I think you need either bigger springs or more preload on your front idler shocks to stop the tracks winding up when you reverse, in the clips you can see the idlers moving backwards and dropping the tension off the tracks. Otherwise fantastic project, reminds me of the bomb disposal robots I drove around for years.
You could use a second set of sprockets towards the center to modify the gearing without changing your front sprockets. You'd just need to move either the motors or the driven sprockets to have the altered alignment so the chains could go towards the center then to the driven axle
I have just been doing your box2D tutorials (which are amazing btw!) and came to check your channel out to see what other crazy things you do. Well, what I see here is not surprising whatsoever and makes me very happy!!! I am so glad you have found success in TH-cam and thank you for those amazing tutorials. Keep up the amazing content!
Thankyou for plowing all of YOUR money into these rad projects....I can't afford to do builds like this without REALLY wanting to save for a while and having a verified BoM in advance. You've been ahead of me from my multi wii days thru to today, making tricky things seem less daunting and even DOABLE for the over cautious like myself. Keep up the great work!
This thing is great! Hoverboard motors are pretty neat. I converted a Razor Dune buggy go kart to direct drive with 350w hoverboard motors and ran 84 volts through them. That setup pulled me (driving) and three kids in a bike trailer which was connected to the go kart! Motors never failed me even with all that abuse. They ran warm but we managed 23mph all day!
A very fascinating design, particularly the suspension. Love the CNC work, it like seeing art being made. Impressive performance, climbing hill, etc. I don't think this fantastic project is going to be put aside for long, as it sounded like you're having too much fun. ;) Regarding the gearing ratio, to me looks like the rear cross brace is in the wrong spot to use larger gear. (45:35) It will need to move aft, or forward relative to the axil(s). Don't think there will be much bulge, just a bit in front of and half way to the motor. Using a nearer to 1:1 gear ratio would eliminate any bulge. The more complex option is bigger rear wheel sprockets (to match the larger gear size) and lowering axil to keep flat top. It would not impact suspension clearance height, but might mean moving rearward to allow the rear suspension arm to have full travel (an issue pointed out). A stronger spring on the rear arm might stop the bottoming out.
I smiled from beginning to end like an old crone gurning! Love the CNC drive sprockets etc. I did not think the MDF would be good for more than a few lightly loaded runs... they sure proved me wrong. When reversing, the full torque of the motor is directly compressing the spring of the front sprocket. Don't increase the preload of that spring, but limit the travel of the front sprocket to the point where the belt teeth cannot skip. Something like tube or hose over the front shock absorber shaft should do it. As you said, sorting the gearing is first priority. Plenty of options, but none that really stand out. The amount of grip you are getting from the smooth tracks is amazing! I always assumed you had to have really agressive teeth (Edit: "treads"). This is obviously not a cheap project, but at this point it has relegated the Farm Rover to Fading Ghost status! Love it, thanks.
Lol, I wasn't expecting to see my name on the video, especially from a 3yo comment! I'm thrilled you make the hoverboard tracked vehicle a reality! They truly are a fantastic platform for robotics applications. Just a couple notes & thoughts: 1. Make sure you tune the ADC readings for battery voltage. At least for me, I found that the replacement firmwares were triggering low-battery shutoffs even when the actual state of charge was fairly high. There is a procedure in the docs for calibrating those values. 2. Have you considered designing a replacement wheel/tire for the hub motor itself, allowing direct drive? The hoverboard motors are pretty well sealed out of the gate, and when I built of HB E-longboard I ran it though lots of muddy, puddles, and through the rain without any water ingress. The benefit here would be eliminating the chain/sprocket/drive axles/bearings entirely, allowing a much more easily 'waterproofed' system with less complexity. Great video as always!
Yeah, after both runs the battery was at 35.5V which seems a little high, barely under storage charge. I did notice some decent seals on the shaft, good to hear they are at least reasonably waterproof. After this first test I'm thinking more that direct drive might be the way to go. I'd like it to be fully amphibious so extra measures might be necessary, but it's probably an easier starting point than trying to waterproof a spinning shaft coming out of the hull.
@@iforce2d yeah definitely sounds like the calibration could be off, unless those HB batteries have high ESR and sag a lot under load. If it's helpful at all, I have discovered those HB main boards and motors run fantastic under a 13S battery. I have a cart I built using two sets (4WD) powered from a 13S Lion data center backup power pack rated for 100A cont. Discharge. Haven't killed a motor or main boards on it yet, so I think it can handle it safely. My HB longboard is a custom 12S battery and it runs great with that too. Last note about the waterproofing, you might want to pop the seals off the shafts and grease them up. Every motor I have gotten has had the seal there, but be bone dry defeating the purpose and maybe even prematurely wearing out the seal. The grease will definitely help them.
Midway in the video I though it was going to be way bigger hehe. I have 4 of those motors on my skateboard. If you rewire them from star to Delta connection inside they become beasts! You loose a slight amount of torque but gain a multiple amount of power! So Delta connection with down gearing turns these motors into something usable in 1:1 size application. 4 of these motors would actually be enough power for something that you could sit on and take you many miles. My skateboard is more powerful than a moped and takes me about 15miles on asphalt. I run 4 vesc on my setup and 8 Makita 18V batteries at 36V working voltage. Giving the motors 56V and really geared down you could make something real useful with 4-6 of em! A ridable 6 wheeler would be awesome! The clue is motor efficiency, Delta is way more efficient and you dont really loose much energy in a gearbox vs twice the twice the winding resistance in the original star winding setup these comes with.
