To see this working, head to your live site.

Did I fry my encoders?

Hi all,

First time long time. Wanted to say first of all what a great project this is, big kudos to Chris and the community here keeping it going.

I'm putting together my first AR-4 using the aluminum kit. Some small challenges along the way but great learning. When I first booted it up, I was able to get motion and limit switches on all except two joints - no problem, I find it's some of my wiring. Among others the exposed J5 limit switch (waiting to tape it after testing) touched the aluminum causing challenges. Strangely at one point jogging J3 caused my whole PC to restart - possibly that short came through the USB and my computer responded?

Anyway, after re-working the limit switch and motor wires, I was able to get all 6 motors jogging and limit switches registering GRD/3.3v. I got the servo gripper together and popped that on as well.

Excited to get into calibration, I started up the software and notice the encoder values don't update anymore! This is bizarre because I hadn't touched any of their wiring. Simply when I run the Test Encoders program, all values stay at 1000 whereas before I could move them up and down by manually moving the joints as expected.

I unwired ALL the encoders and connected J1 encoder directly into the BLK/RED into 5V/GRD and BRN/BLU directly into pin 14/15. Still no encoder reading when I run Test Encoders for J1 as I move the joint left and right.

Unfortunately I don't own an Oscilloscope. With my multimeter I don't read any voltage between BRN or BLU vs GRD, which concerns me. I get ~3.6v between GRD and Encoder GRN or ORG, but I can't get the voltage to change moving the joint manually, so not sure what that reading means. Perhaps my tool can't pick up the microchanges in voltage, or perhaps something is more wrong.

I haven't yet tried to unwire and re-wire the other encoders that run through the Cat6 extensions, though that's my next step in mind. I just figured that going directly to J1 would eliminate some of the potential wiring challenges.

I guess I have a few questions for Chris and the community:

Is it possible that I fried all 6 of my Encoders at once through some boneheaded mistake or short or ground? Without an oscilloscope is there a reliable way to test that they have not failed? (I'm considering buying a scope if needed)
If my Encoders aren't all broken, should I consider an alternate wiring? Or are there encoder changes to consider? I assume the Teensy reading of A+ and B+ makes the most sense, and if those don't read A-/B- and Z aren't going to be a lot of help.
If my Encoders have all failed, any thoughts on replacement Encoders or after-market Encoders (once I lick my wounds about the damages)? Stepperonline doesn't have a lot of documentation on the onboard Encoders or troubleshooting, so curious folks' thoughts on encoders generally for Steppers.
I guess as a corollary, have folks run the AR4 primarily on Open Loop? Any disadvantages other than the obvious lack of clear position detection?

Sorry for the long post, I tried searching the forums and online but have been running into dead ends. Any help is appreciated!

2 comments

Comments (2)

Chris Annin

Aug 01

Hi there - thanks for the detailed post and for the kind words! It’s great to hear you're diving into the AR4 build and already working through some of the real-world learning that comes with it.

The fact that all 6 encoders suddenly stopped updating definitely points to something upstream. People typically have issues with the 3.3v and 5v supply at the 3 terminal block, I would double check that 3.3v is being sent to the limit switches and the encoders red and black wires are powered from the 5v side of the terminal block and that all connections at the terminal block are good. You can run the robot in open loop mode, thats not an issue at all.

sunnykanneganti

Replying to

Thank you Chris, I appreciate the reply!

I went ahead and re-wired the entire Teensy thinking the breakout block might be the issue. It took a while but also gave me an opportunity to de-tangle. I also, for now, removed the keystone jack with the extension since I don't have a J7/8 planed near term.

Thinking the best way to eliminate issues, I started by wiring the drivers and the pul/dir only to pins 0-11. Booting up the software, I could get the joints to jog - great!

Then I hooked up the green limit switch wires to 26-31, orange to 3v3 and orange/stripe to ground. I get good 3.28V readings on the multimeter at the pin when I manually toggle, and I can see the software recognizing the limit switches using "Test Limit Switches" program - great!

Finally I try the encoders. Brown to 5V, Brown/stripe to ground, and the blue and blue/stripe to 14-25. With bated breath I load "Test Encoders" and... notta! The values are set to 1000 and don't move despite me moving the joint by hand with the driver power off.

When I try to jog in the software with test encoders on, the program freezes, I assume because it is expecting an encoder change and is not getting it? If I disconnect / reconnect the Teensy the software unfreezes.

Checking the terminal blocks I'm seeing 4.98v and 3.28v on the side with the connections tinned in. Voltage at the encoder pins (14-23) are all 0v.

Trying one more thing - I hook up J1 encoder (since it's unspliced) to the Nano I was using for the Servo gripper (Red -5V, BLK -GRD, BLU-D2, BRN-D3) and tried to read based on the instructions from stepperonline (https://help.stepperonline.com/en/article/how-to-use-arduino-to-read-the-value-of-the-encoder-1d6c40x/) but wasn't getting any reading there either.

It's possible because I'm new to this I'm still making some boneheaded mistake somewhere or forgetting something, but these steps more and more are making me think something has failed within my encoders.

I've reached out to stepperonline support as well to see if they have any thoughts / ideas. For now I'll run in open loop but will need to figure out a way to get closed loop again, hopefully without re-buying all the steppers and remaking the whole bot haha. I also don't understand what I could have done to cause this situation, when I did have the encoder numbers working earlier...

ANNIN ROBOTICS

Did I fry my encoders?