Floppy disk portable music player

I've just be having a play about with an old floppy drive frm my junk pile, and I've noticed a couple of issues I'd forgotten about.
First off is the slight issue of the spindle motor's speed, it rotates at about 360RPM.....which doesn't help to fit a lot of audio on a limited size disk. This isn't too hard to fix, but it does add another step to be worked out.

Another issue is the need to keep an accurate pace on the drive head, and if it doesn't start the track in exactly the right place at the begining of the disk, it will quickly lose sync with the audio track. Again, not too dificult to sort out, but will require some thinking about.

The other issue is that I forgot about the fact that tape heads have two seperate heads for the left and right audio tracks - floppy drives don't.... They do however have two sides, and two heads per disk, so all is not be lost just yet.


All that got me to thinking about the digital route again, MP3 would solve all those issues. That's when I had an idea - what about gutting a floppy drive, adding the guts of a cheap MP3 player (one that can use some form of seperate flash memory storage - SD cards, USB flash drive, etc), then modifying a floppy disk to hold the flash storage chip! (the floppy is just big enough to hold common flash storage devices inside) The flash device contacts can be hidden behind the floppy disk's metal sliding shield, this slides back as it enters the drive. So from the outside it will look completely standard.
The floppy disk is inserted into the floppy drive, the contacts of the flash storage device line up with the connector on the MP3 player and plug-in as normal, then the player is used in the normal way. Perfect - changing songs is as easy as ejecting the "disk" and sliding in a new one, and it still will look like a bog standard floppy drive and disk.

SS

Thats a good idea SS it wouldnt quite be the same but I might have to resort to that

There are actually a number of cheap MP3 kits that would be ideal for this project, such as the hardware used for this project:
http://www.ladyada.net/make/minty/

The idea of using the floppy disk to hold music is still on the cards, but it's getting a little more complicated - and I'm not sure how much you want to have to start fabricating bits...

SS

Siliconous Skumins wrote: I'm not sure how much you want to have to start fabricating bits...

SS

What kind of bits?

JingleJoe wrote:

Siliconous Skumins wrote: I'm not sure how much you want to have to start fabricating bits...

SS

What kind of bits?

Actually, that may not be the case afterall..... Been doing a little thinking about this, it may be possible to use the disk drive machanism, mostly 'as is'. Assuming I can figure out a way to change the spindle motor speed.....and possibly convert it from CAV to CLV, to increase the ammount that can be recorded per disk (at the stock speed, you would probably get around 30 seconds or less...). Cassette tapes move at 4.76 cm/s, and I will need to get somewhere close to that speed, though a little slower should be OK.

Don't suppose I could possibly interest you in lugging around a 51/4" or even an 8" floppy drive, could I? The extra space on the disk would allow for a much greater recording time...

SS

I would be interested in that but I dont have those big disks nor do I have thier corresponding drives, but any floppy drive will do, the bigger ones have more retro cool points too

SS[/quote]


Don't suppose I could possibly interest you in lugging around a 51/4" or even an 8" floppy drive, could I? The extra space on the disk would allow for a much greater recording time...

SS[/quote]

The old 8" floppy was designed to be a low density, high readablility medium. The design was introduced in 1971 by IBM, and had as a formatted capacity of 79.7KB.
(IBM 23FD was the model). The main use for these floppies was as the microcode load for devices such as 3270 controllers and some early DASD controllers.
The low density allowed for "large" bit tracks, and that increased the likelihood of a "correct" read in adverse conditions (dust, vibration, heat, etc.).
I can remember using these and creating custom load disks for equipment back in my IT early days in the 1970's
Later incarnations of the 8" floppy had high capacites, but even the last model introduced in 1977 (if I remember correctly) had a max capacity of 1.2MB if formatted to FAT, and 980K if formatted in a CP/M format.
It wasn't just the head size that kept tehse large format floppies at low capacity, the oxide film was not capable of higher bit densities, I can remember using magnetic tape developing fluid on one of tehse, and you could see the bit pattern with the naked eye. (tape developer was a suspension of very fine iron particles in alcohol, putting it on a piece of mag tape , the alcohol woudl evaporate, leaving the iron dust to adhere to the magnetized bits of tape. It was a diagnostic tool, in that you could check stuff like the inter-block gap on tape without the need for an oscilliscope hooked up to the drive electronics..)

Cheers
Harold

I would LOVE to see a 5 1/4" MP3 player - You would have to compress the heck out of the file just to get one song on a disk lol

And HAC, once again, your technical knowledge astounds me!

