looking for 1130 plotter interface technical documentation

[Roland Langfeld]

Hi,

we (www.technikum29.de) have a running 1130 with disk, card reader and printer. The system runs fine on DMS2, we are able to compile and run FORTRAN programs.

Originally this system was connected to a Calcomp-plotter via a special interface made by a German company. The company doesn't exist any longer. The interface is connected via SAC to our 1130. We have all the hardware, but no documentation and no software to run this interface. The original 1130 plot routines will not do the job.
Unfortunately our system does not have the original 1130-plotter interface to connect a plotter directly. I expect it is close to impossible to get an original plotter-interface for our machine. So we intend to built a plotter interface by ourselves.

Here is our question: we are looking for documentation, schematics, wiring diagrams etc. for the original 1130 plotter interface.

Any kind of help is highly welcome,

kind regards
Roland

[Michael Short]

You might start by going to bitsavers.org in the PDF Document Archive section.

Go to IBM/1130/INTF. There is manual with plotter information along with interface

specifications. 

Roland

--
You received this message because you are subscribed to the Google Groups "IBM1130" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibm1130+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[peter....@tnmoc.org]

Hi Roland,

I know there was a 1626 interface unit (an external cabinet the size of
the 1442 I believe) and 1627 plotter. I know the 1626 was needed to
connect the plotter to the IBM 1620 system (hence the model numbers) but
I'm not sure if that was also used when connecting the 1627 to the IBM
1130. None of the 1130 configuration diagrams mentions the 1626 so the
1627 may have been directly connected.

The 1627 plotter (or 1626) to the peripheral interface inside the 1130
where the console and paper tape reader connect (not via the large
connectors where the 1442 and 1132 connect).

You will need the necessary SLT cards and cabling in the 1130 and the
internal connector (same as used by the console) to connect to the
1626/1627.

The docs you need for the 1626 and 1627 are under the IBM 1620 system
on bitsavers

http://bitsavers.informatik.uni-stuttgart.de/pdf/ibm/1620/fe/

For the logic see the overview in 1131-B sections XG501 and XG701 which
will take you to the main sections:

http://bitsavers.informatik.uni-stuttgart.de/pdf/ibm/1130/fe/1131-B/1131_ALD_005_Apr68.pdf

and 1131-C sections XG101 and XG111

http://bitsavers.informatik.uni-stuttgart.de/pdf/ibm/1130/fe/1131-C/1131-C_003_ALD_XC-XW_Jun73.pdf

For connections inside the 1130 you can see the physical interface in
this document

http://bitsavers.informatik.uni-stuttgart.de/pdf/ibm/1130/fe/1131-B/1131-B_004_ALD_XP-ZW.pdf
section ZA101

The above should get you most of the way

Note: I don't have a plotter on my Museums system so can't help further.

Regards

Peter Vaughan
TNMOC IBM 1130 restoration lead

[John Doty]

On Oct 30, 2016, at 5:54 PM, <peter....@tnmoc.org> <peter....@tnmoc.org> wrote:

> I know there was a 1626 interface unit (an external cabinet the size of the 1442 I believe) and 1627 plotter. I know the 1626 was needed to connect the plotter to the IBM 1620 system (hence the model numbers) but I'm not sure if that was also used when connecting the 1627 to the IBM 1130. None of the 1130 configuration diagrams mentions the 1626 so the 1627 may have been directly connected.

We had one of those rebranded Calcomp plotters at SUNY Brockport. I don’t recall an additional interface unit: it must have connected to the main cabinet (1131) or the 1133 Multiplex Control Enclosure.

John Doty Noqsi Aerospace, Ltd.
http://www.noqsi.com/
j...@noqsi.com

[Tom Watson]

Some notes on how the IBM1130 interfaces to plotters (the Calcomp 565)
in this case.  The connector on the Calcomp plotter has many
connections, but basically it needs six signals (and ground) to make
anything happen.  These signals are to make the plotter move in +X, -X,
+Y, -Y, +Z (pen up), and -Z (pen down).  You give the plotter short
pulses (positive or negative, depending on the pin used) and the
plotter moves where you want it.  You can combine X & Y to move on the
diagonal.  The timing of the pulses (interrupts/busy for the 1130)
determine the speed of the movement.  The Calcomp 565 (1627 model 1)
can go at 300 steps/second on the X/Y axis, and 10 pen movements per
second.  It is up to the interface to make sure that you don't
"overspeed" the plotter, as the stepper motors could "skip" if you do.
If you look at the interface described for the plotter you will see
that a total of 6 bits are exposed to the interface, corresponding to
the 6 motions described above.

