Current Bots – Someone Else’s Problem

Someone Else’s Problem

My antweight robot, Someone Else’s Problem, has gone through some major revisions over the last few years.  Although it has had a somewhat rough road so far, I like to think that it has slowly been inching toward greatness.  I have learned from my past mistakes, learned to improve my design, and learned how to make the best of the tools I have available.  This is the story of that journey.

The story truly begins with the first weaponized bot that I built called Angry Erector Set.  I built it from Actobotics components, literally just attaching two aluminum channels and stuffing motors and electronics inside.  It was fast to build and fun to drive, but it was never very effective in competition.  Due to some poorly mounted drive motors the wheels and front half of the gear box were liable to fall off after a bit of abuse, and directly mounting the blade to the motor lead to a variety of issues that didn’t make for a reliable weapon system.

Angry Erector Set, the spiritual predecessor of Someone Else’s Problem

After I decided to retire Angry Erector Set, its two wheeled undercutter spirit lived on in its replacement, the first version of Someone Else’s Problem.  After having issues keeping weight down with the aluminum channels, and after acquiring a decent 3D printer, the obvious choice was to try out 3D printing the chassis.  It started out with a design more reminiscent of PP3D, but pushing parts around on the screen led to the typical convergent evolution that underlies most “tomb-clones”. Borrowing the guard design from Jamison Go’s successful bot, DDT, rounded out the basic shape of the frame and added needed protection. 

Someone Else’s Problem v1.0

Because I always want to learn from my past design issues, I changed drive motors over to the more proven N20 gear motors and made sure to solidly mount the motors and protect the gearboxes in the chassis.  To address the issues that I experienced with the direct drive weapon blade, I switched over to a round belt pulley driven system.  This allowed me to use a bigger weapon axle that was less likely to bend and served to prevent the shocks of the weapon impacts from damaging the motor or its mounting. 

The concept for the weapon shaft setup is also borrowed from DDT. A shoulder bolt acts as a dead shaft for the weapon to spin on, with the bolt going through the blade and frame and the bolt head riding on the ground. Bearings in the blade and attached pulley allow the blade to spin, while small spacers above and below the bearings allow the bolt to be tightened to lock the inner races together. Finally, flanged bushings pressed into the frame help to spread impact loads over a larger area to avoid enlarging the shaft hole, as well as helping to clamp the layers together to avoid failure by delamination.

Weapon shaft setup

The bot’s first outing at Franklin Institute 2017 didn’t pit it against any seriously destructive opponents to really challenge the durability of the frame, but it did reveal some design flaws that repeatedly disabled the weapon.  First, the blade was too close to the motor pulley and was flexible enough that a good hit could deflect it to bite into its own pulley.  Second, the impact of the blade on the pulley was enough to break the motor mount at the layer lines where it stood higher than the surrounding frame.  This raised the motor slightly, so the pulley with its low clearance was then rubbing on the frame and ground the whole weapon assembly to a halt.

The next version of the bot (v1.1) didn’t see many significant changes.  Because the weapon motor mounting had been the weakest link, I reinforced it and added a screw through the entire frame to make it impossible to break at the layer lines again.  I also increased the clearance from the blade to the pulley and from the pulley to the frame to try to avoid the same issues with parts touching that shouldn’t.  And because I had some weight left over, I decided to upgrade to a bigger and much stronger weapon motor.  This was a mistake. 

SEP v1.1 – Incremental updates to include a larger weapon motor and reinforce its mounting

The bot next saw action at Motorama 2018, and while the results were entertaining to watch, things didn’t go to plan.  The good news was that my improvements to the weapon largely worked as intended.  The weapon motor stayed where it was, and the improved clearances meant that the blade never bit into the pulley hard enough to stop everything.  The bad news was that that the new weapon motor proved to be excessively powerful. 

The powerful motor was able to spin my blade up to top speed almost instantaneously.  While this may sound like a good thing, Newton’s third law means that when the blade accelerated in one direction the rest of the bot accelerated in the opposite direction.  This meant that every time I turned the weapon on the bot spun out of control, and quite frequently the bot would get tilted enough for the blade to strike the floor and spontaneously launch itself across the box. 

While it can be fun to watch a bot repeatedly fling itself across the arena with a bang, it doesn’t make for a particularly effective winning strategy, and it can be quite hard on the frame of the bot.  This became obvious when the front fell off.  The repeated blows of the blade against the box eventually proved to be too much for the frame to withstand, so the weapon axle liberated itself from the rest of the frame in protest. 

The front fell off. Self-inflicted damage at Motorama 2018

After the explosive showing at Motorama, it was time to redesign the bot from the ground up.  All of the internal components of the bot were changed out.  The drive was upgraded to silver spark motors with bearing blocks to achieve more durability.  The weapon motor was switched out for a much smaller motor with a higher gear ratio, so the weapon spin-up should be slower and more controlled. 

The wheel guards were extended all the way around the bot to better protect the back side.  The blade was upgraded to a custom single-tooth disk cut from AR400 steel, designed to store as much energy as possible, and with the bonus of always being over the drive pulley to make it impossible for the blade to bite into the pulley.

The frame got a major overhaul.  As I had found that infill didn’t significantly add to the strength of the frame I thinned most of the walls so that they contain no infill when printed, and the remaining areas were printed completely solid.  I made sure that everything could fit within ¾” of height so that the frame could be all one height, leaving no easy place for the frame to break at the layer lines.  Finally, I dove into my print settings and annealed the printed frame to try to maximize the strength.  The result is a frame that is significantly lighter but should be more durable than my previous frames. 

SEP v2.0 – Improved design with new everything, now with less fail! (hopefully)

Someone Else’s Problem v2.0 was all ready to fight! Then life happened and it sat on a shelf for two years with no competition to go to. Just because I couldn’t make it to a competition didn’t mean that the gears in my head stopped turning. I released a version of the design on Thingiverse under the name Backlash, and kept thinking about ways to improve the design. The result is that v2.0 will likely never see a competition because I have now moved on to v3.0.

SEP v3.0 – Titanium is the new black

The v3.0 design keeps all of the internal components and the overall shape from the v2.0 design, but replaces the printed frame with a titanium frame. The frame is made with two 1mm waterjet cut titanium plates attached to standoffs embedded in the printed vertical walls, resulting in a design that looks like DDT and AbsurdiTi had a baby. This should hopefully result in a very sturdy frame that is impossible for the weapon shaft to rip itself out. Unfortunately, NERC’s Franklin Institute event has dropped the antweight class for the moment due to the massive influx of beetleweights, so the bot sits on a shelf waiting for its time to shine.

3 thoughts on “Current Bots – Someone Else’s Problem

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create your website with WordPress.com
Get started
%d bloggers like this: