MechanoBlog

New Cards

2008-08-11T10:59:00.001-07:00

We received two new servo cards, which made it possible to get the arm moving and mostly working again. However these cards don't seem to be functioning quite properly. When ever you move a joint that is controlled by one of them, the joint will oscillate about the point it was supposed to finish at. This behavior seems to mean that the cards are not quite compatible with the controller we have. Unfortunately the new cards did not fix the problem with the zero index not being found on joint 1. I've begun testing the wires in order to make sure that there is no problem there but if its not the encoder, card or the wires I really have no idea what else could be wrong.

Repaired! and then it broke....

2008-08-02T15:56:00.000-07:00

A lot of progress has been made since did my last post. The software was not the issue unfortunately, however the memory card was. Once the memory card was replaced, the system was able to load the software and initialize. After initializing the system, I was able to get all 6 joints of the robot to move. After playing around with the robot in teach mode for a little bit I tried to calibrate to system. When calibrating the system the 1st joint was unable to find the zero point. In order to troubleshoot this i first switched two encoders how ever the problem stay at the 1st joint so that meant the problem was most likely in the servo card. While swapping two of the servo cards one of the previously working cards broke. Once the card broke there was nothing more to be done until it is replaced. Hopefully once these cards are replaced everything will work and I can start learning how to program it.

Software!

2008-07-11T16:33:00.000-07:00

This week has gone pretty well. I got a hold of the people at rpautomation, who were very helpful. At their advice i pulled out all of the chips to see if one of the components was not securely fitted. While i did not find any lose components i did manage to find a wire which had broken off. Once i fixed the wire everything started working. After getting everything to work i was incredibly excited to see what the arm could do but no, the software we had received was no longer working and most likely is corrupted. The guys at rpautomation sent us a new copy but unfortunately the disk they used had been mislabeled and was not the software we need so now its more just sit around and wait till we get the software. Hopefully nothing else will be broken buy who knows what actually will be. Things are picking up though and I will get this arm working before the end of the summer.

Chasing Power

2008-06-27T17:16:00.000-07:00

Work this week did not go particularity well, we have decided to try to repair the original control for the robotic arm rather than construct a hacked controller. This would be fine but we have no idea what is broken. All we know is that there is no signal to the monitor which we initially thought could mean two things, 1 that the monitor was broken or 2 that the serial port was broken. The previous owners had wired in a connector so that there was a serial cable that could be connected to a PC. I decided to check to see if the serial card was any good by hooking it up to a PC and then running some tests to see if there was any signal being sent.
These test showed that there was no signal, which meant that the serial board was most likely shot. However before ordering a new board, I had to check the power supply to make sure that there was a voltage being supplied to the circuits. If there was no power being supplied to the circuit boards it would make sense that nothing would work.
I then started checking the power being supplied by the low voltage supply which is the power supply for all of the circuitry. While the fuses are intact no voltage is coming off the board. According to the manual there is supposedly a fuse that if it were blown out no power would be supplied but unfortunately there is no indication of where this fuse is which makes it hard to check if it is dead.

PUMA 500 Series Robotic Arm

2008-06-19T11:43:00.001-07:00

Recently I was hired to set up and if necessary repair a PUMA 560 robotic arm. The arm itself is a six axis manufacturing arm, that is controlled by computer. There are two methods of control with the original system, the first is by programing in VAL (variable assembly language), a series of points and rotations where you want the machine to move too. The second method of control is to use the teach pad with which you can turn any particular joint on its own or move the arm as a whole, once the arm has reached the desired location you can record it onto the computer.
The sequence to start up the arm is fairly simple, all the breakers need to be checked to ensure that they have not been flipped, power is plugged in and the the main break is turned on. When I first flipped the switch i found that the computer was getting power but not loading. In order to troubleshoot this I had to manually check the circuit boards. What i found was that one of these boards was not properly inserted and probably had not been for a while. Reinserting this board was quite difficult due to the overlapping connecting cables that crossed in front of it. When the board was finally reinserted, the computer would still not initialize. The next step according to the manual is to replace the boards which is not an option.
If the system does not initialize, the arm will not activate. So in order to get the system to be useful I now need to come up with an alternative to using the original computer. The two options that i have been investigating so far are connecting a new PC to replace the old one, or creating a completely different control system for this. Most of the information I have found pertains to creating a completely new system.

Test

2008-06-13T18:19:00.000-07:00

test