On 5/24/14 1:07 PM, Paul Koning wrote:
Something like the DSP based tape drive Paul Pierce
built (if I remember that one right)
Paul's drive just has A/D converters and a PC, no hardware DSP
Several years ago, John Bordynuik sent me one of the transports he used in his tape
recovery business. It uses an M4 data transport, 3480 head stack (16 tracks) custom
preamp
and a custom A/D DSP board to capture and convert the data. The preamp board arrived with
one of the SMT resistors in the MR bias supply cracked, and he never sent me enough
documentation
to make it work. I spent several weeks late last year trying to get it to do something.
The whole project has been incredibly frustrating. At the same time, I sent two M4 data
head
stacks to have the capstan wheels redone, and after two months they came back with nothing
having
been done to them because the stacks had parts missing. As far
as I can tell, they never even touched them, and I TOLD them that all I wanted them to do
was
to install new capstan wheels since they would only sell them at the "less
expensive" exchange
price of $150 ea. if they got old ones, which I didn't feel comfortable trying to
remove (they
are a bitch to take out w/o bending the encoder wheel).
This all started when I called them to try to get schematics for the M4 drive motor
control board
which John had added his own interface onto to drive the tape.
Anyway, one of the magic bits of sauce you can do when recovering data with direct A/D is
to monitor
the exact position of the tape wrt the heads by watching the encoder on the capstan. And,
if you're
using magnetorestrictive heads, you can directly sense the magnetic flux level on the
tape. With those
two bits of information, you can get a very good idea of what was the surviving levels of
magentic flux
are at any spot on the tape.