can't flip disk on XF551?

[Marius]

Hi.

 

Today I received an ATARI XF551 diskdrive. In the past I had one, and finally I got the chance to buy an almost brandnew one. It works great, but i can not boot disk B side.

 

I am pretty sure this was possible with my xf551 years ago. I remember it was not possible to format them, but i'm rather positive I was able to boot the B-side.

 

How it is now, I'm not even able to watch B-side of demo's, or play summergames, which is a double sided disk.

 

Is there a fix for this, or is there something else wrong?

 

Thanks

Marius

 

p.s. small update: Inserting a flipped disk, makes the drive spinning eternal. Inserting the A side: the drive spins 1 or 2 secs and stops spinning. Inserting B side: disk spins forever.

 

Does this need a rom update? And if yes... how do I get the right rom? I can burn them myself.

Edited January 29, 2009 by Marius1976

[Urchlay]

p.s. small update: Inserting a flipped disk, makes the drive spinning eternal. Inserting the A side: the drive spins 1 or 2 secs and stops spinning. Inserting B side: disk spins forever.

 

Does this need a rom update? And if yes... how do I get the right rom? I can burn them myself.

 

From the Atari 8-bit FAQ:

 

Chinon-built XF551-cannot read/write/format backside if no timing hole

Mitsumi-built XF551--can read/write backside if no timing hole, if formatted

 

Both versions, unlike the 810/1050, actually use the timing hole... I don't know whether the difference is in the ROMs, the drive mechs, or both.

 

Probably the best person to ask about anything XF551-related is Metalguy66 (who will probably chime in here any minute now...)

[+CharlieChaplin]

Well,

reading the A8FAQ might help...

- Mitsumi drives: will read + write the backside of a(pre-) formatted disk, but will not format unless the disk has two index-holes or a kind of index-hole-redirection...

-Chinon drives: will NOT read+write+format the backside of a disk, unless the disk has two index holes or a kind of index-hole redirection; inserting a b-side of a disk with only one index hole lets the drive spin forever...

 

So, it looks like you have an XF551 with a Chinon drive... Solution: built-in an index-hole redirection !

-Andreas Koch.

[ijor]

Both versions, unlike the 810/1050, actually use the timing hole... I don't know whether the difference is in the ROMs, the drive mechs, or both.

 

In the mechanism and in the digital board logic.

 

The difference in the mechanism is what prevents you to access the flippy side at all in the Chinon units.

The different digital logic is what doesn't let you format the flippy side, not even on Mitsumi units.

 

The ROM is not too much relevant in this regard. None of these issues can be fixed with just a ROM replacement.

[Marius]

Well... that's not so good then. Thanks for your help folks. I shoulg get another mechanism then... And I was so happy with my brandnew xf 551

 

I don't know how to build a index-hole re direction :S

Edited January 29, 2009 by Marius1976

[+bob1200xl]

Does anyone know exactly what is the problem? The controller requires an INDEX pulse in order to format a disk, but not to read... Even the 1050 controller has this requirement, so the 1050 actually dummies up INDEX. Neither XF551 drive is going to give you INDEX, anyway - how could one work and the other not? The drives are 'dumb' for the most part. They just spin the disk and send the data to the controller. Has anyone tried swapping a Mitsumi drive into an XF551 with a Chinon FDD?

 

Bob

 

 

Both versions, unlike the 810/1050, actually use the timing hole... I don't know whether the difference is in the ROMs, the drive mechs, or both.

 

In the mechanism and in the digital board logic.

 

The difference in the mechanism is what prevents you to access the flippy side at all in the Chinon units.

The different digital logic is what doesn't let you format the flippy side, not even on Mitsumi units.

 

The ROM is not too much relevant in this regard. None of these issues can be fixed with just a ROM replacement.

[ijor]

Does anyone know exactly what is the problem? The controller requires an INDEX pulse in order to format a disk, but not to read... Even the 1050 controller has this requirement, so the 1050 actually dummies up INDEX. Neither XF551 drive is going to give you INDEX, anyway - how could one work and the other not? The drives are 'dumb' for the most part. They just spin the disk and send the data to the controller. Has anyone tried swapping a Mitsumi drive into an XF551 with a Chinon FDD?

 

Hi Bob,

 

These "newer" mechanisms aren't so dumb, they are actually pretty smart and have a mini-controller (sometimes called nano-controller) built-in. Among other things, this controller performs the motor stepping mechanism.

 

This mini-controller does detect the index hole, and depending on the drive and jumpers, it might inhibit normal operation if the index is not detected (it would assume there is no disk, or it is not rotating well).

