[FTC]: Tri-Wheel design [Archive]

View Full Version : [FTC]: Tri-Wheel design

ArtemusMaximus

21-12-2015, 12:12

We are thinking to go tri-wheel route. Did anybody made a successful tri-wheel design to climb the mountain?

RRLedford

21-12-2015, 17:16

We are thinking to go tri-wheel route. Did anybody made a successful tri-wheel design to climb the mountain?

We considered doing it, but in the end we were dubious as to whether we could handle the precision and time needed for CADing and machining the the 3-lobed plates to match needed locations & size acccuraies for axle bearings and planetary gear shafts etc well enough for decent function and durability.

http://www.unusuallocomotion.com/medias/images/Epi_q-1-surveillance-robot.jpg

Instead we settled for a four pair dual overlapping "cam-wheels" design pictured below. It performs as well at climbing as any of the ~40 bots we have seen here in Illinois so far at league and qualifier competitions. We will be scoring all 160 mountain points most of the time in a couple more weeks, and we may be hitting the 2nd mountain zone in autonomous by then too. Our goal is for our bot alone to contribute 200 points to our aliance effort once we tune our autonomous efforts.

http://i489.photobucket.com/albums/rr259/RRLedford/FTC2015-16/20151203_202914_zpsfdbqd2jb.jpg

Practice run:
https://www.youtube.com/watch?v=tsOTYCp9CJM

cadandcookies

21-12-2015, 17:21

We are thinking to go tri-wheel route. Did anybody made a successful tri-wheel design to climb the mountain?

I haven't seen it in action, but I know that 5943 had a triwheel design. I'd pm Evan Hochstein about it if he doesn't pop up in this thread on his own.

ArtemusMaximus

21-12-2015, 18:07

We considered doing it, but in the end we were dubious as to whether we could handle the precision and time needed for CADing and machining the the 3-lobed plates to match needed locations & size acccuraies for axle bearings and planetary gear shafts etc well enough for decent function and durability.

http://www.unusuallocomotion.com/medias/images/Epi_q-1-surveillance-robot.jpg

Instead we settled for a four pair dual overlapping "cam-wheels" design pictured below. It performs as well at climbing as any of the ~40 bots we have seen here in Illinois so far at league and qualifier competitions. We will be scoring all 160 mountain points most of the time in a couple more weeks, and we may be hitting the 2nd mountain zone in autonomous by then too. Our goal is for our bot alone to contribute 200 points to our aliance effort once we tune our autonomous efforts.

http://i489.photobucket.com/albums/rr259/RRLedford/FTC2015-16/20151203_202914_zpsfdbqd2jb.jpg

Practice run:
https://www.youtube.com/watch?v=tsOTYCp9CJM

Yeah, I've seen your split wheel design, looks impressive. However, I did see that your wheels stick on upper-mid bars. Did you sort out that problem?

We are rookie team and don't have much to start with. Definitely no CNC for us :(. It does look like even veteran teams are struggling with this season game.

RRLedford

21-12-2015, 23:25

Yeah, I've seen your split wheel design, looks impressive. However, I did see that your wheels stick on upper-mid bars. Did you sort out that problem?

We are rookie team and don't have much to start with. Definitely no CNC for us :(. It does look like even veteran teams are struggling with this season game.

These wheels were not that difficult to make, or that expensive either. We may even still try and improve on this design too.

Even though the curve transition from flat to round was was enlarged to improve shape for better corner release from under the churros as wheels go past them, they still will get somewhat stuck, depending on how well they are. or aren't, kept laterally in sync, but we can always re-sync them on the fly well enough to un-stick them and resume climbing. In autonomous, monitoring the sync status and making sync adjustments may prove too difficult though.

In teleop we always seem to quickly get high enough to place our dual winch hooks with our single joint swing arm though.

-Dick Ledford

ArtemusMaximus

22-12-2015, 10:09

I haven't seen it in action, but I know that 5943 had a triwheel design. I'd pm Evan Hochstein about it if he doesn't pop up in this thread on his own.

