Hi Chris,
I just started building my AR3 recently, and saw that the AR4 was just released. I noted in the manual that the primary update is the use of the teensy 4.1, which I accidentally ordered for my AR3 build. Since I ordered most of the electronics for the AR3, would it be better to order a teensy 3.5, or do you think it wouldn't be too much extra work or cost to outfit the robot with the electronics necessary for the 4.1?
Thank you!
Al
I think you can use the 4.1 with some minor changes. I will be working on a document and teensy sketch to update the AR3 to use the AR4 software this week. The primary difference is that the AR3 uses the CUI encoders so I need to create a 4.1 sketch that has the correct encoder values. The pinout on the teensy 3.5 is nearly the same as the 4.1 so I don't expect there to be much in the way of wiring changes. When I have the sketch and document ready I will post it to the downloads page and make an announcement on the site's blog page.
Is there an encoder wiring change due to the teensy 4.1 being non 5V tolerant? Or does the breakout board have that protection built in?
Hi @Chris Annin, finally had some time away from school to work on the "AR3.5" project, and am currently in a holdup. I'm using a mega 2560 with the teensy 4.1, and as of now, none of the motors jog, though they all are receiving power and remain "rigid" when the steppers turn on. I've been checking through my wiring, unable to find a problem yet.
I may remake my wiring harness for the pul+/dir+ if all else fails, but I'm curious if you think there might be an obvious problem I'm overlooking. I was able to put everything else together, and the encoders/limit switches test programs work great.
Please let me know if you have any insights!
Al
Just realized now, ENA+ and ENA- are not wired, since I was still following the AR3 manual. Will jumper those in and see if it works!
update - so I updated the wiring to include ENA+ and ENA-, and performing the calibration test works only on J4. I am able to jog J4 normally as well. Will continue digging to see if I can find a problem, but right now my thoughts are that the stepper drivers are the main problem.
update 2 - turns out I didn't switch from 24V on the stepper drives. All axes are working now! Though I noticed that J3 seems to be too weak - and skips steps on upward motion, causing the arm to droop and collapse. There is nothing other than the pneumatic actuator attached to the effector- so I'm curious if I have a defective NEMA 17?
i assume the drives amp settings are correct? also double check the A+A-B+B- wiring to the motor.
Figured out the problem. Thought the NEMA 17 motors were the same - found out I swapped the HG10 and HG50. Will swap and rewire, then update later today.
that one caught me aswell! I tested soo many different wirings. Maybe could be a warning in the doc?
Update 3 🙂
Using AR4 software for calibration and movement right now, will be using Python ROS ( https://github.com/ongdexter/ar3_core) and klampt ( http://motion.cs.illinois.edu/software/klampt/latest/pyklampt_docs/) for control in the next month. Hope to show you some ex vivo surgical studies in the next couple months!
