Rebuilding my WASD-controlled tank

Over two months after completing those drive units I finally put them in the tank, fix up the electronics, and make this thing drive.
You can find more of my work here:
/ creo_nova
www.creonova.nl/

[@umitdemir2873]

This is pure engineering

[@antalz]

Thanks a lot mate!

[@dannyperry8070]

Thank you for your time. Nice video

[@antalz]

Thanks for checking out the rest of the channel, appreciate it!

[@dannyperry8070]

@@antalz The pleasure is mine and I appreciate all the time you put in making your videos.

[@erikduvefjard6149]

Interesting project! you cold maybe try to put a MC leadbattery in the back. That will allso give you more power.

[@antalz]

Thanks! I don't think lead-acid or LiFePO will outperform the LiPo battery I have in there. It's 24V, 5Ah, and weighs under 1kg. It can discharge 125A continuous and 250A burst. I think some lead-acids have higher peak currents but certainly not better capacity. They're generally also really have.

[@erikduvefjard6149]

@@antalz But they could help with the wheight for better balance and they normally give better Ah for less money. Most RC tanks i have seen uses them for this reasons. And by my expensive experience you will need much thicker cables for those 250A bursts :-D

[@antalz]

@@erikduvefjard6149 Yeah if I want to go for really long runtime then getting better Wh/$ would be pretty sweet. I'm not sure what to do there yet, adding to much weight to the back gives it a harder time turning, but maybe that's less of a problem on terrain vs this vinyl flooring? And yeah the cables can't handle 250A at all, but the motors only take 26A each at stall. The cables going to the motor controller are 1.5mm2, so that's kinda marginal but good enough I think.

[@dzee9481]

I would agree to have your batteries in the back of the tank. You might also want to shift the electronics more to the back as well. I know it is not much weight but every little bit counts. As far as the wires go for the motors so long as you have 18 gauge or better the resistance would be negligible.

[@DrJeff-]

Antalz this is sick! Amazing stuff

[@antalz]

Thanks a lot mate!

[@MauricioHernandez-de8is]

Awesome tank! A lot of effort, congrats. Looks very robust, hope to see soon a vídeo of 3d printed planetary gears

[@antalz]

Thanks mate, it did take a lot of work indeed, but I learned a lot along the way.

[@platinums99]

half the control board and double stack it right at the back? it looks like a heavy piece of MDF. Have you considered Carbon FIber? :D I hear LED's make things to faster also.

[@TheHypaaa]

Why not print some spacers for the shock absorbers in the front so that they have a higher force, shouldn't that work to raise the front? Or simply adjust them if they are adjustable ones :)

[@antalz]

They are adjustable, if you have a keen eye you'll see that roadwheel station 2 is adjusted to be stiffer than stations 3-5. I think if I print spacers I'm essentially preloading the suspension, which should at some point be enough, but it will also reduce the available travel. My current mental model for springs is that they have a force and a travel, and force*travel = work, which is a number I can't change. So if I want to make an existing spring stiffer all I can really do is make its travel shorter. Maybe that mental model isn't accurate.

[@TheHypaaa]

@@antalz Right of course, if you need all the travel that is

[@antalz]

@@TheHypaaa Good point. They only have ~38mm of travel, and the design intent was to use all of that, but I think the current vehicle can't really get near that far anyway. I think I'll leave it as it is, unless it completely fails apart on the outside testing.

[@aptech2933]

Cool project, What's the body made of?

[@antalz]

Thanks, the body is all MDF. 22mm thick on sides, rear and top, 12mm on the bottom, 30mm lower front, 34mm upper front. Painted with some kind of outdoors paint I got from the hardware store, MDF is vulnerable to moisture so that's pretty important.

[@AwestrikeFearofGods]

Try preloading (extending) the middle wheels a bit more. This will put more load on the middle wheels, and less load (and traction) on the front and back wheels. Track under the middle wheels experiences less scrub while turning in place. By "middle wheels" I mean the row of wheels that are longitudinally closest to the cg of the tank, so it will probably be row 2, row 3, or a bit of each.

[@antalz]

That's a really smart idea. I'm not sure if I can pull it off on this hull, because I would have to lower the suspension anchors, and those are glued in place. I'm currently thinking I'll make new suspension units for the front and then move on. Not putting so much weight on the outside wheels is definitely a lesson for the next robot.

[@AwestrikeFearofGods]

@@antalz You might be able to purchase a spring with longer free-length, that fits your application. Compressing the spring far enough will result in greater preload, even if the spring rate is a bit lower. Preload = Rate x Change of Length. You can also try plastically stretching the length of the springs you have. Alternatively, you can trim the front and back rows, which shortens their free length, but increases their rate.

[@jmcarp0]

Love the tank! Have you considered 3D printed pendulum clocks? Very few out there run for 8 days. I found one that used steel tubes and piano wires for bearings. He claimed an 8 day clock with only 3D printing wasn't likely. What do you think?

[@antalz]

Thanks! I'm afraid I never considered those no. I don't think 8 days is a hard limit per se, it's a matter of how much energy you store and how much friction losses you have. Minimizing friction becomes very hard once you've taken care of the obvious things, but increasing the stored energy is quite simple in principle. It will make your clock bulky though.

[@synterr]

You can use VNC, to see raspberryPI desktop on your computer. There is no need for LCD :)

[@antalz]

The screen is there for if the WiFi connection fails for whatever reason, or if the Pi fails to boot into the operating system. Sometimes it's also handy to check the IP adress of the Pi. Put differently if you can't log in via VNC for any reason, without a screen you're kinda in the dark.

[@synterr]

@@antalz Definitelly true, very often IP can change. Thank You for such great channel. I've learned a lot. Cheers!

[@gregorydrygas298]

Please, add torsion bars and tracks drive sync!

[@antalz]

I did a small test with torsion bars, but the spring rate was extremely low when using a bar of 3D printed ninjatek cheetah, a 95A filament. I think using a rigid filament like PLA would run into creep problems. So I'm still not sure how exactly to do it. If you know more, let me know. What do you mean by tracks drive sync?

[@gregorydrygas298]

@@antalz Plastic torsion bars can be hard to do. You are using 2 motors to drive tracks. It is fine when you are on a flat surface but once your suspension will start to work, you will have a problem to maintain a straight line due to the difference in track speed (actually the distance that track has to travel). This is not a problem when you have a driver who can adjust the speed of tracks accordingly to conditions but with software it might be tricky.

[@antalz]

@@gregorydrygas298 I think fixing the latter should be done electronically. Either by attaching encoders to the shafts, or by using a gyroscope or GPS to drive straight using feedback. I think a mechanical system would be more complex to build. I'm still open to the torsion bar idea, but I'm currently working on some other stuff. Rebuilding the suspension is a lot of work, if I do that I might as well build a new hull to and then I have a second tank.