As noted in another thread http://www.wheelchairdriver.com/board/viewtopic.php?f=2&t=8395, I am not setup to real-world test the analog version of the Roboteq script, and a couple people using analog connections have reported bugs in it. While I am attempting to squash them, doing this "long distance" is a frustrating task. I am also nervous about doing this because I consider the CAN implementation a much safer way to control a wheelchair, as well as having much simpler chair wiring. I don't want anyone to be hurt because of me.
The CANbus system does require a lot more electronics, including SMD soldering several circuit boards, and no one has yet said that they have tried to build one. I have therefore gotten a quote for commercial PCB assembly to see if I can encourage others to go CAN by reducing the amount of work needed.
The complete set of boards contains 4 CAN boards, 2 DC-DC converter boards that turn on when the CAN bus is live, 1 DC-DC converter board that handles ON/OFF/Sleep for the whole system (all quite small and moderate component density), and 2 different larger power distribution boards that handle Roboteq on/off, high-current relay, brakes, lights, current sensors, and two linear actuators. Even using an off-shore assembly house, PCB assembly is not inexpensive and I can not afford it unless there are others interested in covering the cost for several sets. Assembly of five sets would cost about $1,125 (U.S. dollars - $250 per set); somewhat less per set if more were ordered.
Do realize, that there would be other costs and plenty of work needed even with pre-assembled boards. But those costs and that work would be about the same as for an analog setup: joystick pot, contactor relay and current sensor breakout boards (optional) to mount on the power distribution boards, boxes, Arduino Nano-equivalent boards, display screen (if desired), internal wiring and connectors and so on. No long multi-conductor cables, however. If four or more people tell me that they are interested, I will go ahead and place an order and send out sets after I have wired and tested one system that I'd keep for myself.
Lastly, I have no experience laying out boards for production rather than hand assembly. If there are any volunteers, I would certainly appreciate having others look over my designs and suggest changes that might reduce cost or improve reliability or safety.