8 Jan
2007
8 Jan
'07
10:59 p.m.
On 8 Jan 2007 at 14:40, Chris M wrote:
I should have mentioned I wasn't intending to
use
such a monster with a pc. Macs do the gcr thing, but
they don't usually have 5.25" drives. I did a bit or
reading, and am led to ask if the encoding couldn't be
handled in software.
That's why God made Catweasels.
O man what are you telling me. I have a desktop cnc
mill that has a variable speed dc motor. If I want
more rpms I just pump in more emfs. 70, 80, 90 volts.
That kind of juice will get that puny drive to comply
There are three other good reasons to vary the data clock rate and
not the spindle speed (bear with me here). The first is that keeping
the spindle at the same RPM is much less likely to produce problems
with ISV (instantaneous speed variation; an oxymoron if there ever
was one) as the motor changes its speed to match the zone. The
second is that the signal level induced in the read head is
proportional to the square of the linear velocity of the medium.
You're best off setting the motor to run where the signal level-vs-
head characteristcs is best.
The third is that tachometer design is much simpler. You have a
fixed-frequency reference, such as a crystal or a ceramic resonator
to work with, not some software approximation.
And GCR doesn't imply varaible data rate or spindle speed. That
Durango that I posted a web page on used GCR to good advantage
without fooling with spindle speed (almost a 1MB on a 360K 5.25"
diskette). I believe that Microtech marketed a PC that worked
similarly (although I'd have to check to make sure). I think
someone's written a Catweasel driver that fits almost 4MB on a 1.44MB
diskette...
Cheers,
Chuck