Quick Shot SVI-2000 Robot Arm
This robot arm was manufacture by Spectravideo International between 1985 and 1987; Spectravideo going out of business in 1988. It is a four axis robot arm controlled via two, very simple contact switch, Atari style joysticks. It claims to be a five axis with the fifth being the gripper mechanism. There was also an adapter and software for sale that would allow you to control the arm with a Commodore 64 via a serial port adapter.
When I received this unit it had no functionality and was missing the battery cover. I 3-D printing a substitute battery cover and proceeded to dismantle the arm to start trouble shooting. Upon removing the second axis I found that the plastic spindle, that the upper sections of the arm rotates on, was cracked (red arrow Fig. 2). The blue arrow points to where the spring engages to the spindle to help support the weight of the upper section of the arm so the motor doesn't have to overcome the entire weight of the arm. As can be seen in Fig 3 the gear for the second axis had also broken. This was at the end of gear and appeared that at some point the arm had been forcefully extended past its designed range of motion cracking the gear and the spindle. I was concerned that it was ruined at this point but it turned out that super glue bonded just fine to both types of plastic and by being very careful I was able to glue them back together.
Fig. 4 shows the top view of the arm's control board. It is a relatively simple set of transistors and resistors to provide +3.3 and -3.3 for each of the 5 motors depending on the movement of the joysticks. With a couple more to turn the light on when opening and closing the jaw.
- I forgot to document the part numbers on the transistors before reassembling this as I hadn't thought of making this website at the time.
This is my first attempt at creating a schematic using gEDA Schematic Editor for Linux. It is a decent schematic program but there are some problematic limitations. For the purpose of documenting the connections in the above circuit board it was sufficient.
The layout is based as close as was practical to the layout of components on the board.
Fig 6 below shows the center section of the arm once it was disassembled. It appears that during assembly this was one of the last sections to be assembled as the upper arm would have its wiring threaded through its axis of rotation then would be soldered to the board shown here. In order for me to disassemble the arm enough to access the broken sections of the second axis it required desoldering most of the wires from this board. Once the glue had set on the parts and I had reassembled the arm back to this point I had to solder the wires back to this board. Upon finishing soldering them back the arm started to work again. Visually inspecting these solder joints, they hadn't appeared to have failed but it would appear that they had.