17 Oct
2006
17 Oct
'06
1:19 p.m.
I've been trying to puzzle out a board in front of me, loaded with
Intel chips. The CPU on it is a 16MHz 80186 - it appears to be an
interface peripheral board for some Astro-Med product. Externally, it
has a serial port and a GPIB port. What's strange to me is that I
can't identify the parts that appear to be the GPIB buffers - a pair
of 28-pin Intel D8293s. There's also a P82C08, which I can't find any
hard info on.
So... does anyone recognize the Intel part numbers P82C08 or D8293
(yes, I know the P is plastic and the D is ceramic; it's the parts
themselves I'm curious about).
Thanks,
-ethan
17 Oct
17 Oct
1:27 p.m.
So... does anyone recognize the Intel part numbers P82C08 or D8293
I don;t know where you'd find data sheets in electronic form (I think I
have them on paper), but the 8293 is a GPIB buffer (the 8291 was the
talker/listener chip, the 8292, actually a programmed microcontroler
(8042?) was the GPIB contorlelr add-on). And I think the 8208 was some
kind of DRAM controller.
What indormation do you need? If it's just pinouts I might well be able
to tyoe them up.
-tony
1:50 p.m.
On 10/18/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
Indeed... I did look first.
So... does anyone recognize the Intel part numbers P82C08 or D8293
I don't know where you'd find data sheets in electronic form... (I think I have them on paper)
Those could be handy some day if I ever try to re-use the chips.
but the 8293 is a GPIB buffer
That makes sense given where the wires go, but I had just never heard
of the part before.
(the 8291 was the talker/listener chip, the 8292,
actually a
programmed microcontroler
(8042?) was the GPIB contorlelr add-on).
Good call - there does happen to be an 8291 on the board, next to the
buffers, but no sign of an 8292. There _is_ an 8256, but given its
location, I'm guessing it's a UART.
There is also a Siemens 8282 nearby that seems to link into the 8291 -
could that be the micro-controller? It's 20 pins, and has a printed
date code such as one would see on a programmable part.
And I think the 8208 was some kind of DRAM controller.
That makes sense - it's next to a bank of 32 41256s.
What indormation do you need? If it's just pinouts
I might well be able
to tyoe them up.
Primarily, I was curious as to what the GPIB parts were - the board is
out of the scrap pile, and I was contemplating scavenging the
GPIB-related parts in the hopes of someday wiring them to some machine
to talk to Commodore-brand disks and printers.
At this point, I might or might not take the time to desolder them.
Honestly, I'd rather go with more common parts since if I do build an
interface, I'd hate to be at a loss for spare parts or for it to be
difficult to make a second one.
Thanks for the great info and the quick response, Tony.
-ethan
3:46 p.m.
On 10/18/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
Indeed... I did look first.
It wasn't a commonly-used one, but it certainly did exist.
So... does anyone recognize the Intel part numbers P82C08 or D8293
I don't know where you'd find data sheets in electronic form... (I think I have them on paper)
Those could be handy some day if I ever try to re-use the chips.
but the 8293 is a GPIB buffer
That makes sense given where the wires go, but I had just never heard
of the part before. (the 8291 was the talker/listener chip, the 8292,
actually a
programmed microcontroler
(8042?) was the GPIB contorlelr add-on).
Good call - there does happen to be an 8291 on the board, next to the
buffers, but no sign of an 8292. There _is_ an 8256, but given its location, I'm guessing it's a UART.
The number doesn't ring any bells, but I can try to look it up.
There is also a Siemens 8282 nearby that seems to link
into the 8291 -
could that be the micro-controller? It's 20 pins, and has a printed
I doubt it. It sounds more like a bus buffer. Is there any other procesor
on this board? As I said, if this is a peripheral, you'd not need the 8292.
date code such as one would see on a programmable
part.
If you want to do that, you need to make a GPIB controller. You could use
the 8293s and bit-bang the protocol using parallel interface ICs like
6522s (the 8255 has problems, since you can't reverse the direction of
the port lines.....). Or you could use something like a 9914. Personally
I'd stay away from the 8291, it's an Intel chip and we all know those are
unpleasant to use!.
If you're only linking to one or two GPIB devices, you can get away with
using 74LS14's as the receivers and something like 7438s (open-collector
outputs) as the drivers. Yes, it doesn't meet the spec but I can assure
you it'll work. If you're really curious, I have schematics for the first
HPIB interface -- the one that goes in the HP9830. That used no special
ICs at all, just TTL and discretes (I have an idea the output drivers
were discrete transistors).
And I think the 8208 was some kind of DRAM
controller.
That makes sense - it's next to a bank of 32 41256s.
What indormation do you need? If it's just
pinouts I might well be able
to tyoe them up.
