Pre-Munar Landing Anonymous - A support group

[boolybooly]

There is an easy way to guesstimate the altitude to start your deceleration burn which I use.

You simply do a test burn at max speed eg take 50m/s off the descent speed at 650m/s and see how much altitude you lose while doing it. Multiply this up for the total remaining speed add one for luck (ie 1x 50) and then halve it.

eg if it takes 4000m to reduce speed by 50m/s then you will need to start your burn if you are travelling at 600m/s at ...

600m/s ÷ 50m/s = 12

12 + 1 = 13

13 ÷ 2 = 6.5

6.5 x 4000m = 26,000m approx

[DonLorenzo]

There is an easy way to guesstimate the altitude to start your deceleration burn which I use.

You simply do a test burn at max speed eg take 50m/s off the descent speed at 650m/s and see how much altitude you lose while doing it. Multiply this up for the total remaining speed add one for luck (ie 1x 50) and then halve it.

eg if it takes 4000m to reduce speed by 50m/s then you will need to start your burn if you are travelling at 600m/s at ...

600m/s ÷ 50m/s = 12

12 + 1 = 13

13 ÷ 2 = 6.5

6.5 x 4000m = 26,000m approx

This will vary wildly based on the angle of approach. It will work decently for circular orbits I suppose. What I do is start somewhere around around 8KM and then when I get into the 2000M area try to keep my altitude at 10x speed. Ie at 60m/s i want to be at 600m, at 40m/s 400 meter etc. So I always \'have 10 seconds\' to appraise the situation and alter it.

This is only feasible with fairly light/high thrust landers as you need to be able to change your speed rapidly

[Zephram Kerman]

This will vary wildly based on the angle of approach. It will work decently for circular orbits I suppose. What I do is start somewhere around around 8KM and then when I get into the 2000M area try to keep my altitude at 10x speed. Ie at 60m/s i want to be at 600m, at 40m/s 400 meter etc. So I always \'have 10 seconds\' to appraise the situation and alter it.

This is only feasible with fairly light/high thrust landers as you need to be able to change your speed rapidly

I do something similar to your approach, but quite a bit more conservative. I try to keep my descent speed about 2-3% of altitude. (For comparison, in these terms your method would be 10%.) So I guess adjust the factor for your conditions and desired adrenalin level.

[Extemporary]

Looking at this thread, I must\'ve been insanely lucky on my first landing. I had no ASAS or RCS, and hadn\'t actually planned anything other than 'I want to land'. I lost attitude control at 10m and landed on my side, but it was soft enough that nothing exploded. Hadn\'t expected to get back anyway; the thing didn\'t have a parachute.

[DonLorenzo]

I do something similar to your approach, but quite a bit more conservative. I try to keep my descent speed about 2-3% of altitude. (For comparison, in these terms your method would be 10%.) So I guess adjust the factor for your conditions and desired adrenalin level.

Yeah the 10% thingie is about the most I could reliably land it with every time. Now with the new Mun that has considerably higher elevations a bit more prudence isn\'t a bad thing. As long as you\'re landing in unexplored areas at least

[Chattox]

I\'ve been playing KSP for a fair while (long time lurker on this forum; first post!) and I still have serious trouble escaping Kerbin orbit. I don\'t know what I\'m doing wrong since I haven\'t really been checking the forums much, and I\'ve been without whatever the preorder version adds since I only bought it today, and I\'m still waiting for the payment to be authorised or whatever.

The only time I\'ve gotten anywhere was when I totally overkilled it and managed to get into an orbit around the Sun. Good times.

[Mustangrande]

I have not yet successfully landed on the Mun. I think my next project will involve placing a space station in Munar Orbit, though.

[VincentMcConnell]

I\'ve been playing KSP for a fair while (long time lurker on this forum; first post!) and I still have serious trouble escaping Kerbin orbit. I don\'t know what I\'m doing wrong since I haven\'t really been checking the forums much, and I\'ve been without whatever the preorder version adds since I only bought it today, and I\'m still waiting for the payment to be authorised or whatever.

The only time I\'ve gotten anywhere was when I totally overkilled it and managed to get into an orbit around the Sun. Good times.

Get into a 100KM orbit. When the Mun rises over the horizon, burn pro-grade until your AP hits about 12,100. There. Now just WAIT. When you get captured by the mun, practice lunar orbit.

[Kreuzung]

Get into a 100KM orbit. When the Mun rises over the horizon, burn pro-grade until your AP hits about 12,100. There. Now just WAIT. When you get captured by the mun, practice lunar orbit.

I just went to 11.300km (100km below Mun) and warped a looooong time \'cause I didn\'t know what exactly to do at moon rise... >

[boolybooly]

This will vary wildly based on the angle of approach. It will work decently for circular orbits I suppose. What I do is start somewhere around around 8KM and then when I get into the 2000M area try to keep my altitude at 10x speed. Ie at 60m/s i want to be at 600m, at 40m/s 400 meter etc. So I always \'have 10 seconds\' to appraise the situation and alter it.

