When building your robot, try to decide how big you want it, but don't think so much about exact size; think more generally. Imagine the desired size as a box; the goal is to keep your robot small enough to fit within that box. With LEGO robots, it is very easy to get carried away with their size. Adding new parts is easy, so things can quickly get out of control.

Remember, even though LEGO parts are relatively lightweight, their weight will add up quickly, and soon, you'll find that you have built something that is going to require a lot of motor power to move around the game field.

When I speak of power, I'm talking about the strength of your robot. Some robots need to push heavy things or even pull some objects. If this is the case, your robot is going to be required to be very strong. Even though we're dealing with LEGO robots, if you gear one correctly, they can produce a great deal of torque, thus giving your robot a lot of power. You do need to be careful, because if you overdo the power, you will break some pieces. I've seen many LEGO gears and axles twisted and snapped from being having too much torque applied.

No one wants a slow robot; we all love build things that go fast. In FLL, you only have 2.5 minutes to complete as many missions as you can, so speed is important. But remember: in gaining speed, the effectiveness of other design principles, such as power and size, will be sacrificed. It's not easy for a big robot to go fast. Also with speed, the risk of mistakes can increase; it's easy for a speedy robot to miss goals. Many times, it's hard for a fast robot to be accurate, so it's a good idea to start out slowly and then increase the speed as you practice your missions.

Even though I didn't list batteries as one of the design aspects, they do have an effect on all three of the principles—size, power, and speed. Obviously, the faster your robot runs, the more battery power it will consume. The same goes with motor power; producing more torque means consuming more battery power as well. These are things you need to keep in mind with your robot design. With LEGO MINDSTORMS robots, you only have a few battery choices, and only two have an effect on the actual size of your robot design.

You can use replaceable or rechargeable AA batteries. If you choose to do this, you will be replacing batteries quite often during practices and competitions. Keep this in mind for your chassis design, and give yourself easy access to the batteries.

Even though the FLL game is only 2.5 minutes long, you will want fresh batteries for each run to ensure consistent performances from your robot. Also, be sure to always use the same type of batteries, running on Alkaline batteries at the beginning of a season and then switching to Lithium batteries will cause big differences in your robot's performance, most likely unexpected and undesired behaviors. So be consistent with your battery choice.

The easiest thing to do for batteries is to use the LEGO MINDSTORMS rechargeable battery pack. This is included with the FLL robot kit and can be purchased separately from LEGO Education as well. Even though the rechargeable battery packed actually runs at a slightly lower voltage than AA batteries, it will produce a more consistent power level for a longer period of time. Remember, consistency is a good thing for LEGO robots.

Be sure to keep the size and location of the recharger port on the battery pack in mind when building your robot. The battery pack will add about a quarter inch to the thickness of your NXT brick, and this could affect the design if you were originally using the changeable AA batteries. The recharger port needs to located in such a way that you can plug in the charger with out having to remove the battery or the NXT from the robot. You'll also want to be able to see the charger indicator lights on the battery pack so that you know you're connected and getting a good charge. Nothing is worse than leaving the robot charging to only find out later that that plug wasn't connected properly and it didn't charge at all.

Spur, or straight-cut, gears (see Figure 2-4) are the simplest of gears; they are what most people envision when you talk about gears. When you are simply trying to transfer power from one location to another along a straight line, these are the types of gears that you would most likely use. The teeth are straight and aligned parallel to the axis of motion. To work correctly spur gears must be combined with other gears on parallel axles.

Figure 2.4. 24-tooth spur gears inline together

Crown gears (see Figure 2-5) have teeth that are raised on the edges, giving them a crown-like appearance. They are normally used when axles are meeting at a right angle, but they can also be used in the same manner as a spur gear if needed.

Figure 2.5. A crown gear meeting a spur gear at a 90-degree angle

There are two types of bevel gears: straight-tooth and spiral-tooth. LEGO only makes straight-tooth bevel gears (see Figure 2-6). Bevel gears have a slightly conical shape and are useful when axles meet at a right angle in a small space. The bevel gears are much smaller than crown gears and work well in tight situations; they also cause less friction that a crown gear. Angles other than 90 degrees are possible, but using bevel gears at those angles runs the risk of slippage. Also bevel gears, unlike crown gears, can only be used with other bevel gears.

