Pearadox 2021 "reveal" and motor in roller details

A student put together a recap/reveal video combining some of our footage from 2020 build season, and some of our more recent practicing in preparation for the Texas Cup - you can find it here:

The skills challenges pushed us to regain some of our auton abilities in 2019 that we didn’t implement in 2020, so we now have 5 autons ready for Texas Cup. We’re looking forward to putting them to use!

One thing we wanted to do for this event was investigate putting the neo550s in the rollers. The prior placement caused issues (balls would get sucked up by the bell, power transfer along a flexible lexan plate is suspect). There were numerous ways we could have solved this (supporting the plate along the power transfer path, moving the motors further into the robot and extending the power transfer path, reducing to just one motor and only doing floor pick up, etc), but we wanted to use this as an opportunity to do something new.

CAD has been kind of rushed and not totally representative of real life, but here’s a section view of the set up:

With a couple of links to real life implementation:
Google Photos
Google Photos

CAD here: Onshape

Some lessons learned

  • Nylon isn’t affected by loctite. I wish I had corrected this bit of false narrative I had in my head a lot sooner
  • putting the motors in the roller leads to a lot of vibration. Since the rollers are polycarb, we bought some thread locking bolts like these to go into heat set inserts on our parts McMaster-Carr
  • Even though we designed in some clearance, there’s enough slop in the bearings and such that the polycarb still rubs occasionally on the motor housing. Also, the motors get real hot when we practice and run them non-stop. So prior PLA motor housings melted, had serious damage. We switch to Onyx and haven’t had any problems since. Our plugs are all either PLA or PETG (depending on what I had available).
  • if I were to print a couple more, I would put access holes to the bolts for the UP gearbox in the motor housings. That was missed in this version.
  • The thrifty inserts were great at preventing the plugs from rounding out.
  • Our plugs are thicker to help keep them aligned and rub less on the motor housing. First revision was only a half inch or so, and the rollers had a lot of warble to them.
  • Swapping out spare rollers is a breeze. Repairing a roller that had bolts back out suck. So we have several spares ready to go.

Looking forward to watching the Texas Cup! Good luck teams!


Cool designs! What is the reduction in the UP?

We ran a ~1:3 belt reduction for our intake of similar diameter but I’m really curious what the results would be like just bonding the PC tube directly to the bell…

@beter spent the most time looking at them. I believe it was also a 3:1, I seem to recall changing to a 5:1 at one point when we were concerned about some balls getting jammed but it felt too slow. (EDIT: I’m now second guessing myself - I know it was one stage, but maybe it’s actually 5:1 and we added a 2nd stage and it felt to slow)

I think going directly to the bell might be too little torque. Now using the bell as a sun gear and the roller as a ring gear…

1 Like

we used 5:1’s inside i haven’t run the math but if we ran 3:1’s on the old intake the surface speed would probably be similar enough to not see a noticeable impact on intaking because we had to make the tubes larger to account for the Up and the mounting.

1 Like

Always love seeing motor in roller intakes. Glad you got it working :slight_smile:

1 Like

Let me know how they look post competition :slight_smile: You guys might be one of the first teams running them in some actual FRC matches.

Cool design, thanks for sharing!


our PLA-only plugs definitely were rounding out so they were seeing some good forces (probably changing directions when going from floor loading to loading bay, or if a ball got a little jammed). I’ll post some pictures post-comp!

looks a lot better than the crude one we made in 2014 :slight_smile:



They can definitely be elegant pakaging solutions.


In looking for some inspiration to steal I was talking to some mentor friends, and @Akash_Rastogi came across this post from 2004. Anyone got any earlier examples?

Seems like with the size and power of the 550 and ease in 3d printing, it’s more approachable than ever to shove a motor into a roller. Something we’ll keep in our back pocket for sure going forward.

Looks like the link didn’t embed, but yeah this appears to be the first mention of a motor-in-roller setup that I was able to find.

Pretty neat it’s from 177!


The double wide pickup roller in 2017 was totally a mistake, but doesn’t stop me from being excited by motors inside rollers.

edit: Also, our 2021 robot pickup did use 3dp gearbox parts for coaxial driving each roller, but it didn’t measure up to this better use (both for using the 550s) and also the geared part ended up larger than the roller, but within the pickup area, which was jam prone. Something like yours would have been nice, but we didn’t converge there starting with in stock NEOs vs NEO550s. Part of it all is having large rollers (like 2017) to fit the motors, which might constrain some designs, but otherwise it is a great idea.

1 Like

whoops! Can I

:rotating_light: Boomer Alert :rotating_light:

myself? Just forgot to paste the link entirely.


Would definitely add a current limit to this, I’m guessing your theory is correct, our low ratio intake only sees about 10-15 amps when intaking but will easily spike to 60+ without a limit during initial spin up.


Oh yea, we put a current limit in all of our 550s. Too easy to burn up at the default limit of 80

1 Like

We also tried motor in roller intake in 2014 (Pic HERE). It worked great and I’d love to revisit the idea with new motors like you have!


This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.