Perfect thank-you for sharing the info and all the files. I wanted to build a hexapod in the year and i just ordered the same servos! Now onto printing.
Pretty sure I'm gonna end up building this thing. Amazing job, can't wait to see it finished. Bed slinger are as good as core xy, the only issue is speed but to be honest, it not that big of a difference if you don't go for a voron type of craziness. My ender 3 still do an excellent job with klipper and will continue to do so in a few years to come.
Bought myself this printer a while ago. Did some research and concluded that I rather spend a little more initially to get a good start and put the time on 3D modelling instead. BUT NOW, with prime day 30%off discount, highly recommend👍👍
I am starting with Fusion 360 and I am also Blender user. Thanks for this video! I'm going to watch your streams now and study EVERY. SINGLE. SECOND! 😄
Blender has a great mesureing tool plugin wich makes rulers following the lines you make. Of course the part were fusion addaps the rest of the design. Is something i have not seen in blender. So good choice!
There is absolutely no good reason to use Blender for any parts which are supposed to be functional except if you are only ever doing it as a one-off and don't want to put effort into learning a real CAD software. And I am not even specifically endorsing Fusion 360, even thought it is what I personally use. Basically any parametric CAD software will be an improvement over Blender even FreeCAD and I wouldn't recommend that software to anyone. The plugins don't make Blender a good choice, they make Blender a less atrocious choice.
Awesome peoduct review! Anker's got their 3D printer on a killer discount for Amazon Prime Day? That's awesome! Gonna snag one and start printing out some wicked cool stuff!
Are all of the hexapod legs the same as the one modeled? If so, you can use a circular pattern and pattern the "leg" component around the center axis so that 5 other copies are created for the other legs. The nice part is that all of the legs should update properly when changing leg lengths or other parameters- the same way it updates properly now with 1 leg. Hope this helps!
They are, but they are not in a circular pattern and manually placing them in isnt really an issue since they will still update automatically. Honestly i just havent bothered placing the remaining 5 because im lazy lol. I should though so i can get some nice renders.
Hey, like your video. Since the first day of using Fusion i love it. It is not perfect however for prototyping and hobbyists like us it is perfect + it has extensions for manufacturing. I come from a aerospace engineering backround and we design everything from a master model. I also recommend the use of assemblies and even subassemblies and also i would add a assembly which contains only fasteners, bearings etc..and last but not least dont forget to give your parts a practical name. Dont just leave them as body1 or part1.
This hexapod will never walk on a wall, its way too heavy 😂 But I am open to designing something that could, that sounds really cool. what if each leg had a fan on it to suck it to the wall?
Hi, I made hexapod year ago, it was heavy because of sheet aluminum and without inverse kinematics, I already have all the parts. Your work inspired me to try again but make it even better now. I will be using Arduino and pololu maestro servo controller, got rds3225 servo ( robotics one that have shaft on both sides and they are 10 dollars a piece just like you said in the video). For sure I will be looking more at your STL file to accommodate to my hardware because this is truly the best design i've ever seen (including phoenix from zenta)
@@AecertRobotics yeah it was laser cut 2.5mm thick aluminum frame and legs, tibia was also bent to give it more stiffness but that hexapod was heavy, if I remembered correctly about 8kg. I 3d printed feets and mounting for lipo battery and arudino
Hi do you think 24kg RDS3225 servos will work well? They have shaft on both sides. I already printed your design with couple of twist for my hardware. I will be using PS2 wireless controller to control Hexapod. Its like 10 bucks with adapter :D
With the first layer issue using prusa slicer. Check the g-code start sequence in both programs, should be somewhere in the printer settings. Soemtimes it lacks one line of code that tells the printer to use the scanned mesh of the bed created during auto-leveling. Without this line of code your printer prints as it would have perfectly leveled bed which is a rare case.
Have you considered different types of springs? Like wave disc springs or leaf springs. Wave springs might not print well, but could fit in the existing design. Leaf springs seem like they could print well, but would require a redesign if it's found to be worth using. Perhaps the entire outer leg / tibia (?) could be a leaf spring. Big ups to Dusty.
@@AecertRobotics The wave spring is supposed to provide more force with less height, presumably due to it being similar to two coils. A leaf spring wouldn't need to be printed thinly and can distribute the load across a wider area, allowing more bend overall but less bend in each specific location. This is all in theory.
Heyyy!! I remember this robot from one of the pages on reddit but then I recognized your name as well! I did not know you are a UA-camr man :0 Great job with the Hexapod! I think it is really cool. I built one years ago as well but it was nowhere near as good because of limited parts availability as well as budget :D Anyway, I offered help back then on the reddit page and I still offer my help if you need it :D Best, G
I started with existing hexapods and went from there. As a general rule the longer the leg segments the fast it can move, but the more strain on the servos. And yeah after that it was trial and error but really the lengths I went with initially just worked. Tbh youd have to go pretty extreme for it not to work.