Edited January 30, 2009 by ijor

[+kheller2]

They use "stock" 360K PC floppy drives. Stock as in circa late 80s. I never bothered to test different mechs though. I always assumed the XF mobo was identical and the drive mech was different, because I've used panasonic and other drives on the XF board.. I just never remember using flippies.

[ijor]

They use "stock" 360K PC floppy drives.

 

They are stock PC drives, but not all PC drives were exactly alike. PCs don't have the "flippy" issue, so being able or not to access the flippy side doesn't make it more or less PC compatible.

 

Btw, almost all PC 360K drives I've seen have jumpers that let configure the drive for accessing the flippy side. It would be very unusual that these jumpers are missing in the Chinon. But I assume it was already double checked that the Chinon was an exception?

[+Guitarman]

The 'timing hole' usage on a XF551 is for formatting purposes, here is a excerpt from an article from Antic.

 

For years, Atarians have been making what we call double-sided disks on our 810 and 1050 drives. This is done by notching the opposite side of the disk, flipping it over, and formatting the reverse side of the disk. Each side of this disk has its own separate directory, and each side is logically considered to be a separate disk.

 

The proper term for such a disk is "flippy", since you must flip the disk over to access side two.

 

The Atari XF551 has two read/write heads, top and bottom. When a disk is formatted Double Sided, Double Density by Atari DOS-XE, SpartaDOS, or MYDOS, it is a single disk, logically as well as physically, with only one directory which charts BOTH sides of the disk.

 

What's the difference, you ask? Well, there's no more disk-flipping. You can access a full 360K of data (both sides) without turning the disk over! The DOS and second disk head make access to side two of the disk completely transparent.

 

It is true that side two is written in the opposite direction, relative to side one, but you couldn't use this disk as a flippy anyway, since there is no directory information on the second side. Assuming it was written in the "same direction" as the flip side of a flippy, there would still be no reasonable way to access the data. Users have been making flippies for so long that Atari built some "protection" into the new XF551 drive mechanism. To prevent accidental formatting of the second side of a disk which is alreaay double-sided, the XF551 refuses to format the back side of a flippy disk--even if it is notched.

 

The idea is to break a bad habit and protect your data. It is a good idea, but can also be annoying.

 

The XF551 can read and write to the flip side of flippy disks as long as they are formatted elsewhere (on a 1050 or 810 for example).

 

I have never had a problem reading flippy disks from my 1050 drives on my various XF551's and according to this article, you shouldn't have.

[ijor]

The 'timing hole' usage on a XF551 is for formatting purposes, here is a excerpt from an article from Antic.

 

...Users have been making flippies for so long that Atari built some "protection" into the new XF551 drive mechanism. To prevent accidental formatting of the second side of a disk which is alreaay double-sided, the XF551 refuses to format the back side of a flippy disk--even if it is notched. The idea is to break a bad habit and protect your data. It is a good idea, but can also be annoying.

 

The XF551 can read and write to the flip side of flippy disks as long as they are formatted elsewhere (on a 1050 or 810 for example).

 

Guitarman,

 

With all due respect to whoever wrote that Antic article, I think he had no idea about what he was talking. Besides it is clear that the XF551 he had was a Mitsumi unit, and then he (wrongly) assumed they all would be the same.

[+Larry]

The 'timing hole' usage on a XF551 is for formatting purposes, here is a excerpt from an article from Antic.

 

(snip...)

 

I have never had a problem reading flippy disks from my 1050 drives on my various XF551's and according to this article, you shouldn't have.

 

Hi Guitarman-

 

Can you provide the date of the issue that you quoted from?

 

Thanks,

Larry

[+Larry]

Well,

reading the A8FAQ might help...

- Mitsumi drives: will read + write the backside of a(pre-) formatted disk, but will not format unless the disk has two index-holes or a kind of index-hole-redirection...

-Chinon drives: will NOT read+write+format the backside of a disk, unless the disk has two index holes or a kind of index-hole redirection; inserting a b-side of a disk with only one index hole lets the drive spin forever...

 

So, it looks like you have an XF551 with a Chinon drive... Solution: built-in an index-hole redirection !

-Andreas Koch.

 

Hi Andreas-

 

I have several of these, and have looked into this before. I've always come to the conclusion that this is not so easily done. Electronically, it's pretty easy, but mechanical installation is not (IMO).

 

Perhaps you know of a fairly simple way?

 

I do have a pretty rare XF551 with the CSS Enhancer that restores "flippy" function. But other than that, I've found the simplest solution if to make my own "flippies." This is also the reason that I think the best use of an XF551 is to make a 720K drive so there is no issue with the "B" side of a disk. The "Hyper-XF" (720K) is quite a nice mod!