This is only feasible with fairly light/high thrust landers as you need to be able to change your speed rapidly

DonLorenzo, dear fellow, there is only one angle of approach, straight down. Its most efficient to take all lateral movement off at a high altitude by lateral thrust parallel to the horizon, this give you lots of time to get it right (see map view) and wastes no fuel as lateral velocity will not increase due to gravity. If you get lateral drift to near zero and allow the ship to fall as fast as possible to the surface and burn at the last moment you spend as little time as possible hovering and wasting fuel fighting gravity and you get straight into the fine adjustments for landing ASAP. It may seem counterintuitive but the faster you approach the less time you are accelerating so the less fuel you use. The later you leave the vertical component of deceleration the more fuel efficient the landing.

[DonLorenzo]

Are you sure about that? I agree about descending as fast as possible, but you\'re going to have to kill the same amount of energy no matter how you do it, surely.

[RavingManiac]

Attempt number 5!

[rkman]

Are you sure about that? I agree about descending as fast as possible, but you\'re going to have to kill the same amount of energy no matter how you do it, surely.

There are more efficient ways of killing delta-V en less efficient ways.

The least efficient would be to hover and thus eventually use all your fuel (and not even land).

A little more efficient is to descent very slowly and thus use a lot if fuel because it takes forever.

The most efficient (least amount fuel used) is to descend as fast as possible.

[DonLorenzo]

Yeah I get that. However booly said it\'s definately more efficient to kill lateral velocity high up and then descend, building up vertical speed and then killing that. I think it makes no difference at all, with the notion that killing lateral velocity very low can be hazardous

[maltesh]

Yeah I get that. However booly said it\'s definately more efficient to kill lateral velocity high up and then descend, building up vertical speed and then killing that. I think it makes no difference at all, with the notion that killing lateral velocity very low can be hazardous

It is definitely more efficient to kill lateral velocity higher up, simply because you\'re moving slower at that point. And as you descend, you\'ll be moving faster (and thus your lateral velocity will increase), and orbital mechanics will bend your hyperbolic path (also increasing your lateral velocity.)

If you absolutely know what you\'re doing, it really is most efficient to stabilize for a vertical descent as far out as you can, then do a last-second max thrust burn to a stop just at ground level.

If you are learning to land on the Mun for the very first time, it is a very unforgiving landing route.

[DonLorenzo]

yeah ok if your goal is to kill lateral velocity. But it isn\'t, it\'s landing.

What I\'m saying is that suppose you\'re in a 10kmx1000km orbit it will cost the same amount of dV to kill lateral velocity at 10km and then do the perfect vertical deceleration-landing burn as it does to kill lateral velocity at 1000km and then do the perfect vertical deceleration burn. I\'m not sure that\'s correct, but my gut feeling tells me it is.

[khyron42]

Here\'s the explanation. Gravity is acceleration over time. Every second you are near the surface, you gain 1.68 m/s downwards speed.

Let\'s say you drop from 500 m/s to a complete hover at 20 km, then kill throttle and fall. You just killed 500 m/s velocity, but over the next 2 minutes you fall 12 more km and regain 200 m/s speed.

If you had fallen the same 12 km without braking at 20 km, you\'d cover that distance in 23 seconds and only sped up from 500 to 539 m/s. That\'s a lot less speed you have to kill in total!

You slow down higher to have margin of error, but it\'s most efficient to brake only once.

[VincentMcConnell]

While in English class (after test) I got bored and drew this up. I decided to transfer it a little more neatly into Paint and then upload on this page. It\'s a trajectory flight path showing all the Mid Course Corrections you should make, the T&D sequence (We don\'t have docking yet...) and the maneuvers to get into and out of Lunar Orbit. It\'s just kind of a drawing of the big picture.

[willitstimothy]

...several Munar modules running out of fuel whilst still a thousand feet off the surface, requiring an abort; and at least eight command pods left in various locations on the Munar surface when their spacecraft toppled over.

You should (If you are not already), start your approaches at around 20k (It doesn't really matter, just so long as you have enough landing fuel to make all the needed burns) or so, if you are having trouble with them, carrying out a burn retrograde to your motion until your vertical component of velocity is cut to ~=zero and your horizontal velocity is in the vicinity of 100 m/s or so, you can then adjust your final landing spot by using vertical burns to slow your descent, and thereby sending your landing spot further down range, or you can do burns that are normal to your heading to move your landing spot left or right (Northerly or Southerly). When you've gotten your landing spot lined up, and you are roughly over the top of it (within a couple km), you should do another retrograde burn to bring your horizontal velocity nearly to zero, or at least low enough to where you will no be moving primarily downward. You should then switch to a vertical attitude and descend without power until you reach 5000 meters above the surface. At this point you should do another burn to bring your vertical velocity down to 30 m/s or so. Descend at this rate, using short periodic burns to keep your velocity down, also using RCS (you should bring plenty) to make final adjustments on your landing spot and to bring your horizontal velocity to zero (move the the retrograde bug to center on the 90 degree dot). At around 500-1000 m AGL, you should slow your descent to around 15 m/s, slowing continuously until you are within a few meters of the surface and moving at no more than about 3 m/s or so. You should be just fine. I've successfully landed 50 m rockets (whole) on the Mun, as well as a large Mun 'station.'

