So here is my problem: I'm trying to make some custom changes so I broke the wire connections, added some length to them, male and female pins, and then reconnected it all. I verified everything is plugged into the correct port, then tried running Dobot. It turns on, connects, and moves around "fine". The J1 axis (base rotation) seems to be accurate at tracking the position but is offset like 30 degrees for some reason. I can't tell if the photodisk is off. My real problem is with the J2 and J3 axes. If I try to home Dobot from where I interpret (45,45,0) degrees to be, it rotates around the base about 135 degrees, comes back and then the J2 and J3 axes retract and it rams the base and grinds the stepper motors. I don't know why. In the software itself, it usually comes to a point after that where it thinks J2 is at about 225 degrees and J3 is at about 0 degrees which doesn't make any sense. I mess with settings and move the robot back to (45,45,0) degrees and tell it such, it maintains this off-setting of degrees. It doesn't seem to know where it is in manual movement from me moving it. If I press the X +/- or Y +/- buttons, it'll move and J2 and J3 will immediately jump from 225 to, say 75, and from 0 to, say 50, respectively. I can't tell why it is, I don't know why it is that it won't listen to me (firmware, software, hardware, etc.) and have expended all options I can think of, so I would very much appreciate feedback. Thanks!
First of all, this is probably not the best place to discuss magician as this forum was created to discuss only the first version of dobot that was shipped as kickstarter reward.
That said, I have very little info on magician and probably won't help you much, but I'll try.
I haven't disassembled magician and have no idea what is supposed to sense the initial base positioning, hence can't comment on the 30 deg offset.
From the problems described I would suspect that it can't communicate with the accelerometers installed on the arms. Afaik they are sitting on the SPI bus, which may not be working well with your extended wires. Keep in mind that SPI isn't supposed to work on distance more than about 3 meters, but what's more important is the interference from motors - the harmonics extend far beyond the frequency of square signal that drives the motors, and it is quite strong compared to the signal in the control the lines (e.g. SPI).
I'd suggest disconnecting motors and see whether accelerometers start to produce correct angles.
If still not, then I'd look carefully at the accelerometer connections in the part that you extended.
If they do start to produce correct angles, then the problem is indeed in the interference from motors. You'd have to deal with that by replacing the SPI extension cable with something that isn't that prone to interference and/or applying some other measures like ferrite beads (see github.com/maxosprojects/open-dobot/wiki/4.-RAMPS)