TR-2L Ruggedized Wheel

[RocketPilot573]

Does anyone know how to use the new wheels from .20? I tried to make a rover with them, but it seems that every time the suspension uncompressed ( For example, when going down a slope ) , then the wheels would start bouncing slightly, thus causing the rover to shake violently and immediately lose control. This seems to happen on every planet and with every rover design using 0.20.2. So, does anyone know a way to fix this without just putting a 50 ton weight on top of the rover to keep it on the ground. It took at least a hundred quickloads just to travel 30 km on Tylo.

[RocketPilot573]

Anyone?

(10 Characters)

[chemieglennie]

I've been using the new wheels on two new rovers without a problem. I've started stabilising my rovers by adding lots of the OKTO2 probe cores (about 25 on my medium size rover).

Can you post a picture?

[RocketPilot573]

Here's one of the rovers effected by this issue:

[King Jareth]

I get random bouncyness with them too. I made a half sized version with reduced mass and they are unusable on the Mun and never landed in the persistent file.

[MBobrik]

You have to be using a different TR-2L than me. My TR-2L stick to the surface like they were made from velcro. U-turns at full speed, driving downhill at 55 m/s, driving with physics warp 4x, no problem.

Edited June 15, 2013 by MBobrik

[Tex_NL]

The first .20 version had problems with 'vibrations' in the Ruggedized Wheel, it was solved in the update. If you're still experiencing this than you're probably not using the latest update. If you are then there must have gone something wrong and you might want to re-install the latest version.

[RocketPilot573]

The first .20 version had problems with 'vibrations' in the Ruggedized Wheel, it was solved in the update. If you're still experiencing this than you're probably not using the latest update. If you are then there must have gone something wrong and you might want to re-install the latest version.

Hmm, well I might try re downloading (I already have .20.2 ). But that's kinda hard to do because of my horrible internet.

[Vanamonde]

All rover wheel types will do this, to one extent or another, depending on your speed the steepness of the slope you're on. Just steer into the rotation and wait for it to steady up, as braking will cause all kinds of mayhem. Hitting "reverse" rather than the brake will slow you less violently, but also less effectively.

[RocketPilot573]

All rover wheel types will do this, to one extent or another, depending on your speed the steepness of the slope you're on. Just steer into the rotation and wait for it to steady up, as braking will cause all kinds of mayhem. Hitting "reverse" rather than the brake will slow you less violently, but also less effectively.

I'm not sure what you're talking about, I can definitely confirm that this only happens with one kind of wheel. Also, while the wheels are "bouncing" the rover cannot steer of brake, pretty must all control is lost.

[OUScooby]

You have to be using a different TR-2L than me. My TR-2L stick to the surface like they were made from velcro. U-turns at full speed, driving downhill at 55 m/s, driving with physics warp 4x, no problem.

Could you please share some pics of your rover design?

[Vanamonde]

I'm not sure what you're talking about

I am talking about the bouncing/skittering issue you raised in your first post.

I can definitely confirm that this only happens with one kind of wheel.

I can confirm that it happens with the others.

Also, while the wheels are "bouncing" the rover cannot steer of brake, pretty must all control is lost.

While the wheels have interrupted contact with the ground, steering and braking are reduced in effectiveness, but not nullified. Steer into the skid and hit reverse to at least help prevent rollovers until the wheels find purchase again, or don't. I'm not going to hold a gun to your head.

[Spartwo]

I call it wobbly wheels and it exists in 20.2 everytime terrain changes it has a one in three chance of occurring.(in my experiences)

[Tiron]

The first .20 version had problems with 'vibrations' in the Ruggedized Wheel, it was solved in the update. If you're still experiencing this than you're probably not using the latest update. If you are then there must have gone something wrong and you might want to re-install the latest version.

Redownloading does not fix it. I've had this problem since I started trying to use the TR-2Ls in 0.20.2. The suspension pulls out of its sockets and causes extreme instability. The only thing that really affects it is terrain. The super smooth terrain at KSC doesn't set it off, anywhere else does. Regardless of speed, steering direction (or lack thereof), going uphill or downhill, phase of the mun, or any other variable I can discern other than being or not being on KSC's terrain. Big trick is that EVERY time it happens, all four wheels do it at once.

Redownloading, twice, did not affect it. A clean install with no mods at all still had the issue.

Either the alleged 'fix' never worked at all or 0.20.2 does not contain the fix.

Edit:

Every other rover design I've tried with TR-2Ls has had the same issue, including one just now that consisted of four PB-NUKs, Four TR-2Ls, and an Okto-2 strapped to a FL-T800.

