Topics

Interesting HP-IL to HP-IL Interface

Mark Bielman
 

Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)

As I recall, the 75C mfg line created a system where 75's could send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.

Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???

Mark B

 

Mark,

 

What kind of machines were sending/receiving these messages via HP-IL along the production line?

 

HP-110?

PCs with HPHP82973A HP-IL boards?

HP-75C

 

 

--bob

 

From: hp75@groups.io [mailto:hp75@groups.io] On Behalf Of Mark Bielman
Sent: Wednesday, November 29, 2017 10:08 PM
To: hp75@groups.io
Subject: [hp75] Interesting HP-IL to HP-IL Interface

 

Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)

As I recall, the 75C mfg line created a system where 75's could send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.

Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???

Mark B

_._,_._,_

Paul Berger
 

My guess would be 75 to 75 since the 75 will only operate as the controller but I suppose it could be others but since things like the 71B and 85 can operate in device mode something like this would not be necessary.

Paul.


On 2017-11-30 12:06 AM, Robert Prosperi wrote:

Mark,

 

What kind of machines were sending/receiving these messages via HP-IL along the production line?

 

HP-110?

PCs with HPHP82973A HP-IL boards?

HP-75C

 

 

--bob

 

From: hp75@groups.io [mailto:hp75@groups.io] On Behalf Of Mark Bielman
Sent: Wednesday, November 29, 2017 10:08 PM
To: hp75@groups.io
Subject: [hp75] Interesting HP-IL to HP-IL Interface

 

Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)

As I recall, the 75C mfg line created a system where 75's could send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.

Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???

Mark B


Paul Berger
 

Thinking about it a little more, on one side you could have a long loop with several of these devices in it, connected to a  machine that the individual users would receive messages on.  I wonder if the GPIOs are just wired up back to back  or if there is some intelligence between them? you could make a usefull but somewhat slow network with devices like this.

Paul.


On 2017-11-30 12:13 AM, Paul Berger wrote:

My guess would be 75 to 75 since the 75 will only operate as the controller but I suppose it could be others but since things like the 71B and 85 can operate in device mode something like this would not be necessary.

Paul.


On 2017-11-30 12:06 AM, Robert Prosperi wrote:

Mark,

 

What kind of machines were sending/receiving these messages via HP-IL along the production line?

 

HP-110?

PCs with HPHP82973A HP-IL boards?

HP-75C

 

 

--bob

 

From: hp75@groups.io [mailto:hp75@groups.io] On Behalf Of Mark Bielman
Sent: Wednesday, November 29, 2017 10:08 PM
To: hp75@groups.io
Subject: [hp75] Interesting HP-IL to HP-IL Interface

 

Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)

As I recall, the 75C mfg line created a system where 75's could send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.

Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???

Mark B



Mark Bielman
 

Yes - Paul is correct on the latest. Everyone had 75's (one side of the loop interface); the other loop was literally looped with more of these IL-2-IL boxes. One
of them was connected to a "master controller" (best I recall) like a model 220 or 9816.

The Mostek controllers are directly connected on the PCB inside the interface box - nothing between.

I'm a newbie here so I will attempt to load photos.

Mark

Mark Bielman
 

The 9816 (as I recall) must have had an HP-IL/HP-IB converter as I do not recall in HP-IL interface for the 200's (9000's).
The Integral was not invented yet. And it was NOT an 80 series...

Mark

Mark Bielman
 

Also software on the 75's was "HPILCMDS" (memory?) lex file with in-house Basic prgm.
Lots of SENDIO and ENTER and low-level commands too. Wonder if I have it?

 

Mark wrote...
Ran across this item in my junk box - an HP-IL to HP-IL interface
(no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)
As I recall, the 75C mfg line created a system where 75's could
send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.
Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???
I asked Raan Young about it, and he replied:

The device is the Loop-de-Loop, which was never released. I sent
you the "manual" (LDL.pdf) a while back.
I remember hearing that the 75 production line was built using
75s as controllers, "networked" together -- but I have no details...
I did have the file, but I can't find it on Groups.io. Not sure
how it fell through the cracks, but I've uploaded it now (you should
have received the upload notice by now...)

