[MOC] RWD Car with Mechanical Traction Control

[896gerard]

Four wheel drive on Lego Technic cars with small wheels is hard to realize, as they become relatively too wide (which is ugly) and the ground clearance is also compromised.

The problem then is: how to get the same grip back with only rear wheel drive? Then you need to push something 'extra' in the back of your already crammed, small nice-looking vehicle...

This concept is an approach to keep this 'something extra' as small as possible while giving awesome and unexpected grip to the car. (At least, I was astonished by what it achieved!)

Instead of typing a lot, I've put it all into a video with the theory of how it all works at the back of the video to let the 7 billion minus 1000 viewers that are not interested into more advanced Technic stuff experience how it works in practice on a 2 cm (that is: inchy) thick ice floor.

As for the car, it is built from Power Functions spares as all my L-motors (which have quickly become the main workhorse for every Technic builder) are ehm... in use. That is why you find me using a geared up XL motor, which gave me memories of the good old times when the poor knob wheel parts splattered around.

Sharp images of the car and mechanism explanation can be found on: http://www.brickshel...ry.cgi?f=564319.

On MocPages (beware, there is a piece of code there that simply destroys anything such as image resolution on .png files...) http://www.moc-pages.../moc.php/428881

Video:

I finally have found PovRay so (with many thanks to all those who make that software) I can present a high-quality render:

Edited May 17, 2016 by 896gerard

[Jurss]

brickshelf says - This folder is not yet public.

[suffocation]

This is brilliant! I'd love to see a complete model with your mechanism built into it.

[nerdsforprez]

brickshelf says - This folder is not yet public.

The mocpages photos are no too bad. I went there for the pictures. I love the site for some things.... for others, hmm... not so much.

@896gerard - wonderful project. The video accurately show how the mechanisms work.......and IMO, the showing of the difference performance when the mechanism is and is not engaged is very impressive.....

[Vectormatic]

Cool solution, but this is more diff-lock then traction control right? Would it be possibly to build a mechanism which locks the diffs is the speed difference between both output shaft becomes too much?

[Victor Imaginator]

Wow, incredibly cool mechanics) Thanks for sharing.

[Didumos69]

If I understand correctly you are forcing a fixed speed ratio between the left and right rear wheel. This works very well on ice, but on a surface with sufficient friction (a road) bypassing the differential this way will cause tyre scrub and windup.

The idea is nice, but it would be even better to make a system that controls torque ratios rather than speed ratios. You could do this by combining a limited slip differential with a 3-speed gearbox on one of the output axles. The gearbox ratios should be symmetrical, for instance 3:1, 1: 1 and 1: 3. Now if the limited slip differential confines left-right torque ratios to [2:1 - 1:2], then you can force this to change to [6:1 - 3:2] or to [2:3 - 1:6] using the gearbox.

Edited May 17, 2016 by Didumos69

[MRP_]

So it's like an adder/subtractor? That's cool, I've never seen one used in a car before... Interesting!

[896gerard]

If I understand correctly you are forcing a fixed speed ratio between the left and right rear wheel. This works very well on ice, but on a surface with sufficient friction (a road) bypassing the differential this way will cause tyre scrub and windup.

The idea is nice, but it would be even better to make a system that controls torques ratios rather than speed ratios. You could do this by combining a limited slip differential with a 3-speed gearbox on one of the output axles. The gearbox ratios should be symmetrical, for instance 3:1, 1: 1 and 1: 3. Now if the limited slip differential confines left-right torque ratios to [2:1 - 1:2], then you can force this to change to [6:1 - 3:2] or to [2:3 - 1:6].

I like your idea very much... but with the coming of that gearbox goes the simplicity and the smallness of my solution.

And Technic builders like smallness as a lot because they fit other systems as well, and these usually need to be exactly 1 micrometer above the rear differential...

In every corner, a car has a fixed speed ratio between the inner and outer wheels. Normally, the differential sorts that difference out for you but on ice that is gone. So if you can replicate this exact ratio with the limited Lego gear range, there are no problems at all.

