How to Make Custom Wheels for Your Robot - Molding/Casting Polyurethane
ฝัง
- เผยแพร่เมื่อ 20 พ.ค. 2024
- Today’s video is about the process I went through to make custom molded wheels using Polyurethane rubber. On Bloodsport, we used the exact same type of rubber that I’ll be experimenting with and the casting process we used is very similar to what I’ll show here. The main difference is that Bloodsport uses machined aluminum hubs which are quite expensive, and those wheels are much bigger than what I’ll be making. Here, I’ll be showing how I cast rubber tires onto lightweight 3D printed hubs you can make for a couple dollars apiece, and with a reusable custom 3D printed mold, have a wheel of exactly the size and hardness you’d like!
Links to Items in this video (Note: Amazon links are affiliate links)
Vytaflex 30: amzn.to/3dyF2EP
Vytaflex 40: amzn.to/2Jt0kIt
Smooth On Mold Release: amzn.to/34VK7mG
NEW RECOMMENDATION- Mica powder pigment: amzn.to/3GuZcfT
Vaseline/Petroleum Jelly: amzn.to/3pvKtrL
Sainsmart Carbon Fiber Nylon Filament: amzn.to/2WVM3XT
Gold PLA Filament: amzn.to/3mVann5
Epoxy Liquid Pigment Dye: no longer recommended as the Mica powder doesn't interfere with curing. As of late 2021 there is a massive global urethane supply shortage that has greatly impacted availability of these products. I added a link to Vytaflex 30A which I successfully used in competition since 40A was unavailable.
Just 'Cuz Robotics Website: justcuzrobotics.com/
Just 'Cuz Store: justcuzrobotics.com/shop
Instagram: @JustCuzRobotics
/ justcuzrobotics
@bloodsport_battlebot
/ bloodsport_battlebot
Bloodsport Facebook Page:
/ bnsrobotics
0:00 Intro
0:40 Why Use Rubber
4:03 Molds and Hubs
5:42 Rubber Compound
7:09 Pour Walkthrough
8:35 Mold Release Application
10:15 Bloodsport Mold Design
12:27 Dyes
13:27 Attempt 1 - Mixing
14:40 Attempt 1 - Checkin
16:55 Division Hubs/Mold
17:27 Attempt 1 - Demold
18:17 Attempt 1 - Results
21:58 Lighter Draconid Hubs
22:27 Attempt 1 - Wheel Weight
24:39 Attempt 1- Inspection
25:14 Attempt 2 - Mixing
25:51 Attempt 2 - Demold
27:19 Attempt 2 - Problems
28:11 What Went Wrong?
28:35 Attempt 3- Mix
30:06 Attempt 3 - Checkin
31:36 Attempt 3 - Demold
33:46 Attempt 3 - Results
34:13 Proper vs Improper Cure
35:22 Outro
#CombatRobotics #Battlebots #Hobbyweight
Music provided by Argofox:
TheDiabolicalWaffle - Excalibur
• TheDiabolicalWaffle - ... - วิทยาศาสตร์และเทคโนโลยี
While I'm not planning on building a combat robot, I have been kicking around the idea of casting a part for a different project so this video had bonus value.
Keep making these type of videos and am learning a lot! I'm working on a Beetle Weight for Norwalk! Thanks!
Awesome! Hope to see you there!
I am as well. Hoping to show up for the September event. Maybe earlier if I get fully vax'd. Good luck getting your bot built.
That’s so cool! I’ll definitely be doing this for my beetle weight
Awesome video! I saw Robert Cowan's video about molding the wheels for Copperhead and was really wishing he had one covering doing this for the lighter weight robots. But now that I found your video I don't need anything else! Now I have to get to work designing wheels and molds.
I'm going to make custom wheels for a line follower robot and this video definitely helped!
love the video, really great editing, good sound and i love that you show what you tried so people can learn from your experience. I'll be doing Mountainboard wheels for the street (200x50) for my electric board and can use any intel on PU Molding and release agents so thank you for your video. Ps.: I'll be using Flex85 for those.
Thanks for the video! i will try this out! i'd like to see how these are compared to fingertech wheels?
Video quality is getting a lot better!!
awesome content thanks!!
