Quote:
Originally Posted by marshall
Now this is cool! What kind of testing have you done on these to ensure robustness in FRC-type applications?
|
Thank you! I've worked with both custom shafted encoders pretty much identical to this one and the AS5047P chips fairly extensively at this point. I've found that the former is very robust as long as your prints have sufficient wall thickness (which they do) and that the threads are undersized. Although I haven't made this particular model yet, I think I've developed a sufficiently strong formula for design that should make this far stronger than your typical Us Digital encoder, from my experience. I would run an order before selling normally, but I wanted to get them to teams at CC in a couple weeks, and I don't have time for 2 runs in that time.
The shaft assembly feels stronger than that of the S4s and MA3s, although I haven't purposefully tried to physically break one yet. If anybody is interested, I could make a video showing me destroying one of those.
Because the AS5047 is a contactless magnetic encoder like the Talon SRX encoder, it's incredibly robust. You may remember this baby:
https://www.chiefdelphi.com/media/photos/43840
A few of them have been used in a mimicry robotic arm with great success. I'll see if I can post a pic here later, but essentially it's a sketchily made wooden arm that uses potentiometers as shafts, with the sensor magnet glued onto the tip. Despite being handled by kids who apparently have no idea how wiring works (I've been working with them on this for a year and they still don't know how to keep 5V and GND where they're supposed to be) and dropped and mangled numerous times, the encoders have lived on.
With this newest encoder, I'm hoping that this batch will face much heavier testing from teams.
The absolute positioning appears to flucuate around +/-5 counts (0.03%), although that may be just from the vibration of the hands.