[EV3] Big and Light Tank Climber #12

[oracid]

My goal was to make a robotic structure based on a tank frame.

This tank should be sufficiently large and especially light enough to climb stairs, whatever they are, and even those with nosing. These imposing a great height, see my previous videos.

With a length of 85cm and a weight of 2.5kg, I feel I have reached my goal.

Furthermore an additional charge of 1kg is possible and will allow additional batteries or any component for a particular application. This 1kg load is a maximum as you can see in my video, it struggles to climb stairs.

For the tracks, I used D joint (window joint) as shown in my previous videos.

Thank you for your feedback.

Settings and detail components with their weight:

Length 85cm

Width 48cm

Height 22cm

Chassis Weight 1300g

EV3 + 165g battery 122g

4 large servos 82g x 4 = 328g

Infrared receiver 20g

Cables 44g

Track with D joint 272g x 2 = 544g

Front tank angle 35°

[syclone]

Wow, it's really enormous!

[Attika]

Impressive size. I guess you've been experimenting enough. Did you make an attempt with a geared down version? I assume the plastic itself sets the limit of the weight it can carry. Amazing work. Grat.

Edited August 29, 2016 by Attika

[oracid]

Impressive size. I guess you've been experimenting enough. Did you make an attempt with a geared down version? I assume the plastic itself sets the limit of the weight it can carry. Amazing work. Grat.

I am not sure to understand.

You can see the tank going down, but backward, at 0'58.

Yes, it can going down forward. But not easy to make short video and to show everything.

I don't think the problem is the plastic, but servo power.

[Attika]

I think I wasn't clear enough. Sorry. I meant if you were using not a straight connections from the motor to the wheel but a pair of gears (8/24 or 12/20) what slows it down when giving more torque. This way probably it could carry more weight on the way up. I just wasn't sure if the material itself is capable to take more without suffering any damage. I never built in this size so I have no experince. Don't get it wrong it isn't criticism but curiosity.

[oracid]

I think I wasn't clear enough. Sorry. I meant if you were using not a straight connections from the motor to the wheel but a pair of gears (8/24 or 12/20) what slows it down when giving more torque. This way probably it could carry more weight on the way up. I just wasn't sure if the material itself is capable to take more without suffering any damage. I never built in this size so I have no experince. Don't get it wrong it isn't criticism but curiosity.

Yes of course, the power can be increased by reducing the speed, it's a choice.

We can do it with some more gears or use a smaller sprocket, but I do not know if it works, I've never tried, I do not have it, http://www.bricklink.com/v2/catalog/catalogitem.page?P=57520&name=Technic%20Tread%20Sprocket%20Wheel%20Small&category=%5BWheel%5D#T=C

For the size, I think if there were larger beams sizes this could solve the gamme (may be not the good word, sorry) tied to the number of parts.

Thank you for those excellent remarks.

Edited August 29, 2016 by oracid

[MRP_]

very impressive! if you''re having issues with strength, I recommend using studded beams, I've found them to be very strong, plus you can add plates to them for even more reinforcement! just if you want to go even larger...

[TheMindGarage]

Nice! Doesn't look that strong, but it really is!

Impressive size. I guess you've been experimenting enough. Did you make an attempt with a geared down version? I assume the plastic itself sets the limit of the weight it can carry. Amazing work. Grat.

Probably not. I've managed to make an EV3-ised 42043 set with four driven axles, all driven by motors. Used loads of gears in the drivetrain (plus U-joints), but drove fine despite weighing about 5kg.

[Kristof]

Impressive!

Do you have any plans with it or is it just experimental design to find out whether it's possible or not to build such climbler? Would be wonderful if this evolved into some robot that it able to operate on multiple floors :)

Dedpending on your budget, I think that the most effective option to increase the performance is to double up the propulsion motors - I have seen guys doing that with these monstrous mindstorms projects, like full size wheelchair or go cart.

[oracid]

just if you want to go even larger...