Everett

======================================================================
Everett Kaser Software Logic & Puzzle Games for computers
PO Box 403 http://www.kaser.com
Albany OR USA 97321-0117 Phone: 541-928-5259 8am-8pm PACIFIC TIME
======================================================================

J-F Garnier
 

On Wed, Nov 29, 2017 at 09:14 pm, Mark Bielman wrote:
The 9816 (as I recall) must have had an HP-IL/HP-IB converter as I do not recall in HP-IL interface for the 200's (9000's).
The Integral was not invented yet. And it was NOT an 80 series...
A HP-IL interface board (HP 98634A) for the series 200 was announced (OEMNews 1983 April page 14) and almost immediately discontinued ( OEMNews 1983 August).
I wonder how many of these units actually existed.

J-F

J-F Garnier
 

On Wed, Nov 29, 2017 at 07:07 pm, Mark Bielman wrote:
Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)
On Wed, Nov 29, 2017 at 10:14 pm, Everett Kaser wrote:
I asked Raan Young about it, and he replied:

The device is the Loop-de-Loop, which was never released. I sent
you the "manual" (LDL.pdf) a while back.
I remember hearing that the 75 production line was built using
75s as controllers, "networked" together -- but I have no details...
I did have the file, but I can't find it on Groups.io. Not sure
how it fell through the cracks, but I've uploaded it now

Thanks mark for these interesting and (up-to-now) unknown device, and thanks Everett/Raan for the documentation :-)

Some comments:

- It is a kind of mailbox system between two HP-IL loops. A similar system can also be built with two HP82166A converters connected back-to-back using - for instance - the evaluation panel provided in the HP82166A prototyping kit. .
Maybe this is the reason of the non-release of this device since the functionality is already provided by an other product.

- The manual says "The Loop-de-loop is designed to solve the drawback of HP-IL which becomes unoperational when a device on the loop is dead or removed or powered down. ...  The only restriction on communicating between two HP-IL devices
using Loop-de-loop is that there must be a controllor on each loop. This requirement is under consideration to be taken away. Any input is welcomed."
Indeed, it would be quite useful to have only one controller on one side and an isolated sub-loop that could be bypassed when unpowered or open; I may even know how to do it, but I'm afraid it's too late to submit my input :-)

- The manual says the device ID is "HP82000" and accessory ID is 44h; Mark if you could check the actual IDs it would be interesting. You can use the DEVID$ and DEVAID functions from the HP71 HPIL module or the DEVID$/DEVAID$ from the HP75 I/O ROM; or you can use the ILCtrl software with your next-to-come PIL-Box :-)

And of course a few photos will be welcome !

J-F

Mark Bielman
 

Hello J-F,

Yes, the 'manual' has interesting comments. Used (basically) exactly as stated.

I opened it last eve (no pix yet) and mine must be a very early unit as it has no DIP switch
and only one LED. (power?) It does have the discrete HP-IL "modules" rather than individual
resistors, etc.

The box that held it had one 10m HP-IL cable... ONE? What was I thinking!

Mark

Mark Bielman
 

Responses from my unit:
DEVID$ = "HP82165A"
DEVAID$ = "@" (40h)

Clearly set up as back-to-back GPIO interfaces.

Paul Berger
 

Looking at the picture from the front and back that Everett posted I don't think that is a 3780 processor in there, I never heard of NS second sourcing 3780, 3780 cannot use external program memory ( there is a version that has a piggyback EPROM) pin one is connected to one of the DIP switches on the 3780 that is a clock pin so it would not make sense.  I think this is more likely to be 8048/49 or the ROM-less versions 8035/39.  But then the picture that Everett shows a different part number for the processor and I cannot find either one in any of the cross references I have.

Paul.


On 2017-11-29 11:07 PM, Mark Bielman wrote:
Ran across this item in my junk box - an HP-IL to HP-IL interface (no official model number). Appears to be back-to-back HP-IL/GPIO interfaces:
2 - 1LB3's and pulse xfmr (etc)
2 - Mostek 3870 controllers (p/n 1820-2809)

As I recall, the 75C mfg line created a system where 75's could send messages back & forth. For example, I was the repair guy, so I would get a message that
a failing board (s/n) with error X was on it's way. Or something like that.

Not one of a kind (mine has #24 on it) but certainly interesting. Might use it for parts or ???

Mark B

Mark Bielman
 

Very different than mine. Here's what I found on the 1820-3218 (yup - an 8048 - very popular at HP)
1820,3218,8048,Programmed for 82164 RS232,

My "LDL" has the Mostek part.
1820,2809,3870,"""Microcontroller 82165 GPIO""",

I snapped a few photos and added as an attachment - hope that works!

Mark B

Paul Berger
 

The unit shown in your pictures sure does look like two GPIOs wired back to back.  The one that Everett uploaded the pictures for looks a lot like the 82169A HPIL-HPIB with the HPIB replaced by a second HPIL interface.  The document Everett posted with it also suggests that it might work like one.