I could attach those wheels on an antweight viper kit
Good job! This video is very detailed. I learnt a lot of tips and I am planning to make myself some roller skate wheels. Very nice video
Glad it was helpful! I love hearing people find my videos useful, and its always nice to find their applications outside of combat robots.
I like the design of your hubs. Those little pockets to grip onto the tire are quite clever. I might have to borrow that design if my custom antweight hubs don't hold onto the tires well enough.
Feel free! I wouldn't be showing this stuff if I weren't okay with some of it getting copied. Let me know how it works out 😊
I'm late to the party, but I really enjoyed this video - thank you!
THANKS WE ARE LEARNING ALOT FROM YOU
GLAD TO HEAR IT
sick
Very good info, I always buy my wheels but maybe I would try this method in the future but with a machine hub, also saw the video all way through.
Great to hear!
I’m probably going to use this wheel design on my first robot that I’ll be using at the CIRC Bot Brawl in 2022.
canyoupost the video on your own youtube channel that seems interesting and i would like to see how that completion is run
Great video. I would love to see the various wheels on a bot pushing a scale to compare the grip.
Until I discovered your videos this weekend. I was thinking 30lbs would be the smallest bot I'd be interesting in building. But beetle weight looks like a lot more fun for the dollar/build time. I have everything kicking around I need to build one. Just need to put some time in on Fusion 360 to design it up. I have access to a nice CNC router and lots of HDPE/UHMW scraps to play with at work. Plus a nice FDM 3d printer.
I was thinking this scale might be boring but your fights at Norwalk Havoc are anything but.
I tried using a scale to get some readings but instead of the wheels slipping the belts were so it wasnt a great test. Might try that again if I update the drive to timing belts
Hey, thanks for posting this, super informative! Did you consider using a mold made of flexible filament?
I have considered it, but I've been able to demold pretty complex wheels with basic PLA molds so far. See my latest Shrapnel Mine vid for details. I'm very impatient and TPU prints slowly lol.
On another project on the mold I made a extremely thin mold then I broke/peeled away mold. Mold did not take long to print. Made a mold for each wheel.
Thanks for the detailed video. Aren't the draft angles going to alter the wheels behavior? They are no longer at a 90 degree angle
The draft angle in theory would have an impact if the tires are hard enough to not squish very much. I didn't find much difference in practice. That said since this video (which is a few years old) I've been eliminating the draft angle and as long as you have a good amount of mold release generally a 90 degree wall works fine. I even sometimes print the molds with TPU or Polypropylene which can flex enough to pop them out.
I did a video about casting polyurethane wheels a few months ago. Just want to let you know you need the PU pigment paste for a nice solid color. I used the wrong pigment for my wheels, and they came out transparent :) I call them my Fanta wheels. If you use too much of the wrong pigment the wheels stay soft and sticky.
I've seen different pigments meant for opaque and transparent colors. Maybe just ended up with the wrong type? But as another person suggested I might try the powder based pigments later on
I think the paste type works better than the powder based pigment. It mixes better. How is the shelf life of the PU (A and B component) you use? Mine is only 6 months, so I have to make a lot of wheels in a short time.
Thank you so much for sharing this, very informative ! How did they do in combat ? And are they durable ?
I have found the tires hold up pretty well in combat, they have a ton of grip as well. Depending how your hub is designed there could be a risk of chunks of tire ripping out but with a well designed hub the tire should mostly stay together after a hit unless the hub itself is broken.
@@JustCuzRobotics awesome, thanks from France !
Is there a reasonable way to make sandwich type layers? As in, harder on the outer and inner layers, but softer in the middle (i.e. durability outside, traction inside)?
Should be. All you would have to do is wait for the first layer to cure before pouring the next, and use a harder (60A?) vytaflex for the harder layers.
So here's an idea. You could use a dissolvable filament (i.e HIPS) to create the mold. This would allow you to create tread patterns without having to worry about how to remove the wheel. The PU shouldn't react with it, and once cured, you can just dissolve the HIPS mold and get the wheel out. Obviously, the mold wouldn't be reusable unless you design the tread pattern (HIPS part) as a separate part that can be used as an insert in the mold and print the rest in say Nylon. A spool of 1KG HIPS is about $20-30, so not very expensive even if you can't re-use it.