@@AecertRobotics I'm currently designing an octapod scorpion with the big question of how long do I make the legs. I have already bought DS5180 (80kg-cm) servos with which I'll build at least the first iteration of the robot. At first it'll have 3 segments per leg for simplicity but eventually I want to make it 4 segments per leg. Searching for inspiration and answers for questions regarding the design and your videos have given me a lot to think about. Keep up the great content!
@@Hawx111 I would not make it 4 segments per leg, that is super overkill. As for the length of the leg segments, I would just start with what looks good and go from there. There really is no correct answer because a ton of different things could work.
I think a lot of peoples question/video series revolves around using inverse kinematics to solve their robot/hexapod problem. Like when do you know when to use IK or simple trig. or using inverse kin on a Quad bot. do you think youll be able to make a vid on that specifically?
I'm open to making an IK tutorial video. doesnt matter how many legs it has, just what the motor layout of the legs is. Before i spend time making tutorials i want to get the hexapod kit released first though!
@@AecertRobotics totally understand thanks for the reply. Should've put more thought into the comment because i think you said this is ur first rob proj in a while.
Good start for cad softwares, maybe u can try Solid Edge latest version from Siemens. This software have a good student version with some limitation. But evrithing what u didnt know is no longer a problem. U only need to press F1 and u can find the answer, and thevsoftware also well describe what does what. Maybe my english is not so correct but i learned solid edge and this software have one of the most easy to use user interface. I gues u can try it out, oh and also have training examples in the program when u install it. Good luck for u with cad aoftwares! And well made video dude!👊
Have you seen the serial servos? That will help you simplify wiring since each of them can be daisy chained with each other. They provide Realtime feedback, so you could technically use it to determine if a leg is on a surface or not.. think about uneven terrain. You could have the gait stop at a different z height depending on walking up or down. Also they are serial 3 wires vs ~18x3 of em. I have a Pheonix I bought from lynxmotion back in ~2008 ish. I mentioned in one of my other comments from a couple months ago, planning on using an ESP32.
I have! And i totally agree, but I dont know how to get them working with my arguino mega so i will most likely stick with the pwm based servos for this hexapod, but they do sounds really nice to work with.
Great question. First off, the tibia and femur servos actually carry the most weight, so those need to be strongest. The coxa servo definitely could be weaker/cheaper, but frankly its just easier to buy the same servo in bulk.
Love this project and im currently printing everything to assemble it!! Was just wondering where the metal pieces were from? the ones that go into the femurs (bearing side)?
How is performance in fusion 360, and what cpu do you have? The main reason I use blender is performance but I think I might be using fusion 360 wrong.
It chugs but its usable. Granted the hexapod is a pretty big file at this point. What helps is to make the stuff you arent working on invisible, or even better have things in separate projects. Obviously i couldnt easily do that since all the leg parts rely on eachother.
That is true, and i did look into it but I think i decided against it because it was weird with selling your designs or something like that, tbh i dont remember it was a while ago.
fusion 360 is kinda a pain when just starting out learning it, but it’s 100% worth it when you get the hand of it. plus mechanical simulations in it are incredibly useful. i almost paid around 200$ in parts made from cnc but decided to just try out the simulations thing and if i would have bought the parts it would have been a total waste of money
This is great. I had a similar experience switching from using Blender for DFAM (design for additive manufacturing) but with Onshape. In these parametric web-based CAD apps, I foresee not-too-distant future functionality upgrades will include AI, AR or VR, and mind-computer interfaces (for ultra speed modeling).
Yea... I learned CAD in Fusion360 and am now interested in learning Blender... so I've tried using Blender to do CAD projects that I'd normally do in Fusion... but Blender just sucks for CAD. Seriously... they could learn a lot from CAD workflow. I get it... the whole "ART" workflow is different and all... but the ability set values based on relationships to other values in projects (constraints) even based on things from the past in your timeline... is really something blender needs.
Hahaha yeahhh.... I would not recommend blender for anything like this. I don't think blender needs built in cad support tbh since it really isn't their target audience, although if they added it I wouldn't complain.
I did the coding iteratively. So i started with making the servos move, then i got IK working, then I got IK working for all the legs, then i working on making a gait, then making the gait dynamically change based on user input, then on multiple different gaits.
So what is probaly happening with your first layer is that in Prusa Slicer under printer settings in "Start g-code" there is G28 homing command as default. This command besides homing your printer also disables any bed leveling in the printer itself. U should put a "M420 S" after G28 to restore leveling or use G29 and then your printer will do leveling before each print. You can read more about it in Marlin documentation. I don't know Anker slicer, maybe it's posibble to find start g-code in it and copy it to prusa slicer then it will behave the same...
I don't recall all the details (because I have a Bambu so I'm not planning to actually buy an AnkerMake) but ModBot recently did a video on the M5 and discusses this issue. If memory serves, there's some custom functionality on the M5 that only occurs in the Anker Slicer. His workaround was the same as yours. My impression is that Anker is trying to pull off the same "walled garden" effect as Bambu, but where Bambu based their software on a good slicer and then put a lot of effort into creating a pleasant user experience, Anker based their software on a mediocre slicer and left it all bare bones. Bambu also left a lot of "gates" in their garden wall and sure, the paths out there are unpaved and bumpy but if you don't want to be in the garden, you can still get acceptable functionality out of a Bambu printer. From reports, it seems that isn't true for Anker. Honestly, given that it's a bed-slinger and so limited to low temp filaments, Anker is competing with both Bambu's P1P - which is as reliable, as fast, has a better user experience, requires much less setup, and a lower price - and with Prusa - which is *also* as reliable, not as fast, more expensive (for a Mk4) or less capable (for a Mini+), but open source, an excellent user experience, and a strong corporate history with excellent aftermarket support. As far as I can tell, the only reason for someone to buy an AnkerMake is they don't trust Bambu but do trust Anker, and don't want to spend for a Prusa, and that seems like a pretty niche customer to me.
@@davydatwood3158I haven't watched any video on Ankermake, but I guess it's the same "extra functionality" as in every new fast printer that uses some kind of klipper fork which is basically custom macro. That's why the best idea would be to copy start, end and layer change g-codes to Prusa Slicer from Anker Slicer. I don't think M5 is trying to be competition for P1P. At least in my country it's 25% cheaper so cutdowns like smaller build area and non-metal hotend (and it's targeted to "begginers" so they expect user to print PLA) are expected. Being bedslinger doesn't mean it's worse than than CoreXY printer. My Switchwire can pull much higher accelerations than my Voron Legacy(with modified X axis so it uses mgn12 rail), mostly because of its super sturdy frame.
Fantastic choice on the redesign. Do you have any thoughts on the Pimoroni 2040 as a control platform? Fusion consideration for the future: you can use the "derive" command to isolate specific groups of components or or a single item from an assembly. The component will still be parametrically linked to the original design. The larger and more complex the build becomes, the more difficult it will be to work in the main workspace due to downstream computations and bloat. This is also a very easy and efficient way to work on multiple variations in parallel without disrupting the timeline of your main build. I'm really enjoying your project and while it is much too ambitious for me to take on something this large in scale it has reinvigorated a small project of my own. Thanks!
@@aurelienmontmejat1077 Ali express www.aliexpress.us/item/2251832533583005.html?spm=a2g0o.order_list.order_list_main.10.634a1802xcdNyp&gatewayAdapt=glo2usa Ill be testing them out today
Try using the LX-224 servos instead of the type you're using. With the LX-224's the servos can actually be daisy-chained, ie instead of having the tibia wires run all the way from the tibia to the Arduino, the tibia wires can be connected to the FEMUR servos, and the femur wires can be connected to the coxa servos, and then just the coxa servo wires can be connected to the Arduino. That way, you won't have fragile untidy wires all over the place. Please tell me what you think about the LX-224 servos. Pleeease!!!
I would love to use this type of servo, but I have no idea how to make it work with arduino, the servos im using are so much easier to use. With that said, if and when I make a new hexapod with more advanced electronics I will definitely be looking into serial bus servos.
@@AecertRobotics sounds wonderfull !! Cant wait to Get my hands on the stl files!! Do you know when in the future it will be ready for consumers?? Itching in my hands 😂😂😂
@@uhren95 check out the GitHub link in the description for the stl files right now 😉 As for the official release of the kit, I'm hoping in the next couple of months!! But realistically we are shooting for before the end of the year
How much is the total parts cost? With 44$ for servos and I'm guessing ~10$ for filament, and about 40$ for electronics, is it around 100$? This looks like one of the cooler and more approachable printable legged robots, and I'm tempted to build one.
Hello friend where I can find the files to download them, I would like to try to put it together with my 9 year old son who likes these things, greetings
Why are you plating in Prusa Slicer? Not suer you pointed out what this adds to to workflow other than more effort? The Anker Maker slicer can't be in Beta, it's a reskinned Cura. A slicer that has been out for a long time.
Really good question, ankerslicer is in beta atm and is missing a few features which make it annoying to plate in. But, I believe they are working on a slicer which is based on prisa slicer, so I'm really looking forward to that
Very tempted to build one of these, as a platform for code etc it would be great. Do you have any plans to share the parametric CAD file? Also what servos did you use/would recommend for this? Also completely get that this would 100% be with no help or support!
