Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Shawn & Lara need a Toolbox... Droid?
#31
The Toolbox Emperor


Hanging out at ECCC 2023 in Seattle.

[Image: normal_Droids_at_ECCC_2023_-_BB-82C_D-O2..._L3-G0.jpg]
Kresty's Droids at ECCC 2023
Lego BB-8, D-O, H-O15 Sentry, MCK-Y, Toolbox, L3-G0
Reply
#32
So, at ECCC 2023, someone reminded me that I'd considered magnets for the hubs. Not sure why I thought it'd fail, but the 1st evening of the con, I printed some 4x4 blocks with magnet holes and ordered some magnets.

Somehow, I had *way* too much support on the blocks, which took forever to clean off, but once done, they fit perfectly. They feel a tad loose, but are just about perfect to hold the hub.

Now it comes off and reattaches easily. A second instead of fumbling with it trying to make it stay. Huge improvement! Toolbox could still use some work, but this makes a huge difference!

No clue if it matters, but I made sure the magnets at both ends of the axle were facing the same way. The axle is steel, so the magnetic field should extend to the other side, though it probably doesn't really have much impact.

Each spot has three 15mm magnets 2mm thick. (Those were a size easy to get in the morning).
[Image: normal_20230306_190219.jpg]
Toolbox Droid Hub Magnet
Replaced the hub shim stuff that didn't work with a plain old magnet.
Reply
#33
After the con.... We're toying with the idea of taking him to Celebration London, but we'd have to get it in the overhead.  I'm not sure how practical that is.  Specifically, we don't really have a good provision for getting the "box" off of the "base".

After the con, I'm looking at new batteries for all the droids, probably leaning towards Ryobi 18V batteries (we have a bunch anyway).  The Pololu SMC can handle 18V, so I just need to make sure the step-down for the 5V can.  We're using 30:1 Pololu 12V 37D motors.  Pololu mentioned on the forum that I can "just" run them at 67% to keep the average voltage at safe levels.  So, I guess I'll try setting the upper limit to 67%.

I also think we have the power budget to leave the LEDs on.  We already have a wire going up to the top, so I plan to add a DC stepdown in the tool area to handle the 5V for the LEDs.

I guess getting the door to open is on hold for a while.

The black blocks holding the aluminum rails are held to the middle by one bolt each. But we slid a piece of plywood between the aluminum bits, so there's no real way to pull those bolts out right now. That's why the top is "stuck" and can't be easily removed.

[Image: normal_Toolbox_Droid_Drivetrain.jpg]
Lego Toolbox Droid Top Platform Construction
One of the tracks is on, mostly assembled.
Reply
#34
Hmm, it's "worse" than that.  I really wanted the top off so he has a chance of counting as a "carry-on".  (or two.)

Got the bolts out but it was still stuck, so I went back to the plans.  I forgot that I'd made a notch for the aluminum rails in the plywood, so they can't lift up.

[Image: normal_Toolbox_Lower_Drive_Unit_28229.jpg]

Ended up cutting down from the top of the mounts through the plywood tab and could finally get it off.  Whew.
Reply
#35
Toolbox has always been pretty stately.  Kinda slow.  I forgot I set the motor controllers to ~30%....  

Since he's apart and I was thinking of using our Pololu 18V, I went ahead and started that.  To keep the 12V motors cool enough I had to adjust the PWM to not exceed 12V average.  So I had to reprogram the controller.... and it was at 1000/3200 Smile  

Now I've moved it to 2020.  (Batteries can be higher than 18V, so I did 3200*(12/19) instead of 12/18).  Now he's running at 64% instead of ~30%.  And the voltage has jumped by 50%.

He's much peppier now.  

Haven't stuck the top back on, but the bottom can turn on carpet.  On the wood floor it's hard to avoid spinning in circles.  I'm guessing the top will make him a little more controllable.  Otherwise, I'll have to adjust the mixing on the controller.  

Hoping to get him to Celebration.  Hopefully we won't have lego flying off the treads everywhere with his new speed!
Reply
#36
So, the 300% torque was too much on the D shaft on the motor sprocket.  It stripped them into circles and they slipped a lot.

So, I reprinted those with 2/3ds the teeth, more surface on the shaft, and a set screw. (I had a smaller sprocket designed already, but hadn't used it since the bigger one seemed to be working, but I had to add a spot for the set screw.)

So, now we have only 200% of the speed as before, but 450% of the torque.  It's a completely different beast than before.  Like maneuverable.  Too maneuverable even.

This thing is seriously top heavy.  The tracks can get it over a lot of stuff - as long as it isn't too tippy.  Oh, well. 





Also making a YouTube channel for all the brick stuff.  Having too many droids for each to have their own!
Reply
#37
That droid pretty much proves that tracks right next to each other like that turn reasonably well.
Always fun to see your stuff!
Reply
#38
That is working really well.
Bot Builder
My YouTube Channel: https://www.youtube.com/@PaulsBots
Reply
#39
(03-27-2023, 04:52 PM)savagecreature Wrote: That droid pretty much proves that tracks right next to each other like that turn reasonably well.
Always fun to see your stuff!

Thanks, I think that having the weight in the middle still helps. It turns worse on the carpet, where the tracks dig in more. 

Discovered ryobi charger doesn't handle 240v, so ordered a uk charger.
Reply
#40
(04-05-2023, 02:11 AM)kresty Wrote: Thanks, I think that having the weight in the middle still helps. It turns worse on the carpet, where the tracks dig in more. 

I have seen similar issues with LD-F1, whose tracks are also fairly close together.

Carpet was noticeably worse than tile at Dragon Con, to the point that (before his freshly printed traction pads wore in a little) it was hard to convince him to turn at all.

Sent from my SM-A326U1 using Tapatalk
Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)