# How do teams make fast climbers?

How do some teams make mechanism that can lift a robot so quickly? Our team has made mechanisms that can lift robots, but they have always been pretty slow. Seeing some climbs from this game made me wonder how teams decide to power or design their climbers so that they can both lift a great amount of weight and also be fast.

1 Like

Calculate your gear ratio. We’re able to lift 3 robots with 2 775 pros faster than 6"/second.

7 Likes
20 Likes

I would recommend messing around with the JVN Design calculator.

The short answer: They likely did the math. You should also do the math.

The long answer: Each motor has a known power output. When you are lifting a robot, you are basically doing work (W=F*D), and if you want to do as much work in as short amount of time as possible, you want a lot of power, which is the rate at which work is performed (P=W/t).

There are many ways to increase power, but the simplest is to add motors. If you’re out of motors on your bot, or PDP slots, a common way to add more motors is some kind of PTO, where motors that are typically used for another system are switched over to the climbing system. I’m sure there will be many teams that use PTOs from their drivetrain or another system to switch the power over to their climber.

Another way is to use the power-band of whatever motor you are using. Using something like the JVN, or AMB calculator, you can determine what ratio you need to utilize the portion of your motors output that generates the highest power (or most work in shortest amount of time). Others have linked to that above.

One way that we can help you is if you provide a bit more information about your system. How much does your bot weigh? What motor(s) are you using to climb? What gear ratio are you using with said motor? How big is the spool for the wire/strap you are using for your climber? Can you show us a picture of your system?

9 Likes

We are using 2 775 pros into a 3:1 gear box, which then goes into an 80:1 worm gear. The diameter of the winch spool is about 2.5". We are looking at around 105lb inspection weight.

We used a neo and a 15:1 gearbox to do this: https://drive.google.com/open?id=1Pw9P1Gk1vv5W4-7Dovm8adOF6yWh-oqs

In addition to doing the math, we’ve got 2x 25 lbs constant force springs that pull our elevator down. It means we have to push against that 50 lbs when bringing it up, but it means our elevator motors push 60 lbs up, hook, and then only have to pull 100 lbs down.

4 Likes

Based on the numbers you just gave, I think you are using much too high of a ratio (240:1). Just to help give some perspective, when I ran those numbers through JVN, I got a stall load of 1,930 Lbs, which, as long as your robot doesn’t contain any kind of gravity field generating device, is significant overkill for your application.

I plugged in a total ratio of 60:1 into JVN and got an estimate of you climbing 30 inches in just over 1 second that pulls 31.5 A per motor. That ratio still has nearly 500 lbs until it stalls, so you should still be able to lift your bot easily. Compare that to 3+ seconds with a 240:1 overall ratio.

1 Like

The only motor you will ever need. (Unless you need it right now)

17 Likes

I’m going to suggest using my design spreadsheet for calculations like this. The mechanism calculator should let you determine the proper motor(s) and gear ratio for your application.

4 Likes

Do the math, use a mechanical assist if you can, and go for a bit of overkill.

P = F d / t.

F is your robot weight, so let’s call that 150 lbs = 1100 N. d is how far you want to lift, so for this year let’s call that 0.5 m. If you want to lift your robot that far in 1.0 seconds, then your minimum power needed to do so using motors is:

P = 1100 N * 0.5 m / 1.0 s = 550 Watts.

This ignores friction, which will increase that number, and it ignores things like constant force springs, surgical tubing, or whatnot used for mechanical assist, which will decrease that number. Of course a lighter robot will decrease that number as well.

So you’ve got 1100 N you need to lift. Suppose you’re using a .5" hex bar for a winch shaft – that’s got a radius of 0.00635 m, so the torque you need to barely pick up the robot is 1100 N * 0.00635 m = 7 N*m of torque.

At 40 A, the Falcon provides about 0.6 Nm of torque, and two of them provide 1.2 Nm of torque, so the minimum gearing you need is 7 / 1.2 N*m = 5.83:1 to barely lift without exceeding 40A.

So if you’re Team 1551, you slap two Falcons in a toughbox mini geared at 10.71:1, slap a pieces of tiedown strap on the output shaft, and call it done.

4 Likes

Hi I am a member of the same team as the person who created this thread and I was one of the people responsible for the design of the climber. While I agree that the 240:1 ratio is a bit overkill but in our tests it still struggled to lift 120ilbs so I am not convinced of those numbers. But, the way our bot is designed means that we have to route the climber cables (nylon rope) around two pullies before it reaches the climber which means some of the force is translated to those rather than lifting the bot so we needed to have a higher ratio to compensate for that loss.
Note: we know we could increase speed and reduce gear ratio if the winch had a direct line to the climber but with the way our manipulator mechanism and electronics board are designed this is simply not possible.
Do you have any suggestions?

Pictures…do you have any pictures? they really really help us to be able to see what you have, so we can make useful suggestions.

The 240:1 ratio doesn’t really provide enough information on it’s own; we’d also need to know what motor(s) you have on the mechanism and what the diameter of your spool is.

1 Like

That was shown here:

“We are using 2 775 pros into a 3:1 gear box, which then goes into an 80:1 worm gear. The diameter of the winch spool is about 2.5”. We are looking at around 105lb inspection weight."

Losses from the pulleys, assuming they have bearings, should be minimal, at least compared to the geartrain, so those numbers should hold true in the general sense. What exactly do you mean by it is still struggling? Do you have a video, or even a picture of the mechanism? I find it hard to fathom that two 775’s with that high of a ratio are struggling to pull, well, anything.

Something to consider is that worm gearboxes have very bad efficiency. Changing the efficiency in my calculatons to 50% (from 80%) still results in a sub 1.5s 30" climb.

Here is a video of our proto-bot climbing it is only around 50ilbs but it should give an idea of what this looks like. And by struggling I mean took significantly more time to lift it with weight than without and we tested with a variety of weights.

1 Like

Do you have more information about the worm gear you’re using? Some are very inefficient, and you might be losing a lot of your power due to friction there.