LEGO Technic - Caterpillar 226B Skid Steer Loader

For this project, I wanted to build a remote controlled, wheeled skid steer loader. I used the Caterpillar 226B as a reference, but the idea was to have a functional remote control model rather than an accurate scale model.

I decided on using the 62.4 tires with yellow rims and used that as the reference for the scale of the model. It quickly became clear that the space available for the functionality was quite small. 4 motors were needed: 2 for driving the left and right wheels, 1 for lifting the arms and 1 for tilting the bucket. It was going to be a challenge to fit everything in. With an overall width of 15 studs, the hull between the wheels needed to be 9 studs wide, with only 7 studs available for the mechanics inside.

I put together a compact drivetrain using 2 M motors. The reason for using M motors was that I wanted the drivetrain to be low enough so that the cabin could sit on top. That seemed to work alright, and I continued with the other two functions.
I wanted to use the white clutch gears for lifting the arms and tilting the bucket, to not stress the linear actuators. I learned quickly that this the clutch gear takes up a lot of space, and I did not have much space to spare. After trying several variations, I settled on having two L motors, mounted vertically, with the output at the bottom. The clutch gear was mounted directly on the motors, offset vertically by 1 space. They each drove a pair of 24 tooth gears which took the motion up to the linear actuator and the arm where it was needed. I was super happy at this point as it looked like everything was going to fit and work.

lego-technic-skid-steer-initial-layout-3

This is where the first rebuild of many started. On my first try, everything looked to be going fine, but I realized that everything was going to end up too high. So, first rebuild was to try and mount the L motors so that they were flush with the bottom. This worked, and I made a rough bucket, some arms and put it together. The two M motors in the hull drive the wheels. The two vertically mounted L motors are for the other two functions: The rear one drives the arms up and down, and the forward one controls the bucket tilt.

I moved on to the side panels. Initially, I wanted to only use technic panels and beams for the sides. However, I could not get any reasonable shape without making the whole thing wider than 15 studs. So, finally I ended up designing the panels with normal bricks. Even though this was quite a simple piece, I really liked how they came out, and they really remind me of the original Cat 226B.

I managed to cram all the functionality in, but the center of gravity does not work. Back to the drawing board...

Since the whole thing was so compact and relied on the L motors for a lot of the structure, it was not really possible to rebuild parts of it without taking apart pretty much the whole thing.

I rebuilt a lot of stuff to move the 2 L motors, battery box etc. forward by 1 stud. Immediately, I had clearance issues with the 24 tooth gears of the front L motor hitting the 24 tooth bevel gears on the rear axle. I could not find any spacing where the gears cleared each other. At this point, I had two choices: I could redesign the whole thing, or use 12 tooth gears on the axles, losing some gear reduction. I tried out several different ideas at this point, but I could not come up with anything that came even close to the compact efficiency of the original idea. Therefore, I went with the latter, putting 12 tooth gears on the axles and losing a lot of gear reduction.

This is how it looks after switching to using 12 tooth bevel gears on the axles. Notice how close the 3 24 tooth gears are to the axle.

After rebuilding the drivetrain with the 12 tooth gears, I put together the rest. Getting the geometry of the arms correct, having the linear actuators work properly etc took me quite a bit of time. I had assumed that the way I linked the back L motor to the linear actuators could be more or less copied for linking the front L motor to the arm for the bucket tilting. That was indeed possible, but the linear actuator took up a lot of that space when it extended to raise the arms. Also, there was a lot issues with interference with the gears transferring the motion up to the arms, and the arms themselves. After much iteration, I found a solution that worked, and was more or less structurally sound. Now, adding the battery box still caused it to be very rear heavy, but much less than before. Anticipating this, I had already ordered a AAA battery box and was waiting for it to arrive. As it would be half the weight, using that would hopefully cure the balance with this configuration.

In parallel, I was working on the cabin. Due to the arms, the width left for the cabin was 7 studs. This was probably a stud narrower than it should have been, but there was no other choice without making the whole thing wider, which I didn't want to do. I also wanted to have the whole cabin pivot up for maintenance, just like in the real thing. The first version of the cabin itself was straightforward. The difficulty was making it strong enough, as some of the connections on the back were quite small, making it fragile. Eventually, I was able to make it reasonably stable, given the geometry. I also went with a brick built bucket, both to have more weight up front and also a more realistic shape.

At this point, I was thinking that I was on the home stretch. However, as soon as I tried out the driving, I knew that the decision to switch out to 12 tooth bevel gears was not a good one. The model was far too jumpy and almost impossible to drive off without doing a wheelie. The rear heavy weight distribution, low gear reduction, and the digital control of the power functions system came together to make a wheelie machine. I investigated again whether it was possible to go back to the 24 tooth bevel gear, but there was simply no space. I tried to fit the 2 m motors in with more gear reduction, but I could not find a solution without using the 24 tooth gears, which did not fit into the 7 stud wide space in the hull. In the end, I decided to skip the white clutch gear and use the combination of 8 - 16 - 8 gears to get the same distance and gearing as the 24-24 combination for the L motor for the bucket tilt. This allowed me to go back to using the 24 tooth bevel gear for the drive, which worked well.

As the clutch was now removed, I added a simple linear clutch to the axles powering the bucket tilt to protect the small linear actuators.

At this point, I had the M motors for the drive oriented with the output shaft to the front, and cable out towards the back. I had selected this orientation, as the cable was barely long enough. After trying out some more cable positions, I was able to find a way for the cables to reach when the orientation was reversed. This meant that I could move the motors forward by 2 studs, which helped the center of gravity. I know this is a ridiculously small detail, but I needed every little such change I could get on this model, in order to get the center of gravity to work.

With the AAA battery box now in my hands, I assembled everything, including the first version of the cabin. It was still doing some wheelies, but not too badly. As the cabin was still missing some pieces at the front, I thought that it would be stable when the cabin was completed. I had used the 2x2 black panels on the sides of the cabin which of course were lighter than if I did them in bricks. I rebuilt the cabin several times, adding a few details and consciously placing more pieces thus weight at the front. I also filled in the empty space at the front of the chassis, below the cabin, with some beams, to place some more weight forward.

Even at this point, the balance was still not really ideal. The front wheels still lost contact with the ground when starting from a stop. Some of this was due to the digital nature of the PF IR system and the jumpy start. I could not think of anything else that I could optimize further to place more weight at the front, so I started looking at the back. I lightened the brick built side panels slighly, but it was a marginal gain at best. It did not look like I could save much weight at all on the rear mechanisms, so I left that alone not to weaken the whole thing structurally. The only other significant contributors to the weight at the back were the AAA battery box and the IR receivers. I was so desperate not to start from scratch, that I was literally thinking about every gram at this point. I realized that the IR receivers weighed 24 grams each. This is where I realized that I could replace the IR receivers with an Sbrick (only 12.9 grams) and hit two birds with one stone. This would allow me to save some weight at the back, as well as drive it more smoothly. Having the AAA battery box along with the Sbrick improved the balance enough to make the model drive acceptably.

Overall, I'm very happy with how this model turned out. The packaging of all of the functions while keeping the model looking realistic and working well was the biggest challenge, keeping the center of gravity forward enough to have a balanced model was the other. During the build, I considered several times whether I should scrap it and start over with a bigger scale, and at the end, I'm very happy that I didn't. If I were to tackle this again, I would probably consider widening it by 2 studs, which would make things a whole lot easier.

This was such an interesting challenge that I will probably try it again in the future, with a different design. I'm also thinking about building a tracked version, to have some company for this guy.

Technical details