@@AecertRobotics hey, completely unrelated, but i was wondering what kind of filament you used to print your parts. I'm basically doing the same sorts of projects as you are but haven't printed any projects yet (other work keeps getting in the way) and heard that servos, steppers, etc can get hot enough to warp PLA. What have you been using? I'm not sure if it was mentioned in the video and I missed it, but from the timelapse print in this video (nozzle at 200 degrees celsius) it seems you're using PLA.
dont try to run (constraints) before you can walk (basic feature drawing), why you people always do it like this, then you stumble in complexity when it explodes when it does not do what you wanted, lol
Haha that's hilarious - i went the complete reverse direction: 1 1/2 years with Fusion and now 2 years with Blender. I'm having so much more fun with Blender I hope I never have to go back. But yeah I can see scenarios where a different tool is better. Really depends on what your project is. Overall i recommend Blender to people wanting to dive into 3D printing and 3D design. Fusion does not have any explosions 🤣
Blender is soooo good for modeling non-function things, its very freeing you can do pretty much anything. The moment you need to make something mechanical though, fusion 360 (or any CAD program tbh) is what you want imo
Hello Sir, I'd like to use this to teach stem in Kenya 🇰🇪. I'd like learners to get started in stem and robotics with ease. This may be well suited for learners in higher grades but it's also an amazing platform to teach 3d printing since most of the parts are 3d printed. The lesser I have to ship electronics the better. I'll be contacting you soon. This is amazing work and impactful as well.
i assume youre already aware, but just in case, do note that the free Fusion license is only free as long as it is for personal and not commercial use. i dont know the exact threshold and i doubt youtube videos will be an issue. however if you plan on selling the files there might end up being some issues.
Yeah, that's the reason that I ended up climbing the steep learning curve of FreeCAD, once you get the workflow down it's an incredibly powerful tool with absolutely no restrictions on what you do with it.
Hi is there any people here can guide me to learn about the programing input to our design? I am really beginner in coding. fortunately I didnt have any problem with 3D design
Dusty is a legend
He really is
Haha, first rule of the internet: Do something wrong, and a bunch of knowledgeable people will show up to correct / teach you 😅. God bless dusty !
Im loving the contrast in the new design, makes it easier to see all the different mechanisms and appreciate the work put into it
Perfect thank-you for sharing the info and all the files. I wanted to build a hexapod in the year and i just ordered the same servos! Now onto printing.
It's always good to see people find the right tools for the job and recognize how much easier the job gets.
Glad to see you happy with all of these. Sponsor, new printer, well printed new hexa and learning fusion360.
Using blender to model your parts for the hexapod is honestly impressive but for all the wrong reasons xD
🤣
The redesign looks really nice.
Thank you so much :)
Indeed I hit the like (thumbs up) button. I enjoyed/liked this video very much and I'm looking forward to see the next one!
Pretty sure I'm gonna end up building this thing.
Amazing job, can't wait to see it finished.
Bed slinger are as good as core xy, the only issue is speed but to be honest, it not that big of a difference if you don't go for a voron type of craziness.
My ender 3 still do an excellent job with klipper and will continue to do so in a few years to come.
Bought myself this printer a while ago. Did some research and concluded that I rather spend a little more initially to get a good start and put the time on 3D modelling instead. BUT NOW, with prime day 30%off discount, highly recommend👍👍
Thank you for your consistent support! - Kiki Chen from the AnkerMake Team
I am starting with Fusion 360 and I am also Blender user. Thanks for this video! I'm going to watch your streams now and study EVERY. SINGLE. SECOND! 😄
Wow! Such creativity! The design looks awesome!
Great to see the progress.
Looks great, can't wait to build my own!
Awesome project! Really enjoyed this video and maybe is time to get my husband a fun "toy" this Prime Day - AnkerMake M5 3D printer soooo cool :P
Very cool video, love the design!
I've seen many "print your own robot" videos, but this is somethign else. I think this will be one I actually try!
Blender has a great mesureing tool plugin wich makes rulers following the lines you make. Of course the part were fusion addaps the rest of the design. Is something i have not seen in blender. So good choice!
Blender even has a cad plugin which I was trying to use for a bit but yeah fusion 360 just makes more sense. Thank you!
There is absolutely no good reason to use Blender for any parts which are supposed to be functional except if you are only ever doing it as a one-off and don't want to put effort into learning a real CAD software. And I am not even specifically endorsing Fusion 360, even thought it is what I personally use. Basically any parametric CAD software will be an improvement over Blender even FreeCAD and I wouldn't recommend that software to anyone.
The plugins don't make Blender a good choice, they make Blender a less atrocious choice.
@@AlexusMaximusDE 💯
Awesome peoduct review! Anker's got their 3D printer on a killer discount for Amazon Prime Day? That's awesome! Gonna snag one and start printing out some wicked cool stuff!
Dude its like 250 dollars off which is crazyyy. Let me know how you like it!!
Are all of the hexapod legs the same as the one modeled? If so, you can use a circular pattern and pattern the "leg" component around the center axis so that 5 other copies are created for the other legs. The nice part is that all of the legs should update properly when changing leg lengths or other parameters- the same way it updates properly now with 1 leg. Hope this helps!
They are, but they are not in a circular pattern and manually placing them in isnt really an issue since they will still update automatically. Honestly i just havent bothered placing the remaining 5 because im lazy lol. I should though so i can get some nice renders.
Any inserted components can update when the original file updates.
Cant wait to make one of these!
