Yes, I've considered that, but . . . there's a delay involved that wouldn't
work with WD's chips unless you accumulate the pulses and disable ready.
Then mask the delays by extending the head-load delay. Remember, every
drive is different.
Dick
-----Original Message-----
From: Allison J Parent <allisonp(a)world.std.com>
To: classiccmp(a)classiccmp.org <classiccmp(a)classiccmp.org>
Date: Friday, February 04, 2000 6:12 PM
Subject: Re: WD1770 help needed
>"In the case of the 1771 and 179x series it's possible to build a really ne
><circuit I've seen but never tried to match, which uses the /TEST pin on th
><
><Sounds like an ideal candidate for reimplementation in a single chip
><microcontroller.
>
>Did that many years ago using 8748, worked very good too as I could count
>the pulses then issue them out as a ramped rate so that the head was
>accelerated and decelerated to the track. Didn't make any difference for
>a few track move. Head movement that was about 15 tracks or more it was
>very noticeable. CP/M does a lot of seek to the directory track (usually
>track one or two) every time it writes an extent or closes/opens a file so
>decreases in seek time payed nicely.
>
>Allison
>
It could be a microcontroller project, but it's easy enough if you use a
relatively small CPLD. What's critical is that you need a VCO or some
digital analog to it which causes the counters to run more quickly while the
beginning track to ending track difference is large and ramps up slower than
it ramps down. I'd not depend on that, since each drive will be different.
It's best done, IMHO, if one uses a drive-resident circuit with slew rates
tuned to the individual drive. Thos of us who can't remanufacture our
drives NEED this in order to make our 8" drives stay well. I
microcontroller could handle the job fine if it just could be adjusted with
a pot rather than having to have parameters experimentally determined and
then fit via cut-and-try.
Dick
-----Original Message-----
From: Al Kossow <aek(a)spies.com>
To: classiccmp(a)classiccmp.org <classiccmp(a)classiccmp.org>
Date: Friday, February 04, 2000 2:51 PM
Subject: Re: WD1770 help needed
>"In the case of the 1771 and 179x series it's possible to build a really
neat
>circuit I've seen but never tried to match, which uses the /TEST pin on the
>FDC to cause the device to put out its pulses much faster, allowing them to
>be accumulated externally in a counter, which, drives a DAC which drives a
>VCO, which drives the counter as it downcounts the number of steps, thereby
>slewing the head assembly. This could lead to an interesting but lengthy
>discussion.
>"
>
>Sounds like an ideal candidate for reimplementation in a single chip
>microcontroller.
Hello all
Sadly, the WD1770 fcd IC in my home-brew fd i/o board finally gave up the
ghost. Fourteen years of service is admirable -- when it finally croaked, I
felt like weeping.
Additionally, in violation of my policy of *always* keeping spares of
discontinued components which I use -- I have no spare.
Does anyone know of a source for these? Or the 1772 (I believe Tony Duell
once mentioned that the 1772 could replace the 1770 in most cases -- but I
could be wrong).
Also, I understand that the C64 floppy drives (1581?) used the 1770 but I'm
not a commie and can't immediately verify this. Anyone out there with junk
C64 drives who wants to liberate this IC -- to a good home??
Any help at all on this will be greatly appreciated.
Glen Goodwin
0/0
<In going through and sorting my piles of classic ICs, I ran across this Wes
<Digital WD2793A chip. What is it?
Take a 1793 and put some of the stuff needed to make it useful on the die,
call it 2793. Basically it's a 1793 with data sep and precomp logic built
on.
Allison
"Oh, one BLINDING error. Wouldn't that hook up to a VAX? I passed on one
about a year ago, and a lot of parts. Though I might have gotten some
software for it in the mess of DEC stuff that I did get.
"
..or a Unibus PDP-11. What kind was it? I have some docs on the FPS-100.
I keep meaning to stop over at the Physiology Department of the Medical
College of Wisconsin to find out what ever happened to that 11/44 and
FPS-100 that I used to take care of..
"I've often thought that a PAL or small CPLD would do the job. Has anyone
written the appropriate state machine or equations?
"
>From memory, didn't the Furgusson 'big board' use a digital data separator
that was a bipolar prom state machine? I know I've seen this same design in
several late 70's floppy controller boards (maybe it was even in the WDC
data sheets)
drive and monitor for. The "monitor" looks like it started life as a 8"
<security monitor. But I have to say that this is the first monitor that
<I've ever seen that was connected with a coaxial cable with N-connectors!
N??? more likely SO239 (matching plug is PL259). I still ahve a panasonic
that was used for an apple][ /8" Monochrome TVscan rate/.
<a general purpose S-100 computer. All the cards in it including the back
<plane are marked "AV". There are warranty stickers in it dated 1978 and i
<has an 8080 CPU. Does anyone have docs and software for it?
Can't help on the computer from the description. Likely it's just another
8080 S100 box and therefor with the right floppy and boot proms quite
useable for CPM (assuming enough ram).
Allison
"In the case of the 1771 and 179x series it's possible to build a really ne
<circuit I've seen but never tried to match, which uses the /TEST pin on th
<
<Sounds like an ideal candidate for reimplementation in a single chip
<microcontroller.
Did that many years ago using 8748, worked very good too as I could count
the pulses then issue them out as a ramped rate so that the head was
accelerated and decelerated to the track. Didn't make any difference for
a few track move. Head movement that was about 15 tracks or more it was
very noticeable. CP/M does a lot of seek to the directory track (usually
track one or two) every time it writes an extent or closes/opens a file so
decreases in seek time payed nicely.
Allison
This posting is a follow up to my Mac Portable battery post:
The cost of the individual 2V 5.0 Ah "Cyclon" size "X" is $7.25 (USD).
The total for 3 is $25 (USD), not including s&h charges.
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
