[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: How much of a ][ is needed to twitch?
Alex Freed wrote:
> "Tristan Mumford" <xtristan.xmumford@xgmail.xcom> wrote in message
> news:45d1994e$0$52907$c30e37c6@lon-reader.news.telstra.net...
>
>> I'm fairly sure the logic of the controller is compatible with the floppy
>> drive and shouldn't need anything like pullups. Can someone (dis)agree
>> with
>> me on this please?
>
> I did something similar to your project some 20 years ago - interfaced a
> generic "industry standard" disk drive to A2.
> The main things to do:
>
> 1. Convert motor phases to "step" and "direction".
> I used a couple of chips to do it, but it *MAY* be possible to directly
> use one ODD phase as "direction" and OR EVEN phases for step.
> I didn't do it this way, but it looks like RWTS always have one odd
> phase active when an even phase gets active.
I'm doing it a different way via a uC. It's using an odd kind of double
lookup table and a couple of tests for certain conditions. It's not quite
right yet though. Ie too wasteful of cycles. But that's not my main
concern. With the uC running at a decent speed, and the much shorter step
time of the 3-1/2" drives it looks to be quite possible to use an
inexpensive uC to control the head.
>
> This will NOT provide half-tracks.
My method may. Unless I'm wrong about one aspect of the PC floppy drives. I
believe that one strobe of /STEP (ie high, low, high) results in a single
track step. When it's held low it does a half step. Of course I may be
wrong and it may be a full step, and the strobe does two. It's hard to find
low level detail on the workings of the drives.
I was kind of dissappointed when I discovered that the 3-1/2" drives use
bipolar steppers though. It would have been dead easy to add a simple bit
of drive to make it work. I have a darlington array that would even do it
(overkill though).
>
> 2. STEP should be disabled when TRACK0 is active from the drive.
True. Easy done. Actually already in the code to stop the chance of
headbanging.
>
> 3. Write data line will need a one-shot circuit.
Really? How come?
>
> I could have forgotten other issues.
> And the last point is: don't even start doing anything till you buy/make
> a logic probe.
Going to make one later on in my standard overdone style. Although first I
might shove a bare led tester in one of the many dead 12 pen style
testlamps around the place.
--
-----> http://members.dodo.com.au/~izabellion1/tristan/index.html <-----
===== It's not pretty, it's not great, but it is mine. =====