Hi Sam, I ran into the same issue(s) when I attempted to calibrate the robot for the first time. In short, my solution was to reverse the black and red wire on each limit switch - the manual is incorrect or I introduced an extra wiring error. Here's what I did: -> After building the robot, manually positioned the robot to be roughly in the "rest" position, programmed the Teensy 3.5 and established communication with the Teensy in the AR3 control software, and with the robot control box fully powered on, I navigate to the calibration tab. -> In the calibration tab, I selected the "Calibrate J1 Only" button. At this point, the machine rotates clockwise, stops, rotates counterclockwise slightly, then clockwise again a little before completely stopping with an error about either the limit was tripped or the joint is out of calibration. From my understanding, the machine should rotate towards the limit switch making contact, reverse direction to release the switch, then re-establish contact with the limit switch before stopping with calibration achieved for that joint / motor / axis. The issue repeats on all joints / motors / axis. With a multimeter, I read approximately -5V when pressed (ground reference is the ground pin to the GX16-2 connector) and 0V when not pressed. The solution for myself was to reverse the black and red wires for all limit switches so that I read approximately 0V (pressed) and -5V (not pressed). From here, I can calibrate each axis individually as before, with success.

Mfr. No: CUI-3132-1FT That’s what I ordered. The idea is to use the assembled molex connector ends and solder additional wires to gain the required length for the specific joint that the encoder is assembled too. As you noted, it’s more expensive to buy the individual components, then assemble it all. The prefabricated cable reduces cost and assembly. robotshop.com is another supplier for these cables, if available in your location.

Hi Pierre, "...and the Teensy should be powered only via the USB connection to your PC." From my understanding, you have two options to power the Teensy. 1) USB connection to PC. 2) Wired either to the GX16-2 connector (provided the connector is wired to the DIN rail for 5V), or wired directly to the DIN rail for 5V. Also, I believe you can do both simultaneously and the Teensy will automatically decide which one but don't quote me on that. To me, this makes sense since you might have it powered independently from the USB cable but at some point need to reconnect the USB cable to upload a new program. Spec's for the 2 boards: Teensy Vin: 3.6-6 volts (link) Mega 2560 Vin: Limits = 6-20 volts Recommended = 7-12 volts (link) And this is where I have an issue using the 5V power supply for the Arduino. The Arduino operate at 5V but has a low limit of 6V with 7V being recommended. This is why I'm looking at adding a second power supply for the Arduino. Other items related to wiring: I noticed that J4 driver (DM320T) Opto pin is powered via the 3.3V output pin from the Teensy. The datasheet (link) says the typical voltage is +5V but fails to say the minimum or maximum. It's a guess that 3.3V is sufficient. This same voltage is connected to the PUL/DIR signals for the other drivers (DM542T), however, the datasheet for this driver (link) says 4-5V. Maybe I'm being picky, but a weak signal could cause potential issues. On this issue, I'm tempted to use the 5V power supply for these connections instead of connecting it to the Teensy. The other change I'll be implementing in my build, is running separate power supply lines for each individual driver, rather than daisy-chaining them as noted in the datasheets. I don't mean to be nit-picky. I know builds of this nature can have bugs to be worked out and perhaps this information will be beneficial to someone in the future when they run into confusion or issues.

Hi Pierre, I noticed the same wiring error as you have while I've been reviewing the manual for the AR3, as I prepare for my own build. I also searched these forums for a possible answer but I did not find one. The following posts remained unanswered: https://www.anninrobotics.com/forum/robot-builds/ar3-building-wiring-completed https://www.anninrobotics.com/forum/questions/5v-psu Here's my understanding: The pictures from page #238 to #241 shows two pairs each of black and red wires coming from the 4th pair (black) and 5th pair (red) of the DIN rail terminals. This is where the manual is missing the instructions to add this second pair of wires, including the instructions to solder said wires to the GX16-2 plug on the enclosure door even though the pictures show the installed wires. If you then jump to page #265, this is when you solder another pair of 22awg wires (black and red) to the GX16-2 plug on the enclosure door. The other end of these wires need to be soldered, with heat shrink, to a pair of jumper wires, which are then connected to the Teensy 3.5 appropriate pins as instructed on page #265 (bottom image). This is how the Teensy 3.5 derives it's power. From 5V power supply unit, to the DIN rail, to the GX16-2 plug, to the Teensy 3.5. Please note, that it is my opinion that the red jumper wire mentioned on page #258, is the same red jumper wire that will be soldered and connected to the Teensy 3.5 Vin pin on page #265. We're missing the follow through on that instruction, including the making of the black wire. As you can tell, there are missing steps in the manual. Perhaps the community will help with the next revision? So to answer you're question, from what I can tell, yes to either of your methods asked. The power for the Teensy 3.5 is ultimately derived from the 5V connection point on the DIN rail, which as you know, is sourced from the 5V power supply unit. What's the best method regarding common grounding techniques is another question I choose to side-step for now ;) Hope that helps Pierre so you can continue with your build. Would you mind sharing whether you're building the aluminum or 3D printed version? Separate but related to your post, I'm confused by where the missing +V side of power for the Arduino Mega 2560 is coming from. I can only assume the power will then be derived from a direct connection via the USB cable to a PC and the -V side for establishing a common ground. According to the datasheet, "The board can operate on an external supply of 6 to 20 volts. If supplied with less than 7V, however, the 5V pin may supply less than five volts and the board may be unstable." As instructed in the manual, we're not using 5V from the Arduino for the relay module, however, it's the "board may be unstable" part that leaves me uncomfortable. Are they referring again the stability of the 5V output from the Arduino, or the boards ability to function properly? I'm thinking the latter but I would like to hear other's opinion on the matter. Regards, D.D.