Resources

Here are things that actually helped us improve. We didn't do them all at once, and we don't do the same things every year. We try to focus on what really matters to our robot that year, instead of testing just to test. We don't test things only to get points on a rubric. We test things when we don't know the answer to something. It's a great way to learn! Thinking up a good way to test something can be hard, but it also can be fun.

We divided up this resources page into TESTS that your team could try, TIPS for building attachments and robots, and LESSONS that we have packaged up for sharing with other teams or new team members that join our team.

If you have any questions or you want to suggest things to us, leave a message. We're always learning!

TESTS

Wheel Tests

A lot of times you don't want friction in your designs, but for the rubber of your wheels you definitely want friction to prevent slipping. A robot that slips is not reliable at driving.

To test, we created a ramp that you can change the angle up until the wheels start to slip. You need to have a way to measure the angle of the surface, and once the robot slides, you write down the angle.

We built the skeleton of a robot that allowed us to quickly change the wheels to try lots of different types. Then you have to pin the wheel so it can't rotate. You can see the axle sticking out of the frame of this robot that went through the hub of the wheel. This keeps it from rolling down the ramp when you tilt it up.

The steeper the angle, the more friction that wheel has, and the less likely it is to slip on the mat during a competition.

Center of Gravity Test

A lot of times robots will tip forward when they stop or start. Some call this "bucking." The reason it can be a problem is that it might turn a little bit or skid forward and be out of position for the next move that you have programmed. If this happens to you, you might want to test where the center of gravity is on your robot.

If you have 2 drive wheels and 1 castor ball you have 3 points of contact with the ground. You can imagine drawing a triangle between these points, as shown in green in the pictures here. The center of gravity (CG) is the point where all of the weight of the robot is balanced. It is shown as a red circle here. If this CG is within the triangle then the robot will stand upright. If for some reason the CG falls outside the triangle, then it will tip.

There's another concept called a stability margin. The closer the CG is to the edge of the triangle, the less stable it is. This matters because a larger stability margin protects you against more aggressive forces. If you accelerate or decelerate quickly that might be enough to tip a robot where the CG is dangerously close to the edge. This was actually the case in the first picture here.

Then we flipped the wheels around and moved the castor ball to the other side of the robot. So our front became the back, and the back was now the front. You can see in the second picture that now the CG was safely in the middle of the triangle and this made a big difference in our driving.

OK, so how do you figure out where your CG is? Use a ruler or some other very thin but strong beam. You want to carefully lift the robot just an inch or so off the ground and try to get it to balance. We are showing it here as a red line, so you could imagine that this was actually the ruler. The point where it balances will be right on this line. It might be tricky, but you can play with it a bit and move around the ruler until you get it to balance.

You could repeat this in the other direction, too. We didn't, because our robot was pretty symmetric from left to right. So we didn't expect much difference and just assumed the CG was centered. But if your design has more weight on one side or the other then it would be different. If that matters to you, then repeat the test with the ruler pointed in the direction of the wheels and see where that line is. Where the 2 lines intersect is exactly where the CG is on your robot.

ROBOT BUILDING TIPS

Frame around wheels

We have a team rule for building. Every axle needs to be supported by 2 or more points. It is never OK to just have 1, or the axle can bend too much. You can see this in the top picture. If instead an axle is supported by two points then it doesn't bend as much. You can see this in the bottom picture. This can really help you if your attachments have a lot of force and the gears are slipping.

This rule also applies to the wheels on the robot. If you just have wheels hanging off the end of an axle that is connected to your motors, the axle can bend a lot. This makes your wheels not straight and the robot can wobble as it drives.

The solution is to have a frame that goes around your wheels. Another great reason to do this is that it protects your wheels from hitting the wall. Now you can rub right up against the wall if you need to and it won't make you spin out or get stuck.

Motor Choice for driving wheels

We took some data from the actual LEGO website back when they were selling EV3 products and Spike Prime had just come out. This was a huge question for us at the time. In general, people always said, "large motors are strong, but medium motors are faster." This was definitely true for EV3, but it turned out to be much less true for Spike Prime.

We would still recommend that if you use either of these 2 sets, you use the large motors for your driving wheels because you will get more torque. If you are worried about the speed, you can always change the wheel size to get more speed.

Drop On Attachements

The whole point of having drop on attachments is to have faster changeouts. If you are only going to solve a few missions, you probably don't need to worry about how to quickly change attachments. You might even be able to make a generic lifter arm and solve enough missions without changing any attachments. If you are going to try to solve a lot of missions, then you definitely need to change attachments and that will take you a lot of time if you have to make and break connections for motors.

We try not to use pins to hold things in place when its own weight will work. In the past, we used to have an attachment system that snapped on to the front, as you can see in the first picture. Heavy attachments would sag off the front and sometimes require pins to hold them in place. This took extra time in base to get setup and so we moved to a more efficient system.

Now we drop the whole thing on top of the robot. The weight of the attachment sits on top of the robot itself, so you don't need to worry about anything sagging off the front of your robot or dragging down onto the mat and rubbing as you drive. You do still have to keep the attachment from sliding around on top, but that is a lot easier to handle with beams and gears.

You can see from this second picture that we built dog gears and attached them to medium motors to transfer rotation to our attachments. You could also use gears but sometimes that creates a really tight connection that makes it harder to drop the attachment on fully.

Finally, you just have to plan out where to mount the motors and how to get the gears from these motors to connect to the gears from your attachment. Really think this through. Having a good plan can save you a ton of time throughout the season as you make the starting point easier for each of your attachment builds.

Castor Balls & Color Sensors

We talked about the top of the robot in the drop on attachment section, now here's some ideas for the bottom of the robot.

If your robot has 2 driving wheels, then you need at least l more point of contact with the ground to be stable. You could have more than that, but this adds friction. Friction between your drive wheels and the mat is good to help you steer and control your robot. Friction between your castor ball and the mat is bad because it causes forces that resist your robot's motion. It may just be a little resistance that you can overcome, but it could be irregular and cause your robot to turn or wiggle when you don't want it to. So minimizing this friction is idea, this is why we usually have 1 castor ball, and we use the plastic one instead of the metal ones from EV3. Both of these are better than wheels as that really has a hard time turning.

You may also notice that we have mounted 2 color sensors in this last picture. If you use color sensors for detecting things on the mat, we highly recommend shielding them from light. This will increase the consistency from mat to mat. Sometimes your practice field may have different lights than the competition field and that could make a difference, so we try to keep it dark underneath the robot in the area right around the color sensor.

LESSONS

Advanced Robot Design

This was the video chat we held to help FLL teams get ready for the MN state tournament in 2026. [VIEW SLIDES PDF] We discussed how we think about:

Mission Strategy
Robot Design
Attachments
Coding
Testing

Page updated

Google Sites

Report abuse