That was what I was thinking at the first test .. stiffer springs..like triple the stiffness... at least. all the motors energy gets dumped into the springs at slightest velocity change... luckily you pointed that out, so I don't need to add to the "useful comments list" At this point quickly let me thank you very much for pissing away your money, you make really fascinating videos.. love them, keep going ! :)
"If only I had a couple of mobility scooter motors. Icould install a chair and ride it up and down the hills to retrieve errant aircraft". Very nice, one or two things to be fettled but the principle is fine.
hey we muck around building and testing amphibs on the manukau harbour, really keen to see how this goes on the mud flats & mud-to-water threshold zone being so light. sing out if you need somewhere to test it when you get to that point.
9:30 I really enjoyed these segments across the video. I never really gave it much thought using a CNC router to etch a template for welding a frame. I have another tool added to my arsenal. Thank you very much.
Very impressive!! Great design work and your CNC is awesome. Something I need in my shop. I will throw in my $0.02 here. I think if your plastic track center pieces were made from solid rubber instead of 'U' shaped plastic, your tracks would not jump or slip. Anyway, love your work. Thanks for sharing.
Hey Chris, this could cost you a lot, the tracks alone. But I see you had some nice design challenges to solve. Re-gearing it would be better. Going backwards though, there seems not enough guidance from the top part of the track, maybe that's why the teeth are jumping at times? When the tension is the lowest it happens. That bump stop will help for sure, it looks very stable as well. Thanks for this one! Cheers 👍💪✌
A lot of real life tanks have the drive sprockets in FRONT and the tensioner wheels in back? Another thought is, you might be able to use Roller-blade skate wheels instead of the MDF ones, they are made of tough material and have precision bearings already? The track stretch seems like it will be a struggle until you get a better material to make tracks from, but you are onto something for sure! CHEERS!
when i left that comment i started hoarding those motors and control boards from almost free broken hoverboards but never got around using them, im pretty lazy.. now they piling up in my basement but thats exactly the project i had in mind lol big and strong enough to carry my groceries home
Brilliant build, I've built a six wheeled cargo robot of a similar size using hoverboard motors. Top tip (and I wish I knew it before I built mine) is that the width between the tracks needs to be wider than the tracks are long. This gives you a *lot* more turning torque, tanks get around this with brute force which is why they can be narrower. You could also increase the preload on the suspension so that by default the front or rear is higher off the ground
I’m also building ex tanks but smaller and high speed. Achieving good yaw control is critical. I’m currently experimenting with gyros. I also think sensored motor control is also needed.
That chain sprocket setup really held that back I would go with direct drive... And I do believe the shocks would work better if they were a little less angled I suggest 10° more upright 5° may work.... To get it to go both ways over obstacles it might be good to have every other swing arm and shock opposite
Wondered if you considered using chain as the drive to which one could attach the "now flexible and stretchy" rubber. Important just where one locates any "idler tensioner " as well, usually only on the off drive side of any pulley arrangement. Difficult /well-nigh impossible to do when one has reversing drive.
Well dang. I've been looking to do almost exactly this only I didnt think of the hoverboard. The best I'd gotten so far was scooter motor hubs, which are a lot more expensive for one than for whole hoverboard. -edit not for nothing, but if you want more high speed power at slightly (significantly) more cost, you can remove the rectification diodes from an alternator and power it with a regular sensorless ESP. Then you've got the belt attachment ready to go. Also, obviously, you have to power the rotor field coils, which is annoying.
the hoverboard firmwares will put out debug info/telemetry on a serial port depending on the config you use. You can add a single ESP32 to glue together RC (crossfire)/bluetooth (wiimote)/hoverboard protocols as well as drive an OLED display for data, minor data logging, and a web interface for configuration if necessary. you get a lot of bang for the buck for just 1 ESP32. You also probably don't want to be using PWM because those boards are extremely noisy and the hoverboard firmware does not handle PWM noise very gracefully.
Theres a star hal sensor mod for these motors a youtuber made ...his was a bike and it almost doubled his speed . . a yellow bike.......I hope thats enough.......Thanks for your share admire your work.bless
To improve spring strength, could you nest the old springs inside the new ones? For suspension, could you make half of the swing arms swing back and the other half swing forward, so you have them preloading the same amount on the tracks in either direction of travel? That looks like a bit of work to re-weld the frame though. Those tracks look surprisngly good! Though there may be ways to improve them without adding *much* cost, could you use two layers of rubber, instead of overlapping them to join, you just offset the layers by half the length for the most engagement between the layers. You could also possibly put some kevlar cloth between the two layers to prevent stretching of the tracks, that stuff is damn stiff in tension! I hear you can improve power dissipation in those motors by either filling them with a light oil, so there is better thermal conductivity to the outside casing, or machineing vent holes in the side plates so they pump air through themselves
The springs are all the same size unfortunately. I could make some spacers to increase the preload though. If you want any reasonable speed over bumps, you can't have the swingarms swinging forward.
I had some issues with lag on the same transmitter so I wouldn't be surprised if that was the cause of the lag. One thing I would definitely do is increase the tension on the tracks to make sure it can actually reverse. You can see it loosening up quite a bit every time
The 'lag' is coming from the 'velocity' control method which uses the hall sensors to measure how far the motor has turned. When I let go of the stick, the momentum carries the tank a bit further, the controller senses this and corrects it, with a slight delay. You can see this at 32:19 where it even oscillates twice (right where I'm saying 'whoa..')
22:50 - just have the same adapter plate for the small holes, but put the big holes inside the small, right? As long as the small sprogit fits over "the bumpy thing", that should work I reckon.
The chain is about 8mm wide and needs to overhang the aluminium adapter plate, is the awkward part. I could make a second (small) adapter plate to attach to the first and allow the overhang, but I think it would then collide with the motor mount clamp. Might need to change the motor mount completely...
I think we're seeing parts of your house we haven't seen since the cats video? Cool project! Regarding CAD, didn't you design a computer aided physics design system? Probably doesn't prepare one very much for general cad use. I liked that program. Anyway, keep these videos coming! cheers /m
I had the same problem machining a slot in steel with a high speed spindle. What worked for me was to use a smaller end mill and do a trochoidal toolpath, full depth.
