Jonathan Hancock wrote: > Hi Jon. Thanks for your response. I hope this helps: > > > > The Puma560 has 6 joints all having 40v dc brushed motors. The 3 lower > joints are rated at 160W and the final 3 at 80W. > > > > The encoders are sin/cos types with differential outputs and an index > output. I can convert this to A/B square waves if necessary. You'd better check the resolution of these encoders. Some (not all) analog output encoders have low resolution and use interpolator boxes to develop the needed resolution. These Rho-Theta style arms need to have pretty high resolution in the joints or there are certain positions it can get into where tiny movements of one joint can cause big errors in end effector position. Each motor > also has a pot on the back for the control system to find its home > position at power-up. There are no limit switches. > Hmmm, no limit switches, huh? Well, there is no standard arrangement in EMC2 to use pots for setting joint limits. You could, of course, build some electronics with pairs of comparators to detect a joint at the travel limits. You would at least need some kind of sensor to set a home position for each joint. > > > The tree lower joints have a break that needs to be energised by 24v to > allow movement of the motors. > > > > I'm not sure what the control signal to the existing amplifiers needs to > be as I have very little documentation on my system. I am happy to > replace them for a more modern solution if it's easier. There is however > a 'high power discharge' board in the original to soak up the high > voltages created by the inertia of the arm. I guess I can use the > original psu to provide +/- 40V. > It could be +/- 10 V analog velocity commands, or torque commands. But, unless they are a standard brand (Servo Dynamics, Westamp, Kolmorgen or such) it might be hard to get docs on them. If you have tachometers on the motors, then it is likely to be a velocity servo, and you may want to preserve the servo amps if they are still functional. If no tachs, then I wouldn't worry too much about saving them. > > > It would be great to get this working properly as it's been a long term > ambition. I've found some university sites that have made changes to the > original controller but none that have started again. And then there's > the software....! But, EMC2 makes it fairly easy. You have to determine the destance between joints precisely and set up the reverse kinematics for those measurements, and EMC handles it pretty much from there. You do have to work out the correct homing sequence so that things don't collide during that movement.
Jon ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
