Dying XF551

[NuY]

My XF551 appears to be dying, hoping someone might be able to assist...

 

Two days ago, turned my XE on to play on my new Bombjack cartridge. Worked fine, decided to play some old disk games. Wouldn't load.

 

Long story short after dismantling the drive and head, and giving the head a good clean with white spirit, it is now sort of working.

 

I currently have the case off, and after a lot of experimenting and rebuilding, I've found that in DOS (any, but I use SpartaX 4.x) by trying to copy files to the RAM drive, it will copy a few files and then stop, with the Atari emitting an SIO bleep every 2 seconds or so. When it does this the drive spins the disk but won't seek. If I let it stop here, I end up with either a Device NAK or SIO framing error.

 

However, if when it stops, i manually move the drive head back to the start of the disk (either pushing the head towards the back of the drive or twisting the stepper motor to get the same effect) the drive will start reading again most of the time, and will proceed on with the copy operation. If I try another file operation once I have received either of the above errors, all I get is the same result; SIO bleeps every 2 seconds or so for about 10-15 seconds, and the error message. The drive won't perform any file operation until the head is sent back to the beginning of the disk.

 

I'm unable to boot a disk for the same reason; if the drive starts to read at all, it will get so far and then choke by doing the same thing.

 

Obviously having the drive open and moving the stepper manually is not a good situation. I'm no solderer, and I can take things apart and put them back together again, but this one is a bit beyond me. I've only disassembled the drive as far as getting the case, top shield and drive head off at this moment.

 

Anyone able to assist on a possible fix?

 

Edit: Forgot to mention, the drive is a Mitsumi mech, if that makes any difference.

Edited March 18, 2010 by NuY

[tregare]

My XF551 appears to be dying, hoping someone might be able to assist...

 

Two days ago, turned my XE on to play on my new Bombjack cartridge. Worked fine, decided to play some old disk games. Wouldn't load.

 

Long story short after dismantling the drive and head, and giving the head a good clean with white spirit, it is now sort of working.

 

I currently have the case off, and after a lot of experimenting and rebuilding, I've found that in DOS (any, but I use SpartaX 4.x) by trying to copy files to the RAM drive, it will copy a few files and then stop, with the Atari emitting an SIO bleep every 2 seconds or so. When it does this the drive spins the disk but won't seek. If I let it stop here, I end up with either a Device NAK or SIO framing error.

 

However, if when it stops, i manually move the drive head back to the start of the disk (either pushing the head towards the back of the drive or twisting the stepper motor to get the same effect) the drive will start reading again most of the time, and will proceed on with the copy operation. If I try another file operation once I have received either of the above errors, all I get is the same result; SIO bleeps every 2 seconds or so for about 10-15 seconds, and the error message. The drive won't perform any file operation until the head is sent back to the beginning of the disk.

 

I'm unable to boot a disk for the same reason; if the drive starts to read at all, it will get so far and then choke by doing the same thing.

 

Obviously having the drive open and moving the stepper manually is not a good situation. I'm no solderer, and I can take things apart and put them back together again, but this one is a bit beyond me. I've only disassembled the drive as far as getting the case, top shield and drive head off at this moment.

 

Anyone able to assist on a possible fix?

 

Edit: Forgot to mention, the drive is a Mitsumi mech, if that makes any difference.

 

you removed the -head- ??? that is generally enough to muck up the alignment. do you have another mechanism to try in the drive? is the 5 and 12 V power +/- 5%.

[NuY]

you removed the -head- ??? that is generally enough to muck up the alignment. do you have another mechanism to try in the drive? is the 5 and 12 V power +/- 5%.

When it's reading, it reads fine. It just stops reading randomly after file I/O. I had a 1050 do the same thing a couple of years ago, and cleaning the head did the trick.

 

Unfortunately I don't have a multimeter to test voltages, and no other mech either.

[+bob1200xl]

Generally enough? You could win the lottery before you would get it aligned again without special tools. Let's hope he didn't do the head itself. (or the carriage)

 

Anyway, try formatting a disk on the drive. If the drive can read his own disks (ones that he has formatted) but not disks formatted at the factory or on another drive, it is normally an alignment problem - you're pretty much out of luck. (as far as the drive goes - you would need a new one) The carriage (that moves the head) gets slop in it and the head falls off track, which you can sometimes 'help' by putting pressure on the head in its direction of motion. (which may be your case)

 

If the drive can't read its own disks, it may be the controller chip (the 1772). (or, it could still be the drive)

 

No easy answers, but make sure that the head is clean. A Q-tip isn't a really good scrubber. The ceramic on the head is pretty hard - you won't hurt it. Check your disks, too, or you will just propagate the oxide from diskette to head to diskette. You can wreck you whole library that way. Be careful.

 

 

Bob

 

 

 

 

 

My XF551 appears to be dying, hoping someone might be able to assist...

