Just as the title says, what are some of the best ways to actuate intake without pneumatics.
Honestly, pneumatics is the best choice. Intake actuation falls into the sweet spot for pneumatic cylinders.
- It has just two positions.
- Fast motion between two hardstops.
- Cushioned compliance, so when the intake is smashed into a wall at full speed, the intake can collapse without breaking anything.
But, as a thought experiment, let’s consider a simple alternative. One concept which might work and retain some of the desirable properties of a pneumatically actuated intake would be to use a very stiff spring to hold the intake out against a mechanical hardstop. That retains the spring cushioned compliance. Something like a gas spring strut would probably work great for this. Then, when you want to retract, pull it back in with a fast rope winch. Maybe with a limit switch to kill the power when it’s retracted. Depending on how fast you want to deploy, you could just run the winch in reverse. Or if you wanted super fast deployment, use the winch to pull the intake back into a latch. Unspool the winch during a few spare seconds of time, then release the latch when you want the intake to deploy.
Seems simpler to just use pneumatics, though.
7 Likes
You could use gas shocks much like how 148 had done in 2020, they provide the same compliance as a pneumatic cylinder without having to incorporate an air system, the caveat with this is that you cant actuate it back in which shouldnt be too bad if you are driving good and build robust. You would also design this much like a pneumatic cylinder aswell.
Another option is a motor with a high reduction to actuate it up and down for this you get a highly variable and repeatable motion with the only downside being that it takes up a PDP slot. For examples of these I would look at the 2016 game as there were a lot of short small intake like arms to cover the various defenses that needed to be traversed.
1 Like
While I agree that pneumatics is perfect for a lot f intake designs, i also recognize that if this is your only use for pneumatics on the robot, it might be heavier and more troublesome than using a motor driven mechanism.
We have had good luck over the years actuating intakes with motors.
Most intake geometries will involve a long-ish arm type bar somewhere in the intake system that you want to actuate through a short angle of rotation. This would be true if you had a single arm intake (see 118, 148, 179, 1690) or a 4-bar linkage type mechanism (see 125, 1678, 3476, 3940). The movement of this arm can be accomplished with a motor instead of pneumatics (I believe most, if not all of the examples I cited above use pneumatics but the key takeaway is that the system operates by rotating an arm through a small angle).
Here are some tips:
- You will want to have a decent gear reduction from your motor to your gearbox output shaft to create enough torque to move that arm while at the same time slowing down the rotation of the motor enough to have a controllable (but not too slow) arm movement.
- Using either a large gear or a large sprocket attached to the arm allow you to transfer a large amount of torque to the arm with a large bolt circle attaching the gear or sprocket to the arm so that you don’t have excessive loads at any one bolt (you can see examples of a chain and sprocket driven arm actuation on some of the roto-climbers).
- Design the system with a hard stop so that you are not holding the intake at a high back-load position by stalling the motor unless you are very comfortable that you are not going to release the magic smoke.
- Make sure that you remove power from the motor when the intake is at the stop to also avoid another possible cause of magic smoke escape.
1 Like
We designed a choo-choo wheel on a window motor last year, but soon traded it for a rack and pinion.
The eccentric wheel got the mechanism down in half a turn, where the rack and pinion allowed three turns for full deployment.
2 Likes
is “choo-choo wheel” the technical term cuz i sure hope it is
3 Likes
god do i love FIRST
5 Likes
Last year my team used a snowblower motor for the intake arm and it worked great. We tried doing the same thing this year but found that the RPM was too high so we switched to a seat motor. It’s reliable and efficient and for our design, it saves space and keeps us within the frame perimeter.
1 Like
For 4680’s four-bar intake this year, we are using 2 neo550+ultraplanetary combos, then a further belt reduction from the gearbox to the driving links, to drive both sides of the mechanism and deploy and retract our intake.
By limiting the input current to the motors to 20 amps, we also felt comfortable exceeding the recommended UP reduction for neo550 motors after doing some calculations to make sure we wouldn’t exceed what the cartridge output gears could handle (40 N*m, per the info given on the load ratings page).
We’re still working on the controls a bit, but overall I’m pretty happy that we didn’t have to figure out where to put a whole pneumatic system just for this actuation.
1 Like
We’ve used the Bosch Seat Motor to actuate our intake for a few years. This year, we added a home-made friction clutch to it (sprocket with a 13.75mm bore squeezed between two shaft collars, using an acetal spacer as a “spring” with a bolt in the end that could be tightened or loosened). The clutch does a couple things:
- It makes the PID loop easier (using a REV Through-bore encoder on the intake shaft, not the built in encoder on the motor), as we can manage the friction so the intake would droop down if it was raised too far.