Paul.


On 2017-11-30 1:05 PM, Mark Bielman wrote:
Very different than mine. Here's what I found on the 1820-3218 (yup - an 8048 - very popular at HP)
1820,3218,8048,Programmed for 82164 RS232,

My "LDL" has the Mostek part.
1820,2809,3870,"""Microcontroller 82165 GPIO""",

I snapped a few photos and added as an attachment - hope that works!

Mark B

Mark Bielman
 

Hi all. So I connected the box to 2 75's for a test. Seems to work but have an issue I'm sure you can resolve...

75#1 --> ENTER ":L1"; A$ @ DISP A$

75#2 --> OUTPUT ":L1"; "Hello"

The issue is that 75#1 waits/hangs waiting for data until 75#2 sends it.
I'm sure there's a way to test for data prior to executing the ENTER statement
but rather than investigating myself (time=money!) maybe you can assist.

thx

Mark

Paul Berger
 

Mark,

The only way I can think of doing that would be to poll the device and then send the enter when there is something waiting but you would still need a program running on the 75 to do the polling.   I think you need to have an I/O ROM to do polling.

Paul.


On 2017-11-30 11:46 PM, Mark Bielman wrote:
Hi all. So I connected the box to 2 75's for a test. Seems to work but have an issue I'm sure you can resolve...

75#1 --> ENTER ":L1"; A$ @ DISP A$

75#2 --> OUTPUT ":L1"; "Hello"

The issue is that 75#1 waits/hangs waiting for data until 75#2 sends it.
I'm sure there's a way to test for data prior to executing the ENTER statement
but rather than investigating myself (time=money!) maybe you can assist.

thx

Mark

 

Mark,

 

Paul is correct, the I/O ROM supports both serial and parallel polling, and these commands are not in the HPILCMDS LEX file.

 

They are documented in the I/O ROM manual, but the explanation is pretty thin with no real thorough examples. But basically you would just have a loop that runs, checking the HP-IL status on each pass through and when something on the loop ‘triggers’ then you can do the normal ENTER, OUTPUT, etc.

 

You will need an I/O ROM to do almost anything on HP-IL, beyond the very basics, but the good news is the I/O ROM provides a ton of generally useful commands and system extensions. It grew out of the April and August ROMs Raan was developing in the labs (or somewhere) and is pretty much a must have to do useful 75C stuff (other than crunching numbers, in which case, you likely wouldn’t be using a 75C).

 

 

--bob

 

Paul Berger wrote:

Mark,

The only way I can think of doing that would be to poll the device and then send the enter when there is something waiting but you would still need a program running on the 75 to do the polling.   I think you need to have an I/O ROM to do polling.

Paul.

 

On 2017-11-30 11:46 PM, Mark Bielman wrote:

Hi all. So I connected the box to 2 75's for a test. Seems to work but have an issue I'm sure you can resolve...

75#1 --> ENTER ":L1"; A$ @ DISP A$

75#2 --> OUTPUT ":L1"; "Hello"

The issue is that 75#1 waits/hangs waiting for data until 75#2 sends it.
I'm sure there's a way to test for data prior to executing the ENTER statement
but rather than investigating myself (time=money!) maybe you can assist.

thx

Mark

 

 

Robert wrote...
They are documented in the I/O ROM manual, but the explanation is
pretty thin with no real thorough examples. But basically you would
just have a loop that runs, checking the HP-IL status on each pass
through and when something on the loop ‘triggers’ then you can do
the normal ENTER, OUTPUT, etc.
I'm not at all familiar with the 75 programming, but with the IO ROM,
couldn't you use the ON SRQ GOTO xxxx and ENABLE SRQ statements to
have the program automatically branch rather than having to poll it?
Or does arriving data on the IL loop not generate a service request?

Everett

======================================================================
Everett Kaser Software Logic & Puzzle Games for computers
PO Box 403 http://www.kaser.com
Albany OR USA 97321-0117 Phone: 541-928-5259 8am-8pm PACIFIC TIME
======================================================================

Mark Bielman
 

My end goal would be to do Everett's approach. But I need to create a schematic to see how the handshake lines
are connected. As I recall, the connections determine if an SRQ is generated or not. (and the control bits as well.)
I also think the button on my LDL sets the SRQ but have to look into that.

But initially, will do a simple loop, reading the status byte(s) to monitor what's going on.

Sounds like fun!

Mark B