If you actually have a Calcomp plotter, the interface pins are:
1,2,3,4 - Positive pulse X & Y directions.
5,6,7,8 - Negative pulse X & Y directions.
9, 10   - Positive pulse pen up and down.
11, 12  - Negative pulse pen up and down.

I believe that the ground was on pin 15 (I don't remember that well).

Yes, I interfaced a Calcomp 565 plotter to a computer and ported the
IBM1130 software to that platform (it was an XDS Sigma 5).

Many "other" plotters (I understand that Zeta was one vendor) had
compatible interfaces.  The pulses should be around 12 volts and 1
microsecond long to work correctly.  Real Calcomp plotters would work
on lesser voltages, but some of the "compatibles" wouldn't.  This was
demonstrated by a salesman in a most embarassing way.

Much of this is from memory from many moons ago.  Hope it helps.

On Sun, 2016-10-30 at 13:24 -0400, Michael Short wrote:
> You might start by going to bitsavers.org in the PDF Document Archive
> section.
> Go to IBM/1130/INTF. There is manual with plotter information along
> with interface
> specifications. 
>

> On Sun, Oct 30, 2016 at 11:03 AM, Roland Langfeld <rolandlangfeld@gma

> il.com> wrote:
> > Hi,
> >
> > we (www.technikum29.de) have a running 1130 with disk, card reader
> > and printer. The system runs fine on DMS2, we are able to compile
> > and run FORTRAN programs.
> >
> > Originally this system was connected to a Calcomp-plotter via a
> > special interface made by a  German  company. The company doesn't
> > exist any longer. The interface is  connected via SAC to our 1130.
> > We have all the hardware, but no documentation and no software to
> > run this interface. The original 1130 plot routines will not do the
> > job.
> > Unfortunately our system does not have the  original 1130-plotter
> > interface to connect a plotter directly. I expect it is close to
> > impossible to get an original plotter-interface for our machine. So
> > we intend to built a plotter interface by ourselves.
> >
> > Here is our question: we are looking for documentation, schematics,
> > wiring diagrams etc. for the original 1130 plotter interface.
> >
> > Any kind of help is highly welcome,
> >
> > kind regards
> > Roland
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "IBM1130" group.
> > To unsubscribe from this group and stop receiving emails from it,

> > send an email to ibm1130+u...@googlegroups.com.

> > For more options, visit https://groups.google.com/d/optout.
> >

> -- 
> You received this message because you are subscribed to the Google
> Groups "IBM1130" group.
> To unsubscribe from this group and stop receiving emails from it,

> send an email to ibm1130+u...@googlegroups.com.

[Tom Watson]

The interface for the 1627 plotter on an 1130 system DID NOT use the
1626 interface that was used on the IBM 1620.  The interface on the
IBM1130 was contained in the main chassis of the IBM1131 processor
console (where the keyboard is).  As I noted before, it is a pretty
simple interface for the IBM1130.

The CalComp manual (it is for a 566, which is similar to a 565) is on
bitsavers at:
http://www.bitsavers.org/pdf/calcomp/Model_566_Plotter_Instruction_Manu
al_1963.pdf

The connections are indicated on page 49, the connector is "P5".

For how it is programmed, see page 159 (of the pdf) of:
http://www.bitsavers.org/pdf/ibm/1130/functional_characteristics/GA26-5
881-6_1130_Functional_Characteristics_Apr72.pdf


Hope this adds some knowledge.


On Sun, 2016-10-30 at 18:20 -0600, John Doty wrote:
> On Oct 30, 2016, at 5:54 PM, <peter....@tnmoc.org> <peter.vaughan

[Roland Langfeld]

Dear Peter, thankyou for these helpful documents and your clear guidance how to find the important pages in these comprehensive documents. That is exactly what we are looking for !!

Roland

[Roland Langfeld]

And many thanks to all the others for their hints and explanations. I need a few days to digest all the information and I will come back to report on any progress.

Regards
Roland

[Roland Langfeld]

Again thankyou for all your valuable information ! In the meantime we decided to built our own interface. Fortunately we got all required original chips from another collector. Now it would be very helpful...

