pic: Linear bearing rails: the Cheesy RAWC way

[roboticWanderor]

[cdm-description=photo]31607[/cdm-description]

[billbo911]

Pretty, but................... What keeps it from pulling apart vertically?

[roboticWanderor]

Pretty, but................... What keeps it from pulling apart vertically?
take a look at 254's lifter for that answer, i cant explain it that well. or ask Cory ;)

[JesseK]

They had two of these mounted exactly as shown, but mirrored so that the inside rails could be connected via cross members. Then the outside rails were firmly mounted to the drive train frame and cross-braced at the top. It required exact spacing at the top and bottom to be smooth, and I will conjecture that the force that pushed/pulled it up had to be almost dead center as well. If you got to see their model up close, it was smooth as butter and very lightweight.

[billbo911]

take a look at 254's lifter for that answer, i cant explain it that well. or ask Cory ;)

Maybe I'm missing something. Where can I "look at 254's lifter " . Are there any close up pictures posted?

From the CAD, I can see how it is restrained in two dimensions, but not the third.

[JesseK]

Any time anyone uploads a photo to CD Media it auto-tags the photo with 'frc###' (or 'ftc###') where ### is the Team Number in the profile of the uploader.

Try searching 'frc254' and 'frc968'.

[roboticWanderor]

Any time anyone uploads a photo to CD Media it auto-tags the photo with 'frc###' (or 'ftc###') where ### is the Team Number in the profile of the uploader.

Try searching 'frc254' and 'frc968'.

Thing is, there aren't any pictures of it on CD media... ill try uploading a few I have...

[billbo911]

Thing is, there aren't any pictures of it on CD media... ill try uploading a few I have...

Exactly my point. In fact, when you search CD for pis using "FRC254" or "FRC968" you will find pics of your slide, but not theirs.
Please upload what you have. We are always trying to improve our lifts. I would love to see some that worked better than ours.

[roboticWanderor]

Exactly my point. In fact, when you search CD for pis using "FRC254" or "FRC968" you will find pics of your slide, but not theirs.
Please upload what you have. We are always trying to improve our lifts. I would love to see some that worked better than ours.

my pictures are taken with a nice SLR they are over 2.0 mb!! aaa! ill try and crop / downsize them

[Cory]

Maybe I'm missing something. Where can I "look at 254's lifter " . Are there any close up pictures posted?

From the CAD, I can see how it is restrained in two dimensions, but not the third.

This was the little demo we had set up in our pit in Atlanta, that allowed people to play with it. It's not representative of the actual setup.

The outside is a plain 1x2 that makes a U frame, with the bottom being solidly attached to the robot base, and the top the open portion. The top is held at the correct width by a spanner that goes across the two beams. You can see this in pictures of our robot--it's the piece the pulley/flag holder/spring mounts to. The bearings then mount as you see here, top and bottom.

The inside is a rectangle made of channel on the left and right sides, and 1"x2" on the tops and bottoms. it rides inside the bearings, as shown.

[billbo911]

This was the little demo we had set up in our pit in Atlanta, that allowed people to play with it. It's not representative of the actual setup.

The outside is a plain 1x2 that makes a U frame, with the bottom being solidly attached to the robot base, and the top the open portion. The top is held at the correct width by a spanner that goes across the two beams. You can see this in pictures of our robot--it's the piece the pulley/flag holder/spring mounts to. The bearings then mount as you see here, top and bottom.

The inside is a rectangle made of channel on the left and right sides, and 1"x2" on the tops and bottoms. it rides inside the bearings, as shown.


I think this is a "Picture paints a thousand words" situation.

[Cory]

I think this is a "Picture paints a thousand words" situation.

http://thumb7.webshots.net/t/66/666/5/82/65/2702582650055556207hJeknB_th.jpg (http://entertainment.webshots.com/photo/2702582650055556207hJeknB)

This is the 2007 robot, by the way. The only difference is aesthetically the bearing mounts on the 2008 are prettier, and the 07 doesn't have the bearings at the top like the 08. To help with racking side to side in 07 we used teflon pads on the inside of the 1x2 uprights. For 2008 we added the bearings and removed the teflon.