 

Two days ago, turned my XE on to play on my new Bombjack cartridge. Worked fine, decided to play some old disk games. Wouldn't load.

 

Long story short after dismantling the drive and head, and giving the head a good clean with white spirit, it is now sort of working.

 

I currently have the case off, and after a lot of experimenting and rebuilding, I've found that in DOS (any, but I use SpartaX 4.x) by trying to copy files to the RAM drive, it will copy a few files and then stop, with the Atari emitting an SIO bleep every 2 seconds or so. When it does this the drive spins the disk but won't seek. If I let it stop here, I end up with either a Device NAK or SIO framing error.

 

However, if when it stops, i manually move the drive head back to the start of the disk (either pushing the head towards the back of the drive or twisting the stepper motor to get the same effect) the drive will start reading again most of the time, and will proceed on with the copy operation. If I try another file operation once I have received either of the above errors, all I get is the same result; SIO bleeps every 2 seconds or so for about 10-15 seconds, and the error message. The drive won't perform any file operation until the head is sent back to the beginning of the disk.

 

I'm unable to boot a disk for the same reason; if the drive starts to read at all, it will get so far and then choke by doing the same thing.

 

Obviously having the drive open and moving the stepper manually is not a good situation. I'm no solderer, and I can take things apart and put them back together again, but this one is a bit beyond me. I've only disassembled the drive as far as getting the case, top shield and drive head off at this moment.

 

Anyone able to assist on a possible fix?

 

Edit: Forgot to mention, the drive is a Mitsumi mech, if that makes any difference.

 

you removed the -head- ??? that is generally enough to muck up the alignment. do you have another mechanism to try in the drive? is the 5 and 12 V power +/- 5%.

[NuY]

The reason I think the head isn't out of alignment is that it will read files from all the disk I've tried, both ones formatted on the XF and others formatted on my 1050. It will read any file on any of my disks, but only if i set the carriage back to the beginning of the disk physically, otherwise after a few file I/Os, it just bleeps and gives the errors I mentioned. Also, all file copies that I've tried work fine, the destination files have absolutely no errors on at all.

 

The head seeks for a few file I/Os, but it's as if the drive is hanging, and the only way to reset it is to physically move the carriage back to the start of the disk.

[+bob1200xl]

When you move the head to the start of the disk, you are returning it to track 0, a physical constant on the carriage. There, it can read because it is now aligned on the center of the track. Once you seek off of that track, you may lose control of the head alignment out in the middle of the disk. The drive is responsible for stepping from center-to-center of each track. The controller just says "seek one step". If the drive gets lost, it's broke...

 

Could something be interfering with the stepper motor or carriage?

 

Bob

 

 

 

 

The reason I think the head isn't out of alignment is that it will read files from all the disk I've tried, both ones formatted on the XF and others formatted on my 1050. It will read any file on any of my disks, but only if i set the carriage back to the beginning of the disk physically, otherwise after a few file I/Os, it just bleeps and gives the errors I mentioned. Also, all file copies that I've tried work fine, the destination files have absolutely no errors on at all.

 

The head seeks for a few file I/Os, but it's as if the drive is hanging, and the only way to reset it is to physically move the carriage back to the start of the disk.

[NuY]

When you move the head to the start of the disk, you are returning it to track 0, a physical constant on the carriage. There, it can read because it is now aligned on the center of the track. Once you seek off of that track, you may lose control of the head alignment out in the middle of the disk. The drive is responsible for stepping from center-to-center of each track. The controller just says "seek one step". If the drive gets lost, it's broke...

 

Could something be interfering with the stepper motor or carriage?

 

Bob

There doesn't appear to be anything physical in the way of it, but what i find weird is that it can seek right to the end track 39, read a file, go to the middle of the disk, read another file during copy, and then it will stop. I don't get it...

[+bob1200xl]

When it fails and is just spinning, try putting just a little bit of pressure on the carrige - push it either in or out. See if it finds the track with a little help.

 

Mechanical things rarely just break in half and die - they wear out and get intermittent. The lubricant evaporates. Dust and dirt collect. Even IC failures may be mechanical, on a much smaller scale.

 

There used to be a card machine at IBM that if you filled it with concrete and shoved it into the ocean, some would swear that it would sink, intermittently...

 

Bob

 

 

 

When you move the head to the start of the disk, you are returning it to track 0, a physical constant on the carriage. There, it can read because it is now aligned on the center of the track. Once you seek off of that track, you may lose control of the head alignment out in the middle of the disk. The drive is responsible for stepping from center-to-center of each track. The controller just says "seek one step". If the drive gets lost, it's broke...

 

Could something be interfering with the stepper motor or carriage?

 

Bob

There doesn't appear to be anything physical in the way of it, but what i find weird is that it can seek right to the end track 39, read a file, go to the middle of the disk, read another file during copy, and then it will stop. I don't get it...