That's a good suggestion, though it seems like basically the same result as smashing the mold open with a hammer. I had thought about tread patterns and actually that dissolvable support wouldn't be necessary at all provided the tread pattern is properly thought out. Straight vertical lines or even slanted ones that allow the wheel to be pulled/twisted out helically would work perfectly well. I wasn't super confident in the mold release before trying it this way, but may look into that.
It is also perfectly doable to just make a 3 part or 4 part mold that comes apart in sections even with a complex tread pattern which would still be reusable but that becomes a pain to deal with as it needs a way to be fastened together and leak proof etc. Many people do this for things that are not wheels for industrial molding with the sections clamped together or bolted together.
I didn't want to mess with that right off the bat. Several others have done it successfully. I basically wanted in this video to show the most basic and easy way to do this to show it doesn't have to be complicated or terribly expensive. In the future I may attempt a complicated multiple part mold. But I really like the concept of reusable molds rather than braking or dissolving the mold.
@@JustCuzRobotics Good call with a multi part mold. That should work even with a complex tread pattern. Would like to see how that works out.
Here is a good example. Kinda what I was talking about. Like I said I'm not the first to do this, but I went for Urethane over silicone partly due to the choice of harnesses and partly due to the lack of fumes. Silicone gives off a strong vinegar smell as it cures. th-cam.com/video/46umFYdI8uQ/w-d-xo.html
What are you using for a 3D printer? I have my eye on a few just curious to see what you recommend.
My workhorse printer is a Prusa Mk3S which is basically stock with a custom enclosure built around it. Anything that isn't abrasive I print on this machine. I start prints and walk away and come back 12 hours later to a flawless print 99% of the time. That said I have tuned every setting in Prusaslicer to dial in Nylon and use a special bed adhesive for it, but PLA/PETG are perfect with default profiles out of the box.
The printer I used for the CF-Nylon parts is a horrendous monstrosity I created by replacing basically everything except the main board and motion system of a Tevo tornado which I got for free without a functional extruder. I have a video on that as well (Just Cuz Pro 3D printer) but it was a project I would only undertake if you have watched hundreds of 3d printing videos and are comfortable digging into Marlin. All you really need is a knockoff bond tech and knockoff E3d V6 plus a hardenned steel nozzle to get it to work for filled nylons. I tried adding inductive bed probing which turned into sortof a nightmare and I wouldn't recommend it. The touchscreen is Hella nice though.
I built a dry box to feed Nylon and TPU to my printers while keeping humidity below 20% for months at a time.
Would you ever make a video about how much it takes to create and maintain bloodsport/250 pound bots for battlebots?
I discussed what the repairs and maintainance is like in this video. th-cam.com/video/4D0Bba4Jb04/w-d-xo.html
Your dye has a really low percentage of pigment(the actual colourant). Most cheaper dyes will have a higher percentage of filler than the pigment. You might try using some powder colourants as well.
Yeah I think you're exactly right about the cheap dye. Someone who did the same process but used Smooth On's own dyes said he never used under 2% by weight and it was always fine. But those were $15 each compared to $13 for 15 of them.
Apply the Vaseline with a acid brush to get an even coat without major build up.
Where is part 2 with hub design? How do you plan on attaching the hub to the shaft?
Somehow almost 2 years later I haven't found the time to make part 2 of this... Maybe it'll happen now that I have been experimenting with different materials and durometers.
Attaching to the shaft in this case doesn't happen. I press bearings into the hub and it runs in a dead axel. Live axel setups are easy too but require a different hub design of course.
What do you use to fix the shaft to hub on division? It looked like a round bore with no holes for set screws
Good question! It's a D shaped bore that widens to a circular bore. So torque transfer is handled that way. The bore is also intentionally undersized to create somewhat of a press fit but plastic won't deal with that well as it tends to creep so I also hot glue them on. They come off really easily with a heat gun, but it works quite well. Not exactly ideal though.
@@JustCuzRobotics thanks! I guess this works better the wider the shaft is bc the join is under less stress
For a small bot, you could grab a clamp hub from Servo City and then design the wheel/printed part to mate with the hub.