All four wheels will, at intervals, jack themselves up past full extension, pulling the screw/spring/things completely out of their sockets, at which point the rover bobbles about like crazy, in random directions. Any attempt to turn the rover in this state (which frequently occurs while driving straight), will cause an extreme overreaction in that direction. If the rover's CoG on the current slope is at all prone to flipping, it'll flip for sure. When this happens there's no 'skid' to 'turn into', ALL FOUR wheels simultaneously behave this way.

At times it almost seems like some kind of timeout or error condition or something, because it will suddenly go from having sat on full compression for quite some time to that, but it will do it from almost any other state as well. It's characterized by the suspension fairly rapidly jacking all the way up from the current position, at which point it will stay at more-than-full extension for quite some time. I've never seen it happen on a subset of the wheels, only all four at once.

Edit2:

I made a version with Rovemax 1s instead of TR-2Ls, and to my surprise it did in fact do the same thing...sort of.

The suspension would jack up and cause the rover to bobble about, but not as much, and the instability was greatly lessened. It might actually be an issue with the entire wheel suspension system in general, and the TR-2Ls are just more affected...if I had to guess, because they have better grip.

Edited June 16, 2013 by Tiron

[MBobrik]

Could you please share some pics of your rover design?

sure

.

.

.

or probe cored

.

.

or two rovers docked together, one wide one long, for extra stability when carrying colony parts.

.

[MBobrik]

All four wheels will, at intervals, jack themselves up past full extension, pulling the screw/spring/things completely out of their sockets, at which point the rover bobbles about like crazy, in random directions.

.

That happens when you hit an invisible imprecision in the surface collision mesh. Happens even at terrain detail high, just not so often. AFAIK It has nothing to do with the suspensions. And when your rover is not stable enough to handle this kind of bumps, you will get into tumble. Either add probe cores+SAS to your rover, or build it wider. Or both as I do.

[RocketPilot573]

.

That happens when you hit an invisible imprecision in the surface collision mesh. Happens even at terrain detail high, just not so often. AFAIK It has nothing to do with the suspensions. And when your rover is not stable enough to handle this kind of bumps, you will get into tumble. Either add probe cores+SAS to your rover, or build it wider. Or both as I do.

I'm not trying to start an argument, but it does have to do with suspension. I can see how invisible parts of the surface can trigger it, but visible parts also cause it. Unless maybe every slope change has an invisible part to it...

[Tiron]

I'm not trying to start an argument, but it does have to do with suspension. I can see how invisible parts of the surface can trigger it, but visible parts also cause it. Unless maybe every slope change has an invisible part to it...

Well he's calling it an 'Imprecision' so it's likely that they're all over. I have managed to get it to do it on KSC's terrain now, but only barely.

It acts partway between floating and firmly planted (the bobble behavior seems like it's floating, barely touching the surface), but the extreme over-responsiveness to turn inputs suggests it's not.

And add weight? Heresy!!! Seriously though, don't use SAS units, they're dumb and will try to hold the exact orientation they had when you activated them...even when you're later on a slope (meaning it will try to lift one side of the rover off the slope to stay at the same angle). Or try to turn. You'll be pushing F lots.

We almost need a rover SAS unit that's missing the Ki component like the pods and the avionics. But enough probe cores can do that.

[MBobrik]

Seriously though, don't use SAS units, they're dumb and will try to hold the exact orientation they had when you activated them...even when you're later on a slope

Uh, no. That's what ASAS does. SAS w/o ASAS attached will merely resist rotation.

[Tiron]

Uh, no. That's what ASAS does. SAS w/o ASAS attached will merely resist rotation.

The SAS uses the exact same logic, and the same settings even, it just only uses its own torque. ASAS uses Rotpower, RCS (if on), and Control Surfaces, as it HAS no Torque. An SAS unit added to an ASAS unit allows the use of the SAS torque as well as the normal control systems.

Normal SAS is just as dumb as the ASAS, and DOES in fact hold attitude rather than merely damp rotation. The only things which damp rotation only and won't hold an attitude are the Avonics Nosecone (a specialized ASAS with altered settings, including a Ki of 0), and the built-in SAS of Command Pods and Probe Cores (which have no Ki setting at all).

Both SAS and ASAS have a Ki setting of 1.

Edited June 16, 2013 by Tiron

[Tiron]

Wow. Or not. They changed the settings on the SAS, it must've been very, very recently because I looked at these cfgs not that long ago.

Instead of sharing the ASAS's settings of Ki = 1, Kp = 0.6, Kd = 1, it now has Ki = 0, Kp = 5.0, Kd = 3.5.

So I guess it's now fine for rovers, but almost useless otherwise. Lovely. They used to be great for docking.