Thoughts on .22 "suspension" updates

[Don Quixote]

[...] actual numbers and logic against generic statements pulled out of thin air about what is or isn't a "ridicilous amount" of landing legs in Kerbin gravity, where your primary option of landing won't be engine + legs in the first place.

If we're talking about actual numbers here, let's take a look at the landing legs. They're basically hydraulic cylinders with feet. Next we need to determine their actual size:

As seen in the picture six legs fit on one regular 1.25m fuel tank, so that would mean the cylinder of one medium leg is roughly 200mm in diameter. Let's say there's a 140mm piston inside, so that's about 154 cm2. Now the big unknown factor is the pressure supplied to the cylinders, but let's assume it's rather low, say 50 bar, which is a realistic value. That's 5kN/cm2, and since F = A*p we have a maximum lifting force of about 77 kN per cylinder. That's 7.7 metric tons for ONE landing leg. In other words, four medium legs would be more than enough to lift your average 25-ton-lander.

Now of course, physics in KSP are not real-world-physics, so we might as well disregard anything written above.

[John FX]

I like the new suspension. It carries a stock Apollo lander on four legs on Mun with no sag. If I am building something really big I might now consider girders, possible with heavy landing leg suspension which would have been impossible before...

[SkyHook]

Well I think that testing an exoatmospheric lander's capabilities in an atmosphere that also has several times the gravity of the target body is a bad idea if your landing legs sag like that on Duna or Mun, then you would have a valid argument

[zarniw00p]

How exactly is the landing leg suspension supposed to improve gameplay or lander design? The only effect it seems to have is that you have to increase your part count because you now have to add more of them. Is this really what improvement looks like?

[Brotoro]

Well I think that testing an exoatmospheric lander's capabilities in an atmosphere that also has several times the gravity of the target body is a bad idea if your landing legs sag like that on Duna or Mun, then you would have a valid argument

The atmosphere doesn't enter into it. But if you are going to land ships on Laythe, Kerbin, Tylo, or (heaven forfend!) Eve, that you expect to get back into space, they are going to be HEAVY ships, and the current legs are not up to the task. So we either use something else (such as girder or I-beam structures, which sadly can't be folded), or we need heftier legs (either by messing with the cfg file, which I'd rather not do since I like to play stock, or get some heavier-duty landing legs from a mod or perhaps from the devs).

Just because the legs were released this way in version 0.22 does NOT mean that the developers intended them to work this way. I could easily believe that the developers did not test these legs in all situations (or even all common situations), so this behavior may well have been a surprise to them.

What I would LIKE is a simple statement from one of the devs. Either something like:

a) "Ooops... no, this is not what was intended. We'll get back to you with a fix eventually." or,

"Yes, this is exactly what we intended. Get used to it." (Optional: "Maybe we'll give you beefier legs as a new part later.")

I could handle either situation just fine.

Edited October 31, 2013 by Brotoro

[Starwhip]

I haven't seen this happen yet, except when they are deployed at the launchpad. Anywhere else they are exactly the same as before.

[hobbsyoyo]

I experienced significant lean on a rover that weight less than 6 tons with 4 landing legs. That's unacceptable and what's worse is that it was an unrepeatable lean - sometimes it happens, sometimes it doesn't. That says bug to me.

[Brotoro]

I haven't seen this happen yet, except when they are deployed at the launchpad. Anywhere else they are exactly the same as before.

No, they are not "the same as before" anywhere else. This is not an issue with the launch pad.

Below is an almost empty first stage landed back at KSC...this used to work all the time. With the improved legs, it falls over:

SpaceX is planning on doing this...so if it's physically impossible, I think somebody should warn them real fast.

[Guest]

I've been using the airplane style landing gear for awhile now in different configurations. They're tough and light and have their own lights. I'm pretty sure that their mass is not included when you go to launch the craft. I've seen them bounce off of the surface of the Mun and Minmus at 20 m/s. I didn't even realize the new legs were bugged. All of the craft I launched early in the career mode have been really light one man landers and they never really acted up on me.

[Brotoro]

If we're talking about actual numbers here, let's take a look at the landing legs. They're basically hydraulic cylinders with feet. Next we need to determine their actual size:

As seen in the picture six legs fit on one regular 1.25m fuel tank, so that would mean the cylinder of one medium leg is roughly 200mm in diameter. Let's say there's a 140mm piston inside, so that's about 154 cm2. Now the big unknown factor is the pressure supplied to the cylinders, but let's assume it's rather low, say 50 bar, which is a realistic value. That's 5kN/cm2, and since F = A*p we have a maximum lifting force of about 77 kN per cylinder. That's 7.7 metric tons for ONE landing leg. In other words, four medium legs would be more than enough to lift your average 25-ton-lander.

Now of course, physics in KSP are not real-world-physics, so we might as well disregard anything written above.

Your numbers look reasonable. Some info I found online for ole struts used in aircraft landing gear indicated that pressures of around 82 atmospheres are typical.