Hey, like your video. Since the first day of using Fusion i love it. It is not perfect however for prototyping and hobbyists like us it is perfect + it has extensions for manufacturing. I come from a aerospace engineering backround and we design everything from a master model. I also recommend the use of assemblies and even subassemblies and also i would add a assembly which contains only fasteners, bearings etc..and last but not least dont forget to give your parts a practical name. Dont just leave them as body1 or part1.
Configuration Management oh yea!
Wow your project is fabulous, hehe your cat was curious observing the new invent
A Spider and octopus 🐙 are one of the favorite creatures of the robotics department.
Octopus and Spider are my first and second favorite animals :)
So beautiful!
Awesome. Nice cat.
Constraints are a game changer once you understand them
All I saw was that sweet kitty with thumbs in the end :D
there is also Onshape, it is also free, and cloud based.
True! I decided not to use it personally because I wouldnt be able to sell my designs if I wanted to.
apparently they are adding parametric sketching into Blender so you'll be happy to hear that.
Holy shit seriously? That would be amazing
Awesome man!
Try to make it climb on a wall. standford has a robot that walks on walls with a certain silicon 3d print. I think it this very doable.
This hexapod will never walk on a wall, its way too heavy 😂 But I am open to designing something that could, that sounds really cool. what if each leg had a fan on it to suck it to the wall?
@@AecertRobotics electric suction cups would work. some people use them for wall climbing
Hi, I made hexapod year ago, it was heavy because of sheet aluminum and without inverse kinematics, I already have all the parts. Your work inspired me to try again but make it even better now. I will be using Arduino and pololu maestro servo controller, got rds3225 servo ( robotics one that have shaft on both sides and they are 10 dollars a piece just like you said in the video). For sure I will be looking more at your STL file to accommodate to my hardware because this is truly the best design i've ever seen (including phoenix from zenta)
Sheet aluminium wow. I'm so glad to hear that you're giving it another go!
Thank you so much man that means a lot ♥️
@@AecertRobotics yeah it was laser cut 2.5mm thick aluminum frame and legs, tibia was also bent to give it more stiffness but that hexapod was heavy, if I remembered correctly about 8kg. I 3d printed feets and mounting for lipo battery and arudino
How much infill % did you use on main frame? i got it on 20% and my Ender 3 giving me like 20 hours for bottom part.... :D
@@pawezietek127 60% cubic! Oh man 20 hours is quite a long time 😂
Hi do you think 24kg RDS3225 servos will work well? They have shaft on both sides. I already printed your design with couple of twist for my hardware. I will be using PS2 wireless controller to control Hexapod. Its like 10 bucks with adapter :D
With the first layer issue using prusa slicer. Check the g-code start sequence in both programs, should be somewhere in the printer settings. Soemtimes it lacks one line of code that tells the printer to use the scanned mesh of the bed created during auto-leveling. Without this line of code your printer prints as it would have perfectly leveled bed which is a rare case.
Looks like that free version now only supports CAM files at a limited state
love your cat! Isn't it afraid of the hexapod walking around?
12:16 the double paw of the cat
Have you considered different types of springs? Like wave disc springs or leaf springs. Wave springs might not print well, but could fit in the existing design. Leaf springs seem like they could print well, but would require a redesign if it's found to be worth using. Perhaps the entire outer leg / tibia (?) could be a leaf spring.
Big ups to Dusty.
I just went with the first solution that I could successfully 3d print. What benefits would I get from these different spring designs?
@@AecertRobotics The wave spring is supposed to provide more force with less height, presumably due to it being similar to two coils. A leaf spring wouldn't need to be printed thinly and can distribute the load across a wider area, allowing more bend overall but less bend in each specific location. This is all in theory.
put a cat seat on the spider. have it carry the lil kitterkat!
How insane would that be though 🤣 I really want to do this
Really great design. I would love to learn from this. Where can I find the open source files so I may build my own?
Amazing ☺️
can't wait to see the code, really appreciate
Heyyy!! I remember this robot from one of the pages on reddit but then I recognized your name as well! I did not know you are a UA-camr man :0 Great job with the Hexapod! I think it is really cool. I built one years ago as well but it was nowhere near as good because of limited parts availability as well as budget :D Anyway, I offered help back then on the reddit page and I still offer my help if you need it :D Best, G
At the time of my reddit posts this channel didnt exist!
I appreciate it dude, thank you! 🙂
I worked 5 years as an engineer for automotive and the first thing you learn, is to set intelligent constraints.
How did you find the lengths of the leg segments? Did you do some calculations or did you just check what worked and what didn't?
I started with existing hexapods and went from there. As a general rule the longer the leg segments the fast it can move, but the more strain on the servos. And yeah after that it was trial and error but really the lengths I went with initially just worked. Tbh youd have to go pretty extreme for it not to work.