Nice ! But, the fixation of the damper make me a bit nervous : these "long" tubes on the thin square tube must have so many stress. A little diagonal bar would make it a lot stronger, specially if you have oil...
When the hull is finished there will be bars going across the top to connect those upright sections, so they will not look so vulnerable. Also keep in mind they are not taking direct shock from bumps in the ground, only 'slowed' shock via the spring.
Next goal after the modifications: Camera for FPV. Final step: Wrap it in black fur and add a speaker to it to play some snarling sounds. Enter: The doberman tank. 😉
For the welding u need to make a welding table so when u do some welding u clamp it down nice and tight to stop the warping from heat and for the tig tip file it down to a 60 degree angle and clean the alum with acetone or iso alcohol cause it needs to be a clean surface
I don't really understand this logic. If this thing weighed just 1kg do you think it would be less off-road capable? One of the advantages of tracks is you can spread weight over a larger area, the goal being to keep ground pressure low. More weight means more traction is required, seems like it would just cancel out (and use more battery for the same result). If the ground surface traction can be improved by compressing it (eg. snow) I suppose more weight might help.
10:52 Fireball Tool has a great explanation on how to remedy this for anyone wanting to avoid this in his video titled 'Why Do Good Welders Get This Wrong?'. Timestamp would be 5:25-11:41.
Don't weld like this _||_ , weld like this _/\_ if you want a strong butt weld. Stretchy track is also very bad and stronger spring tension is needed. Overall a good proof of concept
It seems as if the tracks get slacker as the spring are compressed more. Could you have some idlers on the top edge connected to the arms fixed to the bottom idlers such that the tension on the track is constant?
That was one of my earlier designs actually. In the end I wanted to keep the tracks straight along the top so that any kind of attachment could be bolted on without worrying about tracks getting bounced up into it... as much.
That's an interesting idea... I've never seen anything like that in practice, but it makes a lot of sense as long as there is clearance above the top of the track.
great build but i suspect that the batteries might be bad, they should last longer or it just pulls a lot of current, i bought a lot of old/broken hoverboards (for the motors to use on all kind of projects i wanted to make, like a go-cart, compact boat motor, electric scooter, ect..) but in my experience the battery packs were always bad, like a few cells in the pack were dead or out of balance, and since they are all in series that ruins the overall performance of the battery, they don't have balancers the bms is usually just overvoltage/undervoltage and overcurrent protection
Did you make a boddy for the tank. Ive got one just about finished (240kg), just looking around for some flysky information. From Lepperton New Plymouth.
I have been working on a hoverboard powered lawn mower for months and cannot find the right combination to get a working product. I'm currently using the hoverboard motherboard with firmware for pwm control. Using the right side rx tx wires for channel one and channel two on my RC receiver, (ELRS PWM) The best I can get it to do is partially work going forward and backwards, but it is very uncontrolled and....scary.. Any tips? I've read, watched 100's of videos multiple times and still haven't found that one thing I must be overlooking.
You need to mix the throttle and steering channels, usually each motor takes 50% of throttle, plus either 50% or -50% of steering depending on which side the motor is on. That will give you a skid steer behavior, which is also the same result as 'elevon mixing' on aircraft so search for that term for more details. You'll need a radio that can do channel mixings, which typically a pistol grip radio for cars will not have.
Привет, чтобы гусеница не проскакивала на ведущей звезде, нужно убрать амортизатор переднего колёса. Также в целом мне не очень нравится подвеска танка, там стоят слишком слабые пружины, их было бы неплохо заменить на более сильные. P.S. досмотрел до конца, увидел, что у тебя в планах уже есть замена пружин ;)
I think you should have built the box section chassis from steel. Cheaper, and you could MIG it together avoiding distortion. Aluminium extrusion is always in a stressed condition, welding it is never a good idea. ...it will distort...The extra weight would have helped traction too...Great effort though, a lot of work....🤔😳😏😀🇬🇧
Actually the aluminium is a few dollars cheaper than steel, at least at the two places I buy from. I like that aluminium is not prone to rust, it's clean and easy to cut/drill etc with wood tools and my CNC router. Not sure why more weight would help, the extra traction required would pretty much cancel out I think. One advantage of tracked vehicle is lowering ground pressure, not increasing it. In any case adding weight is easy, I would rather be adding more battery than just steel, and if the frame is lighter I can control the CG easier.
@@iforce2d Oh, I know why you chose Aluminium, you've machined it. In steel, your router would not have stood a chance. The difference in chassis weight would not have been huge using thin section box along with triangular gusset braces. You could have still used Aluminium Axle supports, but bolted them on instead. Same with the suspension swing arms...You need a pretty decent TIG welder to do aluminium as you know and could have avoided purchasing it. I'm surprised by the smooth rubber tracks working at all TBH. Usually tracks have treads on them for purchase going up inclines. I'm a retired mechanical Engineer, I used to design Laser Cutting machinery. The gearing on the twin motors needs changing a bit as you suggested, too much load on them. Should get more mileage then. Of course this tank needs FPV on it. Digital would be great if you used a directional VRX on a mast with all those hills you have....Interested to see where this project goes now. It's quite an achievement in such a short space of time...😮😏😏👍😀🇬🇧
It is better to take wheels for industrial equipment. Motor from e-bike wheel hub, with polyurethane gear (two-component polyurethane compound) - check first prototype th-cam.com/video/rK9K9qpkvLI/w-d-xo.html and more good chassis - th-cam.com/video/Bok7zft1sCA/w-d-xo.html
i just always dream to make a such project. I wandering , why you don't integrate in the back idler the hub motor (kind of direct drive) , this way no need anymore all gears and chains. the motors will be just fine, and fix your problem with staying under "floor" level.