 

-Larry

[sup8pdct]

I do have a pretty rare XF551 with the CSS Enhancer that restores "flippy" function. But other than that, I've found the simplest solution if to make my own "flippies." This is also the reason that I think the best use of an XF551 is to make a 720K drive so there is no issue with the "B" side of a disk. The "Hyper-XF" (720K) is quite a nice mod!

 

-Larry

 

I have one of those too. very handy.

It uses the AC input to do the timing, so there are 2 versions, 50Hz and 60Hz.

It shouldn't be to hard to build one.

James

[Marius]

Yesterday I 'killed' a few of my Floppy Disks by cutting a part out of it, to create some kind of 'eye' in it, so the index hole is visible at the 'right side' .

 

Those disks work now indeed on the B-side. So this is the final prove (for those who did not believe what was written in this thread) that my XF551 is using that index hole for reading too.

 

I have been thinking about this issue, and I think I could best get myself another mechanism. And if i do not succeed on that, I will store all b-sides of disks on a fresh new disk, only using the A-side. I have plenty of 1050's, so I could do that. But that solution stinks ofcourse.

 

The only reason I bought this brandnew XF551, was the idea that an XF551 does not use a drive belt, so it should last theoretical longer than a 1050. So I wanted to be 'safe' in future. Hmm only for 50% for now The A-sides are safe in future.

 

Is this a right idea btw. will the XF551 last longer? Or is this just an idea?

 

Marius

[Urchlay]

Is this a right idea btw. will the XF551 last longer? Or is this just an idea?

 

I don't know, but I bet it'll be easier to find replacement PC drive mechs than 1050 ones, a few years from now (right now the 1050s are easy to get, Best/B&C stocks them)

 

Even though the 5 1/4" PC drive mech is considered obsolete, they were way more common than the 1050 variety, and you'll probably still be able to find new-old-stock ones 10 years from now. That, or go to any computer salvage yard or used computer store and you're likely to find a few that work, that have been sitting quietly for years...

[+Guitarman]

Larry,

 

Here is the Antic info and it also links to the article. It is actually a review of the XF Enhancer.

 

ANTIC VOL. 8, NO. 1 / MAY 1989 / PAGE 6

 

I guess I got lucky with my XF551's!!!

 

Hi Guitarman-

 

Can you provide the date of the issue that you quoted from?

 

Thanks,

Larry

[MEtalGuy66]

I have 10 of these drives.. One has the XF enhancer in it. All of mine have the factory "NTSC timing" roms in them.. But I have tried several replacement EPROMs that I burned from images I got from the net (eg. Hyper-XF, etc.).. Dont any of these "alternative" roms attempt to adress the "flippy issue" to some degree?

Edited January 30, 2009 by MEtalGuy66

[Tempest]

So the issue isn't that the XF551 can't read/write to the flip side of a disk, it's that it can't perform the initial format of the second side? According to that article, if you have a flippy that was created on a 1050 or 810 it should work just fine with the XF551.

 

Tempest

[Marius]

So the issue isn't that the XF551 can't read/write to the flip side of a disk, it's that it can't perform the initial format of the second side? According to that article, if you have a flippy that was created on a 1050 or 810 it should work just fine with the XF551.

 

Tempest

 

That article is not complete!

 

Read the post of CharlieChaplin.

 

The 'youngest' XF551 drives have another mechanism. The Chinon mechanism! These drives are not even able to READ the backside. So you can NOT swap a disk. Not for reading, not for writing, not for formatting!

 

I'm really surprised atari did release such drive. There are lot's of original titles using A and B side like two different disks.

 

The other mechanism (Mitsumi) is able to read and write the backside of the disk, but it can not format.

unfortunately my xf has this Chinon mechanism.

[Tempest]

So the issue isn't that the XF551 can't read/write to the flip side of a disk, it's that it can't perform the initial format of the second side? According to that article, if you have a flippy that was created on a 1050 or 810 it should work just fine with the XF551.

 

Tempest

 

That article is not complete!

 

Read the post of CharlieChaplin.

 

The 'youngest' XF551 drives have another mechanism. The Chinon mechanism! These drives are not even able to READ the backside. So you can NOT swap a disk. Not for reading, not for writing, not for formatting!

 

I'm really surprised atari did release such drive. There are lot's of original titles using A and B side like two different disks.

 

The other mechanism (Mitsumi) is able to read and write the backside of the disk, but it can not format.

unfortunately my xf has this Chinon mechanism.

 

Hmm... I'll have to check what kind of XF551 I have then, otherwise it's back to my 810.

 

Tempest

[MEtalGuy66]

The 'youngest' XF551 drives have another mechanism. The Chinon mechanism! These drives are not even able to READ the backside. So you can NOT swap a disk. Not for reading, not for writing, not for formatting!

