My RM-101 ARM robot

Rsadeika

Since I was taking pictures, here is my RM-101 robot. This will also be a project, not sure how I will do this one though. As you can see, in the first picture, it has a parallel connector. That is the only way you can talk to the robot.

I think the Propeller could handle a parallel connector(printer port), but I am not sure what the circuit would look like. And of course I would be writing a program in PropGCC, so Iam not even sure what the C code would look like to do this. I know of a web store that sells parallel port break out boards, then I would need some wires going to a breadboard, which would connect to the Propeller pins. Since the parallel port is sending one way, I wonder if it needs a common ground, plus I am assuming that I would be dealing with just eight wires plus ground if necessary. lot to think about with this project.

Ray

jones

It's been forever since I've used a parallel printer port for I/O, but I can say that yes, you will need a common ground between the Prop and the robot. The other thing you'll need is the strobe line to latch the parallel data. You have to set the data up on the data lines, let them settle, then pulse the strobe line. The robot docs should give you a timing spec, or failing that generic port info should be easy to find online.

The first reference I picked up with the Centronics connector pinout shows the strobe as pin 1, then eight data bits on pins 2-9 (LSB on 2) and ground on pins 18-25. You'll have to look at the robot docs to see if it uses any of the other pins. You'll also need to level shift, since the parallel port expects TTL levels.

Some applications kludged the printer port for bi-directional communications using the handshake lines, so you need to be sure you know how the port on the robot is set up. Definitely read the manual. If you don't have that, I found this project online where they did the hard work of disassembling the robot's Z80 firmware to figure out the commands for you: http://shackspace.de/?p=3284

Rsadeika

Thanks jones. Now that I started to think about this, I think I still have my KayPro computer somewhere. If I remember correctly, I think it has a centronics printer port. In fact, that just might be the computer that I used to work with the RM-101. I bet if I dug even deeper, I might be able to find some of the testware that I had developed for the RM-101. 5 1/4" floppy disc, CP/M, who in the heck remembers how to work with CP/M let alone write a new program for the RM-101, in pascal, I think that was readily available.

Also, I do have the RM-101 original manual. What am I getting myself into, next thing you know I will want to start up a company, and buy some more robots, ... , time to go lay down and take a nap before I ...

Ray

jones

I think it might be easier to get the arm going with a Prop and a little glue logic. OTOH, getting the Kaypro working might be an interesting exercise in its own right.

Good luck with the project, or at least enjoy your nap!

Bob

erco

A beautiful arm and what a great project. Must be vintage if it's parallel port, so I like it even more. What year is that, and are you the original owner? Most arms I've seen have either zero wear (almost never used) or excessive. Very little in between. Surely a Prop is up to the task, dare I say overkill just for motion control. But you'll have plenty of room for expansion to get into robot vision.

Many years ago, some 'expert' described robotics something like this:

'Tie one arm behind your back, smear vaseline on your glasses, and assemble something complicated using chopsticks.'

Rsadeika

I think it was 1983 when it was purchased, and it had minor usage. Basically it was used for practice and doing some demonstration runs. It is a very solid piece of machinery, metal gears, and metal enclosure, it feels like weighs in at maybe thirty five pounds.

Actually what you need for this is to be able to setup a repetitive state. If recall it does not have any memory of its own, so you have to have an external device to that feeds it a program, but I could be wrong about this. I will have to read the manual pretty soon and figure out exactly what the capabilities are.

Ray

Duane Degn

Rsadeika wrote: »

I know of a web store that sells parallel port break out boards, then I would need some wires going to a breadboard, which would connect to the Propeller pins. Since the parallel port is sending one way, I wonder if it needs a common ground, plus I am assuming that I would be dealing with just eight wires plus ground if necessary.

Often devices which use Parallel ports have many shared pins. I think there were at least six (I'm remembering nine) ground lines used by the CNC controller I have.

I cut an old parallel cable and soldered some normal 0.1" headers onto the wires I needed.

You should be able to see the 2x8 black connector in the upper center of this photo.



There's a right angle 2x8 male header on a small piece of proto board just to the left of the black female connector. The male connector is wired to a Propeller Proto Board.

The 3.3V logic from the propeller didn't control the stepper driver. I would need to use level shifters to make this hardware work.

I ended up using a different stepper driver which worked fine with the Propeller's 3.3V logic.

If you don't want to cut your parallel cable, you might be able to salvage a mating connector from on old printer. Of course it's been a while since printers had parallel ports. I still have two parallel port printers which still work (though I don't use them) but I've gone through several USB printers which are now sitting in one of my junk part piles.

As you noted, there's always the option of purchasing the appropriate connector.

Does the documentation list the pin outs?

JonM

That is a cool looking robo arm especially for it's age.

There were some folks that took the same RM-101 and disassembled the code from the EPROM and control it from a script. They did have to hack the parallel port since it was not the original and documented the data so that might be of interest to you. This was posted on Hackaday as well
shackspace.de/?p=3284

The other option is to get a USB to Parallel cable to control it via a USB port.
https://amazon.com/IOGEAR-USB-Parallel-Adapter-GUC1284B/dp/B00018RT1E

Oddly enough, I recently worked for one of the largest Flash memory companies on the planet which is based in Japan and I was surprised to see their main SSD test fixture was controlled via a parallel port. I guess you use what you know.

Rsadeika

Back in those days I did have an opportunity to get my hands on a RM-501 robot, that was a light industrial version. I had to talk extra nice to the Mitsubishi sales engineers to get them to let me "evaluate" it for a one week or maybe even two weeks, free of charge.

I did find a Mini-ITX board with a centronics port on the board, in the basement, I think the board still works. I will have to try too get it running, maybe load up a small Linux distribution, and see if can get access to the parallel port. I am supposed to be retired, with these side projects it is starting to feel like work again.:-)

Ray