True, but that would be a lot more weight and money for minimal benefit overall. I had glued on wheels for the last tournament and never lost one.
@@JustCuzRobotics I've used hot glue as a masking agent when sandblasting parts. When I'm done I pour a bit of ISP 70% or better over the part and wait a minute or three. It doesn't dissolve the hot glue but the glue peels right off. I thought the guy who told me about that was nuts.
Is there a .stl file for your wheel hubs, if so where would I find them? I have a 3D printer, lots of filament and ready to go. Thanks
I haven't released anything from my 12lb bot yet, but I am open to it. The wheel dimensions I chose were specific to my robot so I didn't know if they would really be useful for anyone else. It's not too hard to design your own hub for your own bot similar to mine though. I may do that later tonight just to put it out there. I did upload files for Shrapnel Mine's molded wheels here for reference: www.prusaprinters.org/social/31323-team-just-cuz-robotics/prints
@@JustCuzRobotics Thanks, I'll take a look at it and maybe mod it for my 3lb. bot
Use a small brush or would Pam work?
Pam absolutely does not work
Did you ever think of carving some tread grooves into them for extra grip?
Yes, I have grooved tires on my 3lb bots Shrapnel Mine and Division. I'm not convinced it offers any extra grip but it looks cool! The tricky part is you need to ensure the grooves are spiral/slanted grooves not straight along the mold direction or the wheel will be more like a gear and just bounce all over the place and get awful traction.
@@JustCuzRobotics it has to be hard to maintain grip when that plywood gets shredded up. Good luck this season 👍
What is the coefficient of kinetic friction for the 40A polyurethane?
I can't answer that directly since it will depend on not just the wheel but also the other material it is in contact with. I did a test shoving into a scale on the wall with the 30A and got approximately 0.75 CoF on a vinyl floor in my apartment. It's probably a little lower than that with 40A but maybe higher on wood.
I would guess no higher than 0.80 for almost any real wheel IRL once coated in arena dust and debris.
@@JustCuzRobotics Awesome! Thanks so much, I was getting really suspicious of a couple values I was seeing that were giving me a CoF around 1.5. I wanted to ask you to make sure, since I am still fairly new to the building side of the sport.
Curious if Pam cooking spray would work....
Someone on reddit said they tried that and it was a disaster
did you feel like you got alot more grip?
Compared to the O rings absolutely. I couldn't make an objective test since the belts driving the wheels now slip before the wheels break traction. But the Norwalk competition I used Draconid at after making these wheels, I had no issue pushing another wedge/lifter bot around
Ah I realize my question didn’t make sense really. Gonna comment anyway for that good TH-cam interaction
PLA should be fine for the rims correct?
PLA will work in the sense that the urethane will probably stick to it. But it's super brittle and if your wheel gets hit by another robot it will probably shatter like glass. I have been using TPU printed hubs for a bit now and they have proved very durable so far.
@@JustCuzRobotics how well does polyurethane epoxy hold up in these thing
I'm using a polyurethane casting rubber not a rigid epoxy. I have not tried making structural parts with epoxy. This Vytaflex urethane is a rubbery soft and grippy material. There are different hardnesses, the harder wears down slower but provides less grip.
@@JustCuzRobotics do I need to enclose and ventilate my printer for tpu
Nope it prints fine without warping typically on a Elmer's glue stick (PVA Glue) on most bed surfaces. Just need to print slower than usual as it's flexible and difficult to feed into an extruder and be careful about retractions.
In case anyone is wondering, I tried using Pam as a mold release on my wheel molds and IT DID NOT WORK, no matter what a reddit forum tells you.
Has anyone told you that you sound like Nathan Fielder from Nathan for you?
Lol no that's a new one. I have watched that show and I really don't think we sound that alike when I'm using a proper mic rather than the on-camera one.
:3
Plz text me your liquid bottle name ...
Are you talking about the vytaflex? It should be linked in the video description
why your voice like robot lol
I think that you are likely noticing audio distortion from the noise reduction I applied to deal with the fan noise in the background. Had a fume extractor running in case the urethane fumes were dangerous.