 

I'm really surprised atari did release such drive. There are lot's of original titles using A and B side like two different disks.

 

Im not surprised at all.. XF 551s dont even work reliably with NTSC 130XE machines unless you remove some (or all) of the filter caps from the SIO port in the XE.. Early NTSC/PAL XF551 ROMs are different and wont work reliably on systems of the opposite "standard" due to differences regarding a timing kludge in the firmware.. The SIO port connectors on XF551 mainboards ALL need to be resoldered/reinforced (All 10 of mine had micro-cracks in the solder joints surrounding the connector solder tails.) I can go on and on about stupid mistakes in design and/or build quality on the XE generation of hardware. ATARI did some really careless sh!t in the mid-late 80s..

[+bob1200xl]

Yes, that may be the case. I was wondering just what exactly did the drive fail to do if it didn't have see INDEX. Perhaps it can be fixed. I can't see anything that the drive could do to prevent reading except de-gate the data to the controller. That would be a little weird...

 

The one thing I do see is the 'm' flag - spin-up delay. If that is set in the controller command, the controller waits for six (!) INDEX pulses before it reads. In that case, you would not be able to read a 'flippy'. I know the early drives did not use spin-up because a 3.5 drive will write bad sectors if you do not - and, they did. On a Read, however, it matters not.

 

So, if we swap a drive that cannot read flippies into an XF551 that does read flippies and the drive can not now read them, it's a drive issue. If it can read flippies, it's probably spin-up.

 

Bob

 

 

 

Does anyone know exactly what is the problem? The controller requires an INDEX pulse in order to format a disk, but not to read... Even the 1050 controller has this requirement, so the 1050 actually dummies up INDEX. Neither XF551 drive is going to give you INDEX, anyway - how could one work and the other not? The drives are 'dumb' for the most part. They just spin the disk and send the data to the controller. Has anyone tried swapping a Mitsumi drive into an XF551 with a Chinon FDD?

 

Hi Bob,

 

These "newer" mechanisms aren't so dumb, they are actually pretty smart and have a mini-controller (sometimes called nano-controller) built-in. Among other things, this controller performs the motor stepping mechanism.

 

This mini-controller does detect the index hole, and depending on the drive and jumpers, it might inhibit normal operation if the index is not detected (it would assume there is no disk, or it is not rotating well).

[+CharlieChaplin]

And err,

if you are asking where Atari got the idea to use Chinon drives instead of Mitsumi in the XF551... the Chinon drives were common for the Commodore Amiga and at some point/time they were cheaper to have than the Mitsumi drives, so that is why... -Andreas Koch.

 

P.S.: Maybe that index hole problem is the reason why Rainbird put their adventures Guild of Thieves, Jinxter and the Pawn on two diskettes with only diskside A being used (on both disks)...

Edited January 30, 2009 by CharlieChaplin

[ijor]

Yes, that may be the case. I was wondering just what exactly did the drive fail to do if it didn't have see INDEX. Perhaps it can be fixed. I can't see anything that the drive could do to prevent reading except de-gate the data to the controller. That would be a little weird...

 

It is not too wierd. From "their" point of view, it is wierd to access the flippy side The idea, again, is to forbid accessing a disk that is not rotating, or not rotating well. It makes sense. Besides, it is fully documented and described in some drive's datasheets.

 

Yes, probably all that it does is to disconnect some signals from the bus. Mainly they should be READ,WRITE, WGATE and conceivable WPROT in some cases. It might be possible to fix it. As I was saying, almost any DD drive I've seen have jumpers to configure this "READY condition". Conceivable they are there on the Chinon as well but not populated.

 

The one thing I do see is the 'm' flag - spin-up delay ... I know the early drives did not use spin-up

 

No XF ROM (except perhaps ones dedicated to 3.5 upgrades) uses the spin-up delay.

 

The 'youngest' XF551 drives have another mechanism. The Chinon mechanism!...I'm really surprised atari did release such drive.

 

I'm not sure the Chinon are the youngest units. I understand that most drives marketed in the US had a Mitsumi mechanism, and that most Chinon units were sold overseas during some period.

 

Dont any of these "alternative" roms attempt to adress the "flippy issue" to some degree?

 

Some third party ROMS have a better handling of the flippy side, but no ROM replacement can fix the two main hardware issues (not without some hardware mod).

 

XF 551s dont even work reliably with NTSC 130XE machines unless you remove some (or all) of the filter caps from the SIO port in the XE..

 

I think this is true for most other drives that use high speed SIO, not just the XF551. The filter caps would limit the max frequency on the SIO data line. That's ok for the "standard" 19,200 rate, but causes troubles on higher speeds. Some enhacements, like the Happy, even document that these caps sometimes must be removed.