[vexx32]

I've split the discussion about moderator actions to here: http://forum.kerbalspaceprogram.com/threads/56215-Moderator-actions-and-moderators-status-as-players

Please continue that discussion there if you wish, but keep things here on-topic, thanks.

Edited November 1, 2013 by vexx32

[Luis]

The problem is that the legs behave as if they were simple springs, rather than actively managed hydraulic cylinders. Absorbing the shock loading of impact is part of the job of landing legs but it's only a part. Legs also need to use their hydraulic cylinders to attempt to keep the craft upright by adjusting the pressure in each leg dynamically. If you land on a slope with legs that use passive spring suspension, then the downslope legs will receive more load than the upslope ones and the springs will compress more. This makes the craft lean more, which makes them compress more in a feedback loop.

What we actually need is legs that passively compress at the moment of landing, to absorb the shock load, and then dynamically adjust the pressure in each leg to pump them back up again as far as possible, while still keeping the ship upright. I would be completely fine with this functionality coming via an additional component (that detects the tilt or controls the hydraulics) that needs to be researched, but I really think it would be an improvement to the game.

Simply adding more legs or adjusting the stiffness in the cfg file until the craft doesn't lean isn't a fix. All that does is take you back to the previous model, where the legs barely flexed at all. If you're going to do that, you might as well use structural girders. You want the legs to flex. You just don't want them to stay flexed.

[Whackjob]

"If you want landing legs done right, do them yourself."

[vexx32]

That is both terrifying and amazing. Looks like the foot of a giant robot

[magnemoe]

Not had an issue with it yet, except the Eve lander who had to take off again to release the rover at the bottom, however this might have happened in 0.21 too as I had a pretty narrow margin.

However with its 20 ton it was far the heaviest I had landed in 0.22, no bases or kethane miners.

[Ravenchant]

18 ton lander only needs 2 legs on Mün to land without compression, tell me how that's unbalanced.

Which legs were you using? My 11-ton lander has 4 LT-2 struts and they definitely compress even on flat munar surface...

[bsalis]

LMAO

I cannot believe people are defending the legs in this version as working-as-intended. They were made squishy so novices could slam down at 5m/s and survive in one piece, instead of the more realistic sub 1m/s as would be more realistic and expected. It reminds me of people saying SAS should not lock on a heading because it makes it more challenging, even after Squad developers explicitly said it should be locking.

Anyone who thinks that the change was to make craft lean over as an additional challenge to elite players is clearly not on the same page as Squad. You may want to skip forward a chapter or two. The fact that pro players using structural parts and aircraft landing gear as a *better* alternative to legs should be a sign.

[Whackjob]

The fact that pro players using structural parts and aircraft landing gear as a *better* alternative to legs should be a sign.

I use structural parts, and I'm not a pro. I'm just some idiot that builds things the rest of you have the common sense not to.

[karolk]

Simple solution to please everyone: either an action group to lock suspension or a slider for adjusting the hardness of it.

Just a simple action group controlled function to toggle between suspension and stiff shouldn't be too hard to implement.

If we want to get more in-depth however - how about suspension that automatically adjusts to the mass of your craft and the gravity of the body in which's SoI you are?

Edited November 1, 2013 by karolk

[John FX]

The simple solution is to take options away from the player and as has already been suggested, make the suspension active like hydraulic not passive like springs so it tries to keep your craft vertical by decreasing the stiffness of some legs and increasing the stiffness of others (to a degree)

[vexx32]

I just hope the devs get them working more functionally before I get up to sending things to Eve in my career save...

[NASAFanboy]

When six LT-2s cannot support a poodle, a half-size tank, a three-man pod, a decoupler, and a parachute, that's some serious problems.

In total, the eight is around 9T.

The legs still bend

[lipatden]

The simple solution is to take options away from the player and as has already been suggested...

Yes Yes!

...make the suspension active like hydraulic not passive like springs so it tries to keep your craft vertical by decreasing the stiffness of some legs and increasing the stiffness of others (to a degree)

Simple solution to please everyone: either an action group to lock suspension or a slider for adjusting the hardness of it.

Just a simple action group controlled function to toggle between suspension and stiff shouldn't be too hard to implement.

If we want to get more in-depth however - how about suspension that automatically adjusts to the mass of your craft and the gravity of the body in which's SoI you are?

No no

This sounds easy to code. Dynamically determining the CoM to adjust hardness and compensate for lean/unbalanced craft/lateral velocity/uneven surfaces/different gravity/wow maybe it's not easy...

Active Suspension in the parts themselves is difficult from a part design view (in-game, and real-life, a very expensive option in sports cars), and will add weight and/or a power draw (pumps, logic etc). Sure, go ahead, but if it doesn't come with a heft weight/power penalty then balancing is off.

My car is sitting in a parking lot right now with suspension that is partially compressed. I don't see the concern for legs that aren't fully extended at idle, just as I don't expect my car to look like a monster-truck when I get out. Sure, some over-compress and engine strikes occur, but that's more likely a problem with your design. Any design in any field is compromise, and if it doesn't work find an alternative (sepratrons? ants? expendable lower parts to act as a cushion?)