Squash some Ants and Beetles CAD


The little guys are fun to build. Relatively cheap to do so as well. This will be my first post about the design and construction of a robot. And luck you, you get a twofa! I’ve been working on an update for AmbiSinister in the hopes of dealing with the pesky clumsy ass bot issue, damn thing living up to it’s name. Beyond that, I’ve begun CAD work on a beetleweight, a knock off of Hazard I’ve named Ordnance. Here’s hoping I can get both done prior to the next MassDestruction event and have them up and running. I’ve also got a 12 lb sportsman idea in my head and may begin CAD work on that feind soon as well, but that’ll have to be a later posting. 

 

So which do we start with? The next iteration of AmbiSinister. Or the next two. Jeeze, hopefully this doesn’t become a theme of this post and everything needs two. That’ll be quite a bit of typing. And I have a…. ok, I don’t have a social life. So off we go! The first, next iteration of AmbiSinister is to create a better spinner that can stay on it’s wheels. The older versions, if you’ve read AmbiSinister: The Early Years, you’ve already seen that it’s had massive issues with stability. I’ve been told it’s at the very least partially due to creating a very light carrying structure and a heavy bottom ring to actually hit with. What I thought was a great idea is actually dynamically unstable because as it hits or otherwise begins to oscillate vertically, it’s not balanced top and bottom. So one there’s not an equal force being applied top and bottom. At least that’s what I took away from the explanation. So, I began with trying to figure out what kind of shell I could produce without precision tooling of any kind at the house. I came upon what I thought was a pretty sweet idea. I would have someone, probably Shapeways, print a matrix of arms for me. This would splay out from the center, contain the bearings, and attach to an aluminum ring. This ring would be the entire height of the bot. Around the plastic arms would be steel teeth with bolts running through all three materials. This should mean that when it hits forces will be pushed into the aluminum and hopefully spread more evenly amongst the plastic arms and let them survive. The steel teeth should protect from direct hits. And this benefits by having very minimal weight difference top and bottom. I decided on three teeth as well to try and minimize bite a bit so that the individual hits would be smaller and be less likely to crack the plastic. 

And there it is. The TriShell in all it’s glory. Now I did have the arms 3d printed in Onyx from a local place and had the teeth waterjet cut and welded together by a friend. Also bought the aluminum tubing to make the ring. But in the end, I don’t think I’ll be actually using this design. I don’t know if I can trust the arms, as they still seem to flimsy and I can’t spare them much more weight. The aluminum ring creates a heavy burden for a tiny antweight. But I do think this design has potential and if I was willing to keep at it, I could produce something that would work very well. But I’ve made a different decision that we’ll get to later in this post. Bear with me.

The other thing that I knew about AmbiSinister was that once it flipped, it was flopped. No coming back. The very first version did self right if I just kept the throttle pinned. I rewatched some of the initial videos and think I figured out how. The weight on that design was literally all on one side. There was a L bracket that the motors and battery mounted to with a hole for the center dead shaft. Then there was another tab forward with the weapon motor. So no weight was opposite these things. I think that strict bias made the bot have enough of an offset vibration to come back to it’s wheels if the motor could spin the bot inside the shell. With this next iteration I decided to give that another go and see if it was the solution. 

The center shaft is still a dead shaft. The new shell not being a dome means I can more easily use a pulley drive system for the shell. The batteries are 2 2S Thunderpower batteries that will be run in series for a total of 4S power. Insanity proceeds. I like it. It’s also the Pololu micro gear motors and the little bay in front for the electronics. A top plate comes over and secures with four strips of short nut strip. You can see one on the right side and behind both drive motors. The very front with the circle is a small rounded head bolt to ride on the floor and stabilize AmbiSinister as it moves. The weapon motor is out the back. The drive motors are out the back. Half of each battery is at the center line or towards the back. This was the weight distribution I was looking for. Of course a flat topped shell meant I’d have to run a long dead shaft so as to not just be turtled every time I ended up upside down anyways. Might be a moot design point. But I’m happy with it.

A side view and a front view. You can see the front bolt there to support the bot and keep it level. And the pulley system along with the lower bearing that would be bolted to the shell. Four bolts would sandwich the shell and pulley between two bearings and hopefully let it spin freely and viciously. You can also see the holes for the nut strip to attach the top of the frame.

CAD

Here’s the full CAD minus the shell of AmbiSinister

But, this version most likely won’t see the light of day. The hours I poured into it leading up to MassDestruction 6: Watch City Wreckage and consequently MassDestruction Se7en once I wasn’t able to get it ready for either event had me itching to talk to builders about it. I wasn’t able to really stick around for 6 at all, stopped in to say hello and watch a couple matches before having to head back home for other reasons. But 7 I was there all day and actually helped judge! It was a blast and I was seated next to Charles Guan of Team Test Bot who was my compatriot in judging the bots. He gave out fantastic names for the bots when he couldn’t remember what their names actually were. It was glorious I tell you! But, in some of the lulls, I began to talk to him a bit about the design. What I was looking to do. And, whilst we were talking I was even luckier because Jamison Go of Team G3 Robotics was there streaming the event while he competed in it. He did an excellent show, but it was my great fortune to have two of the leading knowledge powerhouses about 3d Printing right beside me. We spoke a bit about the shell and they both had tips for me. Then, magic words left Charles’ mouth. No, I’m not changing how that sounds. It stays. He mentioned that it’s just too bad instead of a plastic arm system to just use a ring spinner system.

