hero-image
HOME
hero-image

High Hang Elevator Robot - VEX Robotics High Stakes

hero-image
Carson Schmitt

Project Timeline

Jul 2024 - Nov-2024

OVERVIEW

In the 2024-2025 VEX Robotics challenge, High Stakes, there was a 6 foot tall ladder that not many teams wanted to attempt climbing up. Because of the rules, you had to "ladder" climb it, making it a challenging problem that my team and I decided would be a super fun design challenge. As the lead designer on the team, I spent 2 months in CAD iterating, developing, and testing different solutions before finally coming to a final solution. We then built it, tested and improved it, and finally ended up creating a working solution. We ended up being the first team in the world to do the hang and inspired many other teams for the remainder of the season.

HighlightS

  • First team in the world to accomplish high hang
  • Won the innovate award at the 2024 Haunted Signature Event
  • Used custom laser cut parts to build linear rails

SKILLS

RoboticsCADInnovationComplex Mechanical DesignProgrammingSensor Integration

Additional Details

Problem Statement:

For the 2024 VEX Robotics Game, High Stakes, teams could climb up a 6-foot ladder with 3 sections. This was a part of the challenge most teams did not want to tackle, but I am always up for things that seem too difficult to others, pushing my design skills. As the lead designer on the team, I had the task of brainstorming, researching, and developing a solution in CAD to then build and implement for our tournament in November.


Brainstorming and Research:

Looking into the climb for this game as shown below:

Pasted image

We realized that the ladder is very similar to the 2013 FRC game. Because of this, we started looking into robots from that game and came across an interesting but very efficient solution

This robot utilizes an elevator lift with claws that can pull up on the bars and then transfer to another set of claws mounted to the base of the robot. This way, they can lift back up to grab onto the next rung of the ladder.

There are a few very important things to note about this design however:

  • FRC has less limitations for materials and mechanical parts meaning we would have to replace a belt powered lift and create custom linear slides
  • There motors are way more powerful than ours, meaning we will need to be attentive to mechanical advantage
  • The tower is angled in, while our tower is straight

Considering these things, we are able to begin finalizing a design.


Design Criteria

I settled on the following things I want out of our hang mechanism:

  • Power Transmission Off (PTO) from the drive system to add 6 motors of power to the hang
  • Winch system with small diameter for a strong speed to torque ratio
  • Hinged, hard mounted hooks to make transferring between stages smoother
  • Custom linear slides using laser cut plastic and custom rollers
  • Continuous string lifting system to replace belt

I am now able to begin designing with these criteria in mind:


Version 1

Pasted image

At this stage in the design, I had figured out the chassis and towers/slides, but as I was sketching out the positions of the other parts of the robot, I realized there may be a spacing problem.

Criteria
Meets Criteria
Does Not Meet Criteria
Notes/Changes
Fast Drive
X

Includes space for all subsystems

X Intakes and electronics mounting didn’t have enough space.
PTO positioned under lift for easy string routing


Robot failed at subsystem space, so the CAD was not completed.
Center of Gravity


Robot failed at subsystem space, so the CAD was not completed.


Version 2

Pasted image Pasted image Pasted image

For this version of the design, I changed up the tower mounting to try and make more space. I also modified the lift to be a 3 stage elevator so that it had enough reach to get to the next stage.

Pasted image Pasted image

Early on in this design, I wanted to figure out PTO, battery, and brain placement so that we could maintain good COG while also ensuring there was enough space.

Pasted image

Due to the placement of the brain and the battery, I had to shift the PTO forward:

Pasted image

As seen by the sketch, this causes the string to need to be rerouted around a pulley and cut an almost 90 degree corner, which is really bad for mechanical advantage.

Criteria
Meets Criteria
Does Not Meet Criteria
Notes/Changes
Fast Drive
X

Includes space for all subsystems
X

PTO positioned under lift for easy string routing

X PTO required complicated string rerouting with poor mechanical advantage
Center of Gravity


Robot failed at PTO position so CAD was not completed


Version 3

The biggest issue in the previous two versions was spacing, because of this, I am designing around a longer drivetrain by utilizing a chain drive. This allows the gears for the PTO to be put wherever they need to as well as opening up more room for subassemblies.

Pasted image Pasted image Pasted image Pasted image Pasted image

After the initial structure, I was pretty confident in the new improvements spacing wise. Next I focused on designing the PTO.

Pasted image Pasted image Pasted image

Pasted image

Next I worked on the intake design.

Pasted image Pasted image Pasted image

Pasted image Pasted image

Now that the intake design is done, everything seems to be coming together, the next step was the mobile goal clamp.

Pasted image Pasted image Pasted image Pasted image

Finally, I added a very similar 3 stage elevator as what I had already designed for version 2.

Pasted image Pasted image Pasted image

Final Design:

Pasted image


Assembly

Pasted image

We spent lots of time building, testing, and tweaking this design to get it to work. More information on that process can be found in our design notebook:
https://heyzine.com/flip-book/7005792d0e.html#page/400


Code

As the lead programmer on the team, I also had to create and code a macro for the hang. I utilized a rotational sensor to keep track of the position of the winch. I also utilized a gyro sensor to make sure the robot was level at each tier.

Pasted image

Pasted image

Pasted image

This macro allowed us to hang consistently and reliably.


Performance



Final Thoughts:

This project was super rewarding to work on. It was so cool to share this complex mechanical design with other teams from around the country and even the world. I competed in VEX Robotics for all 4 years of high school, for other examples of my work on the team, check out our YouTube page:

5203G Gremlin YouTube

lowinertia
Portfolio Builder for Engineers
Created by Aram Lee
© 2025 Low Inertia. All rights reserved.