A thought for redundancy with the tracks, if the crown on the drive wheels is replaced with a groove and an O ring or two, if the track comes off, the wheel will still have some grip.
You might’ve already done this, but I’d recommend using the “expo” setting on your transmitter joysticks for better low speed control. It’s a pretty nice feature for combat robotics, from my experience
Expo won't help. I use these N20/N30 motors on micro crawlers and the cheap tiny DC speed controllers are often inherently bad at low speed control. You get nothing before about 30% throttle even if you raise it super slowly. The reason for this is that the motor's coils act as an inductor and the motor controllers are set up for efficiency and quietness at high speed, which means they run a PWM frequency above hearing range (20kHz+). If you're well versed in analog electronics, you'll see that the inductance and resistance of the motor form an RL frequency filter, reducing the voltage across the armature at low motor speed. Adafruit released a whitepaper on this and it was a revelation to much of the hobby community, including me. So if you're building a crawler or robot and want low speed control but don't mind a slight lack of smoothness, a sub-audible PWM frequency like 20Hz is far superior. All the good quality crawler ESCs are set up to change the PWM frequency as well as duty cycle based on throttle. That way you get maximum power across the speed range.
9:19 That disc shaped robot has some really interesting physics going on. Self balancing single wheel that's jumping!? I doubt it standing on its edge is the intended configuration, but really cool nonetheless.
Hi, that one is my robot. It's a 'ringspinner', a ring shaped weapon spins fast around the outside of a normally stationary robot. The gyroscopic effects of this can be weird, and in that case it was flipped up with the inside spinning within the ring stalling in the ground.
@@jameskirby9981 how effective are spinners at that scale vs other ant weight bots? im curious about its ability to actually do damage in a weightclass where thats rare
@@teslacoil4335 Not James (but I was running one of the other spinners in that melee). Spinners in 150g bots can seem to do nothing, or deliver massive hits that throw/destroy the other bot (or themselves). It depends on how much "bite" they get on a hit, which depends on weapon diameter weapon speed, engagement speed, and probably several other factors. This is true in any weight class, but in 150g bots the power to weight ratio is crazy, but similarly the weapon speeds are probably faster than they need to be (because of the high kV values in small brushless motors) so they tend to "grind" on the other bot, up until they actually get good engagement.
@@slightlyevilrobotics4271 interesting, i was looking at making some custom brushless motors for a 150 gram bot, power to weight ratios arent something i considered, i was originally just going to get the rpms to something crazy but ill tone it down then, thanks for the info!
About the name: it also reminds me of some insects. But less a Spider, than an "Antlion" or a "Praying Mantis". Especially considering the name of the weight class, I think "Antlion" is a nice fit.
Crowned pullies are really interesting, first time I looked inside my bandsaw it was confusing but when tuning it seeing the bands self align was really cool.
I’ve been following the channel for about a year since I got into 3d printing and needed noob tutorials. But your attitude,presence and wealth of knowledge keep me coming back! My name suggestion is “Lord Forkwad” 🤘🏼🤘🏼
(10:08) "using design history, I can just go back and change the shell thickness". Yes, you _can_ do that, but I've found it a lot simpler to just declare a parameter and use it in the course of modeling the object. You can start by setting "shell_thickness" to (say) 3.0, and then when you model your shell just type in the parameter name as the value. If you want to change it later, you don't need to track down the history element to change its thickness: change the parameter directly and it will be applied in the entire design history (which can potentially use this value in a lot more than just one step!). It works in sketches too, in case this feature of the object was derived from sketch lines.
Straight up when I figured this out it started saving me so much time on designs it was crazy You can even have parameters that are functions of other parameters. For example: You have "wall_height" set to 6.0mm and a circle cut out in the middle that needs to have specific spacing from the top and bottom, you can have the size of the circle be the parameter "wall_height - 2.0mm" now the circle will always have 1.0mm between it and the top and bottom.
Good point. I use parameters for just about everything now in Fusion360. I just wish the parameter functionality worked better - global parameters, search by name, namespaces, and making the parameter window dockable would make it a better product IMHO.
@@Gormadt Yes, being able to use them in formulas is great, I use it all the time too. I think there's only _one_ place where I've had trouble using them, and that's in schemas when drawing an N-sided polygon, where N would be the parameter. For example if you start with an hexagon centered on a point and with a certain radius, and want to change that to be an octagon by updating the parameter from 6 to 8. I just couldn't get that to work, even with a "unit-less" value and all, I never managed to get F360 to accept the value. If there's a workaround I'd love to know!
