Dane.Kouttron

[1.2.10] IBM 7575 Arm Development

Update:   Check out the 3D-Print-Arm Project that's based on this platform!  
What?


Aquired an IBM 7575 Robotic arm from a lab-cleanout, without controller.  

Its awesome, but far from operational.
The following page documents upgrading motors to high-torque steppers, building controllers and feedback systems in attempts to get the arm controlled by microcontroller.



DISASSEMBLY DOCUMENTATION
STEPPER MOTOR INSTALL DRIVER HARDWARE IMAGE DIRECTORY

Disassembly Documentation
Descriptor Image
The following describes the disassembly and replacement of a DC brush motor drive, and swap in of a 4 wire stepper motor on an IBM SCARA 7575.

Shown left is an image from the begining of the Arm disassembly, the theta 1 motor assembly is visible. Theta 1 is driven from a DC brush + encoderer'd motor and a belt drive, to a 1:3 planetary gearset

A sketch of the system is shown right, its an image from IBM 7575 manual excerpt, borrowed from (LINK)
Front z axis assembly image. Arm is driven by Brush DC motor thru wormgear setup. Optical encoder is used for positioning feedback.

  IR beambreaks are used for the end of travel sensors
Dc Motor for Z axis and encased optical encoder assembly in case. Brush points are encased by outer black covers. Red / Blk wires are directly wired to brush DC motor.

I like how the whole system is connector-ized. As this is an industrial machine, I was expecting long wires -soldered connections to reduce mtbf. Connectors made swapping in replacement hardware a 'snap' 
Shown right is the z-axis motor assembly. The cover for the encoder assembly is removed. Using the feedback lines i was able to scope out the square-wave manchester position encoding.

Note optical disk assembly and visible LED for optic feedback.

Motor assembly is attached via belt to worm-drive for z axis movement. Damping device appeared damaged and was removed. Stepper motor 'locking' will be used instead
Theta 1 drive point, including drive motor (far right)  belt assembly for harmonic drive.

Swapping In Stepper Motors
Descriptor Image/ Media Image / Media
Adapting stepper motor to shaft for main belt assembly. The stepper crufted from RPI Eclub was slightly smaller in shaft diameter than the existing belt cog, a sliced up piece of copper tubing was used. Tightening the spline and making small modifications allowed for a cheap collet.   

Removing the older DC brush motor and adding in an adapting in a plate for the stepper motor.
Long threaded bolts were used for tensioning the stepper motor to the robot.
Finally the stepper motor was added in to the z axis drive. Note the 'snug' fit.  This was the largest 'scavenged' stepper motor i could find at the time. Unfortunatley it lacked any formal datasheet and holding torque estimations were used to determine the operating characteristics.

To test holding torque of an unknown 4 wire stepper, the motor drive was commanded to hold a position, and mass was added to a lever attached to the motor shaft. The moment on the motor axis  and the mass require for it to miss a step was used as the oz-in estimation for the motor.  Unfortunatley, using an unproven stepper perf-board motor drive with current limits to measure the torque results in the measurement of either the motor torque, or the stepper driver's max holding current, whichever is the lesser
: )


Early Stepper Motor Driver Hardware
Descriptor Image/ Media Image / Media
To test drive the stepper motor that me and anthony added to the robot, i crafted a perf-boarded Stepper motor drive, intended to drive two stepper motors. This is a power-resistor pull-up style single ended driver (pull, not push-pull)

This drives the Z Axis stepper and Theta 2 Stepper. Copper Heatsinks fabricated for thermal dissapation (namely when locking steppers)

Brush DC Motor Encoder Hardware Debug
Descriptor Image/ Media Image / Media
The encoder signals displayed at 50us / Div. DC brush motor operating at 5V (instead of 12) and the output signal is between yellow and ground. The yellow-ground signal is plenty of the purposes of the initial testing, and will be used in the interrupt / brush dc control scheme.

(Note, im @ home, thus the crt-scope)

DISASSEMBLY DOCUMENTATION
STEPPER MOTOR INSTALL DRIVER HARDWARE IMAGE DIRECTORY

From the sponsors:


If you have questions or comments, shoot over an email.

Comments:
HTML Comment Box is loading comments...



Dane.Kouttron
Rensselaer Polytechnic Institute 
Electrical & Electrical Power
631.978.1650