After graduating I took some time off from robotics, however I will be home over winter break and will be helping mentor my team during their design phase. Over the years I have helped build many robots and systems, but I have never built a elevator/lift. My team is very capable but do not have access to high precision work. They have a small mill and a team member has his own lathe. What would the CD community suggest as a method for building a lift that would have been competitive in the 2011 game? Basically I do not think they have the machining ability to build 254/1114esque lift, however it would be nice to have a solution in my back pocket in case their early designs seek out something like this.
How small a mill? We made a full 60" tall “254 style” elevator on a 14" travel knee mill.
It was actually pretty easy on the fabrication side.
It is a SMITHY CX-329 http://www.americanmachinetools.com/benchtop_milling.htm
Do you have any advice/documentation on your process? I guess my concern was with tolerances on a 254 style system. Had you built a version of this system before the build season?
The tolerances aren’t as critical as you would think. If someone knows how to use the mill, their tolerances are likely good enough.
Check out our 2011 prototype CAD (posted in my sig) to see the elevator. Where we did gussets, and a waterjet rack gear, you could do completely different. The overall layout and bearing setup is decently easy though.
We mill 12-14" sections at a time, and then scoot the part down and rezoro off the last hole drilled by just shoving the drill bit in the hole (more specifically, we just leave the drill bit in the last hole drilled, open the vice, full travel X, reclamp vice). We call it “table jumping”. We do it for any tubing longer than table travel.
Do you have any examples of an adjustable system like you mentioned? I would be worried about things like adjustment screws loosening, but perhaps others have had success here?
There are many fastener locking solutions out there, but steer clear of split lock washers. I prefer nylock nuts and Nord lockwashers, personally. There is always safety wire and castle nuts for the supremely cautious.
You can always brute-force it and re-tighten critical fasteners after every match.
Bosch Rexroth extruded aluminum with Ecoslide carriages served as our (winch-driven) cascade lift in Overdrive and our (lead-screw-driven) robot lifter in Breakaway.
It was almost shameful how little manufacturing capability we needed to create either device, and both worked exactly as we designed them to. (Mind you, we’ve learned some about design in both cases, and would make them even better now – but still might use these products.)
In 2011 2168 used 80-20 for our lift and it work out fairly well. With the resources we had it made the most sense at the time. Moving forward we will likely look into something like what 67 did in 2011 (Look under the 2011 Build Season image in this Gallery)
The 80/20 and similar extrusions are a solid low resource way to make an elevator, but it is a great weight hit.
I’m not saying don’t do 80/20, but just hoping people explore the “254 style” of custom aluminum w/ bearings and realize that it doesn’t have to be an insane amount of work. It easily can be a day of manually milling (as ours was about a half day of manually milling for a single robot) by someone who knows their way decently on a mill.
In the end, whatever gets the most robot done in the time it needs to get done is the better option, but I hope people don’t assume that because “254 did it…” or did something similar, that it is impractical for another team to do it.
Why steer clear?
Often times on our robots, due to our standardization on #10-32s and our love of lightweight manufacture, we can never even get close to proper preload on a bolt before we destroy the parts clamped. In these situations the spring force of a spring lock washer is helpful.
I agree for properly preloaded joints they aren’t the solution, but it’s surprisingly rare that we can properly preload on our robots without going to impractically small fasteners.
When we did our 8020 lift in 2011, we found it to be a very simple, machine-less system. I can’t speak on behalf of all 8020 distributors, but ours was happy to visit our site and work with us on our design and what parts we’d need to create it. We had to do minimal (read: a couple dozen holes drilled) machining on the system.
8020 also extended a FIRST team discount to us.
Our 2011 robot was far and away the simplest and most effective robot we’ve produced in our eight seasons.
Adjustment doesn’t necessarily mean “turn a screw to adjust”. It can be just a plan of attack on how to tweak things when you get down to your final configuration/implementation.
Knowing these were going to be our primary mechanism to eat up the tolerance stack up from the welding of all the assemblies and machining of the necessary brackets, we knew we were going to rivet these things on with countersunk rivets and shim stock behind them. The shim stock was the adjustment for the side to side slop. We only needed to tune it once on our final assembly and we were set for the year.