[tregare]

it could also be cracks in the traces or bad solder joints on the PCB, those seem to be common failures with the XF551 boards.

[NuY]

When it fails and is just spinning, try putting just a little bit of pressure on the carrige - push it either in or out. See if it finds the track with a little help.

 

Mechanical things rarely just break in half and die - they wear out and get intermittent. The lubricant evaporates. Dust and dirt collect. Even IC failures may be mechanical, on a much smaller scale.

 

There used to be a card machine at IBM that if you filled it with concrete and shoved it into the ocean, some would swear that it would sink, intermittently...

 

Bob

i just gave this a try, and no dice - still just bleeps a few times and errors out. I pushed slightly too hard back on one attempt however and strangely, it started reading again as it has, but dropped into low speed mode. Odd.

[NuY]

it could also be cracks in the traces or bad solder joints on the PCB, those seem to be common failures with the XF551 boards.

I'll be honest, I wouldn't know what I'm looking for, other than a gaping hole in a joint! I'll dismantle the whole lot tomorrow and have a look though.

 

It has been a bit flaky in recent weeks, taking a few seconds to load up when booting from disks, occasionally with the Atari saying "No DOS", but hitting Reset has sorted that out. It's been idle for about two weeks before this occurred.

[+bf2k+]

it could also be cracks in the traces or bad solder joints on the PCB, those seem to be common failures with the XF551 boards.

I'll be honest, I wouldn't know what I'm looking for, other than a gaping hole in a joint! I'll dismantle the whole lot tomorrow and have a look though.

 

It has been a bit flaky in recent weeks, taking a few seconds to load up when booting from disks, occasionally with the Atari saying "No DOS", but hitting Reset has sorted that out. It's been idle for about two weeks before this occurred.

 

Pay close attention to the solder joints where the ribbon cable connects to the PCB. 2 of mine failed there and I fixed them. I don't remember the exact symptoms of the failures...

[MEtalGuy66]

Ok.. I got my latest XF551 off EBAY.. The track zero sensor was bad.. Fixed it with a sensor from another REALLY DEAD Mitsumi mech.

 

But This makes me wonder about some things..

 

First off, I've seen the same sensor go bad in 1050s.. And from what I understand, they are unavailable nowadayze without buying the whole mech.

 

Also, PC/XT 5.25" drive mechs are getting more rare/expensive these days than they used to be. Hardly worth buying one just to "part out"..

 

Theres articles on the web talking about people just buying the LED and photoresistor separately, and MAKING a new sensor.. I was wondering if you could get the sensor out of an old 3.5" drive.. Or maybe an old ball mouse. I mean, assuming you could fabricate some little bracket or adaptor to mount the thing, and then get it lined up correctly, and wire it in correctly... Would it work? Are these type of sensors pretty much universal in electrical characteristics?

[Rybags]

I reckon they probably are, maybe the voltages used might vary depending on origin.

 

Fairly sure somewhere I've got an old (probably dead) paper-out sensor off a mainframe printer. They used the same principle, just an optical sensor that returned the presence or otherwise of the light beam (normally hidden by the paper).

 

Would some other alternative work? Like maybe some sort of leaf-switch arrangement such that it's tripped on Track zero. Of course, being a mechanical type thing, it'd probably be less reliable.

[MEtalGuy66]

Oh yeah. Printers have lots of them.. Even ink jets.. One for head travel zero position.. one to tell if the op cover is closed or not.. probably another for paper-out, but some of those are mechanical, depending on the design..

 

Problem with most modern crap is that the damn sensor is surface mounted.. not even thru-hole.. So youd have to make a small pcb to mount it on at a right angle...

 

Hey by the way.. The description of how the drive is acting in the first post of this thread sounds exactly like the way this drive I just got was acting, and I fixed it by replacing the track zero sensor.

 

Heres something to check for:

If you turn the drive off and on, repeatedly, each time you turn the drive on the head creeps forward a little more and does not return back to zero.. If it does that, it's a bad track zero sensor for sure.. 1050s behave the same way.

Edited March 31, 2010 by MEtalGuy66

[+bob1200xl]

The track zero sensor in a 1050 (and, maybe an XF551 - don't know) is a standard optical interrupter with a little different mounting ears. The actual semiconductor is in what looks like a TO-92 mold so you can pull the LED/diode out of any OI and replace the bad one in your housing with a good one. I have even seen them with the 1050 mounting ear configuration. Check around (Digi-Key). As far as regular OIs for parts, I have a zillion spares, if anyone needs some.

 

Bob

 

 

 

Ok.. I got my latest XF551 off EBAY.. The track zero sensor was bad.. Fixed it with a sensor from another REALLY DEAD Mitsumi mech.

 

But This makes me wonder about some things..