.. to get a picture of the original circuit-board (showing chip position and wiring), the circuit board seems to have the internal number 5806223,

and/or

... to get a wiring diagram showing position of the chips and wiring.

Roland

[Carl Claunch]

Hi Roland

The card in question is a double high one, fitting into backplane slots M4 and M5 even though the ALD pages list it only as M4. 

Also, it was common for IBM to use different backplanes when features are installed or not installed - thus you would need to verify that your backplane had connectivity from the pins for sockets M4 and M5 to the relevant signals from the machine (e.g. B reg). Also, they would need to have connectivity out to the of the side connectors in order for the driving signals to the plotter to be wired to the socket at the rear of the machine. The socket and cabling will not be installed if the 1130 did not have the plotter interface configured. 

Another possible complication is whether the DU111 card installed in your system, its backplane and the connecting cables carried the signal Area Code 5 which is essential for your card to be aware it is being addressed by an XIO. Often if a peripheral was not configured in the system, related signals such as interrupts, Area Codes and cycle stealing lines were also not installed. Without those, the plotter card you install wont be activated for XIOs and the card can't present the completion interrupt. 

My 1130 does not have the feature configured, or I would send you the pictures of the card you requested. 

I have addressed this differently, in order to support a Calcomp 565/IBM 1627 plotter. I made use of a different 1130 feature called the Storage Access Channel (SAC) which brings out the various signals needed to build an interface for just about any device. With it, I support disks, printers, card readers and the plotter, because it is possible to get access to the needed input and output signals. If you have the SAC Interface, you could make use of that to attach the plotter, 

It would still be feasible to wirewrap lines and find the various signals you need, if the backplane isn't wired for the card, and probably a smaller task than using the SAC. I would only recommend a SAC based approach if there are other peripherals you need to attach in addition to the 1627 that would warrant the substantial work involved. 

Happy to chat and provide advice as you work on this. You can start with XG101 and XG111, validating that all the signals you need to run to other parts of the 1130 are in fact wired up by using a continuity tester. If this is the case, you need only figure out a way to route the outputs to the 1627 which is a much simpler task. 

Carl

[Roland Langfeld]

Dear Carl,

thank you very much for your advice: we will check carefully, if all the required signals/cables are in place - I will report soon !

Roland

[John R Pierce]

On 11/5/2016 1:20 AM, Roland Langfeld wrote:
> Again thankyou for all your valuable information ! In the meantime we decided to built our own interface. Fortunately we got all required original chips from another collector. Now it would be very helpful...
>
> .. to get a picture of the original circuit-board (showing chip position and wiring), the circuit board seems to have the internal number 5806223,

those SLT logic cards are multilayer with a lot of internal traces, I
doubt a photo of the card would be sufficient to reproduce it.


--
john r pierce, recycling bits in santa cruz

[Roland Langfeld]

Dear Carl,

you said

"I have addressed this differently, in order to support a Calcomp 565/IBM 1627 plotter. I made use of a different 1130 feature called the Storage Access Channel (SAC) which brings out the various signals needed to build an interface for just about any device. With it, I support disks, printers, card readers and the plotter, because it is possible to get access to the needed input and output signals. If you have the SAC Interface, you could make use of that to attach the plotter, "

I assume this solution will not work with the original FORTRAN PLOT commands, is the required software available ?

kind regards
Roland

[Roland Langfeld]

Hi Peter (and all the other helpers here),

I've read carefully all the links you posted here - I'm looking for a very special diagram of the plotter interface, which I could not find there. For explanation I added a picture of our plotter interface together with a picture of the type of schematic we are looking for (the schematic in the attachment is related to another card, NOT the plotter interface). I hope someone can help us,

kind regards

Roland

Roland

[Richard Stofer]

It shouldn't make any difference whether a standard peripheral is connected directly or through the SAC. If it were otherwise, there would have to have been many versions of the drivers and that simply isn't the case.

http://www.mirrorservice.org/sites/www.bitsavers.org/pdf/ibm/1130/fe/SY26-3670-4_1130_Features_FETOM_Dec70.pdf

See Part 5

AFAICT, the SAC works as a wide spot in the road. The exception seems to be that devices attached to the SAC are responsible for dealing with the IOCC word. But that isn't a software issue and, again, AFAIK all devices had the capability.