A few reasons. The motors are 1/4 of the total weight so having them way out on one end might be bad for CG balance. Having the motors inside the drive wheel means there would be a limit on how small the drive wheel can be. The motors might not be waterproof. There would be no possibility for gearing change. How to attach the drive wheel (cog part) to the motor becomes a new problem, not as easy as my chainring method. Eventually the drive wheel will be HDPE so I could not simply weld it onto the motors.
Nice work, looks like something from robot wars 🙂 i'm wanting to do something like this but i don't how to fix the firmware, would you be happy to walk me through it ?
It's a Unimig "Viper TIG 180 AC/DC", seems to be also branded Xcel-Arc in some places. I wanted one that was intended for use on a regular household 10A outlet, which really reduces the options available. Not long after I started looking around, an (only slightly) used one showed up for a few hundred dollars less than the new price.
@@iforce2d what about manufacturer? The alternative firmware site says in its wiki that it's only compatible with certain main boards. Maybe you have a link to it to share?
I have no idea. I just find used hoverboards in online auction sites. I have never seen any manufacturer names, or model names, or numbers, or any other kind of identification anywhere on them. One time, the original box was included, but it also had no maker labeling or information whatsoever.
If you make an account on the OnShape site, this link should let you see my document, then you can clone it to make your own changes..... I think. cad.onshape.com/documents/c2147d8bbeec9d8f003dcec0/w/332726c019b224a2e58cb365/e/8966310bf69c61c66fddb412
just add two rollers per side with spring pressure upwards that the slack gets compensated. that would be my idea (similar to these support rollers: th-cam.com/video/KgCaZiriN5Y/w-d-xo.htmlsi=na8hDTdFFjTlzRsQ&t=157 ) or maybe stiffer springs on the front rollers. very nice job , the idea using hoverboard motors is instantly noted, cheap and powerfull these motors and sick job on the frame!
HARDER SPRINGS MEAN ONE OF THE MOST IMPORTANT ASPECT OF A TRACKED VEHICLE WONT BE UTILIZED. FOR THE TRACKS TO GO OVER STEEP OBSTACLES LIKE THE DOOR SILL, THEY NEED TO GO IN FOR EVERY WHEEL WHILE IT TOUCHES THE DOOR SILL TO ENSURE A LOT OF CONTACT IS MADE BY THE TRACK AND THE OBSTACLE TO ENABLE MOVEMENT. SO YOU EITHER NEED MORE WEIGHT OR SOFTER SPRINGS OR A MODERATED COMBINATION OF BOTH. ALSO YOU NEED TO ADD EITHER RUBBER OR METAL TRACKS ON THE OUTSIDE LIKE ON THE INSIDE TO ENSURE PROPER GRIP. THAT IS ANOTHER ESSENTIAL THING ABOUT TRACKED VEHICLES THAT IS MISSING. I'M SURE YOU HAD OTHER THINGS YOU WANT TO DO TO IT BUT I JUST THOUGH OF POINTING THESE THINGS OUT. I ALWAYS TYPE IN CAPS, MAKES MY COMMENTS STAND OUT.
I wanted an easy way to convert the motors, preferably without permanent change or welding, and only altering the back plate, and have possibility for gearing. Also, the motors are 1/4 of the total weight and I didn't want them hanging way out the back. But it's looking like I might need to change plans, the direct drive is looking more appealing now.
I didn't know you did build requests. In that case, you should build the Iron Man suit. First, we'll need a flux capacitor and some uranium. You start on the flux capacitor and I'll get the uranium. I know a couple of guys, but first I need to stop by a used pinball machine shop. 😂
I think you need either bigger springs or more preload on your front idler shocks to stop the tracks winding up when you reverse, in the clips you can see the idlers moving backwards and dropping the tension off the tracks. Otherwise fantastic project, reminds me of the bomb disposal robots I drove around for years.
His suspension is pointing ass end. This means any speed or traction will reduce the diameter of the tracks.
Yeah, the preload needs to be wound up quite a bit.
Pretty good traction for what looks to be a smooth band.
You could use a second set of sprockets towards the center to modify the gearing without changing your front sprockets. You'd just need to move either the motors or the driven sprockets to have the altered alignment so the chains could go towards the center then to the driven axle
I think it's called a counter shaft or jack shaft. Looks like the tank could use more weight towards the middle.
I have just been doing your box2D tutorials (which are amazing btw!) and came to check your channel out to see what other crazy things you do. Well, what I see here is not surprising whatsoever and makes me very happy!!! I am so glad you have found success in TH-cam and thank you for those amazing tutorials. Keep up the amazing content!
Thankyou for plowing all of YOUR money into these rad projects....I can't afford to do builds like this without REALLY wanting to save for a while and having a verified BoM in advance. You've been ahead of me from my multi wii days thru to today, making tricky things seem less daunting and even DOABLE for the over cautious like myself. Keep up the great work!
This thing is great! Hoverboard motors are pretty neat. I converted a Razor Dune buggy go kart to direct drive with 350w hoverboard motors and ran 84 volts through them. That setup pulled me (driving) and three kids in a bike trailer which was connected to the go kart! Motors never failed me even with all that abuse. They ran warm but we managed 23mph all day!
A very fascinating design, particularly the suspension. Love the CNC work, it like seeing art being made. Impressive performance, climbing hill, etc. I don't think this fantastic project is going to be put aside for long, as it sounded like you're having too much fun. ;)
Regarding the gearing ratio, to me looks like the rear cross brace is in the wrong spot to use larger gear. (45:35) It will need to move aft, or forward relative to the axil(s). Don't think there will be much bulge, just a bit in front of and half way to the motor. Using a nearer to 1:1 gear ratio would eliminate any bulge. The more complex option is bigger rear wheel sprockets (to match the larger gear size) and lowering axil to keep flat top. It would not impact suspension clearance height, but might mean moving rearward to allow the rear suspension arm to have full travel (an issue pointed out). A stronger spring on the rear arm might stop the bottoming out.
