Concept Sketches 1
Concept Sketches 2. Attempting to come up with a dual-motor drive train for different torque/speed options without the use of a gearbox. Ended up scrapping that idea.
Team brainstorming
First CAD Render
First CAD render exploded view. The bearings we are using are not pictured.
First bucket of chips coming up! Machining our first part, the main structural arm/shaft of our robot.
Pre-drilling out the bore on the arm/shaft prior to using a boring bar to bring it to size. For better scale, the drill bit pictured is of 1.5" diameter…
Ryan here likes his drills BIG…
Running finishing operations to bring dimensions within tolerance.
Attempting to entertain ourselves with Lathe Animation while the powerfeed of the manual lathe takes its pass… for the 30th time…
Before and After. This piece is turned down from a single piece of 7075 Aluminum, the original stock measuring in at 5" in diameter and about 8" long.
An extremely expensive part considering the amount of material removal, but as a primary load bearing part, it was necessary to build the flange into the shaft to give it as much strength as possible. Two of these parts were made in total.
Second part! The wheel core of our robot. Starting size with Kevin’s head for scale.
This poor clapped out old lathe probably hasn’t been put under this kind of workload for decades…
The inner bores are meant to have bearings pressed into them, and have been turned down and measured with a bore indicator in order to be exactly .0002" undersized from the nominal bearing OD.
Final wheel core part size, with two goons' heads for scale.
Cutting off the excess stock prior to running the final facing operation on the backside of the part.
Before & After. Each wheel core is also machined from a single 7075 aluminum stock, its starting size at 8" diameter and 12" in length. All diameter tolerances of this part has been kept to +/- .0002"
With about a total of 30+ work hours sunken into a set of both of these parts, thought we deserved to show off some of our machining vanity~
Arm/Shaft & Wheel Core. Top View
Arm/Shaft & Wheel Core. Bottom View
Arm/Shaft & Wheel Core. Side View
Next Up: The Wheel End-Caps. These end caps are made of 4340 steel, and have been cutout to shape using an abrasive waterjet, and reamed at the center hole to be indexed together for further operations.
The endcaps indexed with a pin at the center, and clamped to the mini-mill to drill the remainder of the bolt clearance holes at the same time.
Keving setting up the boring head on the manual mill.
Boring out the weight saving cutouts. Although this feature could have been waterjet to size sine it isn’t a precision feature, we like to make our parts pretty, so decided to spend the hour to give it a nice surface finish.
Broaching the keyways into the endcap that’ll connect to the motor output shaft. Although not depicted, after the broaching process, the two encaps were aligned on the lathe, and their perimeter was turned down to the final part OD.
The steel tube on the far right was the next and last set of parts made on the manual lathe, which was unfortunately undocumented due to forgetfulness… Starting off as a giant tube of 4140 steel, this part was quite the challenge to avoid work hardening and reducing the amount of chatter introduced from the long boring bar we had to use to turn the ID down. The Tube will be used as our primary wheel, is slip fit around our aluminum wheel core.
Weight saving cutout slots being milled into the circumference of the wheel drums. This video is me being silly, fully aware that this endmill had chipped corners, and still attempting to mill through the 4140… enjoy.
The progression of the wheel drum before and after slotting, and test assembly with the wheel core and endcap.
Obligatory picture for our amazing bearing sponsors, Koyo and NSK. For the drive bearings, we decided to go with a pair of needle roller bearings and spherical roller bearings due to their high dynamic load capacity in the direction indicated by our FEA results. The crazy looking bearing on the right provided by NSK is the spherical roller bearing, and is typically used in many mining industry applications, where high impulsive loads are experienced.
Custom gearbox housings being machined on the Haas TM2P
Close up of gearbox housing after one of the operations.
Cutting off excess stock to perform the back side operations on the gearbox housing.
Custom softjaws made to hold these gearbox housing parts for the backside operation. The work coordinate system of the part has been probed off using the center bore of the part made in the first operation, to guarantee concentricity of features
Meanwhile… our controls team assembling a test platform to implement the Melty Brain drive system, which theoretically allows our robot to spin, while translating at the same time.
Abraham and Andrew making electronics do their things with all the electrical pixies and 1s and 0s.