4:00 Fusion has also removed more and more features, from paying customers even. Funnily enough, the simulation talked about is exactly what they removed recently. Now only usable through consumable cloud credits bought only in packs of 1500 usd or more. I use it myself, and it is more convenient than any other option I found, but its very troublesome a feature you might be paying to use today, might be taken from you without any reasonable warning the next. Mixed bag there.
The simulation shown in the video is a static stress simulation, which is still included in the standard Fusion 360 commercial license. Also, 'flex tokens' (formerly Cloud Credits) can be purchased in lots as small as 100 for $300 per lot of 100 tokens. Trust me, as a paying Fusion 360 subscriber, I am NOT happy that modal frequency, thermal steady state, and thermal stess simulations now require flex tokens, but I just wanted to clairfy that for paying subscribers, Fusion 360 static stress simulations do not require flex tokens even though static stress simulations will now be done in the cloud.
well, i once used fusion 360, then found out I couldnt export it as a normal 3d file.... And then got the free blender which is AMAZING (even though you still need a slicer)
Yeah, I won't be using any cloud software ever. If I stop paying for a piece of software I expect it to keep working. Of course, I also expect that a piece of software will at most be a one time fee. There are enough open source and completely free 3D design programs, and they're quite good. Although, if a program came along that was open source yet non-free, I'd consider buying it over ever using Fusion 360.
@@anon_y_mousse : If you pay for a Licence to Use something, I think it is fair enough that it stops working if you stop paying. If you purchase an item however, you should be entitled to expect it to keep working without further payment. I purchased an accounting package, and used it for twenty years with many many thousands of transactions entered into it. After an update, it stopped working, and left me with no usable way back. The only option was to purchase a new version, without even a loyalty discount! Sometimes the original provider is not responsible; I have a nice little Pressure-sensitive graphics tablet that no longer worked after a Windows update. Even more annoying was my genuine MicroSoft webcam that also stopped. There ain't no scruples in the computer industry! ☹👎
@@PiefacePete46 All software you use is licensed, even when it is open source. If I pay money to license something I expect that to be a one time fee. In other words, I would stop paying for it after I've paid once. No large corporations have loyalty to their customers. The closest you might get would be when they open source their drivers, but hardly anyone now or 20 years ago picked hardware for the drivers, even though they should.
It was great to see you and be your Ant's first fight, even if Scottland Forever is more of an art bot (I'll have my fight report out next week) Loving the new ant, the crowned pullies are genius! hopefully, I get to fight it in the future
Make an O-ring groove in your drive wheels that will preserve the crown profile when an O-ring is around the wheel. That way if you loose a belt you can still drive. It won't add much weight and give you redundancy.
the forks might have to be closer to the ground. maybe also redesign them to be sheathed in the bots front, since they're relatively flimsy and your only weapon. I'd suggest making the interfaces angled, so they don't have the tendency to get stuck while lowering them into their sheath+scooping position. oh and then I also wonder if you have enough weight allowance to add some metal spikes on the upper surface of the fork, to keep enemies from sliding off the fork too quickly... maybe just superglue sand on it. it would probably be better to have it grip only in one direction though, so they can slide on easily, but can't just slide off.
@@jermihalia9418 the front of the robot can have an angle too. but also the tips can protrude a bit, just having them completely exposed isn't too good for stability.
Was thinking something similar as far has the forks should be lower to the floor. Saying it's a plastic ant weight I would probably say he's fine keeping them out from the front of the bot since not many people design kinetic robots most are control saying it's only 150g
@@kylecarter7293 hmm that's true and I don't know if kinetic weapons are even allowed there... still I'd rather go safe than sorry, since sheathing doesn't cost much anyways.
@Maker's muse "Praying Mantis" is what you should name it. It even has sort of green arms, but you could paint or reprint them in a better color easily.
Ive been slowly checking on this channel over the last 3 years and watching the battle bots get better and better has been my favorite thing by far. I also watch the other videos despite never having 3d printed before in my life lol
one of the wonderful things about this weight class is how many hilariously easy+cheap to get materials actually work pretty dam well for structure in a bot. here's a great example which do you think has more structural integrity per weight as a side panel: a sheet of titanium or a panel of decent quality pvc foam? even the cardboard from a cereal box ends up being surprisingly viable and both of those materials have the benefit of not transferring shock to the rest of the structure.
Looks like a praying mantis, I'd call it "The Mantis". A big bite in a small package. I also wanted to add that I appreciate the fact that you took time to incorporate your sponsored message within the video itself rather than just a 40 second slot that kills the flow. Usually I tend to skip the ad portion and move onto the content, with the way you presented this I found myself watching through the entire video including the ad. Well done!
This series of videos is seriously helping build my list of plans for using my 1:10 scale UK robot wars arena. If only i had the will power to stop procrastinating and finish axtually building it...