Primarily, I was curious as to what the GPIB parts were - the board is
out of the scrap pile, and I was contemplating scavenging the
GPIB-related parts in the hopes of someday wiring them to some machine
to talk to Commodore-brand disks and printers.
At this point, I might or might not take the time to desolder them.
I'd save them, but I'd not use them in a new design. Keep them for the
time you have some rare device that uses them and needs to be fixed.
If you can find them, I'd go for a TMS9914 GPIB interface chip and 75160
+ 75162 buffers. They are fairly easy to find I think.
-tony
4:29 p.m.
On 10/18/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
As I've just noticed there are 3 of them, that also makes sense.
I've seen several implementations of that - easy enough to do the
hardware - all the work is in the software (and avoiding timing
problems).
Good idea.
The 8291 gives you all the functions you need to make
a talker or
listener -- that is a deviec which doesn't act as an GPIB bus controller.
Gotcha.
The 8292 adds the functions to act as a controller. If
you're making a
peripjheral, you only need the 8291. If you're making a GPIB controller
-- that is a device that can send bus commands, etc, you _also_ need the
8292. The 8292 appears to be a programmed microcontroller, probably an
8042, but I wouldn't swear to that.
Makes sense - this board is almost certainly a peripheral board, not a
controller.
There is also
a Siemens 8282 nearby that seems to link into the 8291 -
could that be the micro-controller? It's 20 pins, and has a printed
I doubt it. It sounds more like a bus buffer. Is there any other procesor on this board?
Yes... I mentioned the 16 Mhz 80186 in the initial message.
Primarily, I
was curious as to what the GPIB parts were - the board is
out of the scrap pile, and I was contemplating scavenging the
GPIB-related parts in the hopes of someday wiring them to some machine
to talk to Commodore-brand disks and printers.
If you want to do that, you need to make a GPIB controller. You could use
the 8293s and bit-bang the protocol using parallel interface ICs... ...like 6522s (the 8255 has problems, since you
can't reverse the direction of
the port lines.....).
Right. Since my first experience with chips of this nature was with
the 6520s (6821) and 6522s in Commodore equipment, I was taken aback
at how clunky the 8255 is to work with. Still, between my SBC6120, my
RB5X, my MC-1N, and several other devices, I'm continually suprised
how common the 8255 is. I guess if being able to pick a direction and
stick with it isn't a problem (dedicated in and out ports and such),
the 8255 isn't so bad. It just lacks a flexibility commonly found in
members of the 65xx family.
Or you could use something like a 9914. Personally
I'd stay away from the 8291, it's an Intel chip and we all know those are
unpleasant to use!.
That certainly all makes sense.
If you're only linking to one or two GPIB devices,
you can get away with
using 74LS14's as the receivers and something like 7438s (open-collector
outputs) as the drivers. Yes, it doesn't meet the spec but I can assure
you it'll work.
Interesting - given how rare (to me, at least) "proper" GPIB drivers
are, that's handy to know... most of the time, for the sorts of tasks
I'm thinking about, 2 peripherals is about the most I'd be talking to.
If you're really curious, I have schematics for
the first
HPIB interface -- the one that goes in the HP9830. That used no special
ICs at all, just TTL and discretes (I have an idea the output drivers
were discrete transistors).
I'm always interested in learning about new GPIB implemenations.
At this point,
I might or might not take the time to desolder them.
I'd save them, but I'd not use them in a new design. Keep them for the
time you have some rare device that uses them and needs to be fixed. If you can find them, I'd go for a TMS9914 GPIB
interface chip and 75160
+ 75162 buffers. They are fairly easy to find I think.
I've seen those buffers on cards for the C-64 and VIC-20, but have no
experience with the TMS9914. ISTR there is at least one ISA design
using the 9914, so if I see one of those cards float by, I could
always extract the chips from that board (and install sockets in case
I want to use the ISA board later).
Thanks for all the good info,
-ethan
18 Oct
18 Oct
4:25 p.m.
If you're
really curious, I have schematics for the first
HPIB interface -- the one that goes in the HP9830. That used no special
ICs at all, just TTL and discretes (I have an idea the output drivers
were discrete transistors).
I'm always interested in learning about new GPIB implemenations. 
17 Oct
17 Oct
4:32 p.m.
On 10/17/2006 at 11:46 PM ard at p850ug1.demon.co.uk wrote:
> >(the 8291 was the talker/listener chip, the
8292, actually a
> programmed microcontroler (8042?) was the GPIB contorlelr add-on).
In one of their rare bits of candor, Intel admits right in the databook
that the 8292 is a preprogrammed 8041A.
I've got a x80 board here that uses the 92/93 to talk to a hard disk drive.
I think I also have the schematics for same and I definitely have the
driver source. There were a couple of "quirks", but nothing too awful.