Edited August 17, 2012 by willitstimothy

[boolybooly]

Actually, my sincere apologies to Don Lorenzo and everyone else.... I do believe I was wrong. Mea culpa. Here is my explanation. There is a more efficient way to land which involves angled descent, but it is quite tricky to fly unless your craft has good manoevurability. For large craft with low turning speed it is not going to be easy to do and for these the long drop is easier to fly and is only slightly less efficient.

As khyron42 said, when I said a long drop was most efficient I was thinking about efficiency of a long drop over a two stage drop (which is what most of my angled descents turn into). As a basic piloting principle helpful for the hard of landing this remains true, see top spoiler. But the expert angled descent does seem to be more efficient than the long drop.... see below and bottom spoiler.

METHOD : I delivered a lander vehicle with full tanks into various orbits using a previous stage which was discarded at the beginning of each experiment, then a quicksave using F5 was made and both landings were run from the same start orbit shown in the screeny by using F9 to reload it. Then I compared remaining fuel on landing graphically. It is not entirely empirical because it depends on my landing skills but if I am biased I still managed to prove myself wrong so its a pretty ineffectual kind of bias!

In both cases I had 5 pixels more fuel left after dropping from AP than from PE. Screen resolution 1920x1080.

I tried two lunar orbit scenarios.

One was AP 2,081,860m PE 841,443m

Second was AP 2,081,860m PE 14,240m

You can see from the images the kind of split I was going for, I landed from AP and from PE in both cases, this is not quite the same as an angled descent. The comparison box compares the top fuel bar from a landed screenshot.

experiment one AP 2,097,983m PE 841,443m

experiment two AP 2,081,860m PE 14,240m

My last experiment showed me that Don Lorenzo is right that the long drop is not the very most efficient method because you can save a small amount of fuel by not removing the lateral velocity at all (and not dropping vertically) as long as you start from an orbit which will meet the planet at about <45° angle at a location 90° prograde from the comparison point (where the vertical drop would start and end) and you leave the deceleration burn to the last possible moment and successfully cancel lateral and vertical velocity more or less simultaneously.

IMHO You get two advantages from this, one you dont waste fuel removing orbital velocity at the start and the velocity you dont remove is also subtracted from the downward velocity at the point you \'touch\' down so you double its value. Second advantage is that you can begin your burn at a lower altitude for a given speed because the angle means you have smaller vertical component, this requires exacting piloting but it is possible to realise the benefits of these advantages with real fuel savings see bottom spoiler. You have one disadvantage which is you are dropping for longer but the advantages seem to win.

I don\'t know the math but I ran some experiments today.

In this experiment I compared a straight drop from approximately AP to an angled drop as shown. I started both approaches from the same game save ran each a few times and picked the best result for each.

The best angled approach landed with 2 pixels more fuel than my best straight drop and it was not a fluke as other pairs of results showed similar but smaller gains.

experiment three AP 2,112,864m PE <0m

Would be good if someone with more math skill could analyse this. If I did my second two stage drop (top spoiler) properly it shows dropping from a PE of 14km was not as efficient as dropping from AP. So the angled drop has a very narrow window of optimum efficiency. Would be good to be able to predict what this window is from theory

[khyron42]

Because we don\'t precalculate it, I usually end up having to fly every new design twice (or more.) The first time, I don\'t know how well it will accelerate and how much fuel it will have left when it\'s getting close to the moon. I either run it out of fuel, or start braking way too late and plow into the surface at 200 m/s.

The second time, I start braking earlier/later and usually land ok with enough fuel for a return trip.

I do tend to err on the side of 'running out of fuel' lately because, worst case, I\'ve built my return section as a separate stage and can use it to abort.

[RyanGt04]

Something tells me this doesn\'t count as a successful landing :\'(

[boolybooly]

Yes it does, the tin can still has live Kerbals in it, that\'s a successful landing!

The return journey might prove a little tricky...

btw it looks like your lander legs might be too high up, so the craft would probably end up landing on its engine.

[vexx32]

Heh. A good landing is one you can walk away from!

Of course, the Mun has no atmosphere, so you won\'t be walking very far, but still...

[Johno]

Okay. We\'ve had one dress rehearsal - all but the actual descent - and I believe that this rocket can do it.

Swordfish 1A is a development on the Swordfish 1. The Swordfish 1 was the same design, but lacking the four solid rocket boosters just below the midline (initial flight trials pointed to a need for more sustained boost at the 2000m - 5000m phase of the flight).

Swordfish 1A is designed to be boosted out of the atmosphere by its quadruple 5-tank Liquid engines. The three-tank vectoring engine should then perform a TMI burn.

Once the craft is approaching the Mun, the next step is an orbital injection performed by the single tank 3rd stage. There is a dedicated lander stage, with a return stage.

Opinions, ladies and gentlemen?