Is 3 f/s raise/lower speed reasonable for an elevator?

I’m designing an elevator for practice, and I’d appriciate help deciding how to gear the 2 neos that will be raising/lowering it - I don’t have a great frame of reference for this kind of stuff - is 3-5 feet per second a reasonable top speed for this thing to move at? To get an idea of what this mechanism will be like, its going to be in some ways similar to 254’s 2019 elevator mechanism. Thanks in advance for any help!

First three questions:

  1. Length of travel?
  2. Approximate weight to be carried?
  3. How fast do you want (2) to go through (1)?

3-5 FPS is maybe a hair on the fast side, but if you’re a 2018 robot targeting the scale it’s not unreasonable. It all depends on what you’re trying to do. For approximate reference: That’s floor to average adult head height in 1-2 seconds.

Now, if you’re the same 2018 robot trying to buddy climb, that’s a whole 'nother story.

1 Like

Acceleration is (almost always) more important than top speed for a FRC elevator mechanism. The vast majority of its travel will be spent accelerating or decelerating.

What’s your expected load on the elevator? Do you need to stall the motors to maintain position? What size circuit breakers are you motors on? These types of questions will often drive your gearing decisions for mechanisms more than top speed.

3 Likes

about 3ft travel, and I’d like to cover it in ~ a second.

Not 100% sure of the total weight yet, but I’d say sub 30 lbs hopefully. If it gets to be a serious problem I suppose I could always resort to counterbalancing with elsasitc or tensions springs or smth.
@Lil_Lavery I think you’re right about acceleration being important - Since the distance of travel is so small, there’s really not much time to be saved from going a bit faster and sacrificing torque. And yea, the motors will need to be able to hold the thing in place.

You might be asking the wrong question here, but understandable for just practicing a design.
Normally, this is determined by working out how many times an action, say scoring a hatch panel, must be preformed during the match. On 330 we usually planned for winning a 1 v 3 (assuming a division-level alliance) and doing the most a single robot can for an extra RP, which drive the # of points scored and specific actions preformed.
You get to the point with a big jumble of variables involving drivetrain speed and power, average mechanism speed, etc and mess around with those until you’re happy.

Or, hear me out, stowed to scoring position in under 1 second.

11 Likes

That’s the only year we considered using a shifting gearbox on an elevator. Power cubes are light and need to go up quickly, robots are heavy and can take a few seconds longer.

3 Likes

The amount of power to raise a lift fast can be very substantial. One way to lower the power requirement and speed up a lift is to use constant force springs to assist. We have used Vulcan springs.

4 Likes

An elevator that is fast enough, will seem like a guillotine when it’s moving down. Nice and scary.

and will be driven by two Falcon motors

This!

As a point of reference, in 2018 we did use a shifting gearbox in 2018 (3 CIM ball shifter). We selected the maximum ratio between the two gear ratios. We geared the low speed gear ratio to be able to lift 500 lbs with only the 1st stage of our elevator (where the hang hooks were mounted) moving about 12" in about 3 seconds (the free speed was faster than this, of course). This design criteria was based on a double buddy lift. Later in the season we switched to a single buddy lift and we were able to do the endgame climb in about 2 seconds instead of 3.

The elevator was a cascade design with the cube intake on the second stage (which was just a shuttle stage running on the 1st stage. In high gear, the elevator was able to go to full extension (about 7 feet) in about 0.8 seconds (using a trapezoidal acceleration, deceleration profile with encoder set points) while lifting the weight of the power cube. So, the first stage was probably running 4.5 - 5 ft/sec at the peak speed such that the second stage was running 9 - 10 ft/sec. The free speed was probably a bit faster than that. This allowed the “stowed to scoring” in less than 1 sec.

If we had decided to do a single buddy lift initially rather than the double or if we were not doing the buddy lift at all, we might have geared a little faster for just the cube lift, but probably not much faster. Control would have become the limiting factor if we had gone much faster.

2 Likes

One way to approximate how fast you can go in theory is to use a simple power calculation.

If your game piece, elevator carriage, and friction (estimated) are 10kg, you’ll need to apply a force of m*g to lift it. g is 9.81 m/s^2 gravity, so you need to apply 10 * 9.81 = 98.1 Newtons of force. To get energy required to lift something, just multiply that by the distance to travel - in your case, height to lift to (say about 1m in this instance). 98.1 * 1 = 98.1 Joules of energy.

Power in watts is in units of Joules/s. So a motor that can output 100W of power constantly (like A NEO or Falcon) can lift that mass 3ft in (98.1 J)/(100 J/s) = 0.98 seconds. If you counterweight the elevator down to a mass of 1kg using constant force springs, suddenly you’ll be capable of going up to ten times faster, and really just be limited by friction and acceleration.

You will spend some time accelerating and decelerating, but it will be a small amount of energy relative to the energy required in the lift if it’s not counterweighted. A back of the napkin kinetic energy calculation says a 10kg elevator will use only 5J to make it to a speed of 1m/s. The reality of more complicated equations involving acceleration and motor power means that those 5J could take a moment to get from the motor, but it’s useful to estimate.

All in all, we have enough power available to use that 3-5ft/s is totally fine for most elevators and loads using Falcons or NEOs. In years like 2018 where the elevator could be moving constantly up to 6ft, I might even go faster (we had about 3ft/s that year and it was too slow for me). For 2019, 3-5 seems fine. There’s usually not a lot of harm in gearing fast IF you have a high factor of safety on your motor stall, and you can control it. Having preset heights for the driver to go to will help more than a fast elevator, and is a must of it’s too fast to manually control easily.

The 11th commandment is: “Thou shall balance all elevators when possible”. It gets rid of that pesky 9.81 number and then all your energy can be concentrated on accelerating the mass. Also the control loop looks the same up as down. Also, no need for a massive brake or holding current… Millions of garage doors (and their tiny openers) can’t be wrong.

This commandment doesn’t apply to elevators used as winches.

10 Likes

It’s also a really fun surprise in the pits when you turn the robot on its side to access the bottom of the bellypan!

15 Likes

180 had a pretty neat solution to this at champs 2019 – I didn’t get any pictures, but it was basically a quick release pin that they could use to hold the stages together when they turned the bot on its side, and then easily pull out when they were done

Sounds like a fun surprise when your elevator can’t raise in an eliminations match.

16 Likes

Maybe this is a use for those big red “Remove Before Flight” tags on airplanes?

3 Likes

Isn’t that why teams are supposed to do pre-match checks / have lists of things to do before every match?

1 Like

Murphy

3 Likes

Just be careful to review first names of people when selecting your pit crew. ez.

4 Likes

We had a stick that held open the Versaplanetary ratchet in part of our climber with a large UP-shaped piece of wood attached that stuck out like a sore thumb.

UP → Upper Peninsula of Michigan. We’ve made a few as awards and I think we had a spare lying around.

image

5 Likes