Sunday, June 26, 2022

First outdoor tests of the screw mechanism!

 Gremlins eradicated - USB hubs are  bad, don't use them:



"Only" ~3.3 meters in this test - not maxing out the motor/mechanism yet, there's still some motor control tuning to do.  And probably another round of transmission ratio optimization once the actual robot mass and inertia get finalized.

7 comments:

  1. Amazing work, thanks for posting Ben!

    ReplyDelete
  2. Hah, yeah, hubs - especially unpowered ones - have caused many a headache. Though on the flip side, on occasion have saved my bacon by letting out the magic smoke before the computer they were connected to did...

    And wow, 9m? Is that a _scale_ 9m, where the actual robot would weigh much more and the peak height much less? If not, that's... a lot of potential energy coming down on, say, a foot/toe... :)

    ReplyDelete
    Replies
    1. This USB hub is particularly bad, it's one of those USB-C to many USB A + HDMI ones, everything acts kind of weird plugged into it. Like my logitech mouse only works with 2 of the 4 usb ports. Took me way too long to realize that was the problem given how many other issues I've had with this hub, but everything almost worked with it...

      Actually 9m seems achievable if I can hit ~750g total robot weight.

      Delete
    2. Oh man that's the most frustrating debug scenario: when things *kind* of work, but only sometimes, and never repeatedly, but only under certain bizarre conditions. Reminds me of the RPi2 Xenon Death Flash craziness...

      If the goal is to hit 9m all the better! If anyone can do it you can ;) Are you designing the control system around the assumption you've got 9m of jump space, or will the reaction wheels have the bandwidth to handle lower jumps as well?

      Delete
    3. Oh, one thought: add a 2x4 or something to distribute the foot load. I'll bet you're losing a non insignificant it of energy to elastic dirt reaction forces/compressibility.

      Delete
    4. Yeah, I certainly want to be able to stabilize standing and small hops. Reaction wheels are a little tricky to package though, I might have to choose one "good" axis with a larger diameter reaction and smaller reaction wheels on the other axes.

      Delete
  3. Wooooooohhhha. Really glad to see such a pleasant outcome after all the work you put in over the last years. It can only get better. This is going to be very awesome! Thank you for taking us along the way!

    ReplyDelete