Any helpful information is much appreciated :)

ehochstein

27-12-2015, 13:14

ArtemusMaximus

28-12-2015, 21:56

Here are a couple of photos of the 5943 design (http://imgur.com/a/FTIAU) - we're using a Lazy Susan from mcmaster (http://www.mcmaster.com/#1413t16/=10f92nk) to allow independent rotation. This is from our current version - we're redesigning the robot currently with a Slewing Ring Bearing instead of the plastic lazy susan. The design works great, however there are some inconsistencies in how we attached the lazy susans and the plastic lazy susans have a "wiggle" in them that causes our chain to skip.

You can download and see our full CAD model on GrabCAD (https://workbench.grabcad.com/workbench/projects/gcFNPXC_uaDE_W4I-Hyo6bx6OAJgF5FxU2yRjK_XslInlS#/space/gcWLRcrBF-pJCXnrT_O4J4jYHIYrjZrcfUCrtD7FVlg031).

It looks like you do drive tri-wheel assembly as well as wheels; which what we I am thinking. Our design is a bit simpler though. I hope we can finish prototype very soon (we are still waiting for some parts. see this thread (http://www.chiefdelphi.com/forums/showthread.php?t=140206)for details)

How well does your robot with tri-wheel climb and how high? also seeing that you are using omni wheels, I assume you don't have problems with turning on the floor?

ehochstein

30-12-2015, 14:38

It looks like you do drive tri-wheel assembly as well as wheels; which what we I am thinking. Our design is a bit simpler though. I hope we can finish prototype very soon

How well does your robot with tri-wheel climb and how high? also seeing that you are using omni wheels, I assume you don't have problems with turning on the floor?

I can't wait to see what you guys have come up with!

We got our robot to climb to the mid-zone once during practice with one of our prototypes. At that point, we were using 3D Printed custom sprockets and we sheared about 30% of the teeth off. Since that point we've encountered a lot of issues with the lazy susan we were using. That's why we are switching to the new bearing system that can be seen in the partner space on our GrabCAD account, see V4 of the design. https://workbench.grabcad.com/workbench/projects/gcFNPXC_uaDE_W4I-Hyo6bx6OAJgF5FxU2yRjK_XslInlS#/space/gcWLRcrBF-pJCXnrT_O4J4jYHIYrjZrcfUCrtD7FVlg031

One new item to note, is we removed powered front wheels. We have an idea of how to bring it back but right now the plan is to have that completely removed as it didn't add too much benefit.

ArtemusMaximus

30-12-2015, 15:05

I can't wait to see what you guys have come up with!

We got our robot to climb to the mid-zone once during practice with one of our prototypes. At that point, we were using 3D Printed custom sprockets and we sheared about 30% of the teeth off. Since that point we've encountered a lot of issues with the lazy susan we were using. That's why we are switching to the new bearing system that can be seen in the partner space on our GrabCAD account, see V4 of the design. https://workbench.grabcad.com/workbench/projects/gcFNPXC_uaDE_W4I-Hyo6bx6OAJgF5FxU2yRjK_XslInlS#/space/gcWLRcrBF-pJCXnrT_O4J4jYHIYrjZrcfUCrtD7FVlg031

One new item to note, is we removed powered front wheels. We have an idea of how to bring it back but right now the plan is to have that completely removed as it didn't add too much benefit.

I also would like to see your design in action. Video would be nice ;)

So, am I understanding correctly that you are planning to remove omni-wheels, so your tri-wheel assembly becomes 3-point star wheel? or you just removing power drive to wheels and they will be free spinning?

We are still waiting for parts from Pitsco since December 5. One of the thing we are waiting to finish this design is a worm gear box to drive tri-wheel assembly.

ArtemusMaximus

30-12-2015, 16:11

Looking at your CAD design, wanted to ask you, where do you get sliders?

ArtemusMaximus

10-01-2016, 01:52

As promised. Here are pictures of our tri-wheel design. That was this morning. It has been taken apart now :S

It showed a lot of promise, however even with 4:1 worm drive powered by AM motor wasn't enough to power this assembly.

http://www.chiefdelphi.com/media/img/a22/a229ddb787fd63ee93a3fd82c2184922_m.jpg

http://www.chiefdelphi.com/media/img/b58/b589e401eb5dc4650b6388f902b1b35d_m.jpg