Cool solution, but this is more diff-lock then traction control right? Would it be possibly to build a mechanism which locks the diffs is the speed difference between both output shaft becomes too much?

Well, there is some 'control', the wheels spin on defined speeds where with an open diff they would spin on undefined speeds.

I like all the ideas here on Eurobricks, maybe a great version 2?

Yes, that would maybe work, but as soon as you use more than 2 differentials the rear end of your 'car' is becoming elephanty. And then it's better to go for the good old 4WD...

Edited May 17, 2016 by 896gerard

[Didumos69]

I like your idea very much... but with the coming of that gearbox goes the simplicity and the smallness of my solution.

And Technic builders like smallness as a lot because they fit other systems as well, and these usually need to be exactly 1 micrometer above the rear differential...

In every corner, a car has a fixed speed ratio between the inner and outer wheels. Normally, the differential sorts that difference out for you but on ice that is gone. So if you can replicate this exact ratio with the limited Lego gear range, there are no problems at all.

You are right. This is a compact solution and very suitable for ice or other slippery surfaces. More advaced solutions that would also work on normal roads would grow much bigger.

Edited May 17, 2016 by Didumos69

[Lipko]

SBrick with contoller app. for calculating proper speeds+one motor per side.

[896gerard]

SBrick with contoller app. for calculating proper speeds+one motor per side.

I know, I know.. the good old proper Technics are being replaced in a high pace by the everything-able-to-do-Sbrick. But let's not get lazy, right?

[Lipko]

I know, I know.. the good old proper Technics are being replaced in a high pace by the everything-able-to-do-Sbrick. But let's not get lazy, right?

That's technology. Gives room for other, possibly more interesting (mechanical) features.

[zux]

I don't see how engaging switch from one side to another makes other wheel spin faster while gears for both sides are the same. Unless that extra axle is driven as well?

[Lego Nerd]

I don't see how engaging switch from one side to another makes other wheel spin faster while gears for both sides are the same. Unless that extra axle is driven as well?

Yes, the other axle is driven, at an increased speed relative to the normal drive axle. (His video shows this)

[zux]

Oh, my bad then. Didn't catch this while viewing the video.

[__________________________]

I don't think I have ever seen this function before. I think this is going to break Eurobricks Technic

[Didumos69]

I can't help thinking about how to make this suitable for normal surfaces as well. If you could somehow put a 24 tooth clutch between each of the diffs out axles and the extra input axle, then the left and right out axles would have a chance to escape windup. That way this might also work on a surface with friction.

[896gerard]

I can't help thinking about how to make this suitable for normal surfaces as well. If you could somehow put a 24 tooth clutch between each of the diffs out axles and the extra input axle, then the left and right out axles would have a chance to escape windup. That way this might also work on a surface with friction.

Well, I didn't just post this to say: 'Now this is the final solution' but more: 'This is the first step, now the whole internet can start improving it'... So I'm fully happy if this actually is improved.

[Didumos69]

Well, I didn't just post this to say: 'Now this is the final solution' but more: 'This is the first step, now the whole internet can start improving it'... So I'm fully happy if this actually is improved.

Something like this (with the top-right axle being the extra input axle):

[Kristof]

^The clutch gear is rather asking for troublesome performance. It doesn't have friction low enough to allow for smooth speed rate change unless you use some ultra grippy wheels or something.

Anyway, I'd look at this as on some auxilary systems you engage if the surface is slipery :) All in all, if the grip is good, rear wheel drive with diff is usually good enough. I think that any sophisticated mechanism that would precisely set the speed ratio according to the turning radius would only add pointless bulk to the cmall car :)

... but it would be really interesting if someone constructed purely mechanical way of doing this. I can't think of any at the moment. The problem is that steering radius is continuous while any classical type of gearbox only provides discrete number of ratios. Maybe some continual gearbox then but yet again, that's difficult to do with lego on its own.

[896gerard]

Something like this (with the top-right axle being the extra input axle):

Nice.. but I never trust clutch gears. Those are huge forces, note that they are used to turn a complete car! And the first things that will give up, are clutch gears (I think they'll also wear out but that is for another topic).

