ve already speculated about a lot of the information below but these are now the definitive facts. Trust me on this!

• HD (1.44Mb) discs have a higher coercivity than DD (720Kb) discs. This means that HD’s are harder (require more magnetism or flux) to write and erase. HD’s are thus classed as ‘less sensitive’ than DD’s since the latter are easier to write and erase.

• When a disc is inserted into a drive, the density detection switch uses the hole (or the lack of one) across from the write protect slider to determine the density of the disc.

• If the drive detects a DD disc it switches to a slow (250KHz) data rate between it and the FDC chip in the computer and it configures the write/erase heads to use weak magnetism - i.e. low head current and hence low flux. Conversely, if the drive detects a HD disc it switches to a fast (500KHz) data rate between it and the FDC chip in the computer and it configures the write/erase heads to use strong magnetism - i.e. high head current and hence high flux. Thus, a disc interface that is fixed in one density or data rate can only communicate with a drive of the same density. The Acorn 1770/1772 DFS for example is limited to DD or slow data rate only.

• On the thorny issue of ‘kidding’ a drive by blocking the hole of a HD disc – if a HD disc has been written to by a true high density drive, the recorded data (magnetic pulses) cannot be fully (often not even partly) erased or overwritten by a drive functioning in DD mode. This why pre-used or new but pre-formatted HD discs will fail in a DD drive or a drive forced to DD by covering the HD disc hole. One’s only chance to partly succeed here is to first erase the disc with a proprietary bulk degaussing device. This is NOT however a solution to cross-mixing disc and drive densities because…

• A DD drive or a HD drive forced to DD by covering the hole of a HD disc will not adequately magnetise the disc coating of a HD disc when writing data. Data so written will, from time zero, immediately begin to ‘fade’ and will eventually ‘disappear’. The time for this decay to cause the data to become unrecoverable will in practice vary from milliseconds to years because…

• Any practical variations experienced which appear to be contrary to the above are only due to variations in the actual achieved surface coercivity of a given disc or the actual achieved head flux of a given drive. If we take discs and drives from reputable manufacturers that meet the intended specifications within the expected tolerances then cross-mixing disc and drive densities will simply fail. Partial successes can only be attributed to discs and drives not fully meeting the HD/DD specification tolerances.

• During the evolution of drives and the 34-way interface that we still use, the (abridged) story of Pin 2 is interesting (to me anyway) and eventually relevant. When HD discs and HD drives came about, the previously unused pin 2 was brought into play and initially acted as a ‘Select Density’ input to the drive from the PC. This allowed the user, via the PC OS, to select the appropriate drive operation for the media in use. However, because PC manufacturers were slow to standardise, the media notch or hole was introduced to allow the drive to self-select the correct HD or DD mode. As a result, pin 2 slowly changed to an output indicating to the host computer which density media was in use and on some drives, the function of pin 2 was even link selectable between the two uses. Today, in many cases, pin 2 has become dormant again as FDC interfaces have become more intelligent and flexible whereby the PC can determine the density from the interface data rate although some drives still provide the selected density output.

So, that’s the DD/HD disc story and I think that everyone who has played in this area will find an explanation for their experiences, anecdotal or personal, somewhere above.

It's a bit late now so in another post I will explain how I've got one of my Beebs to happily use a 3.5” HD drive in HD mode with HD discs (used or otherwise) and, due to my DFDC mod (not essential though), why it only took me about 30 seconds to achieve! The full mod will be slightly more involved depending on how flexible one wants to be with combinations of drives and discs but it’s still quite straightforward. Fascinating stuff, especially if like me, you still enjoy using discs.

Martin

User avatar
retroclinic
Posts: 3016
Joined: Thu Jul 03, 2008 1:22 pm
Location: East Riding of Yorkshire
Contact:

Re: high density disks

Postby retroclinic » Thu Oct 16, 2008 8:56 am

Hi.

That's superb Matrin, well done.

Now for your next trick, can you please do the same for 5.25"? They have the issue of different spindle speeds between DD and HD, as well as double the coercivity. Also, no hole for auto detection, so they will rely on Pin 2 from the controller, but the data rate then changes because of the spindle speed. Also, HD 5.25" are not exactly double the capacity of DD, well not on a PC anyway.

I had hoped to fiddle with this on an Arc (A5000) that can run in HD mode for 3.5", so see if it can make 5.25" HD disks reliably, but I've baught 2 external Beebug interfaces from eBay now that both seem to be duff....grr....

Mark.

User avatar
sorvad
Posts: 2172
Joined: Wed Aug 24, 2005 12:13 pm
Location: Back of beyond

Re: high density disks

Postby sorvad » Thu Oct 16, 2008 1:02 pm

Amazing stuff Martin and a fascinating read.


Return to “hardware”

Who is online

Users browsing this forum: No registered users and 7 guests