Workshop engineering at it's best. Awesome video. Subbed.
Nice setup. Been working on my own hover project, just a 4 wheel rover but having fun... looking forward to your next video
I smiled from beginning to end like an old crone gurning! Love the CNC drive sprockets etc. I did not think the MDF would be good for more than a few lightly loaded runs... they sure proved me wrong.
When reversing, the full torque of the motor is directly compressing the spring of the front sprocket. Don't increase the preload of that spring, but limit the travel of the front sprocket to the point where the belt teeth cannot skip. Something like tube or hose over the front shock absorber shaft should do it.
As you said, sorting the gearing is first priority. Plenty of options, but none that really stand out.
The amount of grip you are getting from the smooth tracks is amazing! I always assumed you had to have really agressive teeth (Edit: "treads").
This is obviously not a cheap project, but at this point it has relegated the Farm Rover to Fading Ghost status!
Love it, thanks.
Nice work. I suspect that it just takes the mounting of a Flight Controller to get you a call from Kiyv 😉
Lol, I wasn't expecting to see my name on the video, especially from a 3yo comment! I'm thrilled you make the hoverboard tracked vehicle a reality! They truly are a fantastic platform for robotics applications.
Just a couple notes & thoughts:
1. Make sure you tune the ADC readings for battery voltage. At least for me, I found that the replacement firmwares were triggering low-battery shutoffs even when the actual state of charge was fairly high. There is a procedure in the docs for calibrating those values.
2. Have you considered designing a replacement wheel/tire for the hub motor itself, allowing direct drive? The hoverboard motors are pretty well sealed out of the gate, and when I built of HB E-longboard I ran it though lots of muddy, puddles, and through the rain without any water ingress. The benefit here would be eliminating the chain/sprocket/drive axles/bearings entirely, allowing a much more easily 'waterproofed' system with less complexity.
Great video as always!
Yeah, after both runs the battery was at 35.5V which seems a little high, barely under storage charge.
I did notice some decent seals on the shaft, good to hear they are at least reasonably waterproof. After this first test I'm thinking more that direct drive might be the way to go. I'd like it to be fully amphibious so extra measures might be necessary, but it's probably an easier starting point than trying to waterproof a spinning shaft coming out of the hull.
@@iforce2d yeah definitely sounds like the calibration could be off, unless those HB batteries have high ESR and sag a lot under load. If it's helpful at all, I have discovered those HB main boards and motors run fantastic under a 13S battery. I have a cart I built using two sets (4WD) powered from a 13S Lion data center backup power pack rated for 100A cont. Discharge. Haven't killed a motor or main boards on it yet, so I think it can handle it safely. My HB longboard is a custom 12S battery and it runs great with that too.
Last note about the waterproofing, you might want to pop the seals off the shafts and grease them up. Every motor I have gotten has had the seal there, but be bone dry defeating the purpose and maybe even prematurely wearing out the seal. The grease will definitely help them.
Midway in the video I though it was going to be way bigger hehe. I have 4 of those motors on my skateboard. If you rewire them from star to Delta connection inside they become beasts! You loose a slight amount of torque but gain a multiple amount of power! So Delta connection with down gearing turns these motors into something usable in 1:1 size application. 4 of these motors would actually be enough power for something that you could sit on and take you many miles. My skateboard is more powerful than a moped and takes me about 15miles on asphalt. I run 4 vesc on my setup and 8 Makita 18V batteries at 36V working voltage. Giving the motors 56V and really geared down you could make something real useful with 4-6 of em! A ridable 6 wheeler would be awesome! The clue is motor efficiency, Delta is way more efficient and you dont really loose much energy in a gearbox vs twice the twice the winding resistance in the original star winding setup these comes with.
Beautiful video dude! Wow, so inspiring 😊
That was what I was thinking at the first test .. stiffer springs..like triple the stiffness... at least. all the motors energy gets dumped into the springs at slightest velocity change... luckily you pointed that out, so I don't need to add to the "useful comments list"
At this point quickly let me thank you very much for pissing away your money, you make really fascinating videos.. love them, keep going ! :)
"If only I had a couple of mobility scooter motors. Icould install a chair and ride it up and down the hills to retrieve errant aircraft".
Very nice, one or two things to be fettled but the principle is fine.
hey we muck around building and testing amphibs on the manukau harbour, really keen to see how this goes on the mud flats & mud-to-water threshold zone being so light. sing out if you need somewhere to test it when you get to that point.
9:30 I really enjoyed these segments across the video. I never really gave it much thought using a CNC router to etch a template for welding a frame. I have another tool added to my arsenal. Thank you very much.
Very impressive!! Great design work and your CNC is awesome. Something I need in my shop. I will throw in my $0.02 here. I think if your plastic track center pieces were made from solid rubber instead of 'U' shaped plastic, your tracks would not jump or slip. Anyway, love your work. Thanks for sharing.
Hey Chris, this could cost you a lot, the tracks alone. But I see you had some nice design challenges to solve. Re-gearing it would be better. Going backwards though, there seems not enough guidance from the top part of the track, maybe that's why the teeth are jumping at times? When the tension is the lowest it happens. That bump stop will help for sure, it looks very stable as well.
Thanks for this one!
Cheers 👍💪✌
Great project!