@@AecertRobotics I'm currently designing an octapod scorpion with the big question of how long do I make the legs. I have already bought DS5180 (80kg-cm) servos with which I'll build at least the first iteration of the robot. At first it'll have 3 segments per leg for simplicity but eventually I want to make it 4 segments per leg. Searching for inspiration and answers for questions regarding the design and your videos have given me a lot to think about. Keep up the great content!
@@Hawx111 I would not make it 4 segments per leg, that is super overkill. As for the length of the leg segments, I would just start with what looks good and go from there. There really is no correct answer because a ton of different things could work.
I think a lot of peoples question/video series revolves around using inverse kinematics to solve their robot/hexapod problem. Like when do you know when to use IK or simple trig. or using inverse kin on a Quad bot. do you think youll be able to make a vid on that specifically?
I'm open to making an IK tutorial video. doesnt matter how many legs it has, just what the motor layout of the legs is. Before i spend time making tutorials i want to get the hexapod kit released first though!
@@AecertRobotics totally understand thanks for the reply. Should've put more thought into the comment because i think you said this is ur first rob proj in a while.
@@7dedlysins193 no worries, and you're welcome!
Good start for cad softwares, maybe u can try Solid Edge latest version from Siemens. This software have a good student version with some limitation. But evrithing what u didnt know is no longer a problem. U only need to press F1 and u can find the answer, and thevsoftware also well describe what does what.
Maybe my english is not so correct but i learned solid edge and this software have one of the most easy to use user interface. I gues u can try it out, oh and also have training examples in the program when u install it. Good luck for u with cad aoftwares! And well made video dude!👊
incredible
Hey man, do you know the 3 lengths of the leg from one joint to another? I am building a hexapod inspired by yours and I am afraid I made it too big.
Yeah the coxa is 45mm, femur 110, and tibia 180
@@AecertRobotics
Have you seen the serial servos? That will help you simplify wiring since each of them can be daisy chained with each other. They provide Realtime feedback, so you could technically use it to determine if a leg is on a surface or not.. think about uneven terrain. You could have the gait stop at a different z height depending on walking up or down.
Also they are serial 3 wires vs ~18x3 of em. I have a Pheonix I bought from lynxmotion back in ~2008 ish. I mentioned in one of my other comments from a couple months ago, planning on using an ESP32.
I have! And i totally agree, but I dont know how to get them working with my arguino mega so i will most likely stick with the pwm based servos for this hexapod, but they do sounds really nice to work with.
Why did you choose same servos for all parts, fir example servos at the edge can be smaller and cheaper
Great question.
First off, the tibia and femur servos actually carry the most weight, so those need to be strongest. The coxa servo definitely could be weaker/cheaper, but frankly its just easier to buy the same servo in bulk.
W cause of Dusty, dude helped you out!
I like the cat
Love this project and im currently printing everything to assemble it!! Was just wondering where the metal pieces were from? the ones that go into the femurs (bearing side)?
How is performance in fusion 360, and what cpu do you have? The main reason I use blender is performance but I think I might be using fusion 360 wrong.
It chugs but its usable. Granted the hexapod is a pretty big file at this point. What helps is to make the stuff you arent working on invisible, or even better have things in separate projects. Obviously i couldnt easily do that since all the leg parts rely on eachother.
Onshape is also free (for hobbyists) and I think that is true for other Programms too
That is true, and i did look into it but I think i decided against it because it was weird with selling your designs or something like that, tbh i dont remember it was a while ago.
fusion 360 is kinda a pain when just starting out learning it, but it’s 100% worth it when you get the hand of it. plus mechanical simulations in it are incredibly useful. i almost paid around 200$ in parts made from cnc but decided to just try out the simulations thing and if i would have bought the parts it would have been a total waste of money
Dude I need to try the Sims, you think it would be hard to get it working with my hexapod?
This is great. I had a similar experience switching from using Blender for DFAM (design for additive manufacturing) but with Onshape. In these parametric web-based CAD apps, I foresee not-too-distant future functionality upgrades will include AI, AR or VR, and mind-computer interfaces (for ultra speed modeling).
Yea... I learned CAD in Fusion360 and am now interested in learning Blender... so I've tried using Blender to do CAD projects that I'd normally do in Fusion... but Blender just sucks for CAD. Seriously... they could learn a lot from CAD workflow. I get it... the whole "ART" workflow is different and all... but the ability set values based on relationships to other values in projects (constraints) even based on things from the past in your timeline... is really something blender needs.
Hahaha yeahhh.... I would not recommend blender for anything like this. I don't think blender needs built in cad support tbh since it really isn't their target audience, although if they added it I wouldn't complain.
i want to make something like this but i want to make my own model
how did you do the coding?
I did the coding iteratively. So i started with making the servos move, then i got IK working, then I got IK working for all the legs, then i working on making a gait, then making the gait dynamically change based on user input, then on multiple different gaits.
