My landers slide down a hill - does adding more legs help?

[Magzimum]

Is there some formula or a rule-of-thumb to predict how many landing legs you need to stop a lander from sliding down a hill? Or is sliding down inevitable at a certain slope? 

I am playing around on Laythe, which has little land, of which exactly zero percent appears to be flat... My trial landings (cheat mode) thus far landed me only on slopes of 20 degrees or more. I now accept that I will land on a hill. How many legs do I need to put on a lander so that it doesn't slide down the inevitable >20 degrees slope that I will land on? 

I'm deliberately not (yet) posting a picture, as the question is really about landing legs and nothing else. I would expect that the mass/legs ratio is all that matters, which is 25 tons / 4 legs. At such a ratio, I happily slide down a 27 degree slope at 1.2 m/s. On the 34 degree slope, I slide at an impressive 1.7 m/s (still not tipping over though, which is good). Reducing weight is, although always a good idea, not the solution I am looking for today.

[edit] I am playing 1.2.

Edited November 3, 2016 by Magzimum
Question below - should have mentioned it immediately

[Clipperride]

First question - which version of KSP are you playing?  Since 1.2 I've noticed landing legs are a lot more effective at stopping your craft sliding on touch down.  I've landed on the Mun, on an approximately 30° slope with 4 large landing legs (the lander was medium sized - based on the 4 Kerbal Hitch Hiker Storage Container). 

Prior to 1.2 I had stopped using landing legs completely as the slightest slope seemed to cause the lander to slide (plus they had an annoying tendency to accelerate Kerbals that bumped into them to lethal velocities!)

[bewing]

Agreed that 1.2 should be better than 1.1 for this. But you still get microbounces from legs, and that vibration will always cause you to slide a bit. The easiest way to stop it is to build your craft so you can retract some legs -- this helps to level the craft, and when some other part than a landing leg is touching the ground that helps even more to stop the sliding.

[Magzimum]

@Clipperride, I am playing 1.2. I have updated the OP to include that info as well.

@bewing, I can see that retracting some legs may work on smaller landers, and on the Mun, but my lander (below) may not gain any stability. I gave it a try, but it actually toppled over (sideways) without the legs on the uphill side.  

In the meantime, an attempt with twice the number of landing legs (16 in total) didn't work out so well either. The legs did a lot of "microbounces". So much in fact that it walked itself right off the launchpad when I tested it there. This frequently happens when you attach the legs onto multiple different parts - which means that the legs aren't 100% at the same height, but differ by a pixel or two. (To the naked eye, they look the same). Especially relevant on larger landers with e.g. asparagus staged boosters or stages, like in my lander here.

I am now following two lines to solve my problem: one is to learn about the landing legs. The other is to find an actual flat spot on Laythe, somewhere reasonably near the equator, so the lander can just land safely. It may take a few attempts to nail that spot, but the lander has (minimalistic) airplane characteristics so I may be able to do some flying before landing by parachute. 

 

Not relevant comment 1: I tried solving this by using a space plane - but the sheer lack of good landing spots on Laythe, adn the fact that you have to land the plane with full tanks (it's heavy) did not make that option easy. I failed so many times to land it that I now try the parachutes. 

Not relevant comment 2: It was quite amusing to see that after my lander had toppled over (because I retracted some legs) it started shedding bits and pieces while accelerating down the mountain. The trail of parts looked quite dramatic. Picture in spoiler due to lack of relevance. 

Spoiler

 

[DrLicor]

Haha that's quite a hill where you're trying to land on. I don't know if you have KIS and KAS installed, but if you shoot a harpoon in the ground, you'll won't move anymore. Most of my lander are equipped with 3 or 4 of those harpoon shooting winches. Works great  

[bewing]

Heh! Sorry for laughing, but that debris trail is indeed impressive. The easiest way to handle laythe with a spaceplane is to ditch it in the water, then you can drive it to shore. As soon as some part touches the ground, you are technically "landed" even though you are still a few meters from shore. But you miss a few inland biomes that way.

[Kerbal101]

Don't forget, that the new collapsable heatshield allows to build on water.

[Sharpy]

Adding more legs won't help. That's a really bad slope, so sliding is unsurprising.

Anyway, results of my test on one of steeper mountains near KSC:

- no landing legs, just "your own belly": - accelerating uncontrollably, probably up to explosion or terminal velocity.

Landing gear, default settings: 2.2m/s stable

Friction override: 1.0 is equivalent, I don't have the screenshot but it remained the same 2.2

Landing legs: 1.9m/s stable

Landing gear, friction 2.0: 1.3m/s

I experimentally set brakes to 200. That doesn't change the terminal speed but changes the deceleration rate - braking faster.

Friction 3.3 seems to be the break-even point for this slope. 0.1m/s stable

Friction 3.5: full stop

Conclusion: forget landing legs. Use airplane landing gear and set friction to 5.

[Clipperride]

Looking at the debris field, I bet you're glad that you practiced the landing with cheat mode high fidelity simulations first?

As for bouncing landers, the things I would try are (and thes are just suggestions with no guarantee that they will make things better rather than worse!). Experiment with the spring and dampener settings in the landing legs tweakables. Although if, as you mention, it's a result of minor differences in leg positions this may be of limited use.

One other suggestion is to use something that helps keep your lander in place.  An upwards facing vernier or sepratron (with the thrust turned down) may help "plant" the lander more firmly.  You could even try a single lowered thrust sepratron pointing outwards.  If you them rotate your lander so it is facing down slope, as you near touchdown, it may prevent slippage.

I hope you find a solution and let us know how you did it.