How wonderful, well done you
It's called a standing wave for when you were spinning the chain
Nice work! I know the struggle of things not working the first time. Hope you make another video of it with the new tracks
thats amazing and very cool
Great video. Eagerly waiting for mini hopper.🙃
A lot of real life tanks have the drive sprockets in FRONT and the tensioner wheels in back? Another thought is, you might be able to use Roller-blade skate wheels instead of the MDF ones, they are made of tough material and have precision bearings already? The track stretch seems like it will be a struggle until you get a better material to make tracks from, but you are onto something for sure! CHEERS!
when i left that comment i started hoarding those motors and control boards from almost free broken hoverboards but never got around using them, im pretty lazy.. now they piling up in my basement but thats exactly the project i had in mind lol big and strong enough to carry my groceries home
Brilliant build, I've built a six wheeled cargo robot of a similar size using hoverboard motors. Top tip (and I wish I knew it before I built mine) is that the width between the tracks needs to be wider than the tracks are long. This gives you a *lot* more turning torque, tanks get around this with brute force which is why they can be narrower. You could also increase the preload on the suspension so that by default the front or rear is higher off the ground
Suspension is 100% on the list for the v2...
th-cam.com/video/VlSwjEoBnWk/w-d-xo.html
Very cool!
I’m also building ex tanks but smaller and high speed. Achieving good yaw control is critical. I’m currently experimenting with gyros. I also think sensored motor control is also needed.
That chain sprocket setup really held that back I would go with direct drive... And I do believe the shocks would work better if they were a little less angled I suggest 10° more upright 5° may work.... To get it to go both ways over obstacles it might be good to have every other swing arm and shock opposite
When riveting soft materials use a washer under the rived head.
Impressive!
Wondered if you considered using chain as the drive to which one could attach the "now flexible and stretchy" rubber. Important just where one locates any "idler tensioner " as well, usually only on the off drive side of any pulley arrangement. Difficult /well-nigh impossible to do when one has reversing drive.
Well dang. I've been looking to do almost exactly this only I didnt think of the hoverboard. The best I'd gotten so far was scooter motor hubs, which are a lot more expensive for one than for whole hoverboard.
-edit
not for nothing, but if you want more high speed power at slightly (significantly) more cost, you can remove the rectification diodes from an alternator and power it with a regular sensorless ESP. Then you've got the belt attachment ready to go. Also, obviously, you have to power the rotor field coils, which is annoying.
Amazing
the hoverboard firmwares will put out debug info/telemetry on a serial port depending on the config you use. You can add a single ESP32 to glue together RC (crossfire)/bluetooth (wiimote)/hoverboard protocols as well as drive an OLED display for data, minor data logging, and a web interface for configuration if necessary. you get a lot of bang for the buck for just 1 ESP32. You also probably don't want to be using PWM because those boards are extremely noisy and the hoverboard firmware does not handle PWM noise very gracefully.
Nice project!!! 😃😃😃
Don't forget to put shock oil in those shocks! It will take out the springiness!
Theres a star hal sensor mod for these motors a youtuber made ...his was a bike and it almost doubled his speed . . a yellow bike.......I hope thats enough.......Thanks for your share admire your work.bless
The difference in that track between hot and cold length is probably enough to need more stiff travel in the front pilot wheel.
Can't wait for part 2 with more proper tracks and gearing. This will be a beast. Also please test how much extra load can it support?
It wasn't really meant for heavy load carrying, I would be happy with maybe 4-5 kg or so.
Dude keep working on this thing. Would be interesting to see some skid plates on it and some extra weight
To improve spring strength, could you nest the old springs inside the new ones?
For suspension, could you make half of the swing arms swing back and the other half swing forward, so you have them preloading the same amount on the tracks in either direction of travel? That looks like a bit of work to re-weld the frame though.
Those tracks look surprisngly good!
Though there may be ways to improve them without adding *much* cost, could you use two layers of rubber, instead of overlapping them to join, you just offset the layers by half the length for the most engagement between the layers. You could also possibly put some kevlar cloth between the two layers to prevent stretching of the tracks, that stuff is damn stiff in tension!
I hear you can improve power dissipation in those motors by either filling them with a light oil, so there is better thermal conductivity to the outside casing, or machineing vent holes in the side plates so they pump air through themselves
The springs are all the same size unfortunately. I could make some spacers to increase the preload though.
If you want any reasonable speed over bumps, you can't have the swingarms swinging forward.
I had some issues with lag on the same transmitter so I wouldn't be surprised if that was the cause of the lag. One thing I would definitely do is increase the tension on the tracks to make sure it can actually reverse. You can see it loosening up quite a bit every time
The 'lag' is coming from the 'velocity' control method which uses the hall sensors to measure how far the motor has turned. When I let go of the stick, the momentum carries the tank a bit further, the controller senses this and corrects it, with a slight delay. You can see this at 32:19 where it even oscillates twice (right where I'm saying 'whoa..')
Maybe the motors could be built into the rear drive wheels. Lower gearing, simpler, and would keep the whole deck flat
Looks pretty cool, I am trying to do a remote control lawnmower with a hoverboard motor wheels
22:50 - just have the same adapter plate for the small holes, but put the big holes inside the small, right?
As long as the small sprogit fits over "the bumpy thing", that should work I reckon.
The chain is about 8mm wide and needs to overhang the aluminium adapter plate, is the awkward part. I could make a second (small) adapter plate to attach to the first and allow the overhang, but I think it would then collide with the motor mount clamp. Might need to change the motor mount completely...
Very nice... I started one similar, but using conveyor chain (used in bottle mfg plants) for tracks. I need to get back to it one day?
I think we're seeing parts of your house we haven't seen since the cats video? Cool project! Regarding CAD, didn't you design a computer aided physics design system? Probably doesn't prepare one very much for general cad use. I liked that program. Anyway, keep these videos coming! cheers /m
I had the same problem machining a slot in steel with a high speed spindle. What worked for me was to use a smaller end mill and do a trochoidal toolpath, full depth.
I bought a lathe recently which I will try for this job next time, if there is a next time. The lathe can go much slower.
Nice !
But, the fixation of the damper make me a bit nervous : these "long" tubes on the thin square tube must have so many stress. A little diagonal bar would make it a lot stronger, specially if you have oil...