Figure 2.6. Bevel gears work well in tight areas but can only be paried up with other bevel gears.

The double bevel gear (see Figure 2-7) is a great mixture between the spur gear and the bevel gear. The double bevel comes in various sizes and allows for a smooth meshing of teeth at various angles. It can be used in parallel and in various angles, and the double bevel can be used with various other gears as well.

In today's NXT robots, double bevel gears are the most popular. Not only are the gears versatile but they tend to mesh together better than the traditional spur gears, giving a smoother motion to your robot and less friction between the gears.

Figure 2.7. Pair of Double Bevel gears meshing at 90 degrees.

Worm gears are similar to a screw (see Figure 2-8); they have threading that runs along the outside of a cylinder. When meshed with a spur gear at a right angle, the worm gear can create a very high gear ratio. The worm gear will move one rotation for each tooth on the connected spur gear; this is an n:1 (n-to-1) reduction. A worm gear meshed with a 24-tooth gear will produce a 24:1 reduction. Nothing can beat the shear strength and size of a worm/spur gear combination for power, but be careful because the torque can be too much for some LEGO pieces and cause separation or actually break a piece.

It's important to note that the worm gear must be used as the input axle, since the output axle connected to a worm gear cannot turn the input axle. Basically, a worm gear can turn another gear, but another gear cannot turn a worm gear. This one-way relationship can work to your advantage for locking and holding things in place; it might not be so useful in chassis movement but is great for attachments.

Figure 2.8. A worm gear meshed with a 24-tooth spur gear in a technic gear box

Clutch gears (see Figure 2-9) are special because they are designed to allow the axle to rotate around the gear when the maximum allowable torque is applied to the gear. They contain an internal clutch that will slip and no longer move with the axle once the maximum force is applied. The LEGO 24-tooth clutch gear has "2.5-5 N.cm" printed on the front of it; this is the torque rating for the gear. The clutch gears are useful for saving motors and keeping your robot from breaking or tearing up itself when too much torque is applied to the gear. At this time, LEGO only produces one size of clutch gear.

Figure 2.9. A clutch gear meshed with a 24-tooth spur gear

Pulleys (see Figure 2-10) are unlike gears in that they do not have teeth but have grooves that hold belts in place. LEGO offers a variety of pulley and belt sizes. The ratio of the pulley sizes is similar to the ratios that we get when meshing gears together. Pulleys can be a great choice when trying to deliver power in unusual spaces that gears don't necessarily fit. The drawback of using pulleys is that they will slip when high amount of torque is applied to them, but this behavior can be used to your advantage as an alternative to the clutch gear for creating limited slip mechanisms. Just be careful of relying on pulleys too much in an FLL robot; the belts are not the most dependable for staying in place. You'd hate to have one shoot off during an event.

Figure 2.10. Pully wheels set up with red band

The knob wheel is an unusual type of gear; it doesn't even look like a gear (see Figure 2-11). But if you look at it closely, you'll realize that it is just a simple four-tooth gear. Similar to the bevel gear, the knob wheel is good for 90-degree, low-speed connections and for applying high torque in angles without the risk of gear slippage.

Figure 2.11. Two knob wheels meshed at 90 degrees

Knowing all the different gears is great, but how do you connect them together? It's important to understand gear ratios. The gear ratio is how much the output axle turns based on the rotation of the input axle. For example, if you have an eight-tooth gear on your input axle and a 24-tooth gear on your output axle, for every complete rotation of the input axle, the output axle would have turned one-third of the distance. This would produce a 3:1 gear reduction ratio, as in Figure 2-12.

Figure 2.12. An eight-tooth input gear meshed with two- tooth output gear for a 3:1 gear reduction

You can read this 3:1 ratio as saying that for every three turns of the input, the output will make one turn. When slowing down the rotation with gears, we call this gearing down or gear reduction, and it can increase the torque generated by the input source, such as a motor.