If I made a tank of this size, it is for him to climb any type of staircase and not specifically to make it big.

Impressive!

Do you have any plans with it or is it just experimental design to find out whether it's possible or not to build such climbler? Would be wonderful if this evolved into some robot that it able to operate on multiple floors :)

Dedpending on your budget, I think that the most effective option to increase the performance is to double up the propulsion motors - I have seen guys doing that with these monstrous mindstorms projects, like full size wheelchair or go cart.

My goal was to show people interested in robotics, it is possible to make a real climber with few resources.

Like everyone else, I read the book of Sariel, but if you have a schematic showing me how to use 8 servos with EV3, I will like to know how.

I do not think I will go further in this domain. My goal is reached.

[TheMindGarage]

My goal was to show people interested in robotics, it is possible to make a real climber with few resources.

Like everyone else, I read the book of Sariel, but if you have a schematic showing me how to use 8 servos with EV3, I will like to know how.

I do not think I will go further in this domain. My goal is reached.

You can just double-up the servos - for each servo currently on the tank, add an extra one hard-coupled next to it. You'll need two EV3 bricks though, and you'll need to connect them with Bluetooth. I don't own multiple EV3s, so I don't know how to do it, but I've heard that it's relatively easy.

[Kristof]

My goal was to show people interested in robotics, it is possible to make a real climber with few resources.

Like everyone else, I read the book of Sariel, but if you have a schematic showing me how to use 8 servos with EV3, I will like to know how.

I do not think I will go further in this domain. My goal is reached.

Fair enough. I was just curious :) It's still great feat of technic engineering, the chasis and everything. I wonder what will be your next project.

[oracid]

You'll need two EV3 bricks though...

I think it would be too heavy. I already made a version with 8 XL motors, unfortunately it is also too heavy and quite complicated.

I wonder what will be your next project.

Good question !

I hesitated because I would not disappoint. I am looking for a domain where I could bring my stone.

[TheMindGarage]

I think it would be too heavy. I already made a version with 8 XL motors, unfortunately it is also too heavy and quite complicated.

From your numbers, an extra EV3, 4 Large motors, IR sensor and set of cables would add 679g. The extra structure to support the extra motors and brick can't possibly take this over 1kg. Double the power at the cost of a 28% weight increase seems reasonable as long as the tracks have enough grip.

[oracid]

From your numbers, an extra EV3, 4 Large motors, IR sensor and set of cables would add 679g. The extra structure to support the extra motors and brick can't possibly take this over 1kg. Double the power at the cost of a 28% weight increase seems reasonable as long as the tracks have enough grip.

Good thinking, but I think we can do better with that, http://www.mindsensors.com/ev3-and-nxt/21-multiplexer-for-nxtev3-motors

May be one day, I will do that.

[Kristof]

^^ 55 bucks for a splitter... I mean, I know this is probably very exclusive, though this really feels that I'd rather hack solder my own crude version :D

[oracid]

^^ 55 bucks for a splitter... I mean, I know this is probably very exclusive, though this really feels that I'd rather hack solder my own crude version :D

Do you mean you had or you can make a multiplexer like this one ?

[Kristof]

No, it was rather exaggeration. But realistically speaking, if I really needed to double up the motors, I might have taken the challenge to bastle my own splitter, perhaps with on board power amplifier. Such thing could be rather simple. Programmable multiplexer like this is far from simple though and I'd have hard time building such thing, certainly spending much more than $55 with an insecure result :)

[oracid]

No, it was rather exaggeration. But realistically speaking, if I really needed to double up the motors, I might have taken the challenge to bastle my own splitter, perhaps with on board power amplifier. Such thing could be rather simple. Programmable multiplexer like this is far from simple though and I'd have hard time building such thing, certainly spending much more than $55 with an insecure result :)

Here is a module for connecting 8 servos.

But are not LEGO servos which have the advantage of being very powerful and cheap.

http://www.mindsensors.com/ev3-and-nxt/25-8-channel-servo-controller-for-nxt-or-ev3