This is probably the simplest yet most effective I’ve seen so far. Really nice work y’all
16 Likes
Great work. I think you made an elegant design choice lining up your extension with the angle of the grid, and it paid off in a capable but simple robot.
Thank you!
A little more complexity might be needed to make a gripper that secures cones but doesn’t pop cubes.
I agree. I handled the design of the gripper and made one simple design flaw (and very solvable) that didn’t allow the gripper to close any shorter than 2.5" while we still had 1" left of stroke length. Allowing the gripper arm to close 1 more inch would have gotten a much more positive hold on the cone. You wouldn’t have to worry about popping the cube because the force is the same no matter how far you close.
Did you evaluate whether this can pick up off the substations?
We only had really 5 consistent people working on this bot, so our bandwidth was stretched thin to say the least. We knew we could probably only make the scoring nodes, so we made a decision to do entirely ground pickup. However with that said, later this week we could get a station built and test it. From the CAD drawings, this exchange should work.
It looks like you added a plastic shield to the claw to limit how far in the game piece can go. Was that to protect the cube from a pinch point?
Yes and no. Main purpose was to make sure the cone did not go back too far into the grabber preventing us from scoring the cone on the high node. It doubled as another purpose to prevent popping.
I’m curious about the decision to do an asymmetric claw. Obviously mechanically simple but seems to be a little less intuitive to control.
This was entirely decided upon for mechanical simplicity and materials on hand. With more time and resources a design like you are saying would be very advantageous.
I suspect moving the arm pivot forward will create geometry problems achieving the stowed and scoring positions - maybe moving the motor mechanism below the arm could save a headache.
Once again due to our time constraint we couldn’t get to solving this problem, however the pivot point could be shifted forward about 1.5" and everything would still fit nicely. Close, but nicely. Moving the motor mechanism below the arm was considered, however this can not be done, due to the picking up position resting the arm on the bumpers.
This robot also probably needs some good covers to keep game pieces from landing under the arm.
Under the grabber or inside the control system?
Between the elevator and control system.
Speaking for the whole team, Nothing but positive things to say about the thrifty elevator system. Really easy to use, assembly went super smooth, bearings don’t bind at any point, there is no slop between the Base stage and the support stage, and most of all, very thrifty.
5 Likes
Agreed, a game piece in the control system would put us out of the match.
So how do you feel about the “elevator on an arm” system after this build?
Do you think a narrower elevator is viable?
How about a cascade stage?
1 Like
Are you referring to a telescoping elevator on a pivot. If so I am a big fan. After seeing the capabilities of this system I think there are a lot of possibilities for something like this. Redux Ri3D pulled this off pretty well on one of their robots.
Under this current design, no. We tried to make the elevator as wide a possible to give us the biggest possible grip range, which was still smaller than I would have liked it to be.
Perhaps, however it would not fit at all on this design due to the amount of items inside the base stage to make the grabber work.
1 Like
Awesome build! The claw design is the most intriguing bit by far to me. It’s so simple but it works really well!! What keeps the piston from popping the cubes? I watched the video with the short explanation but i’m not following how the clamping side of the claw is able to close to the cube compression point and the cone compression point without damaging the cone. Very ingenious design, well done!
2 Likes
This robot is very simple, yet achieves a high level of play. Props to the team behind it.
That said, I have a few concerns and suggestions:
- No inner carriage means this design can and should add cross bracing to stiffen up moving stage
- Take advantage of the relatively large overlap and add extra bearing blocks on inner and outer frame
- Better intake, possibly even a chassis intake that hands off to the clamping intake, and have the clamping intake rotate like Ri3D redux so the cone spins downward due to gravity
1 Like
Looks awesome! How did you guys feel about only having dyneema to pull up the elevator pivot and not having anything to hold it down? It looked in your reveal like it could bounce up a little, and I’m wondering how well it seemed like it held its position with gravity. Would you have added some kind of pull-down or closed loop system if you had time?