- It allows for some flexibility in the intake, so the ball can push the intake up a bit as it comes in over the frame
- It lets the intake push in without damage when we run into a wall, and “spring” right back out thanks to the PID loop.
- With a speed of ~24RPM, and a necessary rotation of less than 30 degrees to get to our setpoint, it means we can get there in half a second or less.
2 Likes
do you have any pictures?
Jon - do you have a print or a sketch of this setup? We’re going to a double-chain drive because we broke a chain on our intake in the last competition, but that just moves the failure point up stream to the gearbox. A friction clutch design would be amazing.
Our 4 bar intake uses constant force spring to extend it out of the robot and is then pulled back in by a winch. This allows the intake to freely push back into the robot when we run into something. Another set of constant force springs and pulleys keep the cable tensioned so it doesn’t get caught on anything when the cable slacks.
1 Like
If you are looking for pictures 2767 has some good 4 bar mechanisms (of different designs for different motion profiles) that are motors only. See 2019 intake & 2022 climber. I can elaborate more if needed (lessons learned, real-world performance, controls systems, maintenance, etc.)
Unfortunately, we never got it into CAD - we sort of made it up as we went along. Picture below probably doesn’t help much, but it’s the only one I have showing it (and I won’t be back in the shop until Monday - spring break).
The stack-up looks like this:
- Bosch seat motor on the inside of our 1/2" polycarb side
-
AndyMark hex shaft adapter , with the retaining screw in the back
– Added groove for an e-clip, located just on the inside of the polycarb (so the retaining screw/washer isn’t taking the force of the friction)
– turned down to 13.75 mm (aka thunderhex) outside diameter - two bearings, one pressed in from each side of the polycarb (we may have an appropriately sized spacer in here for added structure, I can’t remember)
- thin spacer to ensure we’re rubbing on the inner race only
- metal shaft collar (not tightened)
- sprocket, with the inside bored out to 13.75-ish mm (aiming for a slip fit without slop… more of a feel thing with the shaft than an actual measurement)
- another metal shaft collar (not tightened)
- Acetal spacer to get us a little past the end of the shaft
- Bolt and washer, with loctite
Typically this sort of thing would have a spring in it, but we didn’t have any on hand and found that by adjusting the tightness of the bolt we could make an adjustment to how strong the clutch was. We tightened it up to the point where it would slip down to where we wanted it and called it good. So the spacer provides the “spring”, and the two metal shaft collars provide the friction (aluminum on aluminum). The force of everything is going against the two bearings, so we don’t worry about stripping out or damaging the motor.
1 Like
Here’s a a snip of a picture (taken by the wonderful Dan Ernst) of our robot as it stood at the Detroit District.
The belt being squeezed together by the yellow zip tie and spacers is mirrored on the opposite side of the robot, and are what is driving the lower links of our four-bar.
Since Detroit, we ditched the zip ties for larger pulleys which seems to have solved our belt slipping issue that the zip tie tensioner was there to mitigate. We also increased to a ~60:1 reduction in the ultraplanetary stack, with an additional 1.25:1 reduction using larger pulleys after the gearbox.
As pictured, I believe our reduction was ~30:1 in the UP stack and 1:1 pulley to pulley. It worked to get the intake in and out, but definitely had some hesitation while retracting. I can snag a better picture of the updated mechanism at our meeting tomorrow if wanted.
1 Like
We’ve never used pneumatics. Never found the added complexity worth it.
For actuating an intake with a motor -
- design it to be able to be folded into the robot if it’s smashed against a wall
- careful using worm gear motors (seat, window, etc) because you can break the worm gear if the intake moves during an impact
- you don’t really need PID if it’s just 2 positions. Use a timed action to Move at 1 speed for x seconds and then at 0.1 speed for the rest of the time between 2 hard stops. The 0.1 holds it in place either in or out of the robot. Use a motor that can take being stalled at low power for the whole match. We usually use bag motors on a 100:1 VP for this, but other motors will work, too. (see the locked rotor testing on vexpro’s website. Don’t use a 775 pro.)
- you can use belts, chain, or linkages to transfer rotation. linkages are easiest to deal with if you have access to fairly accurate machining capabilities
2 Likes
One of our lead mentors has been calling Aren Hill “Choo-choo guy” or something similar for a while, as he popularized it with several robots and teams over the last decade or so.
5 Likes
That’s a really clever concept. Mechanisms don’t always have to be rigid or strongly coupled. Sometimes, being loose and/or floppy can be advantage.
1 Like