You can reduce the gears more by meshing gears of even higher differences in teeth count, such as combining an eight-tooth gear with a 40-tooth gear to produce a 5:1 reduction. For even greater reductions, you can join pairs of gears together. Two sets of 3:1 gears would produce a 9:1 reduction, as shown in Figure 2-13. You can continue this concept to create extensive gear reduction to produce far more torque than you'd ever need for an FLL robot.

Figure 2.13. A pair of 3:1 gears joined together to produce a 9:1 gear reduction

As you can imagine, if you switched the input to the 24-tooth gear and the output to the eight-tooth gear, you would cause the gear ratio to switch as well. The new ratio would be 1:3; for every turn of the input axle, the output axle would turn three times. This is called gearing up and is a way to gain more speed from an input source, such as a motor.

When putting gears together, also note that an even number of gears will reverse the direction of the input, while an odd number of gears will keep the input traveling in the same direction, as shown in Figures 2-14 and 2-15.

Figure 2.14. Two spur gears meshed together will reverse the rotation direction.

Figure 2.15. Three spur gears meshed together will maintain the current rotation direction.

Knowing the circumference of a tire is important when considering which wheels and tires you want on your robot. The circumference will be the distance that your robot will travel after the wheel has made one rotation, or more simply, it's the distance around the circle of your wheel (see Figure 2-16). Knowing this information will be very important when we talk more about navigation in future chapters.

For now, all you have to know is that to calculate the circumference, you simply multiply the diameter of your tire by pi (roughly 3.14), so C = πd. To determine the diameter of your wheel, simply lay the wheel on a ruler and measure the distance across the wheel at the widest outside part.

Figure 2.16. The circumference of a wheel

When mounting wheels to your chassis, it's important to give your axles proper support. If your wheel is mounted close to the chassis, be sure to include a bushing so that the wheel does not get pressed against the chassis. At the same time, you do not want the wheel to be too far from the chassis unless it's supported correctly.

Most robot chassis use a cantilevered assembly to attach the wheels to the chassis. In Figure 2-17, you can see that a single bushing holds the wheel just the right distance to prevent it from rubbing the chassis and is only supporting the axle on one side. This creates friction on the axle as it is being pressed against the chassis. In Figure 2-18, the wheel has been placed further from the chassis with multiple bushings. Because the wheel is further away from the chassis more force is put on the axle and chassis, causing more friction between the two. Also, this setup is going to cause the wheel to camber and give us unpredictable results when trying to navigate. And over time, the axle itself will begin to bend, thus making the robot's performance even more unpredictable.

Figure 2.17. Single bushing spacing on wheel axle

Figure 2.18. Multiple bushing spacing on a wheel axle, moving the wheel further from the chassis causing more friction

The most dependable solution is to have a chassis that fully supports the wheel axle on both ends, giving maximum support the wheel and axle and removing a great deal of friction on the it; see Figure 2-19. We will cover this concept in more details in Chapter 3, and I'll explain the physics behind it. For now, you need to keep in mind that proper support of the axle and wheel are important your chassis design to ensure the reliability of your robot.

Figure 2.19. A wheel axle being supported on both ends, reducing friction on the axle

One of the simplest designs is the two-wheeled robot design shown in Figure 2-20. This consists of two wheels, each attached to its own motor, with some kind of skids or ball casters to help the robot keep its balance. Depending on the game field, these designs can actually do very well in FLL events. Many times, the simple designs perform the best.

The center of gravity can be a concern with two-wheeled robots just because the wheels tend to be close to the front, and once any attachments are connected, this can cause the robots to fall forward. So it's always a good idea to test and retest the design as you add new features to your robot chassis to ensure that you have not shifted the center of gravity off-center in the wheelbase.

Figure 2.20. A two-wheeled robot with skids in the rear for balance

A very common LEGO robot design is the three-wheeled robot, or tribot (see Figure 2-21). The three-wheeled robot is similar to the two-wheeled robot but has a caster wheel instead of skids; the robot will have two wheels, each driven by a separate motor, and a caster wheel in the rear. The passive caster wheel will turn and travel in the direction of the robot. The third wheel is just in place to give balance to the robot chassis and to reduce friction on the game surface caused by skids.