I like the wristbands as treads idea. Probably way more grippy than even the softest TPE that I've seen(85a). That said, I will point out for anyone who is curious that you totally can actually print belts with ok grip with TPE 85a if you dont find the perfect already existing item. The softer, the more grippy usually.
Just want to say that I went to school for mechanical engineering but didn't get really interested in designing my own objects and mechanisms until I found your channel. Your appreciation for elegant contrivances is really infectious and inspiring. I would call the robot fiddleback because you seem to be a fellow musician and I agree that it looks like a spider's defensive pose
To help your sponsor, I've been using Fusion 360 for a pet project and I really, really like how easy it is to 3D print with. Specifically I can load a print straight into Prusaslicer without having to save as an stl. Very minor thing and very convenient!
That was honestly a really nice integration of an in-video ad - normally I hate them, but how smoothly you fit it in and made it relevant to what you were doing anyway made it fine. Not a jarring, off-topic, irrelevant intrusion as you so often see. Kudos! Also a really cool project. I love antweights!
Crowned pulleys are a genius bit of engineering, you'd think you want to make pulleys with the inverse curve like a rope pulley, but for a flat belt this causes the edges to "walk" up the slopes and come off the pulley!
The simple fact that you’ve always used it and that it’s really cool that you got them as a sponsor makes me consider (once again) to use Fusion 360. I learned to use a different program at school, which is why I don’t yet, but you do often manage to get me thinking of using yours in stead.
@@emanggitulah4319 I'll have to give that a shot. I don't think I've ever thought about watertightness with 3D prints, but I do have a copy of FreeCAD that I downloaded last year. Been meaning to check it out.
I would suggest when you go into the competition to have several extra charity bands to change them out if you are allowed to do so that way after each match you can switch them out because I'm sure the floors will leave a lot of debris on the bands which might loose traction
@@TeamPanicRobotics Yeah I know, I saw the video when it came out. Im still impressed a bot made from gift shop toys still managed to actually drive lol
I thought the thumbnail said 'Unbeatable2' rather than 'unbeatable?' so I naturally think you should name it Unbeatable2 Keep this stuff coming! Love watching the creation process of robots being made and you're ticking all the right boxes so far
Love these combat robot videos so much. Considering switching to Fusion 360, but i JUST became pretty fluid in solidworks and I don't want to go back to taking hours to design a simple part because I don't know where the tools are.
Super cool! The arms remind me of those videos of tarantulas hunting from their dens where they strike in an instant. In the us, tarantulas are affectionately called "ranchos," so maybe you could call it the Rancher, since it keeps the other robots in their place XD
Fantastic video! I was intrigued by your use of the sketch function to play with 4 bar linkages. Previously, I had been making bodies/components to test linkages. This saves a ton of time prototyping! To anyone interested, it involves fixing (see fix/unfix function while sketching) the pivot points and using sketch dimensions to fix the bar lengths of your 4 bar linkage.
How about making a third channel where you put up some sweet hardware synth jams. Can’t just tease us with those in the background. :) Awesome video. Especially liked learning about crowned pulleys.
I also enjoy using Fusion 360 for my 3D printing, though I use the hobby license it still does what I ask of it. The name that comes to my mind for that is "Mantis" as in the insect the preying mantis.
I just realized that you are standing in front of a synthesizer. I'm just getting started with my modular synth, so seeing another synth gets me excited.
I think it's worth mentioning that Fusion 360 is FREE FOR PERSONAL USE, with minimal restrictions. The biggest restriction for me has been the "10 editable files", but you can change which files are editable ANY TIME, as often as you need to. The price of entry kept me from using Fusion for a LONG time, because I thought you had to have an educational license to use it for free, and had a really hard time finding the magic phrase "Free for Personal Use". It's almost as if Autodesk doesn't want people to know about that option, and pay for the yearly or monthly subscription when they don't need to. I will probably subscribe to the annual license eventually, but I haven't needed it so far. Honestly, the software is FAR more powerful than my skills in using it so far. As I grow as a designer, my needs may grow as well.
For the name, my suggestion is Lucas. Some time ago there was a YT series "Lucas the Spider" that featured a cute animated spider, and this bot does remind me of that little guy, especially when filmed in macro. As for Fusion, would be an amazing piece of software if it was "buy and download" instead of "rent from cloud". Besides, the "all models you upload now belong to Autodesk" part of their contract seems fishy. At least it did when i looked at their user agreement some time ago while considering using the service. Still, cheers to them for sponsoring this video, it is a good one!
Angus you are one of the most interesting and informative personalities on the social media platforms of today! We know that you’ve worked hard for it and it definitely shows. Keep up the good work! Oh btw , “ SprantisRant “ may be a usable name?