^The clutch gear is rather asking for troublesome performance. It doesn't have friction low enough to allow for smooth speed rate change unless you use some ultra grippy wheels or something.

Anyway, I'd look at this as on some auxilary systems you engage if the surface is slipery :) All in all, if the grip is good, rear wheel drive with diff is usually good enough. I think that any sophisticated mechanism that would precisely set the speed ratio according to the turning radius would only add pointless bulk to the cmall car :)

... but it would be really interesting if someone constructed purely mechanical way of doing this. I can't think of any at the moment. The problem is that steering radius is continuous while any classical type of gearbox only provides discrete number of ratios. Maybe some continual gearbox then but yet again, that's difficult to do with lego on its own.

Well, if the mechanism has to become sophisticated then why not fit simply 4WD instead... Anyway, your method of measuring the error precisely (mechanically) would maybe not succeed, as the wheels have no grip so they do not deliver an accurate measurement. And remember, there's nothing wrong with the speed ratio being a fixed ratio, as the distance between front and rear axle and the turning circle do not change for the complete car lifetime. The problem is: getting it exactly right and that might be difficult to realize with the limited number of Technic gears. Maybe using one more differential would do the job but this is very difficult stuff to think of. And I like building creations as much as doing ideas like this so now it's time for a lot of MOCs again..

[Didumos69]

^The clutch gear is rather asking for troublesome performance. It doesn't have friction low enough to allow for smooth speed rate change unless you use some ultra grippy wheels or something.

Anyway, I'd look at this as on some auxilary systems you engage if the surface is slipery :) All in all, if the grip is good, rear wheel drive with diff is usually good enough. I think that any sophisticated mechanism that would precisely set the speed ratio according to the turning radius would only add pointless bulk to the cmall car :)

Nice.. but I never trust clutch gears. Those are huge forces, note that they are used to turn a complete car! And the first things that will give up, are clutch gears (I think they'll also wear out but that is for another topic).

I agree with krisandkris12 that the whole idea could best be seen as an auxiliary system you engage when the surface is slippery. In that case the clutch gears make no sense, as there will be no windup on a slippery surface. And if you would incorporate clutch gears, I also agree that their operation will be far from smooth.

Edited May 19, 2016 by Didumos69

[Kristof]

Well, if the mechanism has to become sophisticated then why not fit simply 4WD instead...

You tell me :D But no, actually I only ment that it would be interesting to even see whether such mechanism can be constructed. I think it can't be really practical for a lego technic car.

Anyway, your method of measuring the error precisely (mechanically) would maybe not succeed, as the wheels have no grip so they do not deliver an accurate measurement.

Yep, very likely.

And remember, there's nothing wrong with the speed ratio being a fixed ratio, as the distance between front and rear axle and the turning circle do not change for the complete car lifetime..

Not sure whether you are right here - depends on whether your system only engages when you steer at maximum angle. If so, then it's non conflict. But if the system engages earlier (which I think it might), then you have some (although maybe quite small) range of steering angles for which the back wheels turn in the same fixed ratio and that is conflict. because the turning changes within this range.

Either way, this is just about being rigorous in theory. In praxis, your system proves to work and even if there is some slack it's still great piece of mechanics :)

Edited May 19, 2016 by krisandkris12

[896gerard]

Not sure whether you are right here - depends on whether your system only engages when you steer at maximum angle. If so, then it's non conflict. But if the system engages earlier (which I think it might), then you have some (although maybe quite small) range of steering angles for which the back wheels turn in the same fixed ratio and that is conflict. because the turning changes within this range.

Either way, this is just about being rigorous in theory. In praxis, your system proves to work and even if there is some slack it's still great piece of mechanics :)

You are right about the steering angle... but actually my prototype system was designed to activate the extra traction control system only when the steering angle of the front wheels was large enough (this is caused by the slack that exists in every U-joint, and I used 2 of them...). And that is why you found the return-to-center rubber band above the red clutch arm. And I agree: this is quite a binary use of the system but.. it worked! There are so many systems on Youtube that never leave the studded plate or see daylight..