Carl is creating a system where everything except the CPU is a virtual device. He wanted to use the SAC because it provided a single point of attachment to the CPU. His FPGA and surrounding circuitry allows everything to run over USB to some other system. Very clever work!

I built an FPGA version of the 1130 using only the "Functional Characteristics" manual.

http://media.ibm1130.org/E0006.pdf

I chose only devices that used cycle stealing as a DMA channel seemed cleaner than having a bunch of interrupts running around. I found the description of the various XIO functions sufficient to be able to duplicate the functionality and the FPGA version runs the OS unmodified. The only thing I had to add was a 'page eject' subroutine to complete the plots I send to a LaserJet. It's just an XIO with all 6 bits set and the hardware tosses a command to the LJ. There is a microcontroller between the FPGA and the LAN that does all the conversion of motion commands and HPGL sentences.

[Carl Claunch]

Hi Roland

The software you can use is identical regardless of how it is attached - same area code, interrupts, ILSW bits, commands, etc. Acts the same as a direct attach device.

Carl

[Carl Claunch]

Hi Roland

If you can't find a card schematic, you can infer quite a bit of it from the SLT Designers Handbook, documentation of the circuits in the various SLT modules on the card, and similar card schematics. 

For example, there are other cards with one-shot timers that use the same SLT module, for which schematics exist, leaving you a template. Using the designers handbook you can figure out how to set up the timer for intended duration.

Other cards combine the modules in similar ways, allowing you to figure out the resistor values and connections between module pins. 

Even if you had one of the cards, it is hard to work out the wiring and components without a schematic. Your past success engineering a replacement for a memory line driver board, a simpler board with less circuitry, illustrates the work involved. 

Best wishes,

    Carl

[Carl Claunch]

Hi Roland

The picture of the card shows six AI modules (361451), 2 FTX modules (361497), 4 II modules (361479), 2 FDD modules (361459) and 7 361435 which I don't have a schematic or identification for right now. 

The 2 FTX modules provide four transistor drivers each; these are obviously the amplifiers driving the pen, drum and carriage output. 

From the older designer documents available on the web, a FF requires a pair of AI modules plus resistor an resistor/capacitor packs. The six outputs each have a FF, and there is a response FF to trigger the interrupt. These FFs can't be built in that manner, as you only have 6 AI modules on the board and it would take 14. I am going to presume that the mistery 361453 module  is a complete FF in a can, as that lines up nicely with the number of these modules. 

The FDD modules each provide four dual diodes used to add additional inputs for logic gate circuits.  They are also used to build circuits such as the two Single Shots (SS) which use portions of an FDD and of an II. The classic SSA would have parts of a DCI module, plus an adjustment potentiometer, but neither of those parts are on the card. The circuit will also differ from the classic SSA in that these have two inputs, triggering one of two preset timings depending on the model of the plotter used; jumpers on the backplane determine which input is wired into the circuit. 

We don't know the exact wiring for these special SSA, therefore.

The remaining AND and OR logic on the board is accomplished with the AI, II and remaining bits of the FDD modules.

The board type 5806332 has an equivalent 'new type' board number 5800132. I don't have that schematic either but you might search under both board numbers. 

Conceptually you could build a replacement set of logic that exactly matches the XG101 and XG111 diagrams, add some threshold converters to interface it to the SLT logic levels, and cable it via SLT sockets from sacrificial board(s).

Carl

[Roland Langfeld]

Dear Carl,

Thankyou for these helpful hints. We will put together all information given in this thread and decide, how to go on. In the meantime someone offered us to lend us a plotter inferface card - lets see, how this will help us further...

Roland

> Roland

[Johannes Thelen]

I don't know if this coming too late, but IBM 1800 has a build-in interface for the 1627. Here's an ALD for it, look page EK221 and forward: https://drive.google.com/file/d/0B_DR111cK6W-NEdwLVd0M0ZqNUk/view?usp=sharing

Rest of ALDs you can find here: https://drive.google.com/drive/folders/0B_DR111cK6W-UW5qVVZrUGRoaUU?usp=sharing

Johannes Thelen

ennenmikrotietokoneita.blogspot.fi

[Roland Langfeld]

Hi Johannes,

It is never too lat for a hint, we got similar docs from different sources. Actually we are waiting for an original Interface for testing purposes, may be that will speed up our plans.

I wish you and all the other helpful guys in this forum a Happy New Year !

Roland