[IKE]

33 actually used some benchmarking off of 254 for our elevator this year. It was a really cool experiment in the importance of certain dimensional tolerances. While many dimension in this system can be reasonable, many have to be extremely precise. Too tight and the system will die a binding death. Too loose and it will beat itself to death.

Overall it was a very cool project, but tough on a mill without a digital readout. Definitely gave me additional respect for the POOFS craftsmanship.

If you have never built an elevator it is definitely a good off season project to learn what is important.

[billbo911]

http://thumb7.webshots.net/t/66/666/5/82/65/2702582650055556207hJeknB_th.jpg (http://entertainment.webshots.com/photo/2702582650055556207hJeknB)

This is the 2007 robot, by the way. The only difference is aesthetically the bearing mounts on the 2008 are prettier, and the 07 doesn't have the bearings at the top like the 08. To help with racking side to side in 07 we used teflon pads on the inside of the 1x2 uprights. For 2008 we added the bearings and removed the teflon.

As usual, while doing my cardiovascular workout today I had a brain flash. Suddenly I realized what everyone had been saying and it all made sense. A cross member at the top of the slide is what prevents the rails from pulling apart. I was viewing the CAD drawing and not realizing it was rotated 90°.

Cory, thanks for the pictures. They're a little fuzzy but they confirm what I finally realized.
I'm glad that you mentioned you were using Teflon as a glide surface. In some situations it makes perfect sense, this being one of them. It is also much easier to manufacture a Teflon glide that fits properly then it is to manufacture a bearing mount without access to a decent mill.

[Colin]

Actually our glides (254) for 2007 where simply adhesive backed Teflon from McMaster, they where really easy to manufacture, we just cut them out with scissors. :)

[roboticWanderor]

I was viewing the CAD drawing and not realizing it was rotated 90°.
yep, this model is of the demo sliders they had in their pit. They even have "slide me" sharpied on! i am working on figuring out the entire lifter mechanism, coming soon.

[billbo911]

I am working on figuring out the entire lifter mechanism, coming soon.

Excellent! I have seen multiple posts from people looking for lift design ideas. This will definitely be a great addition.

[=Martin=Taylor=]

Tolerances...

The Poof frame is welded, heat-treated, and then powder coated. And yet when it all goes together the bearings rest perfectly on either side of tubing. Amazing? I think so...

They probably took into acount how much the frame would expand and contract after each process to get it that perfect. Having welded an elevator myself this season, I fully appreciate that kind of craftsmanship.

[ejSabathia]

I am working on packing the entire 2008 robot cad, when I have the elevator done I'll post a .stp file for everyone.

[Jeff Waegelin]

Cory,

What kind of bearings did you use for the lift, and where did you get them?

[R.C.]

Congrats Poofs, you guys probably have the smoothest forklift made in the past few years.

I like the colors, blue and black

[roboticWanderor]

I am working on packing the entire 2008 robot cad, when I have the elevator done I'll post a .stp file for everyone.

wait what? sweeet! now i can figure out how you guys fit everything underneath your lifter!n

[ejSabathia]

Cory,

What kind of bearings did you use for the lift, and where did you get them?

http://www.chiefdelphi.com/forums/showpost.php?p=753305&postcount=4

[waialua359]

thanks EJ and Cory for sharing. Pictures, CAD and explanations like these does wonders in inspiring others to either do something similar or kickstart other offshoots.
No wonder our machinist asked me to order a whole lot of channel aluminum. He's been taking "classes" in your pit area getting more ideas.;)

[sdcantrell56]

How big of an improvement was the switch to bearings over the teflon pads. It seems like it wouldn't make much of a difference for relatively light gamepieces and manipulators but for the trackball might have had a bigger effect.

[Cory]

How big of an improvement was the switch to bearings over the teflon pads. It seems like it wouldn't make much of a difference for relatively light gamepieces and manipulators but for the trackball might have had a bigger effect.

That was the reason we did it.

We never had any problems with the teflon pads. Nor did we have any with the bearings. I'd say there was a little less play this year than last.

[sdcantrell56]

Thanks cory. Also how much play did you factor into the spacing for the teflon pads and the channel or did you have it exact?

