HP91xx controllers (was HP-IB disc drive emulation program

Tony Duell ard at p850ug1.demon.co.uk
Mon Dec 28 12:39:50 CST 2009

> And I seem to recall something GPIB from other sources, like
> Fairchild, AMD and NatSemi, as well.

The common HPIB/GPIB chips were the Texas Instruments TMS9914, the Intel 
8291 and the NEC uPD7210. I've come across all of them. There was a 
Motorola 68488 (what an obvious number!) which IIRC was a talker/listener 
(no controller functions) designed to link to the 6800 bus. HP had their 
own Medusa chip (1TL1 IIRC).. And Philips made (or second-sources) a 
curious deveice in the 4000-series CMOS family that was essentially the 
state machines and random logic for a talker/listener, and was designed 
to add a GPIB interface to just about any digital instrument. I forget 
the number of that one, I have a data sheet somewhere but have never seen 
the actuall device.

Doubtless other manufactuerers made GPIB interface chips, not to mention 
all the solutions using TTL logic chips (HP, Epson and several others 
like that have passed my bench) and custom chips.

> >=20
> >>> I still think it would be a useful project for somebody to make a=3D2=
> 0
> >>> HP-drive-emulator. Mot using a PC, but using a microntroller, HPIB=3D=
> 20
> >>> buffers, the minimal (if any) logic you need to handle HPIB, and a CF=
> =3D20
> >>> card. Make it 'open' and I'll build several...
> >>
> >> Or even a SATA, PATA, USB storage interface. Such a stand alone
> >> emulator would be very useful.
> >=20
> > If the design is 'open', then presumaly you could mofify it for whateve=
> r=20
> > storage device you want. And surely it would be better to have a workin=
> g=20
> > desing for _any_ modern-ish storage device rather than debate about whi=
> ch=20
> > device to use.
> >=20
> I was suggesting a bit more along the lines of "It would be nice if a
> developer wouldn't focus solely on one type of storage but rather

Agreed. But I suspect the 'storage' side of things is the easy part, and 
thus it would be easy to modify the design (assuming sufficient 
documetnation) for just about any storage device.


> > If this is a 9133XV with the HP controllers (6809 + 9114, etc), then I=20
> > have unofficial scheamtics. The floppy controller is actually quite=20
> > simple -- most of the logic is on the top board, the lower board contai=
> ns=20
> > the master clock and disk data separator. Apart from the EPROM (which i=
> s=20
> > the sama as in a 9121 IIRC), there's nothing remotely custom in there.
> >=20
> It has the separate HP HD and FD controller boards. It appears to work
> normally except that it is unable to read any floppies. Changing the

DOes the floppy controller respond to HPIB commands? I assume it does 
from what you say, which implies the processor is running

> drive did not change anything about that. Beyond that, no idea,
> although it may be the data separator board. One Of These Days, I'll
> dig it up, thaw it and see what else I can find.

IIRC there's one power line (a -ve one) which is used by the data 
separator and nothing else. On soem even older designs this came from a 
diode/capacitor voltage multiplier driven from the master clock. I think 
by the time of the 9133V, it came from the main SMPSU. It's still worth 
checking it. 

Then see what the read clock and read data pins on the FDC (a 1793-a-like 
IIRC) are doing. It's a pretty simple design, actually.


More information about the cctech mailing list