When the hull is finished there will be bars going across the top to connect those upright sections, so they will not look so vulnerable. Also keep in mind they are not taking direct shock from bumps in the ground, only 'slowed' shock via the spring.
Next goal after the modifications: Camera for FPV.
Final step: Wrap it in black fur and add a speaker to it to play some snarling sounds. Enter: The doberman tank. 😉
At least a few steps forward! I hope, in the end, your R&D will result in a R/C tank that also can achieve considerable speeds.
The current speed is about fast enough for me. I'm planning to prioritize mobility, not getting stuck, and amphibious capability.
For the welding u need to make a welding table so when u do some welding u clamp it down nice and tight to stop the warping from heat and for the tig tip file it down to a 60 degree angle and clean the alum with acetone or iso alcohol cause it needs to be a clean surface
The warping comes from the weld bead shrinking when it cools, you can't get around it just by clamping things in place.
more weight - more off-road capability
Tell that to Suzuki owners
nothing will help them
I don't really understand this logic. If this thing weighed just 1kg do you think it would be less off-road capable? One of the advantages of tracks is you can spread weight over a larger area, the goal being to keep ground pressure low. More weight means more traction is required, seems like it would just cancel out (and use more battery for the same result). If the ground surface traction can be improved by compressing it (eg. snow) I suppose more weight might help.
10:52 Fireball Tool has a great explanation on how to remedy this for anyone wanting to avoid this in his video titled 'Why Do Good Welders Get This Wrong?'. Timestamp would be 5:25-11:41.
Don't weld like this _||_ , weld like this _/\_ if you want a strong butt weld.
Stretchy track is also very bad and stronger spring tension is needed.
Overall a good proof of concept
It seems as if the tracks get slacker as the spring are compressed more. Could you have some idlers on the top edge connected to the arms fixed to the bottom idlers such that the tension on the track is constant?
That was one of my earlier designs actually. In the end I wanted to keep the tracks straight along the top so that any kind of attachment could be bolted on without worrying about tracks getting bounced up into it... as much.
That's an interesting idea... I've never seen anything like that in practice, but it makes a lot of sense as long as there is clearance above the top of the track.
great build but i suspect that the batteries might be bad, they should last longer or it just pulls a lot of current, i bought a lot of old/broken hoverboards (for the motors to use on all kind of projects i wanted to make, like a go-cart, compact boat motor, electric scooter, ect..) but in my experience the battery packs were always bad, like a few cells in the pack were dead or out of balance, and since they are all in series that ruins the overall performance of the battery, they don't have balancers the bms is usually just overvoltage/undervoltage and overcurrent protection
Yeah all these batteries I have are probably not very good, sometimes people sell these thinking they're broken when it's just the battery dying.
Did you make a boddy for the tank.
Ive got one just about finished (240kg), just looking around for some flysky information.
From Lepperton New Plymouth.
No, I haven't done anything further on this since making the video :)
I would love to buy/fab something similar to make moving my laser cleaning machine around easier/safer
I have been working on a hoverboard powered lawn mower for months and cannot find the right combination to get a working product. I'm currently using the hoverboard motherboard with firmware for pwm control. Using the right side rx tx wires for channel one and channel two on my RC receiver, (ELRS PWM) The best I can get it to do is partially work going forward and backwards, but it is very uncontrolled and....scary.. Any tips? I've read, watched 100's of videos multiple times and still haven't found that one thing I must be overlooking.
You need to mix the throttle and steering channels, usually each motor takes 50% of throttle, plus either 50% or -50% of steering depending on which side the motor is on. That will give you a skid steer behavior, which is also the same result as 'elevon mixing' on aircraft so search for that term for more details. You'll need a radio that can do channel mixings, which typically a pistol grip radio for cars will not have.
Привет, чтобы гусеница не проскакивала на ведущей звезде, нужно убрать амортизатор переднего колёса. Также в целом мне не очень нравится подвеска танка, там стоят слишком слабые пружины, их было бы неплохо заменить на более сильные. P.S. досмотрел до конца, увидел, что у тебя в планах уже есть замена пружин ;)
Excellent work! Would you consider making wheels out of PTFE? Would be easy to cut it out of an old chopping board in your router
I'm thinking maybe HDPE eventually, yes would be easy to machine if I can find large enough blanks somewhere.
If I asked for the schematic of the electronic parts, I could not find a healthy employee
Search google for "hoverboard hack feru" or just search here on TH-cam, there are many videos about it.
Amazing job!
Have you tried to use a 6V gasoline engine?😂😂
Many of those problem you had real tenks have as well.
turn motors sideways and use a herringbone gear on the axle to maintain flatness😎
I think you should have built the box section chassis from steel. Cheaper, and you could MIG it together avoiding distortion. Aluminium extrusion is always in a stressed condition, welding it is never a good idea. ...it will distort...The extra weight would have helped traction too...Great effort though, a lot of work....🤔😳😏😀🇬🇧
Actually the aluminium is a few dollars cheaper than steel, at least at the two places I buy from. I like that aluminium is not prone to rust, it's clean and easy to cut/drill etc with wood tools and my CNC router.
Not sure why more weight would help, the extra traction required would pretty much cancel out I think. One advantage of tracked vehicle is lowering ground pressure, not increasing it. In any case adding weight is easy, I would rather be adding more battery than just steel, and if the frame is lighter I can control the CG easier.
