On 7 Dec 2009 at 10:50, Jules Richardson wrote:
Do you have any kind of estimate for when this will be
finished (to
the point of being releasable - sounds like there might be ongoing
firmware patches?). I'd be interested in buying one I think (even
though I try to avoid USB if I can).
Over at VC Forum, I've been working on a low-cost alternative using a
AVR (or pair of them) and the timer "capture" feature at a 16MHz
sample rate. One objective was to design something that anyone
could build, so the SMT (ATMega128) uC that I've been using is out.
I'm migrating the design to a two-uC (ATMega162 and ATTiny2313)
approach. The 'tiny handles all of the "mechanica" aspects of drive
management, such as motor control, drive select and seeking, while
the 'mega handles the sampling and communication with the host.
Although I could have used V-USB to provide USB functionality
directly, I decided to use RS-232C instead, as serial-to-USB adapters
are cheap and plentiful.
The output is essentially Catweasel-style sampling, with the leading
index edge signified by a 00 byte in the sample stream. Full 256-
count resolution is thus available. I/O is a track-at-a-time and
decoding and encoding is handled by the host. There's a 128K sample
buffer, which can be upgraded to 512K by simply swapping the SRAM
chip. Programming of the uCs is in-circuit and can be done with a
simple PC parallel-port adapter.
Thus far, I've done MFM, FM and Apple II floppies in the host
software with very good results.
Pricing should be very good--the Mega is <$6 Qantity 1; the Tiny is
about $1.50; the 128K SRAM is less than $5; other chips are pretty
much commodity items.
This is not to take away from Phil's fine work, but an attempt to
economically hit most of the common formats that a hobbyist is likely
to encounter with something that anyone who can solder can build.
Eventually, I'm looking at porting the design to one of the new NXP
Cortex CPUs which can be had in PLCC. But first things first...
--Chuck