[Magzimum]

Heh, reading all your comments, I guess that I did pick a nice slope to get into trouble. Anywhere else (Kerbin, Mun, Duna etc.) I would just advise to find another landing spot... but Laythe only has so few of them.

@bewing, my lander it really not made for that solution: landing in the water may be possible, but this lander cannot move sideways (yet). It would have to be able to drag itself ashore completely, because I intend to launch one or two planes too - just for fun, btw. It also requires precision landing to land near the shore, which may still be difficult. This solution may be easier with a space plane. Although my lander contains wings and jets (Rapiers), it is more a rocket than a plane. Complete redesign will be considered if all other options fail - and then I will certainly take this into consideration.

@Sharpy, thanks for that analysis. That seems like a very feasible change to the lander. Wheels aren't much heavier than legs. Obvious difficulty is that they are mounted below, not radially. I may be able to find a few free spots on the wings or perhaps on some small cubic octogonal part that is itself mounted radially.

@Clipperride, if the upwards facing thruster can stop the lander, after which I can shut it down, that's an interesting solution. If it must burn continuously to stay in place, it is a no-go. My lander comes with 2 planes, and I intend to stay a little while to enjoy the landscape. I will give it a go.

@DrLicor, I am trying to keep it stock (only KER installed).

My latest strategy is to actually learn to better predict my landing spot. It appears that on Laythe, from a 110x110 orbit, and a burn to lower the Pe to exactly zero, I come down to the ground about 104 degrees later. I am now doing a lot of attempts, to scout out some decent islands for landing. Then I can perhaps pick one, and learn how to do a precision landing from orbit, by parachute. The lander comes with a little extra liquid fuel, and some backwards facing air intakes on my outermost tanks, so I can use my Rapiers (airbreathing) while descending.

Earlier today I managed a landing on a 10 degree slope with only 0.4 m/s sliding, which eventually stopped (without the slope changing, I guess the 0.4 m/s was some kind of residual velocity from the landing). However, the location was horrible for my little planes... so the search continues.

 

Edited November 3, 2016 by Magzimum

[Sharpy]

18 minutes ago, Magzimum said:

 

@Sharpy, thanks for that analysis. That seems like a very feasible change to the lander. Wheels aren't much heavier than legs. Obvious difficulty is that they are mounted below, not radially. I may be able to find a few free spots on the wings or perhaps on some small cubic octogonal part that is itself mounted radially

 

The game only recognizes none-mounted and surface-mounted. Wheels are surface-mounted, same as legs. They just default to a different orientation, but you can rotate them any way you like. Learn to love the move and rotate tools, it took me a long time to learn to use them, but now I'm using them all the time. Wheels extend downwards without the slant, and if you angle them too much, the wheel raycast will stop catching the surface and they will start sliding just as any random non-wheel part, but the same angle as legs should be well within safe range. These wings on your lander look like a really good spot too. Note though that wheels, especially the smaller ones, have significantly lower impact stress resistance than legs. They WILL break if the lander is too heavy for them. Use the right gear size for your needs, test it on Kerbin.

Also, possibly perform a tiny burn right before touchdown; you can generally save a lot of mass on parachutes if you do a powered landing - I see you're using a truly massive number of parachutes; meanwhile one drogue plus one chute is more than enough if you perform a second or two long burn at TWR of 0.95. (you may use separatrons with thrust limiter set to the perfect value with help of KER if you don't trust your throttling skills; just note to set the "end" TWR (the one displayed in braces in KER) to be BELOW 1; you really don't want your craft to take off after touchdown.

The vid below uses "SRB legs" from my mod; the landing is quite hard - but it uses only a drogue, not a real parachute.

 

 

[Clipperride]

2 hours ago, Magzimum said:

Heh, reading all your comments, I guess that I did pick a nice slope to get into trouble. Anywhere else (Kerbin, Mun, Duna etc.) I would just advise to find another landing spot... but Laythe only has so few of them.

@bewing, my lander it really not made for that solution: landing in the water may be possible, but this lander cannot move sideways (yet). It would have to be able to drag itself ashore completely, because I intend to launch one or two planes too - just for fun, btw. It also requires precision landing to land near the shore, which may still be difficult. This solution may be easier with a space plane. Although my lander contains wings and jets (Rapiers), it is more a rocket than a plane. Complete redesign will be considered if all other options fail - and then I will certainly take this into consideration.

@Sharpy, thanks for that analysis. That seems like a very feasible change to the lander. Wheels aren't much heavier than legs. Obvious difficulty is that they are mounted below, not radially. I may be able to find a few free spots on the wings or perhaps on some small cubic octogonal part that is itself mounted radially.

@Clipperride, if the upwards facing thruster can stop the lander, after which I can shut it down, that's an interesting solution. If it must burn continuously to stay in place, it is a no-go. My lander comes with 2 planes, and I intend to stay a little while to enjoy the landscape. I will give it a go.

@DrLicor

Sorry for the mess in the quote box above, when posting on a tablet editing can be a real issue!

 

Hopefully a quick blast from your RCS will be enough to stop any motion and keep your lander where it is.  

Edited November 3, 2016 by Clipperride
Posting from a tablet can cause issues!

[Tidus Klein]

Dont be scaried to play with the suspension settings, a general rule of thumb is to have more dampening then spring rate, i normaly set my legs to 7or 6 spring rate with a dampining one notch higher, your craft will lean more do to tje softer spring but the legs will stay planeted much better. :3 rember the 3 main reasons for suspention soften blows, keep as many of the legs on the ground as possable, and leveling