 

First off, I've seen the same sensor go bad in 1050s.. And from what I understand, they are unavailable nowadayze without buying the whole mech.

 

Also, PC/XT 5.25" drive mechs are getting more rare/expensive these days than they used to be. Hardly worth buying one just to "part out"..

 

Theres articles on the web talking about people just buying the LED and photoresistor separately, and MAKING a new sensor.. I was wondering if you could get the sensor out of an old 3.5" drive.. Or maybe an old ball mouse. I mean, assuming you could fabricate some little bracket or adaptor to mount the thing, and then get it lined up correctly, and wire it in correctly... Would it work? Are these type of sensors pretty much universal in electrical characteristics?

[+bob1200xl]

The 1050 uses a Clairex CLI810W device. Digi-Key has a Panasonic CNZ1023 that may be the same. $1.35... 4,000 of them in stock.

 

Bob

 

 

 

Oh yeah. Printers have lots of them.. Even ink jets.. One for head travel zero position.. one to tell if the op cover is closed or not.. probably another for paper-out, but some of those are mechanical, depending on the design..

 

Problem with most modern crap is that the damn sensor is surface mounted.. not even thru-hole.. So youd have to make a small pcb to mount it on at a right angle...

 

Hey by the way.. The description of how the drive is acting in the first post of this thread sounds exactly like the way this drive I just got was acting, and I fixed it by replacing the track zero sensor.

 

Heres something to check for:

If you turn the drive off and on, repeatedly, each time you turn the drive on the head creeps forward a little more and does not return back to zero.. If it does that, it's a bad track zero sensor for sure.. 1050s behave the same way.

[MEtalGuy66]

Well, the one in the XF551 Mitsumi mech has one of those mylar ribbons.. I didnt even think about trying to pop apart the little black housing.. I'll look at that when I get home.. I still have the old one..

 

You know, i't really too bad ATARI used these fragile Mitsumi mechs.. Some of the earlier half-height drives they put in XT/CLONES a few years earlier were really built like tanks by comparisson. Of course they probably draw a little more current as well, but they use real wires and connectors for everything, even the heads.. And theres more/thicker/heavier metal in them as well..

 

I have one that I got out of one of those old Compaq "luggables".. The ones that are about like one of those old "portable" sewing machines from the 70s.. heh. Anywayze.. It was a half-height mech with a full-height face-plate on it. Its tough as nails.. All metal hardware.. All thru-hole PCBs, big heavy-assed stepper & spindle motors. It's my "test drive" and when Its not being used to test anything, I use it with my BB/Floppyboard.. I Stuck it in an old "Apple Disk II" aluminum enclosure just as a joke, But I kinda like it..

Edited March 31, 2010 by MEtalGuy66

[tjlazer]

Does anyone have a list of compatible 5 1/4" PC Mechs that will work on the XF551?

[NuY]

Oh yeah. Printers have lots of them.. Even ink jets.. One for head travel zero position.. one to tell if the op cover is closed or not.. probably another for paper-out, but some of those are mechanical, depending on the design..

 

Problem with most modern crap is that the damn sensor is surface mounted.. not even thru-hole.. So youd have to make a small pcb to mount it on at a right angle...

 

Hey by the way.. The description of how the drive is acting in the first post of this thread sounds exactly like the way this drive I just got was acting, and I fixed it by replacing the track zero sensor.

 

Heres something to check for:

If you turn the drive off and on, repeatedly, each time you turn the drive on the head creeps forward a little more and does not return back to zero.. If it does that, it's a bad track zero sensor for sure.. 1050s behave the same way.

My 1050 behaves like that, but the XF never has. the drive light comes on very briefly and then off again both on power on and off, it spins, but the head doesn't move, but it's always had that same behaviour. Whether that's normal or not I don't know...

 

I've had a look at replacement 360k drives, but there seem to be none in Europe, never mind the UK. Out of the drives currently available on eBay, could you suggest a specific one to go for?

[MEtalGuy66]

eBay Auction -- Item Number: 120552076730

[+bf2k+]

eBay Auction -- Item Number: 120552076730

 

Is this a 360k drive?

[tjlazer]

Do all 360k 5.25" drives work? Or is it a hit and miss? I tried one from my Commodore A2000 bridgeboard and it works!

[flashjazzcat]

Pity something like this couldn't be fitted:

 

eBay Auction -- Item Number: 370326407214

[puppetmark]

Pity something like this couldn't be fitted:

 

eBay Auction -- Item Number: 370326407214

 

 

Some can, I did it with a Teac SD-505. I had to alter the XF551 ROM and solder a jumber to put the 5.25 drive in 360k mode and it only allowed me to read the Atari single and enhanced density disks. I could not write to them. I could read and write to any other density in 5.25 or 3.5.

Edited April 26, 2010 by puppetmark