So what is probaly happening with your first layer is that in Prusa Slicer under printer settings in "Start g-code" there is G28 homing command as default. This command besides homing your printer also disables any bed leveling in the printer itself. U should put a "M420 S" after G28 to restore leveling or use G29 and then your printer will do leveling before each print. You can read more about it in Marlin documentation. I don't know Anker slicer, maybe it's posibble to find start g-code in it and copy it to prusa slicer then it will behave the same...
I don't recall all the details (because I have a Bambu so I'm not planning to actually buy an AnkerMake) but ModBot recently did a video on the M5 and discusses this issue. If memory serves, there's some custom functionality on the M5 that only occurs in the Anker Slicer. His workaround was the same as yours.
My impression is that Anker is trying to pull off the same "walled garden" effect as Bambu, but where Bambu based their software on a good slicer and then put a lot of effort into creating a pleasant user experience, Anker based their software on a mediocre slicer and left it all bare bones. Bambu also left a lot of "gates" in their garden wall and sure, the paths out there are unpaved and bumpy but if you don't want to be in the garden, you can still get acceptable functionality out of a Bambu printer. From reports, it seems that isn't true for Anker.
Honestly, given that it's a bed-slinger and so limited to low temp filaments, Anker is competing with both Bambu's P1P - which is as reliable, as fast, has a better user experience, requires much less setup, and a lower price - and with Prusa - which is *also* as reliable, not as fast, more expensive (for a Mk4) or less capable (for a Mini+), but open source, an excellent user experience, and a strong corporate history with excellent aftermarket support. As far as I can tell, the only reason for someone to buy an AnkerMake is they don't trust Bambu but do trust Anker, and don't want to spend for a Prusa, and that seems like a pretty niche customer to me.
@@davydatwood3158I haven't watched any video on Ankermake, but I guess it's the same "extra functionality" as in every new fast printer that uses some kind of klipper fork which is basically custom macro. That's why the best idea would be to copy start, end and layer change g-codes to Prusa Slicer from Anker Slicer.
I don't think M5 is trying to be competition for P1P. At least in my country it's 25% cheaper so cutdowns like smaller build area and non-metal hotend (and it's targeted to "begginers" so they expect user to print PLA) are expected.
Being bedslinger doesn't mean it's worse than than CoreXY printer. My Switchwire can pull much higher accelerations than my Voron Legacy(with modified X axis so it uses mgn12 rail), mostly because of its super sturdy frame.
Fantastic choice on the redesign.
Do you have any thoughts on the Pimoroni 2040 as a control platform?
Fusion consideration for the future:
you can use the "derive" command to isolate specific groups of components or or a single item from an assembly.
The component will still be parametrically linked to the original design.
The larger and more complex the build becomes, the more difficult it will be to work in the main workspace due to downstream computations and bloat.
This is also a very easy and efficient way to work on multiple variations in parallel without disrupting the timeline of your main build.
I'm really enjoying your project and while it is much too ambitious for me to take on something this large in scale it has reinvigorated a small project of my own.
Thanks!
What servos are you using exactly? Are those DS3230 Pros? 180 or 270 degrees?
Where did you buy your servos? Do you know if they ship to Europe? $44 for 4 servos seems like a good price!
They ship everywhere, and i agree it is an excellent price! I really hope they work well ill be testing them tomorrow.
@@AecertRobotics Do you remember on what website you got them? :)
@@aurelienmontmejat1077 Ali express www.aliexpress.us/item/2251832533583005.html?spm=a2g0o.order_list.order_list_main.10.634a1802xcdNyp&gatewayAdapt=glo2usa
Ill be testing them out today
@@AecertRobotics thanks!
Try using the LX-224 servos instead of the type you're using. With the LX-224's the servos can actually be daisy-chained, ie instead of having the tibia wires run all the way from the tibia to the Arduino, the tibia wires can be connected to the FEMUR servos, and the femur wires can be connected to the coxa servos, and then just the coxa servo wires can be connected to the Arduino. That way, you won't have fragile untidy wires all over the place.
Please tell me what you think about the LX-224 servos. Pleeease!!!
I would love to use this type of servo, but I have no idea how to make it work with arduino, the servos im using are so much easier to use.
With that said, if and when I make a new hexapod with more advanced electronics I will definitely be looking into serial bus servos.
Is there a way for us viewers to Get our hands on this hexapod? I want to make this one myself!!
Not at the moment, but there will be!!
@@AecertRobotics sounds wonderfull !! Cant wait to Get my hands on the stl files!! Do you know when in the future it will be ready for consumers?? Itching in my hands 😂😂😂
@@uhren95 check out the GitHub link in the description for the stl files right now 😉
As for the official release of the kit, I'm hoping in the next couple of months!! But realistically we are shooting for before the end of the year
How much is the total parts cost? With 44$ for servos and I'm guessing ~10$ for filament, and about 40$ for electronics, is it around 100$? This looks like one of the cooler and more approachable printable legged robots, and I'm tempted to build one.
That 44 is for 4 servos! You need 18 of them! So its around 400 ish. And ty, i hope you do end up making it 🙂
I couldn't help but notice at 11:46 that your cat seems to have extra toes on its front paws. Did I see that correctly? Sorry for the offtopic.
