Landing gear loading

[MaxPeck]

Okay, after having yet another airplane (using the awful fixed landing gear) dance like an impatient Labrador retriever on takeoff and then ASPLODE on touchdown, I'd like to ask if we could possibly be allowed to know the loading limits of the landing gear?

Seems like most threads about landing gear have someone who will very helpfully tell you that you're overloading the wheels and that's why they don't work, but I cannot seem to find any reference to an actual max ramp/takeoff/landing weight.

Anyone?  Please?

[GoSlash27]

19 minutes ago, MaxPeck said:

Okay, after having yet another airplane (using the awful fixed landing gear) dance like an impatient Labrador retriever on takeoff and then ASPLODE on touchdown, I'd like to ask if we could possibly be allowed to know the loading limits of the landing gear?

Seems like most threads about landing gear have someone who will very helpfully tell you that you're overloading the wheels and that's why they don't work, but I cannot seem to find any reference to an actual max ramp/takeoff/landing weight.

Anyone?  Please?

MaxPeck,

 There isn't a max load value given in the .cfg file. There are values for max deflection, max deflection rate, and max impact velocity. As a practical guideline, I wouldn't exceed 5t per main gear for the fixed type. Landing speed should be 40 m/sec max and you should keep it under 5 m/sec sink rate.

 It's usually the max deflection that gets you. The fixed gear have very little suspension travel and if they bottom out, they explode.

HTHs,
-Slashy

[MaxPeck]

@GoSlash27That does help, and thank you, but it would be nice to have real values to work from. This is true for landing legs as well... how are we supposed to design landers if we don't know the weight/force under which the legs will fail?

This should either be calculable from known values, or a community testing protocol needs to be organized so we can have some real values to work from.

if there is a max impact velocity, there should be a critical mass to go along with it (i.e. 100 m/s @ 10t). I get that KSP is supposed to be slipshod engineering and guesstimating, but it's frustrating when you do a 40 minute GAP contract-a-ganza only to have your airplane explode on touchdown. 

 

Edited November 5, 2016 by MaxPeck

[bewing]

I did a fair amount of testing during 1.1 and 1.2 betas -- the important load limit is seen during landing, so you have to land a lot of planes.

for version 1.2, the current max loading for the basic wheels is:

LY-01 = 2.5 tonnes

LY-05 = 2 tonnes

LY-10 = 7 tonnes

I don't do ultra-massive planes, so I didn't bother testing the heavier gear.

 

 

[MaxPeck]

26 minutes ago, bewing said:

I did a fair amount of testing during 1.1 and 1.2 betas -- the important load limit is seen during landing, so you have to land a lot of planes.

for version 1.2, the current max loading for the basic wheels is:

LY-01 = 2.5 tonnes

LY-05 = 2 tonnes

LY-10 = 7 tonnes

I don't do ultra-massive planes, so I didn't bother testing the heavier gear.

 

 

What were the landing conditions, do you know?  What was your touchdown forward and vertical speed?

[bewing]

15 minutes ago, MaxPeck said:

What were the landing conditions, do you know?  What was your touchdown forward and vertical speed?

Forward speed about 35 m/s. Vertical speed is as close to 0 as I can get it. I come in over the grass with the engines off, flying level at about 70 meters, and wait for the plane to land itself when it can no longer stay in the air.

BTW, I put the 1.1 landing leg info in the wiki months ago. But they got beefed up a lot for 1.2, so they need a retest.

And it's not always about guesstimating. It's about doing some testing in advance based on your flying style, so you know what works. That's one of the main points of sandbox mode, I say.

Edited November 5, 2016 by bewing

[MaxPeck]

@GoSlash27, @bewing

Okay, I think I've got a lead on this.  I built a mockup aircraft using modular girder segments, attached the fixed gear in a normal configuration, with the mains just behind the CG, and then mounted a fuel tank just above the CG.  I put it on launch clamps and raised it to a height of 20m and started dropping it until wheels exploded.  What I found was that the wheels started blowing up when the mockup weighed more than 9t.  At 9 tons it would sustain a straight drop from 20m and bounce.  So using the formulas v=sqr(2gh) and Favg = (1/2 mv^2)/d, I figured the impact force was about 176KN.  I went back and looked at the stats on the LY01, which has a mass of 0.022t or 22kg and a supposed impact tolerance of 125m/s.  Assuming a 1m drop, and using the same formula, I come up with ~172KN of allowable impact force.  That's probably a better figure to use, so now the trick will be to develop a table of approach speeds at various vertical speeds to determine the max allowable landing weight and/or best speed for a given weight.  That's a lot of math for a guy that doesn't do a lot of math.

Also, just for fun, I used the mockup to figure out the ramp failure weight, and what I found was that the gear fails 100% of the time when it's starting ramp weight is >8.725t.  What's interesting is that the gear would take a straight drop and bounce at 9t, but would fail at start at 8.725t.

I'm going to see if I can make a spreadsheet tomorrow that will let me calculate this all out for each landing gear/landing leg in the game, and then develop a reference that can be used to figure out how many or what type of gear you need for a given mass on Kerbin, which could then be scaled to other bodies according to their own gravity.

Anybody want to check my numbers or try to reproduce what I did to see if they get similar results? 

[GoSlash27]

MaxPeck,

 This is going to be a *very* useful resource. Excellent work!

 a 20m drop test seems pretty extreme to me. I'm surprised the gear survived it at the heavy end. Did you have a probe core on the mock "aircraft"? I think that may make a difference.

 Also, remember that the spring rate and damping can be tweaked. raising the spring rate should make heavy aircraft a little more survivable.

Best,
-Slashy

Edited November 5, 2016 by GoSlash27

[bewing]

I tried some drop tests myself, but found the results not quite so useful as doing an actual landing. The wheels do truly have an extremely high impact rating. So if you drop a plane onto its wheels, all the joints on the plane will tend to fail before the wheels do. So all the wheels fall off, and the rest of the plane explodes, and you end up with a bunch of wheels laying around.

If you just go by the drop test rating, and then try an actual landing -- I tended to find that everything fails at about half the load as you got with the drop test. But as I said above, it depends on your flying style.

 

[MaxPeck]

25 minutes ago, bewing said:

I tried some drop tests myself, but found the results not quite so useful as doing an actual landing. The wheels do truly have an extremely high impact rating. So if you drop a plane onto its wheels, all the joints on the plane will tend to fail before the wheels do. So all the wheels fall off, and the rest of the plane explodes, and you end up with a bunch of wheels laying around.

If you just go by the drop test rating, and then try an actual landing -- I tended to find that everything fails at about half the load as you got with the drop test. But as I said above, it depends on your flying style.

 

The funny thing about this is that I did my initial drop tests on the launch pad and the pad itself exploded before the wheels would fail.  Apparently the launch pad is made of glass.

Now that we know the max force that the wheels will sustain, it's just a matter of variables to figure out the rest.  If you add a forward velocity vector, it's going to necessarily subtract from the vertical velocity, so that the faster you're going forward, the slower your rate of descent must be to keep the net force equal.  It should be relatively easy to develop a table for 1t, 2t, 4t, and 6t aircraft at given rates of descent that shows the maximum forward speed for that configuration.  As soon as my kids will give me some free time, I'll get into excel and see if I can't flesh that out into a graph.  Unless some other math wizard wants to jump on it first.