WIP - RbtKpfW II Ausf. B - Help!

[Lauwenmark]

Note: After extensive testing, the OKL - OberKommando des Lego - concluded that the RbtKpfW II Ausf. B prototype was "unsuitable for the battlefield" and withdrew the project. The following is thus kept there only for historical purposes.

Hello!

The Robotkampfwagen II is my first attempt at a 'major' homebrew Lego design since more than 15 years. And, well, it seems that I'm not skilled enough to successfully complete it...

The original design goal of that model was to create a sort of "explorer bot", a remotely-controlled robot with a rotating "sonar" sending the trace of obstacles around to the computer performing the control.

Some pictures of the model in its current stage can be found here. In particular:

- Full side-view;

- Full top-view (Wheels turned);

- Detail view: read gearing train;

- Detail view: front gearing train;

- Underside view: 'gearbox' assembly;

- Underside view: compressor and front differential;

- Rear view: rear direction transmission;

Current design notes:

- Suspension: Spring-based. Each wheel got an independent suspension.

- Motorization: I have used a single Power Function XL motor as the main movement engine. Each pair of wheels is connected with a hand-made differential. The main design idea regarding the engine is that the XL Motor is never stopped - instead, I use a 'gearbox' to select between the two possible directions and the 'zero move' position. This is because I wanted the same motor to be connected to the pneumatic compressor, hence it was important to allow it to turn even when the robot was not moving.

- Movement selector: the 'gearbox' to select the direction of the current movement ('advance', 'rear move', 'stop') is controlled by a single NXT motor. This is a rotating wheel on which transmission gears are fixed. By rotating the control wheel by 180°, the movement is reversed. Any intermediate position of the control wheel disengages the gears, so the car stops moving.

- Direction: An NXT motor controls the direction. It is rack-and-pinion-based. The pairs of wheels are connected by bars, and gears transmit the movement from the motor to the bars.

- Compressor: The initial plan was to connect the compressor to the XL motor and use it to drive an arm that could be used to grasp and carry small objects around. Currently the compressor is mounted, but not connected to anything.

The main issue with this model is that it is way too heavy! The power transmitted to turn the wheel is not sufficient, and the gears of the movement selector tend to slip, because the resistance is too high. I'm unsure on what I should do now: should I scrap the model completely and restart from scratch? Maybe somebody here has a better idea on how I could make the model functional? I'd like to get advices about this from more experienced people than me.

Anyway, thanks in advance for any help!

Edited July 21, 2010 by Lauwenmark

[Zerobricks]

Your driveline is waaaay too complicated and sloppy. 8 tooth gears arent good for trnsmitting high torque, the worm gear has too much room and no bracing, so it can slip... You need to simplify things.

[DLuders]

Sariel's excellent "Gears Tutorial" ( http://sariel.pl/downloads/ ) may be able to help you GEAR DOWN the single Power Functions XL motor so that it can move your robot. If you have more internal bracing to keep the axles aligned, maybe the gears won't have a chance to get out of line and slip.

[Sariel]

I would add that using the 24t crown gears is a bad idea in general, and especially without strong bracing which is your case. The worm gears tend to push against gears, so if you locate them in the middle of a long axle, the axle will bend and the worm gear will snap. Moreover, seeing your suspension I doubt that the wheels will stay properly aligned while turning (there are some weird angles the links are attached at) and I have no idea how you'll make these soft light-gray shock absorbers support the apparently heavy vehicle. It really looks as if you wanted to make it as complex as possible. This may not be the best approach.

An another note, I like your compressor ;)

Edited July 21, 2010 by Sariel

[Lauwenmark]

I would add that using the 24t crown gears is a bad idea in general, and especially without strong bracing which is your case. The worm gears tend to push against gears, so if you locate them in the middle of a long axle, the axle will bend and the worm gear will snap. Moreover, seeing your suspension I doubt that the wheels will stay properly aligned while turning (there are some weird angles the links are attached at) and I have no idea how you'll make these soft light-gray shock absorbers support the apparently heavy vehicle. It really looks as if you wanted to make it as complex as possible. This may not be the best approach.

An another note, I like your compressor ;)

Well, I didn't try to make it as complex as possible intentionally :). I started by developing the base chassis with the four wheel suspension. I only owned a single differential part, so I built two out of multiple gears instead. But those differentials have the drawback of being larger than the pre-made ones... And so the chassis was set as a rather large one.