Just an example of how you can build adjustment into a design with part of your assembly. It doesn’t need to be as elegant as bolt adjustments and the like, but of course pulling those off can be cool as well.
Want to 2nd, 3rd and 4th what many have already posted. We (272) have used 80/20 as the primary structural component on our lifts for years. 2004, 2007 and 2011. It has always served us well and we have little to no advanced machining capability. If you would like drawings (STEP files), pictures and/or videos illustrating our design please feel free to send me a private message and I will dig those up for you. GOOD LUCK!!!
Split lockwashers can cause bolts to work loose from vibrations sooner than normal. I’m coming from mostly automotive experience on this one, and have had miserable luck with split lock washers. I now avoid them on principal and haven’t had an instance where I wish I used them (car, robot, or otherwise).
A belville or wave washer (or even an o-ring under the bolt head) might work for what it sounds like you’re doing. But if you’ve never had an issue it’s hard to argue with empirical evidence on FRC robots.
Wildstang had one of the simpliest to make elevators in 2011 and ended up World Champions. It was basically many C channels. You should contact someone on their team about some pictures or tips and tricks on that elevator.
How was the durability of the pads on this? Did you notice that the tower was swaying when all the way up high?
When we made our lift with 8020, it was fast and came together nicely. However, it was heavy. One issue that was critical was that when the tower was at its maximum height, the tower would sway. We found that this was because of the play that would develop in the plastic slides (in all fairness, those slides are meant for compression loads). In short, would not do again.
One solution that I found nice was use the iGus products. If I had to do it again, I’d have to track down exactly the types to use, but IIRC, 1771 simply used some super-light c-channel lined with the iGus products. Worked like a charm.
For the elevator, you could probably even use PVC pipes and just stack them inside each other, like a telescoping device on each side. Or you could get the special c-channel PVC, which I think you can get on mcmaster.com, and make something similar to Wildstang’s design. This would probably be strong and save weight (since it’s not metal) but our team has never actually made a forklifting device before so I’m not that aware if there would be any major problems of the design.
We did an 80-20 lift in 2008, and in many regards, it was our best robot in recent memory. (We almost made it to division finals, I believe.)
The first stage was 2x1 80-20 with the standard nylon slides. It was slightly lightened (the back half of the 80-20 was replaced with welded on c-channel), but very similar to a system you could put together in an afternoon from regular 80-20. The second stage used v-shaped rollers that fitted into the grooves in the back of the 80-20.
While that system was fairly simple to design and execute, I wouldn’t go with it again. First, it was very heavy, even with the lightening on the first stage. Second, it was poorly counter weighted. And third, the nylon sliders introduced a lot of friction into the system. You really have to pull on it to get it to lift up, and it took two CIMs to raise. I never noticed any sway, but then again, we haven’t competed with it in a while.
For the OP: 80-20 is a good solution if you don’t want to do bearings riding on the outside of 2x1 tubing (like 254 did in 2007 and 2011). However, don’t count yourself out of that design too soon. Almost any design can be modified to fit your machining abilities, so take a look at 973’s CAD to see how much machining is really required.
Team 841 built an elevator system with a drill press and chop saw in 2011. The elevator uses 2x1 square tubing as the outer section, 2x1 c-channel for the center and a rolling device in the center (consisting of 2x1 square tubing connected with flat bars).This elevator design worked very well for us and will definitely use it again if the need arises.
Everything was measured with a tape measure, center punched by hand and then drilled with a 1/4 drill bit.
We noticed that there needs to be enough room for the parts to move within the system, make sure that the bearings do not apply too much tension to the moving components.
The ecoslide pads are beastly, and hold together very, very well – I had forgotten that we didn’t even use the carriages on Shiela, just the pads (screwed into the second stage of the lift, inserted into the first stage, which was the Bosch stuff). The elevator we made consisted of two rails spaced almost to the sides of the robot, stiffened with a few small crossbeams made of 1/2" L aluminum. It didn’t sway.
Our robot lift in Breakaway used the ecoslide carriage and lifted our whole robot no problem over and over again – it didn’t sway because it had little weight up top: it was just used to lift a hook, and then drew down to lift the whole robot.