TehZorch wrote:I would LOVE to see a 5 1/4" MP3 player - You would have to compress the heck out of the file just to get one song on a disk lol

Actually, it would be easier to use a 5 1/4" disk - the extra pysical size of the disc media would allow the use of a smaller r/w head, such as that found on the 3.5" floppy drive. That means you effectivly increase the the storage space, and you could probably get around 15 minutes or more, of MP3 playback. However, as HAC points out, the disc media isn't up to it (analogue audio is maybe another matter though).


Mind you, you could always turn the flopy disc into a mini gramaphone record.... Computing steampunk style!

SS

Maybe not a floppy disc, but there is a Gakken kit to construct a primitive recording Emile Berliner Gramophone which uses 3.5" plastic discs to record onto.
It may be possible to re-task an old FDD into a playback unit for these discs, but you'd essentially be rebuilding the drive from the ground up.

Well, so far I have figured out that floppy drive can be wired up in such a way that it will move the head one track step per revolution of the disc, and that the head direction can easily be controlled. Still having problems getting the spindle motor slower - my floppy drive is a more recent one, as such the motor controller chip is apparently a custom / cheap non-descript item that I have been unable to find any data sheets for...
I have a couple of others somewhere, I'll check them out when I find them.

Anyhow, provided the motor can be made slower (or better yet, used in constant linear velocity mode rather than the standard constant angular velocity) this will allow the disk to store much more audio (don't know how much yet), but if stereo audio is dropped and instead a mono track is used, then BOTH sides of the disk are available as a 'continuous' track by simultaneously switching the head direction and swaping to the second head / "B" side (on the other side of the disk), at the exact moment the head reaches the end of the last track on the "A" side. This is easier than it sounds - the drive produces an index pulse once per revolution of the disc (used to time the track change of the head, by the drive controller). This can be used to monitor the head position (track number), the start / end of each track, and as the trigger to start reading the "B" side of the disc, by counting the index pulses.

Basically it will at least double the ammount of recording space available. A small micro controller might be needed to control the drive, but other means may be possible. A simple and cheap PIC chip should have more than enough power to do this task.

Still trying to work out a way to get stereo audio on a single track (I have an idea based on a MUX, but I need to test to see if it's possible).


I can post up the info for what pins control what, on the drive's cable connector, if you wish to experiment yourself.

SS

Siliconous Skumins wrote:...Still having problems getting the spindle motor slower - my floppy drive is a more recent one, as such the motor controller chip is apparently a custom / cheap non-descript item that I have been unable to find any data sheets for...
I have a couple of others somewhere, I'll check them out when I find them.

Anyhow, provided the motor can be made slower (or better yet, used in constant linear velocity mode rather than the standard constant angular velocity)...

I was under the (perhaps mistaken) impression that most floppy drive spindle motors were just stepper motors. That being the case, I also have a dim recollection of a commonly-available driver IC which took an external clock signal to control step frequency. (The L297/L298 combo used in the RepRap might do the trick.)

von Corax wrote:I was under the (perhaps mistaken) impression that most floppy drive spindle motors were just stepper motors. That being the case, I also have a dim recollection of a commonly-available driver IC which took an external clock signal to control step frequency. (The L297/L298 combo used in the RepRap might do the trick.)

Nah, the spindle motor is just a very accurate brushless 'pancake' type motor, the driver chip gets a signal pulse from the motor, which is used to adjust the RPM. The head uses a stepper motor though, an 18 degree / 20 step per revolution motor in 3.5" drives.

Usually you can just reduce the voltage on a brushless motor like those in computer fans to slow it down, but the floppy driver chip stops working if there isn't enough voltage or current.
I just need to find out which pin(s) are used for the clock signal from the motor (I know there is a hall effect sensor in there for the Index signal, but I'm unsure if the same sensor is used for the RPM clock signal), and by altering the clock signal, it should adjust the speed of the motor. Depending on how the motor RPM sensor is setup, it may just be a simple case of adding a second sensor at 180 degress out from the original sensor, thus doubling the clock pulse / slowing the motor to half speed.

I could just use another driver for the motor, but the existing chip would be better due to the fact it keeps such accurate speed on the motor - something thats important for analogue audio - and the fact it keeps the costs down.

SS

Thankyou for doing all this SS, I'd never have known what to do with half of the things you've mentioned here!
I can't express my gratitude enough

JingleJoe wrote:Thankyou for doing all this SS, I'd never have known what to do with half of the things you've mentioned here!
I can't express my gratitude enough

No probs, I'm actually quite enjoying the challange! And you knever know, it might be worthy of a mention on Hackaday (if it works...)

SS