The second issue I had to face was the motorization of the whole. I had planned to use one NXT motor for direction, a second for propulsion, and the last one I own to control extra function (a top arm). Given the weight of the chassis, I couldn't use the NXT motor for propulsion directly - it obviously lacked torque. I then changed plans, and thus decided to use the much more powerful XL, with the NXT controlling its course.

The last problem was powering the pneumatic compressor. The first design of the motorization control circuit used one NXT motor to flip the XL switch. The problem with that approach is that since the compressor could be needed at any time, I had to find a way to 'shut down' the wheel power while maintaining the compressor circuit alive - and thus, I needed a gearbox. The weird rotating design I made for it came from that I needed a synchronized gearbox. Its main flaw is exactly what Zblj underlined: it is too sloopy to drive high-torque functions.

I'm not sure I understand the remark about the alignment of the wheels - I made tests, and the chassis seem to turn well, without breaking down or imposing hard constraints on the parts.

I understand your remark about the compressor - I just checked your site after having read the link to your gears tutorial, and found a model of compressor that uses the same basic structure as mine. The lever part of your design is not the same as mine, so I'll try your variant and see if it isn't sturdier than what I had done myself. You obviously are way more expert in brickology than me! :)

As for the shock absorbers, well, I have no real choice - those are the only ones I own! :) Sure enough, I could try to find more/harder ones. On the short-term, I have made tests to increase the hardness of the suspension by binding the left and right suspension elements with a large rubber band underside the car (I removed those to make pictures). This makes the suspension hard enough for the weight, although of course it would have been better with using only real lego parts.

Now indeed, the result is definitely a clockworker's nightmare, I'm very aware of it :). That's why I'm trying to design a better, sturdier way of powering the wheels and control the direction of the rotation. But I admit that I don't really know how to solve the issue without buying specific parts, like pre-made differentials or 16-clutched gears - but that would be cheating, in a way :P. Replacing the suspension train by a dual pendular system would of course help making the chassis somewhat smaller and lighter, but that would be giving up the 4-wheel independent suspension.

Maybe I expected too much from the brave old Lego? I was probably overly ambitious... :) Probably the best would be to scrap this model and rethink it entierly, *snif*!

Thanks for the comments!

Edit: Added a view of the wheels turned at the maximum acceptable range here. Not exceptional, but good enough for my needs.

Edited July 21, 2010 by Lauwenmark

[Sariel]

I only owned a single differential part

Perhaps this is the essence of the problem - no offence, but it looks like you're trying to build something you don't actually have parts for. Lots of builders try that and proceed straight into a dead-end.

Given the weight of the chassis, I couldn't use the NXT motor for propulsion directly - it obviously lacked torque. I then changed plans, and thus decided to use the much more powerful XL, with the NXT controlling its course.

XL's torque is 14.5 N.cm, while NXT's torque is 16.7 N.cm. Sorry, but under no conditions can XL be called much more powerful.

Maybe I expected too much from the brave old Lego?

That may very well be the case. With the 8297 set's suspension components and differentials you could build a powerful 4x4 with a full independent suspension with very little effort. You would have a compact, neat and strong construction, nothing like the one you got now :)

[Lauwenmark]

Perhaps this is the essence of the problem - no offence, but it looks like you're trying to build something you don't actually have parts for. Lots of builders try that and proceed straight into a dead-end.

No offense taken - that's quite a likely possibility; I just wanted to try a few things with what I got before going the easy way. Even if the result is a dead-end, it was fun trying :).

XL's torque is 14.5 N.cm, while NXT's torque is 16.7 N.cm. Sorry, but under no conditions can XL be called much more powerful.

Aah, that's indeed interesting info for sure! I've also found the philohome's motor comparaison page, which is definitely an invaluable piece of data. That taught me a lesson: check good ol' Internet *before* designing a model :D.

That may very well be the case. With the 8297 set's suspension components and differentials you could build a powerful 4x4 with a full independent suspension with very little effort. You would have a compact, neat and strong construction, nothing like the one you got now :)

Oh, sure, but where would I have spent all those headaches? :D

Just like its glorious(err?) ancestor the Panzer VIII Maus, the RbtKpfW II Ausf. B is an error of nature that deserves the Pit of Oblivion. But for sure I've learned a few things all the way. I'll now whip the design braincells so they quickly come up with a better prototype! ;)

Thanks for the comments!