Use of Lead Screw

[nnfuller]

Our team is currently looking to lift a decent amount of weight in our shooting mechanism, and one of our mentors suggested using a lead screw or ball screw. I said that I as hesitant because of the fact that I cant find a robot that used one as an effective mode of lifting a large amount of force. We are thinking about going with a design similar to Delphi or Robowranglers arm linkage in 2011. I guess my question is why not use a lead screw? What makes it so that so many teams don't use them? According to my math it would have to be 3 feet long.

[Al Skierkiewicz]

Lead screws have been used on many robots over the years. There are pros and cons as with any mechanism. For one they can be slow but at the same time precise in movement. cost and weight can be factors to consider as well.

[nnfuller]

As I see it:

PROS
Precision
Locking
Relative Simplicity

CONS
Often Slow
Weight
Size
Fragile

[metaltech]

A screw as you describe can lift a lot of weight, or more accurately, generate a lot of thrust. So, you have to have strong mounts to react that thrust. IOW, be able to hold the screw without the mounts deforming. In real machinery, there are bearings designed for mounting leadscrews or ballscrews.

Using something like that would be great, but you may be able to get by with less (cheaper) considering the 'bot doesn't need to run for years. Perhaps a block of nylon with a counterbore (pocket) that the end of the screw sits in. This is NOT optimal (won't guaranty that it will work) but it's pretty cheap and easy to make.

[MrForbes]

We use leadscrews when making robot parts...there are leadscrews in the mill, and in the lathe.

But they usually are kind of slow for what we need on the robot. We are considering using one for the small height adjustments on a shooter this year.

[MechEng83]

1741 used a lead screw on our robot for logomotion. We actually used a 5 helix screw that went up 1/2" per revolution. When running our initial tests it was described by several team members as "scary fast" however, we were plagued by our shuttle binding which would burn out the FP motor. We eventually solved the issue by increasing the driven sprocket diameter by a factor of 2.

We had about 32" of travel with a 42" screw that could be traversed in under 2 seconds, but that speed would jam our mechanism at the top and especially bottom, so we used a profiled motion curve to ramp down the speed near end of travel.

While I think our design could be improved upon, my recommendation is to look at another path. With a 6 week design and build process, there are easier ways to lift your shooter.

[Grim Tuesday]

We used a lead screw in our 2011 robot. It had a 'duck bill' that clamped on tubes; this is what the lead screw as used for. It was surprisingly fast; we used one of the Banebots (not a 775, but not the smallest one either). It clamped on a tube in about .5 seconds. The one odd thing was, considering the fact that it is essentially a worm gear, it wasn't terribly powerful. The tubes could still be pried out, and the bill forced up and down.

[martin417]

1771 used a lead screw last year to power our elevator. It could go from the floor to the top rack in 3 or 4 seconds. See early slow motion test here. It worked flawlessly all season long. We purchased a 6' length of 5 start, 1" per rev ACME lead screw from McMaster Carr. The lead screw was not very expensive, but the nut and flange were more than expected. Here is a video of an early test showing the lead screw and assembly, activated by a hand drill. Sorry for the sideways video, and ignore the commentary. I couldn't find any still pictures of the assembly, but I did have this vid.

What was good:

• precise. we used an encoder on the screw to get position, and it was always dead on
  
• powerful. We never had a problem lifting the wight of the second and third stage of our elevator, and didn't have to use springs to assist.
  
• Reliable. We never had problems.
  
• Simple. It was super easy to design and build, no complex mechanisms required.

What was bad:

• Not as fast as other elevators I saw.

One thing you want to look out for is shaft whipping. There are equations out there for calculating the maximum rotational speed for a lead screw. Look at worst case, which is when the traveling nut is at one end or the other, and most of the screw is un-supported. If you exceed the critical speed, the shaft will "whip" or "Whirl" and cause vibrations that can be catastrophic to bearings and supporting structure. The critical speed is directly proportional to the length of the screw, so if your screw is short, you probably won't have a problem.

Good luck, and if you have questions, feel free to PM me.

[pfreivald]

Originally Posted by nnfuller

I said that I as hesitant because of the fact that I cant find a robot that used one as an effective mode of lifting a large amount of force.

Our 2010 robot used a 3' lead screw to lift our robot two feet in the air during the end game. We used a four-start screw, shaft coupler, and thrust bearings from McMaster and a 12:1 reduction CIM (and lots of grease on the screw). It was extremely reliable (failing once at FLR because we used a set screw instead of a spring pin -- never again!), but took a few seconds.

The motion is smooth, but not tremendously fast. Inspired by our success, we used encoder-driven lead screws for our shoulder and elbow on our 2011 robot, and we should have used pneumatics for speed.

[kws4000]

Mona Shores used a ball screw for their gripper: FP Motor, dismantled drill clutch, basic screw included in KoP; pulled a bike brake cable to run the gripper. Fast and simple. The clutch was what got me hyped!

If only my team could've used that instead of our closed-circuit air pump....