@@iforce2d Oh, I know why you chose Aluminium, you've machined it. In steel, your router would not have stood a chance. The difference in chassis weight would not have been huge using thin section box along with triangular gusset braces. You could have still used Aluminium Axle supports, but bolted them on instead. Same with the suspension swing arms...You need a pretty decent TIG welder to do aluminium as you know and could have avoided purchasing it. I'm surprised by the smooth rubber tracks working at all TBH. Usually tracks have treads on them for purchase going up inclines. I'm a retired mechanical Engineer, I used to design Laser Cutting machinery. The gearing on the twin motors needs changing a bit as you suggested, too much load on them. Should get more mileage then. Of course this tank needs FPV on it. Digital would be great if you used a directional VRX on a mast with all those hills you have....Interested to see where this project goes now. It's quite an achievement in such a short space of time...😮😏😏👍😀🇬🇧
It is better to take wheels for industrial equipment. Motor from e-bike wheel hub, with polyurethane gear (two-component polyurethane compound) - check first prototype th-cam.com/video/rK9K9qpkvLI/w-d-xo.html
and more good chassis - th-cam.com/video/Bok7zft1sCA/w-d-xo.html
i just always dream to make a such project. I wandering , why you don't integrate in the back idler the hub motor (kind of direct drive) , this way no need anymore all gears and chains. the motors will be just fine, and fix your problem with staying under "floor" level.
A few reasons. The motors are 1/4 of the total weight so having them way out on one end might be bad for CG balance. Having the motors inside the drive wheel means there would be a limit on how small the drive wheel can be. The motors might not be waterproof. There would be no possibility for gearing change. How to attach the drive wheel (cog part) to the motor becomes a new problem, not as easy as my chainring method. Eventually the drive wheel will be HDPE so I could not simply weld it onto the motors.
Nice work, looks like something from robot wars 🙂
i'm wanting to do something like this but i don't how to fix the firmware, would you be happy to walk me through it ?
There's a link in the description, see the 'getting started' pages there.
ola o projeto ficou muito bom
What're your thoughts on prototyping the belts out of 3d printed tpu?
I think they would break in five seconds :D
It's solid. Could you possibly fit the wheel motor right inside the driven cog and avoid the shafts, chains, bearings, etc?
Quicker, cheaper, lighter, better reliability... as long as the gearing is acceptable, what's not to like?!
Don't be annoyed if it cracks ,best to preheat aluminium 😢
What welder did you choose?
Thanks.
It's a Unimig "Viper TIG 180 AC/DC", seems to be also branded Xcel-Arc in some places. I wanted one that was intended for use on a regular household 10A outlet, which really reduces the options available. Not long after I started looking around, an (only slightly) used one showed up for a few hundred dollars less than the new price.
@@iforce2d thanks
Did you check the shocks for oil, looks like they are dry going by the rebound rate
They don't have any oil in them at this stage... he mentioned it at 4:27.
@@PiefacePete46 ok I must of missed that
Instead of paying that for the track look at what pbcway will charge for TPU printing they may even sponsor the prints
I really would like to know, which exact hoverboard got used
It's a "6 inch" size, as far as I can tell they are all functionally equivalent.I'm not aware of any model names or numbers anyway.
@@iforce2d what about manufacturer? The alternative firmware site says in its wiki that it's only compatible with certain main boards. Maybe you have a link to it to share?
I have no idea. I just find used hoverboards in online auction sites. I have never seen any manufacturer names, or model names, or numbers, or any other kind of identification anywhere on them. One time, the original box was included, but it also had no maker labeling or information whatsoever.
Somebody might have mentioned it, but how about a planetary "final drive"?
Do you share the models of the components?
If you make an account on the OnShape site, this link should let you see my document, then you can clone it to make your own changes..... I think.
cad.onshape.com/documents/c2147d8bbeec9d8f003dcec0/w/332726c019b224a2e58cb365/e/8966310bf69c61c66fddb412
Always v prep a but weld,,, but ya welds aint the worst ive seen😊
just add two rollers per side with spring pressure upwards that the slack gets compensated. that would be my idea (similar to these support rollers: th-cam.com/video/KgCaZiriN5Y/w-d-xo.htmlsi=na8hDTdFFjTlzRsQ&t=157 ) or maybe stiffer springs on the front rollers.
very nice job , the idea using hoverboard motors is instantly noted, cheap and powerfull these motors and sick job on the frame!
HARDER SPRINGS MEAN ONE OF THE MOST IMPORTANT ASPECT OF A TRACKED VEHICLE WONT BE UTILIZED. FOR THE TRACKS TO GO OVER STEEP OBSTACLES LIKE THE DOOR SILL, THEY NEED TO GO IN FOR EVERY WHEEL WHILE IT TOUCHES THE DOOR SILL TO ENSURE A LOT OF CONTACT IS MADE BY THE TRACK AND THE OBSTACLE TO ENABLE MOVEMENT. SO YOU EITHER NEED MORE WEIGHT OR SOFTER SPRINGS OR A MODERATED COMBINATION OF BOTH. ALSO YOU NEED TO ADD EITHER RUBBER OR METAL TRACKS ON THE OUTSIDE LIKE ON THE INSIDE TO ENSURE PROPER GRIP. THAT IS ANOTHER ESSENTIAL THING ABOUT TRACKED VEHICLES THAT IS MISSING. I'M SURE YOU HAD OTHER THINGS YOU WANT TO DO TO IT BUT I JUST THOUGH OF POINTING THESE THINGS OUT. I ALWAYS TYPE IN CAPS, MAKES MY COMMENTS STAND OUT.
DO YOU SHOUT A LOT WHEN YOU ARE TALKING TO PEOPLE?
Why not use the motors to directly drive the tracks?
I wanted an easy way to convert the motors, preferably without permanent change or welding, and only altering the back plate, and have possibility for gearing. Also, the motors are 1/4 of the total weight and I didn't want them hanging way out the back. But it's looking like I might need to change plans, the direct drive is looking more appealing now.
I didn't know you did build requests. In that case, you should build the Iron Man suit. First, we'll need a flux capacitor and some uranium. You start on the flux capacitor and I'll get the uranium. I know a couple of guys, but first I need to stop by a used pinball machine shop. 😂
Not for beginners😢
hope you're melting all the left over alloy???
There is a big pile of it in my garage that says this to me every time I walk past it, yeah probably will some day.