[Travis Covington]

This design isn't much different from your average forklift. The original design used last year was inspired by a walk-behind forklift that was at my previous place of work. If you look at one you will notice they are almost identical in both form and function, bearing placement and all.

The comment about heat-treating and precision tolerances should be noted. We did not heat treat any of the welded aluminum components of the frame or the elevator. Also, the precision and fit that we were able to attain was also due to the quantity of assemblies that were made. When making four complete sets for both teams, we started out with each set labeled as a part of a complete assembly. After welding, the assemblies no longer fit together as desired (somewhat expected). Luckily, we were able to mix and match sets and ended up with 2 sets that were super smooth (competition robots) and 2 sets that were acceptable (practice robots). With more than 5 pieces per robot, with 4 robots, the mixing and matching process took almost 4 hours one evening to get things to the way we wanted. Not as much preparation and precision as some give us credit for. We wish we were that good. ;)

I also feel the need to mention that this elevator and the previous years robots are as much 968's babies as they are 254's. Students from both teams put a lot of time and energy (and heart) in designing and making such competitive robots, and it’s hard to see those efforts ignored. I think most people who know the specifics of the collaboration would agree that both teams contributed equally. Without certain aspects from each team, neither would be as competitive, engaging, or rewarding to the students as they are when they are working together.

[R.C.]

Nicely put

[JesseK]

I think it's actually alot easier than most people think to keep things square and precise while welding. The biggest success we had this year was the fact that we had a perfectly square and flat frame on the first try. Our student build team captain took a bit of time to consider everything that was needed before welding and came up with a plan for manufacturing the frame. We didn't start welding until I and (more importantly) the professional welder at our sponsor approved the plan.

When we welded our drive train frame together, we also brought in threaded rods that matched the axle holes and a bunch of nuts to fit the rod. We then spaced out the frame on the rods and welded the outer rails to the rear cross support. Then we used precisely-cut wooden blocks as spacers, and clamped the inner rail to the outer rail with the spacers in between. We measured diagonals as well as across the gaps before we welded, then after we spot-welded, then again after the final welds. Since the rods and spacers were firmly in place, the frame could not bend/warp.

[thefro526]

I've always looked at the 254 and 968 elevators and wondered how'd they do that. IMO it looks like the best way to do an elevator if you can machine the bearing blocks and if you can weld the aluminum. We used 80/20 the is year an it worked great for use but I noticed that the pads had a tendency to sag at full extension towards the end of the season. Also 80/20 is some heavy stuff (and that's putting it nicely:D ). Did you notice that the teflon pads wore down on your old elevators after a lot of use? I'd really like to experiment with an elevator like this and it seems like the teflon pads are a more feasible option for us.

Thanks a lot for discussing your designs here as well, I know 254 and 968 and even 60? when they were with 254 back in the day have inspired more robots then you'll ever be able to count.

[=Martin=Taylor=]

I think it's actually alot easier than most people think to keep things square and precise while welding. The biggest success we had this year was the fact that we had a perfectly square and flat frame on the first try. Our student build team captain took a bit of time to consider everything that was needed before welding and came up with a plan for manufacturing the frame. We didn't start welding until I and (more importantly) the professional welder at our sponsor approved the plan.

When we welded our drive train frame together, we also brought in threaded rods that matched the axle holes and a bunch of nuts to fit the rod. We then spaced out the frame on the rods and welded the outer rails to the rear cross support. Then we used precisely-cut wooden blocks as spacers, and clamped the inner rail to the outer rail with the spacers in between. We measured diagonals as well as across the gaps before we welded, then after we spot-welded, then again after the final welds. Since the rods and spacers were firmly in place, the frame could not bend/warp.


I think the key word here is "professional welder," I could best be described as a "professional n00b" seeing how all I weld are the team robots... ;)

Thanks for the tips though. We'll keep that in mind if we ever do it again.

[sdcantrell56]

Were the bearings attached with simple bolts or shoulder bolt? If you just used plain bolts then it would be much cheaper.

[Travis Covington]

Regular screws with nylon washers so the outer races could spin freely.

[sdcantrell56]

I was hoping regular bolts was the answer. That will make things much simpler.