My rover doesn't like Physics.

[BagelRabbit]

Hey everyone!

Recently, I made a tiny li'l rover. It was a wonderful thing. It was made out of Cubic Octagonal Struts, with some of the tiny rover wheels on there, as well as an External Command Seat, some solar panels, and a battery. Nothing too out of the ordinary. I drove it around for a while (no problems yet, mind you) but noticed that putting a Kerbal on there raised the CoM enough to make it a wee bit unstable on hills. So, I added a small SAS module to help it right itself...

oh no.

Immediately after being decoupled, the rover began to vibrate (the SAS was not turned on at this time). The vibrations increased to an absurd level over a period of a couple of seconds, before the rover crashed into the pod I had used to load it. One of the wheels came off, and the rover immediately stabilized.

Upon further investigation, here's some things about this strange vibration.

> Removing any one part of the craft via a crash seems to make the wobbling instantly stop. This includes wheels, cubic octagonal struts, or the Kerbal pilot.

> Once the wobbling stops, there is no way to make it start again. The rover remains perfectly stable, and in some cases, can even drive reasonably well.

> The Small SAS unit is the cause of the wobbling, and removing it makes the problem vanish. The SAS unit is not clipping into any other part. Furthermore, it has mass (I think) and thus should be the thing keeping it from wobbling to bits in the first place! (Scott Manley did a video in which a similar problem occurred on a rover with 100% massless parts.)

I'm wondering what's going on here.

Here's a Dropbox link to the rover (called Pico-Pico-Pico Rove on account of its smallness) and I've also made a Xacktar-style video on the problem it faces...

Any help would be thoroughly appreciated. I wanted this to be my flagship small-rover, and it was all going so well...!

Thanks,

Flobbert

[sal_vager]

Oh yeah, I made one simiar to this earlier today, the command seat was clipping into the wheels and making it wobble to bits.

It was fine after I removed the seat, maybe putting the wheels lower so they don't touch the seat at all will fix it

[Vaporo]

I think it also has to do with the fact that you are using almost entirely physically insignificant parts. Those things do some weird stuff to the physics engine. (Scott Manley has a video somewhere where he builds a rover entirely out of physically insignificant pars. I can't find it, though.)

[BagelRabbit]

Oh, c'mon you guys.

Oh yeah, I made one simiar to this earlier today, the command seat was clipping into the wheels and making it wobble to bits.

It was fine after I removed the seat, maybe putting the wheels lower so they don't touch the seat at all will fix it

Not the problem. It runs fine, as long as the SAS module is not on there.

I think it also has to do with the fact that you are using almost entirely physically insignificant parts. Those things do some weird stuff to the physics engine. (Scott Manley has a video somewhere where he builds a rover entirely out of physically insignificant pars. I can't find it, though.)

But adding a part that the physics engine does care about makes it worse! I referenced the very same Scott Manley video in the OP, too:

> The Small SAS unit is the cause of the wobbling, and removing it makes the problem vanish. The SAS unit is not clipping into any other part. Furthermore, it has mass (I think) and thus should be the thing keeping it from wobbling to bits in the first place! (Scott Manley did a video in which a similar problem occurred on a rover with 100% massless parts.)

With all due respect to both of the repliers so far, I clearly stated in the OP that it was a problem caused by adding that small SAS module, and not by the rover's massless-ness or any previous part clipping that occurred. I did this because I wanted replies that were related to the problem, which I have not received so far. (Sorry if this sounds a bit harsh. I just want answers for what the heck is doing this.)

I've done a couple of small experiments, and this is what I've found out so far:

Substituting the SAS module with a Z-200 inline battery pack (massless, and with almost the exact same dimensions) makes the problem vanish.

Substituting it with an Oscar-B fuel tank (greater mass, slightly larger dimensions), the problem occurs to a greater extent.

Substituting it with an OKTO2 (small mass but not massless, smaller dimensions) makes the problem occur to a lesser extent.

It seems that there's some sort of strain placed upon the craft by a part that has mass. I'm going to test out different methods of attachment and see if that does the trick.

To anyone who has read the entire report, do you know what is causing this?

Thanks,

Upsilon

[sal_vager]

Not sure how to help you then, my rover was fine with the small probe core still attached :/

[BagelRabbit]

The mystery deepens.

I initially thought that the SAS unit with mass was somehow 'tugging' on one part of the structure of the rover, rendering it unstable...

...but placing the module (or any other non-massless part) anywhere on the craft causes the problem to occur.

I honestly don't know, anymore.

[Bilfr3d]

Try right-clicking on the SAS module before you launch and turn of the torque. Also do that with your little rover command thing (if you have one). The torque MAY be a problem, but not too sure. This will narrow it down a lil'.