If you're not dealing with very high-speed devices, (e.g. voltmeters, etc.)
and only talking to a single device, a simple bidirectional printer port
can provide most of the functionality of GPIB/HPIB. Sometime in the
mid-80's, I released a bit of code (may still be in SIMTEL) for a DOS TSR
to drive an HP Plotter through the printer port.
It wasn't an original idea--I saw the printer port on the Victor 9000 used
to do the same thing. Victor would even sell you a cable for that purpose.
Cheers,
Chuck
18 Oct
18 Oct
10:56 a.m.
On Tuesday 17 October 2006 07:32 pm, Chuck Guzis wrote:
If you're not dealing with very high-speed
devices, (e.g. voltmeters, etc.)
and only talking to a single device, a simple bidirectional printer port
can provide most of the functionality of GPIB/HPIB. ?Sometime in the
mid-80's, I released a bit of code (may still be in SIMTEL) for a DOS TSR
to drive an HP Plotter through the printer port. ?
It wasn't an original idea--I saw the printer port on the Victor 9000 used
to do the same thing. ?Victor would even sell you a cable for that purpose.
And all of the Osbornes I've ever encountered called theirs a "Printer/IEEE"
port...
--
Member of the toughest, meanest, dealiest, most unrelenting -- and
ablest -- form of life in this section of spac, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
4:33 p.m.
If you're not dealing with very high-speed
devices, (e.g. voltmeters, etc.)
and only talking to a single device, a simple bidirectional printer port
can provide most of the functionality of GPIB/HPIB. Sometime in the
mid-80's, I released a bit of code (may still be in SIMTEL) for a DOS TSR
to drive an HP Plotter through the printer port.
If not, I appear to have a copy floating around here. I've never used it,
though.
It wasn't an original idea--I saw the printer port on the Victor 9000 used
to do the same thing. Victor would even sell you a cable for that purpose.
The printer port on the Sirius (==Vistor 9000) was a GPIB port with a
strange pinout and connector. According to the schematics, it uses a
75160 to drive the data lines and a 75161 to drive the handshake lines
(for example, pin 1, which would be the strobe line on a Centronics
printer is driven by the DAV buffer in the 75161).
The whole thing was driven in software (there's a 6522, not suprisingly,
behind it :-)). The normal device drives talk to the Centronics printer,
for which you use a striaght-through cable. This is the first I've heard
of ther being a GPIB cable (and presumably driver software), though.
Talking of the Sirius, was the user port ever used for anything. That's a
50 pin header on the mainboard carrying power and ground, light pen
input, and all the port lines of a 6522. Well, one of them is also used
as the clock for the sound system, but the rest are totally uncommitted.
And was the sound input (the little header on the mainboard) ever used?
-tony
17 Oct
17 Oct
1:45 p.m.
On Oct 17, 2006, at 4:19 PM, Ethan Dicks wrote:
I've been trying to puzzle out a board in front of
me, loaded with
Intel chips. The CPU on it is a 16MHz 80186 - it appears to be an
interface peripheral board for some Astro-Med product. Externally, it
has a serial port and a GPIB port. What's strange to me is that I
can't identify the parts that appear to be the GPIB buffers - a pair
of 28-pin Intel D8293s. There's also a P82C08, which I can't find any
hard info on.
So... does anyone recognize the Intel part numbers P82C08 or D8293
(yes, I know the P is plastic and the D is ceramic; it's the parts
themselves I'm curious about).
Hmm...The 8208 is a DRAM controller. Dunno about the 8293.
-Dave
--
Dave McGuire
Cape Coral, FL
2:16 p.m.
On 10/18/2006 at 9:19 AM Ethan Dicks wrote:
I've been trying to puzzle out a board in front of
me, loaded with
Intel chips. The CPU on it is a 16MHz 80186 - it appears to be an
interface peripheral board for some Astro-Med product. Externally, it
has a serial port and a GPIB port. What's strange to me is that I
can't identify the parts that appear to be the GPIB buffers - a pair
of 28-pin Intel D8293s. There's also a P82C08, which I can't find any
hard info on.
Tony's got it right--the 8293's a GPIB transceiver--no intelligence in it.
It's in the 1983 Intel Microprocessor and Peripheral Handbook. If you need
pinouts, etc. I can scan it for you.
The 8208's a DRAM controller--I've got the datasheet for the 8207, which is
probably very similar.
Cheers,
Chuck
So... does anyone recognize the Intel part numbers P82C08 or D8293
(yes, I know the P is plastic and the D is ceramic; it's the parts
themselves I'm curious about).
Thanks,
-ethan
6640
days inactive
6641
days old
10 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
ethan.dicks＠gmail.com -
mcguire＠neurotica.com -
rtellason＠verizon.net