@@Kuroneko42 The battery is the simplest item. He talks about the metal geared motors and Botbits receiver in this video. But yeah, no product looks or model numbers. There's more detail in the last Antweight build video. Here : ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-BvNRtRuOALw.html
Just watched ur filament tangle video before this and you just me from a nightmare situation, wrestling with a 3kg polycarbonate roll that I dropped while trying to load in it into an AMS haha. Your content never fails to inspire me to get up and do something productive, thank you for what you add to the world and the 3d printing community!
I love your vids, I actually am doing robotics right now and it's amazing how you can make these robots like this. I think that you are really really smart and I encourage you to keep doing these videos. Love em!
Would be great to add self-righting mechanism functionality to lifter arms. Also might be worth adding shielding and in-fill to tracks to prevent tracks being caught or being cut.
you might want to alter the linkage in such a way that it can flip itself back over to upright it turned upside down, possibly by having the back "wheels" contact the ground in that position if the linkage is extended - that way you might flip back to correct stance with a fast forward jerk.
This little thing is absolutely adorable, though I wonder if the profile can be made even lower by using a different method of attack. Perhaps something like a spring powered forward push? I wonder what the ratio of small springs to batteries is in terms of energy storage per weight
Thanks for making these files public and free. My 7 year old and I were in Jaycar the other day and he was looking at "battle bots" (very loose term). I'll be spinning a couple of these up to have a play over the holidays with him! Would you consider doing a BOM for the other parts? Motor, Servo, receiver etc?
The motors look like n20, probably 600 rpm or so, the esc he mentioned, and you'll need a small reciever. Guessing the servo would be mg90, you could run it on a small 1s or 2s lipo battery, something like 200mah should work. Alternatively get a melinki nano from Bristol bot builders and that has radio and esc built in.
It has two horns and charges forward a lot. More like a bull than a spider. It's very thin as well. like a sheet of paper. Maybe call it the "Bull Sheet"? I think it's great, whatever you call it.
What about using continuous rotating servos instead of the geared motors? They generally have a lot more torque and you can increase the voltage to up the speed as well as changing the wheel diameter. It also makes things a lot more simpler.
Very cool. I'd consider making the lifting arm symmetrical - so that you could lift from the front or the back (or both simultaneously), but I'm not exactly sure how that would work. Kind of like a scissor lift.
I would min/max the electronics. Use motor wire, enamel coated wire. Replace all wiring with it. Remove all pin headers from the boards. Maybe even go as far as to re-design the boards using only the required components and move those components to the new boards. Or, combine the speed controller and receiver into one board.
Bit biased as I've built one too, but you really can't beat a 3d printed, tracked 4 bar! Maximise cool stuff! Nice to see the theory proved with those shaped pulleys, I've suggested it to people throwing tracks but never seen it implemented. Looks like you're getting a great amount of 'flick' from the lifter too.
Reminds me of the butter robot from Rick and Morty, and before someone tries to tell me to "Watch another show!" I've never even watched it, but the memes are inescapable.
Seems like with a bow shaped plastic piece on top of the arm, it would be possible to self-right if flipped over. That could come in handy. Great little machine.
For lifting the forks, a faster firing system would be better. Maybe something like a spring that can be tensioned and released quickly, like a cuckoo clock, or springs like in toy guns that use springs to fire, like Nerfs. This would make it possible to launch the opponent's robot higher and faster.
I has been a while and I might be totally wrong here, but the increased surface area of tracks shouldn't give you more traction, since traction is just static friction and surface area isn't a part of that equation. The reason tracks are used on vehicles is always to spread out the weight of the vehicle over a larger surface to prevent it from sinking in/damaging it. So really the only advantage in traction you should be getting here is from the material being sticky and the "front wheels" being driven compared to having a 4 wheeled setup. I'm really curious how a 6 wheeled version of the same platform would perform with a thin belt/string transferring power from the axles connected to the motor to the other 4 wheels.
@@thomaskletzl6493 Bigger wheels on cars work because they're actually sticky when they get hot. There's also a very slight amount of deformation where the tires are flat on the bottom so instead of pure friction you get a slight bit of pushing(diagonally up but still) So bigger wheels on cars work not because the larger surface area gives you more friction, but because they have other methods of giving you traction that don't rely on traction. For this bot there's never going to be enough heat to get the material to adhere to the floor(nor is that legal AFAIK) and flat deformation aspect of round tires doesn't apply to tracks. Tracks instead get their benefits on loose surfaces where they sink less and their shapes can "bite into" the surface. Neither is applicable here.
Crowned pulleys! I've never heard of those before but it makes so much sense once explained. I'm going to need to remember this for some projects. If you like a spider name, the Peacock Spider. Otherwise, Sir Lifts.
