Lego Racer

This week we stared exploring mechanisms, and in particular exploring drive trains and the speed torque trade off using Legos, to build a cart that would win a race to carry a 1 kg weigh as fast as possible on a 4 meter course on carpet
a written and visual summary and reflection of your design process.
We were given these  building materials:
o Pico cricket & cable (to power the motor)
o Motor board & cable (to control the motor)
o Motor (the old grey motor style, without internal gearing)
o Gears (40-tooth, 24- tooth, 16-tooth, and 8-tooth)
o Wheels: (3 different pairs, one pair of large hubbed with thin tracks, one medium pair with thick grabby tracks, and one small pair with smooth )
o Lego parts: apart from gears and wheels, for vehicle construction.
o 1.0 kg weight

We knew that because the old grey motor did not have internal gearing, that it had a lot of speed but not a lot of torque.
Torque is a measure of the ability to cause an object to spin. Speed is angular speed.
Speed and torque are inversely related to each other.  In fact power = torque*angular velocity , where angular velocity is the angular speed but with a direction.
Rotating large gear using a small gear lowers the speed of rotation by a factor equal to the ratio of the number of teeth on the small gear over the the number of teeth on the large gear, but increases the torque by a factor equal to the number of teeth on the large gear over the the number of teeth in the small gear.
Because the old grey motor did not have the sufficient torque to carry a the cart with the 1 kg weight without stalling form lack of torque, we wanted to create a drive train that would increase the torque.  However, since increasing torque would come at the expense of decreasing speed, we wanted just enough torque for the cart to travel without stalling, so that we would still have the highest speed to win the race. We also knew that the lighter the cart was, the less torque was needed to move the wheel to move the cart, and the faster we could go.

Thus the main thing was to determine the correct gear reduction.  A technique that we thought of was to after buiding a gear train, we placed the 1kg weight on the last gear and see if that weight stalled the gear train.  But the professor suggested that we build the full car and test the drive train driving the whole car which would test for the weight of the car, the weight as well as the significant resistive  force of friction of the heels trying to move on carpeted floors.

Our first try at a drive train had a gear reduction of 1:30.

Since we were told to test it out in real life, with wheels and weight and everything, we had to decide what what type of wheel to use.  As mentioned earlier there were three sets of wheels, we found that the smallest wheel had treads that tended to slip with regard to the axle, which meant that sometimes, although the axel spun, the treads did not spin with it and the spinning axle just spun in place without carrying the car along.  We knew that we should not use this set of wheel for the driving set of wheels.   We thought that if we used the big wheel, that because of the larger radius that we would have a bigger velocity with the same amount of angular velocity.  However, we noticed that the biggest wheel is rather heavy and we thought that would make it so that it has a bigger moment of inertia, requiring greater torque to move, and would increase the weight of the car.  Later on we also saw that the large size of the wheel interfered with our attempts at gear trains.  Thus we settled on using the medium wheel as a the driving wheel that would be attached directly to the drive train.

We tried all sorts of combinations, and we found that using the combination of two small and one middle sized wheel(attached to the gear train) was the fastest combination of wheels. with the worse combination being two middle and one small.


When we tried the first iteration, because the motor would not connect to the first gear of the drive train very well, we used a chain to connect the motor to the drive train.  First of all, the chain connection wasn't very tight, and I found out through later redesigns of the drive train, that it was causing a lot of internal friction in the drive train that reduced the power, reducing both the torque and the speed.
This image shows the drive chain in the fully built cart with small wheels.

Since the cart was going moving well, but rather slowly we tried to reduce the gear ratio to give it a little more speed, since the torque was obviously sufficient.

so we tried a different gear train:

As you cans see, we do not have a chain, which decreases the friction in the drive train.  Additionally we decreased the gear ratio and the number of gears used, which also reduces friction.
Whereas our first iteration had a gear ratio of 1:30, this itieration's gear ratio was 1:25.
Consequently it runs much faster.  The fastest that this iteration runs the track was about 12 seconds, compared to 17 seconds for the first iteration.
This is a picture of how everything fits together.

We tried to see if we could lower the torque yet further to increase the speed.  I tried a 1:15 gear ratio, but that produced a torque so small that the cart often stalled and didn't move.
During Xixi's office hours, my partner Jiaming came up with a revolutionary redesign of the drive train that used more Legos to utilize a 40 wheeled gears at a level above the chassis to produced a diver train with a gear reduction of 1:16.67 which was barely enough to move the car while theoretically producing the most speed.

This shows the elevated motor that enables the motor to directly drive a 40 tooth gear.

 Despite the drastic change in gear reduction from 1:25 to 1:16.67, the third iteration of the cart didn't actually go much faster than the previous iteration.  The new iteration takes about 11.25 seconds, as opposed to the 11.7 seconds of our previous reduction.  We think that this may be because the third iteration required more gears and more legos to construct than the second iteration which may have negated the additional gains we might have expected from the change in gear ratios.

Other notes: It was remarkable how flexible the lego designs were building drive trains.  the 8, 24 and 40 tooth could be used all in one plane with the regular holes provided.  the 16 tooth could not be used in the same plane with the regular provided holes in the lego bricks, bit it could be used to transfer to a different plane of lego bricks.  Getting the right gear reduction with the right alignments was difficult, but I imagine it could have been a lot harder.

Therefore we ended up creating two iterations that were about the same speed as each other.  


When we raced our little carts, I was fascinated that the winning cart was built essentially upside down from all of the other carts, with the lego nobs facing downwards rather than upwards, which allowed the cart to use a particularly unique drive train, that would have been impossible in the normal direction. Sometimes, all it take is a simple paradigm shift that lego knobs can face down as well as up to come up with the optimal design.