1 Like
Depending on what your robot is doing this actually might be a good thing. Ramming into a goal or field object would just cause it to pop up rather than damaging the whole thing. Kind of a nice built in feature to me. I also bet there are some beefy torsion springs out there that would help balance or offset the arm motion at the pivot point so the rope does even less work.
I think with some refinement on the gripper end and a lot of driver practice this would be a top 8 bot at many events.
6 Likes
I really like this design as well. I am just afraid of what my drivers could do to it trying to pick up game pieces on the other side of the filed.
Have you tried the full speed joust into a wall test with a horizontal thrifty elevator yet?
1 Like
I don’t really have a wall I’d feel comfortable yeeting a robot into here at the house. But if @RSimpson and gang have a wall they can do that to and feel comfortable with it I’ll replace that elevator if anything gets damaged. Worst case if you’re extended it’s just going to push the elevator down the track back inside the perimeter. Not sure how that pivot joint would fair though.
9 Likes
Once we have the driver station wall up, I’ll give them an opportunity to try.
The pivot is made from 1/2" grade 5 bolts. I think the pivot will be fine, but the 10 screws on the mounting blocks might not like the impact force.
4 Likes
It does indeed have some bounce, but not enough to cause any issues, at least in our testing. In addition, it doesn’t feel slow or anything when going from stowed to picking up. As someone who’s driven a ton of robots throughout the years, the speed going down felt totally reasonable, as far as FRC things go.
Like @Ryan_Dognaux mentioned, the fact it has give is actually a kinda nice feature to have, when potentially running into things. We did consider adding surgical tubing across the top, to give it a little positive pressure when it was in the “stowed” position, but you certainly don’t need anything to get the arm to go down nicely or anything, the weight alone does that fine.
As far as a closed loop system, we did consider it, but we decided against it, until we could show that it was 100% necessary. Knowing what we do now, I think we made the right decision; it’s really good as-is, and making a closed loop system is much harder to implement and has some other drawbacks as well.
3 Likes
CAD IS HERE!!
Few notes
CAD does not represent actual robot 100%. I have implemented changes to the CAD to improve the effectiveness of the robot. These changes include:
- Reduced gripper depth from 9” to 5”. We found the 9” depth was not needed and moving the depth to 5” gets another 4 inches of extension which should allow the robot to score cones in any orientation.
- Moved pivot point forward 2.5”. This makes the robot legal, by moving the gearbox and motors into starting configuration during the stowed state. This was done by using a new 3-D printed part. This part is untested, but I believe a markforged part should hold up to normal operating forces, but may not hold during hard impacts. There are many ways to reinforce this pivot point.
- Changed the mounting of the pulley system for the grabber assembly. This allowed me to change the brackets which allows for 1” more of grip range. This will allow for greater hold on the cone.
- Changed the gripper piston from a 1 1/16” Bore 9” stroke to 3/4” bore 10” piston. The extra force from the larger bore is not needed for grabbing the cones and cubes. This is because the most important thing for grabbing the cones is actually how far you can close and not how hard you can grab. So moving to a 3/4” bore will speed up the piston while still having enough force to grab the cones.
Also not represented in the CAD:
- Chain for the elevator
- Rope for the pivot system
- Surgical tubing return for the gripper
If you have any questions feel free to reply in this thread or message me directly.
3 Likes
Great concept! How high was the elevator’s pivot from the top of the drivebase?
From your experience, do you think experimenting with different gripping parameters (more cone-shaped grippers, larger gripper surface area, different rubbers, etc.) would result in a significantly better grasp on the cone?
Your grip on the cones was definitely good enough to score, but we’re trying to determine if it could be tuned in enough to do full field sprints at 16 ft/s and take some impacts from defense.
1 Like
3"
1 Like
The biggest thing is going to be closing far in on the cone. If you can get your gripper to close to about 1.5"-2". Then using the barnacles that come in the KOP would be more than sufficient for what you’re looking for.
2 Likes