You noticed correctly :)
There some CAD bleder add-ons
True, and i did have them, but at that point i had no experience with cad and i didn't get very far with them
Hello friend where I can find the files to download them, I would like to try to put it together with my 9 year old son who likes these things, greetings
Why are you plating in Prusa Slicer? Not suer you pointed out what this adds to to workflow other than more effort?
The Anker Maker slicer can't be in Beta, it's a reskinned Cura. A slicer that has been out for a long time.
Really good question, ankerslicer is in beta atm and is missing a few features which make it annoying to plate in.
But, I believe they are working on a slicer which is based on prisa slicer, so I'm really looking forward to that
What Servos are you using currently?
Is there a bom?
HEy i Love ITTTTTTT!!!! I've been here on all streams and every video! Can you give a tutorial? I wanna make it for my MIT portfolio!
GGS SPONSOR
what about Onshape?
Based on my research i would have issues selling my design if I wanted to. Now of course I could be mistaken but better be safe than sorry.
Very tempted to build one of these, as a platform for code etc it would be great.
Do you have any plans to share the parametric CAD file? Also what servos did you use/would recommend for this?
Also completely get that this would 100% be with no help or support!
Great video 💪👌
Thanks 🙂
when is the kit release
Dusty is an amazing dude lol
Dude ikr
@@AecertRobotics hey, completely unrelated, but i was wondering what kind of filament you used to print your parts. I'm basically doing the same sorts of projects as you are but haven't printed any projects yet (other work keeps getting in the way) and heard that servos, steppers, etc can get hot enough to warp PLA. What have you been using? I'm not sure if it was mentioned in the video and I missed it, but from the timelapse print in this video (nozzle at 200 degrees celsius) it seems you're using PLA.
This design is awesome will you be making it available?
Check the github link in the description 😉
Explain how to print and install code!!!!
Bro can you plz drop the whole assembled cad files
maybe cadsketcher in blender when it gets grid snap functionality
you dont have to constrain, those will explode unpredictability, by hand cad sketch drawing that you control yourself is much better, in general ***
the primary feature of cad is the measured sketch drawing, not the constrain automation
dont try to run (constraints) before you can walk (basic feature drawing), why you people always do it like this, then you stumble in complexity when it explodes when it does not do what you wanted, lol
I think the f360 "pro" imposed the constraints on you
k.i.s.s. pro's
Is it possible to download STL files?
You should of added parametric controls to blender 3d.
Haha that's hilarious - i went the complete reverse direction: 1 1/2 years with Fusion and now 2 years with Blender. I'm having so much more fun with Blender I hope I never have to go back. But yeah I can see scenarios where a different tool is better. Really depends on what your project is. Overall i recommend Blender to people wanting to dive into 3D printing and 3D design. Fusion does not have any explosions 🤣
Blender is soooo good for modeling non-function things, its very freeing you can do pretty much anything. The moment you need to make something mechanical though, fusion 360 (or any CAD program tbh) is what you want imo
Hello Sir,
I'd like to use this to teach stem in Kenya 🇰🇪. I'd like learners to get started in stem and robotics with ease. This may be well suited for learners in higher grades but it's also an amazing platform to teach 3d printing since most of the parts are 3d printed. The lesser I have to ship electronics the better. I'll be contacting you soon. This is amazing work and impactful as well.
i assume youre already aware, but just in case, do note that the free Fusion license is only free as long as it is for personal and not commercial use. i dont know the exact threshold and i doubt youtube videos will be an issue. however if you plan on selling the files there might end up being some issues.
Yup i know, I plan on paying for it, I just dont want everyone else to have to!
@AecertRobotics ah okay i see, just wanted to make sure you didn't end up on any radars. Keep it up then, looks amazing so far
Yeah, that's the reason that I ended up climbing the steep learning curve of FreeCAD, once you get the workflow down it's an incredibly powerful tool with absolutely no restrictions on what you do with it.
subscribing for the fantastic content ( |~ drake avoiding meme) subscribing for the adorable cat ( --~ drake pointing meme)
Isopropyl alcohol is a saver too fyi
Alguien me podría compartir algún stl para imprimir en 3d ? O me dice alguna página
Cool
11: 47 why does it have thumbsssssss??????? :(
nice project man keep it up
Shes a mutant 😼
Ty :)
Hi is there any people here can guide me to learn about the programing input to our design? I am really beginner in coding. fortunately I didnt have any problem with 3D design
I'm engaging because nobody knows a better way to get the recommendation algorithm to work.
hhahah ty ty. Watching the entire video all the way through + liking is the best way to help 😉
Great video ! #dusty
Watchdogs moment
TIL blender has CAD features.
Your cat is biting the wires. When animals chew on plastic, its usually a sign they need vitamins. At least that is what ive heard
she likes to just chew on anything string like 😂 Im not too worried
11:47
😸