Figure 2.21. A three-wheeled robot with a caster wheel on rear

Four-wheeled robots are much more stable that a two- or three-wheeled ones. Having a wheel on all four corners of the chassis provides a large area for the center of gravity, thus making the robot very stable and able to avoid losing its balance. Using differential steering on four-wheel robots is a bit more difficult than with two- or three-wheeled robots, but with proper programming, that issue can be overcome. In Chapter 4, we will talk about this more.

There are two types of four-wheeled robots: two-wheel drive and four-wheel drive.

Two-wheel drive works very similar to other robot designs in the fact that two of the wheels are each driven by single motors; the other two wheels are passive and just tagging along for the ride to add stability for the robot. Steering can be a challenge with two-wheel drive, since one wheel on each side of the robot is going to be skidding while the robot turns. If these wheels have good traction on the mat, they can cause a lot of resistance during the turn. Since these passive wheels are not being powered, they really don't need to have a tire (see Figure 2-22); you can run the wheels without tires to allow them to skid freely on the mat when turning.

A four-wheel drive robot is a bit more work to build; each pair of wheels on the side of the robot needs to be driven by a single motor. In most cases, this is going to require some type of gearing setup. Four-wheel drive is useful when you need to have a strong robot or one that can climb steep inclines. A four-wheel drive robot will steer the same way other differential steering robots do, but instead of having just two turning, two wheels will turn in one direction and two in the other. There will be some skidding by the wheels, but turning all will minimize this. Also, the closer the wheels are to each other, the less resistance they will encounter when turning.

Figure 2.22. A four-wheeled robot with two-wheel drive

Six-wheeled robot chassis are great when designed correctly. The wheel selection in the LEGO MINDSTORMS FLL kit is a bit limiting for such designs, but if you have access to some of the huge variety of other LEGO wheels, a six-wheeled robot can be done quite nicely. There are really two types of basic design ideas for a six-wheeled robot.

The first design option is very similar to the four-wheeled robot but with an extra set of wheels either in all-wheel-drive mode or still just two-wheel drive. If you need power, six-wheel drive works well but is not always the best choice for FLL challenges. Steering can become a bit challenge as well as navigating tight spaces. So even though using six wheels makes for a cool design, this is often not the best choice.

Now, the other six-wheeled design involves middle wheels that are larger or set lower that the wheels on the ends; this is sometimes referred to as a West Coast chassis. The idea is that the center wheels are the drive wheels, so when the robot turns, the balance of the robot will shift as needed to either the front or the rear wheels. However, the amount of friction during the turn is limited, since the balance will fall mainly on the center wheels.

For this design to work properly, your robot chassis needs to be balanced over the two middle drive wheels. When you do this properly, you can have a very nimble robot chassis that can still be of a substantial size, as shown in Figure 2-23.

Figure 2.23. A six-wheeled robot

A tracked robot can have a similar wheelbase to a wheeled robot, but of course, it's using a set of LEGO rubber tracks. The tracks give the robot chassis a very stable stance and a naturally low center of gravity (see Figure 2-24).

A tracked robot chassis is very helpful when tackling obstacles that involve crossing over an opening or uneven surfaces. The 2009 Smart Moves challenge featured a dyno-meter that the robots could cross to access part of the field; the spinning nature of the dyno-meter caused many of the wheeled robots great headaches, while the tracked robots tended to do better at making the crossing.

Unfortunately, the drawbacks in using tracks tend to outweigh the advantages. Tracked robot chassis have a hard time traveling in a straight line and tend to hop on a smooth surface. Also, keeping tension on the tracks can prove to be challenging at times. Often, a third wheel inside the tracks will help relieve any issues with tension but will not help the tracks to remain predictable during missions.

I don't want to discourage a team from using tracks on its robot, so try them if you like the idea of using them and they meet the requirements of your mission task. As with everything, experiment and see what works best for your situation.

Figure 2.24. A tracked robot