I wasted a weekend then on this. Why? Well, quite frankly because it made sense. I kept looking for any and every way of lightening the central support structure on a shell so that I could have weight for motors and what not. In the process, I was making compromises I wasn’t fully satisfied with. Would they hold up? With proper designing, quite likely. I’m no engineer and don’t have a great way of iterating on the shell with printing more. I have no printer myself. But this way eliminated that pesky structure I wanted nothing to do with anyways. What’s the point in not migrating the design over?

So I will. 99% because I really, really, Really, REALLY REALLY REALLY wanna hear this thing whine. I will upload the video of its first spin up as NSFW. Everyone will love the sound it makes. Why?

Gear Driven. The above picture is a version I was still working with, but it clearly shows the ring that holds the teeth is an aluminum ring gear. It’s 16 pitch and I plan on having it waterjet cut along with the inner gear that fits outside the weapon motor. It’ll be a friction fit on the motor can. But this thing will howl at full speed. a 4S battery with the 2150KV means a motor RPM of 31,820. Ha. HAHA! Yes. The smaller circles are rollers and the full CAD above shows the 3D printed runners that will attach to the top and bottom of the ring to give an angle for the runners to hold against. Four rollers in the final version means four through bolts literally contain all the insanity. The rollers have needle roller bearings inside that let them spin on the bolts. The teeth then sit against flat spots on the ring and have a bolt that holds them on. Might not be the best method, but I want to try it out before sticking with just a bolt head sticking out. 

I have not made any effort currently in getting the materials together for this bot as of yet. But I have run weight and I’ve got some good wiggle room of about an ounce. And I have a list of where to get every part so whatever I don’t currently have, I’ll be picking up in the near future and getting this ready for the 8th MassD. Once I begin to build, there will be more pictures and reasoning and delving into the design.

 

ORDNANCE

 

The second bot! Yes, I’ve also been spending more time at the computer CADing up a Beetleweight. I was glad to purchase blades off of Kyle Singer. I bought one of the 7″ Bull Horned blades, one of the 10″ two tooth and two of the 10″ Single tooth. My aim was to take the 10″ blades and make a Hazard clone. So I had to pancake another design and try to get it as tiny as possible. This time though, I needed to fit a 3 lb bot in a tiny space. 

Whelp, it’s something! I wanted to keep the blade under 1.5″ so that I felt I had a good shot at hitting the other robots. Undercutters could still pose a problem if they can get under that height, but hopefully the front wedges would be strong enough to take a hit or two and perhaps direct them up into my spinning blade after they’ve slowed a bit. Then again, I might end up mangled. That’s the beauty of this sport!

The front wheels are not powered, so I doubt I’ll win any shoving matches. But if it gets that far, I’m hoping to have mutilated my opponent or else I’m dead in the water as is. This shaft is again a dead shaft with the bearing being inserted into the hub of the blade. The frame walls here are 3mm aluminum, along with the front wedgelets. This may change as I’ve only got about 2 ounces of wiggle room and that makes me nervous on a 48 ounce bot. I’d like 4 or so for miscellaneous little bolts and wiring. 

What components are in it? A multistar weapon motor, KitBots 22MM drive motors, and a 5S 800Mah battery. That’s the other reason I’d like to step down in the weight of the framework, I’d love to run a second battery. I think a single is pushing the limits of the batteries. The space is there, the whole top bay is the same size as the lower and was initially designed with two batteries to be used. The weapon will be friction driven from a 1.5″ wheel attached to the top of the weapon motor that rubs against the 2″ blade hub. Gives a little reduction and isolates the weapon motor from shock a bit. Also keeps things low.

The rear has 3d Printed guards behind the drive motors that secure the outer rails to the inner rails. And then the little half moon guy in the very back to give better mounting points for the top and bottom plates. The bottom plate does mount the weapon motor, but they will be Garolite. I’m thinking 1/16″, but not sure that’ll be quite enough to get the job done. I may very well go for a thicker base plate. The top plate can be duct tape for all I care because if something is hitting it I’m screwed anyways!

The plates are all waterjet cuttable, and the wedge could be made easily from some C channel and a couple templates. Then it just bolts together with the Garolite. Ease of construction is certainly a fantastic thing for CAD to offer. Changes are relatively easily made in the program and you can prepare everything and have it ready to go once the final product is in production.

——————————————————————————————————————————————————

So there you have it. Both AmbiSinister and Ordnance have their CAD’s pretty  much wrapped up and they’re ready to get to building. Again, I’m most likely going to hold off another couple weeks on buying things to make sure my bank account gets plenty big enough before I splurge for all this again and drain it right back out. And once I get to building, I’ll take more pictures to give a more detailed diary on how it’s all going together and what my reasoning was. And soon enough MassD8 will occur and we’ll get to toss them in the Cochran Combat Corral and see if they can hold their own. They make such pretty renders…

There are no tags associeted with this post.