Mid-year evaluations on what still needs to be done to be competition ready… with guest star Kelsey Sakamoto. Wait… do I read… HEAT SHRINK BEARINGS?!?!?
Attempting to heatshrink the needle roller bearings into the wheel core, using kitchen appliances… The aluminum wheel core is heated to expand, while the bearings are chilled in the freezer to contract. Theoretically, this should allow them to slip into each other if done quickly, and should be a tight shrink fit when the temperatures normalize.
What should I have for dinner today… Oooh i almost forgot my daily dose of aluminum!
It fits, it sits, and all it required was some oven mitts.
Now for the real challenge… Making Andrew let go of our giant slab of 7075 so it can be machined into the two body halves for our robot…
Shout out to Speedy Metals for sponsoring us with materials donations!
Yet another poor machine abused past what it’s meant to handle… with a girth of 6"x10", this badnsaw took a solid hour to cut through the block…
Size matters. A big job calls for a big tool, like these brand spankin new 1" diameter carbide endmills.
Post facing, drilling, and the first layer of material milled off.
All machining operations are being performed at full depth, and the Haas TM2p only complains occasionally about it. What a trooper.
Being babysitted by my own mentee so I don’t fall asleep at the controls while running this part at 2AM in the morning… Thanks again Ryan.
Pocekting out the center of the body half that will house all our electronics. As one might be able to see, NO we don’t have an automatic chip conveyor…
All the operations for one side of one of the halves done! Setting up the other stock to begin the mirrored second body half. Hopefully those jig saw features will line up…
AND THEY DO! Had to run a a couple extra finishing ops to shave off a hair of material to keep the jigsaws of the two body halves from binding.
Flipside operation done as well, and it’s perfect!
ASSEMBLY TIME!
The Endcap, Wheel Drum, and the Wheel Core assembled to create the final wheel assembly.
As seen in the previous picture, the needle roller bearing is heat shrinked into the wheel core, and the spherical roller bearing is test fitted to ensure that it is a nice slip fit.
The arm/shafts are slid in from the inside of each body half, and bolted onto the body. Then the two halves are joined, aligning the jigsaw pattern, and bolted together from both sides.
Side view of assembly showing bolt patterns. Although unseen in the pictures, we can see that the motor and gearbox assembly has been assembled and inserted from the back side of the arm/shaft, with the output shaft sticking out.
The spherical roller bearings are then pressfit onto the bearing surface step of the body halves. This area will be the primary load bearing surface, and is designed to resist extremely high impulsive loads when the robot collides with another.
AND THE WHEELS SLIDE ON WITHOUT A HICCUP. The beauty of precision machining~
Oooooh shiny~
Packed and ready to go to competition!
Dangercone makes friends with Darwin at Robogames! Hopefully he doesn’t have to face him in the ring…
Codename: Danger Cone
For my last year on the Battlebots team for UCLA’s student chapter of the American Society of Mechanical Engineers (ASME), I wanted our team to build a robot that would push our designing and fabrication skills to the limit, while trying out something cool and never before done. The obvious caveat to that is the likelihood of us building a competitive robot was pretty low, BUT HEY, you never know until you try. While this project was an opportunity for me to have fun implementing the impractical design concepts I have, it also my last effort to pass on the torch. Everything I knew about design, machining, robotics, and etcetera, I wanted to leave it behind for the next generation of students to learn, so they too can puruse their wildest project goals, and continue to teach their peers as well.
When the team finally came to a design agreement, Dangercone was born. The premise behind Dangercone is that it is a 60lb (lightweight) combat robot, that is shapped somewhat like a pill, and has two wheels that spin in opposite directions to allow the entire robot to spin about it’s center. As the robot spins, its controls system (called the melty brain drive. Read more about it here! http://www.nothinglabs.com/openmelt/), pulses each motor in order to create a speed differential, allowing the robot to translate on the horizontal plane. TL;DR: It spins and moves like a top, crashing into anything in its way and hopefully destroying it!
Role |
Project Manager & Lead Machinist |
For |
UCLA ASME: Battlebots Team |
Date |
April 20th, 2017 |
Type |
Competition Team Project |