Re: A2 5.25" drive on a PC?

[pausch@saaf.se (Paul Schlyter)]

In article <438333d4$0$20848$9b4e6d93@newsread2.arcor-online.net>,
Linards Ticmanis  <ticmanis@gmx.de> wrote:

> Tim Haynes wrote:
>> We all know that a PC cannot read/write an A2 disk because of basic 
>> MFM/GCR differences.
>> 
>> What about the possibility of putting together a specialized cable that 
>> would adapt a standard A2 5.25" drive to a PC's serial port?
> 
> You need a data rate of 255,682 bits per second, and you need to keep 
> that rate quite exactly; that's beyond the capability of standard PC 
> serial ports. Plus, the drive can't deal with the start/stop bits which 
> the serial port expects. Plus, the voltages are wrong.
>
> On the paralell port, maybe; but at least write timing will be 
> problematic, and still some voltages will be wrong.

By adding some transistors, one can get the voltages right.  But that
extra electronics would probably need its own power from somewhere.
 
>> The cable would also have to pull power from somewhere - possibly from
>> a molex connector hooked into the PC's power supply?
> 
> If you don't mind that the drive will always spin, this will be easy;
> otherwise you need to build something to switch the motor power on and 
> off. Standard PC ports can't source enough current by far to run the
> drive motor on a switchable line without external electronics.

When adding some transistors to get the voltages right, one could as
well add a few more transistors to let some suitable pins on the
parallell port switch the drive's motor on and off, and also control
the stepper motors.
 
>>  And then we'd need some specialized PC software to actually do the
>> reading/writing.
> 
> Right; and it would in all probability run only in pure MS-DOS mode. 
> Otherwise correct timing is probably impossible to get.

Not impossible - just a bit harder.  One would need a kernel mode
device driver running on Win32 (or Linux).  And one would also need
to turn interrupts off at least while writing to a sector (a sector
read can always be retried if an interrupt destroys the timing).

If we add some transistors to that cable, perhaps we could as well
add a microcontroller which can read or write (and buffer) the
nibble data of one sector on the Apple II floppy.  That microcontroller
should have all interrupts turned off to ensure accurate timing in
the instructions.  Then the software on the PC side could be written
at the application level and wouldn't need to be kernel mode device
drivers.

And if one really wanted to be fancy, one could make that microcontroller
board for the Apple II disk drive connect to the PC through a USB
cable rather than a parallell port.  That would of course be even
harder, but more convenient to use when it's all finished.
 
>> Does this seem at all feasible?  Any thoughts?
> 
> Maybe it's doable, but it wouldn't be too easy.

I fully agree on that.... :-)

> -- 
> Linards Ticmanis

-- 
----------------------------------------------------------------
Paul Schlyter,  Grev Turegatan 40,  SE-114 38 Stockholm,  SWEDEN
e-mail:  pausch at stockholm dot bostream dot se
WWW:     http://stjarnhimlen.se/