Robot using already existing servo amplifiers

More
11 Mar 2014 18:47 - 11 Mar 2014 18:50 #44662 by dimech
I am a student of Mechanical Engineering Department in National Technical University of Athens.
My final year project requires me to reprogram the MITSUBISHI RM-501 robot through linucCNC environment. The project requires use the already existing servo amplifiers of the drive unit (zero budget project)

I am wondering if it is that possible. :dry:

In my search I was not able to find any datasheet for this drive unit except for a user's manual of the model. Unfortunately, this manual does not contain the information I need regarding the hardware.

Servo type: Sanyo Denki Super Mini 24V DC
Thank you in advance.
Last edit: 11 Mar 2014 18:50 by dimech.

Please Log in or Create an account to join the conversation.

More
11 Mar 2014 19:06 #44663 by ArcEye
Hi

In my search I was not able to find any datasheet for this drive unit except for a user's manual of the model. Unfortunately, this manual does not contain the information I need regarding the hardware.


What about this one?

At 353 pages long you would think something would be relevant

www.datasheetarchive.com/dl/Datasheets-SW18/DSASW00346856.pdf

regards
The following user(s) said Thank You: dimech

Please Log in or Create an account to join the conversation.

More
11 Mar 2014 19:49 #44667 by dimech
Thank you for the immediate answering!

The robot is really old ('86 I think),so doesn't match a lot with the new technology. My main concern is, if the boards of the servos, which are located in drive unit, are capable do whatever mesa board does. Also I have not so much experience on electronics, so I would be grateful if you could suggest me a way, how to figure what signal is for without having datasheets. I am new in this field!
Excuse my English…

Please Log in or Create an account to join the conversation.

More
11 Mar 2014 23:05 #44676 by PCW
I think the trick will be figuring out the servo motor drive interface wiring (the encoder wiring should be pretty simple).
It looks like the RM-501 is a small teaching robot with quite tiny DC servos. I suspect the motor drives are simple PWM
driven HBridges so should be easy for LinuxCNC to drive, but lacking a hardware manual or schematics, you will need
the help of someone familiar with electronics to decipher the drive connections.
The following user(s) said Thank You: dimech

Please Log in or Create an account to join the conversation.

More
11 Mar 2014 23:21 #44679 by andypugh

I am a student of Mechanical Engineering Department in National Technical University of Athens.


Is Vassilis Kefalas still there?

Please Log in or Create an account to join the conversation.

More
12 Mar 2014 02:48 #44695 by dimech

I think the trick will be figuring out the servo motor drive interface wiring (the encoder wiring should be pretty simple).
It looks like the RM-501 is a small teaching robot with quite tiny DC servos. I suspect the motor drives are simple PWM
driven HBridges so should be easy for LinuxCNC to drive, but lacking a hardware manual or schematics, you will need
the help of someone familiar with electronics to decipher the drive connections.


Thank you! I am not familiar with the expressions PWM and HBridges, but I am going to study!

Is Vassilis Kefalas still there?


Still Here! He is Assistant Professor of Engineering at the Department of Applied Mathematics and Natural Sciences right now.

Please Log in or Create an account to join the conversation.

More
20 May 2014 02:18 - 20 May 2014 03:23 #47062 by dimech
Here I am after two months and 8 days!

I just finish the schematics , in an attempt of reverse engineering, of driver motor boards.

The boards are using flags to activate each motor section. A DA conversion with a resistor ladder controls the velocity and up down counters are in charge for the control and direction. Still after schematics is not very clear how to manipulate the boards.

I am attaching the sch. It's one of the three boards. (The other two have got, pretty much, the same orientation)

I need your knowledge and your experience. Have you ever see something like this? Is it possible to make them functional with a step/dir communication?



Thank you in advance!
Last edit: 20 May 2014 03:23 by dimech.

Please Log in or Create an account to join the conversation.

More
20 May 2014 05:36 #47064 by andypugh

Is it possible to make them functional with a step/dir communication?


I can't claim to understand the schematic, but I think that it will be easier to make LinuxCNC talk to the drives than to make the drives understand step-dir.

Can you work out what input the drives are expecting?

Please Log in or Create an account to join the conversation.

More
20 May 2014 09:09 #47073 by PCW
can you post a link to the actual image file? the google image is unreadable to me

Please Log in or Create an account to join the conversation.

More
20 May 2014 15:41 - 20 May 2014 15:42 #47083 by dimech

Can you work out what input the drives are expecting?


I am trying to acquire signals, but they are really noisy. A lot of spikes and mixed lines on oscillator's screen

There is an 20pin interface for each board, which communicates with the cpu.
pins 1~ 3 carry the information
pin 4 is a clock for a flip flop
pin 5 is a trigger for a buffer which controls the signal of hard sw
pins 6~12 they are something like flags
pins 13,14 outputs (?)
pin 15 reset for all flip-flop and up-down counters
pin 16 rectangular signal for clocking
pin 17 signal for the grip
pins 18~20 no connection

can you post a link to the actual image file? the google image is unreadable to me


SHLD L/R WRIST BOARD

Thank you
Last edit: 20 May 2014 15:42 by dimech.

Please Log in or Create an account to join the conversation.

Moderators: PCWjmelson
Time to create page: 0.183 seconds
Powered by Kunena Forum