Not too many people will argue that Robot Arms aren’t cool. [Dan] thinks they are cool and purchased a LabVolt Armdroid robotic arm on eBay for a mere $150. Unfortunately, he did not get the power supply or the control unit. To most, this would a serious hurdle to overcome, but not for [Dan]. He opened up the robot and started probing around the circuit board to figure out what was going on.
Since there was a DB9 connector on the outside of the robot arm, he assumed it was a standard RS-232 controlled device. Good thing he checked the internal circuitry because this was not the case at all. There was no mircocontroller or microprocessor found inside. [Dan] painstakingly reversed engineered the circuit board and documented his results. He found that there were SN76537A chips that drove the 6 unipolar stepper motors and SN75HC259 latches to address each individual motor.
Now knowing how the robot works, [Dan] had to figure out how to control the robot from his computer. He started by making a custom Parallel Port to DB9 cable to connect the computer to the arm. After a series of several programs, starting with simply moving just one arm joint, the latest iteration allows manual control of all joints using the computer keyboard. A big ‘Thanks’ goes out to [Dan] for all his work and documentation.





[Max] designed all of the mechanical parts himself. After weighing the advantages and disadvantages of different materials, he decided that the frame would be made from 5mm acrylic sheet. The main body of the robot has acrylic ribs that are spaced apart by threaded rods. Twelve RC servos make up all of the joints, 3 in each leg. Notice in this photo how there is one servo that immediately rotates another servo. To support the other side of the rotating servo, [Max] epoxied on a T-nut, stuck in a short length of threaded rod which is then supported in the frame by a ball bearing. Simple and effective! The upper portions of the legs are also made from acrylic sheet and the lower legs are from a cheap camera tripod. Rubber feet ensure a slip resistant stance.
Like any project, there were some hiccups along the way. First, several revisions of the gripper were necessary to get the correct tooth profile that resulted in smooth and tight movement. Also, while making the shield the spacing between banks of headers came out one header too close! On this first board [jjshortcut] just bent the pins so they would fit into the Arduino. You can’t let some minor snafu prevent forward momentum of a project!