Efficiency - Mun Descent

[Osprey]

Hello all,

So I've been landing on the Mun for a while now, my best effort to date being a 1.5m/s touchdown in a Midland Crater, felt properly smug for a while.

My question is about efficiency. I currently land my command pod and fuel for return on the surface, and blast off to get back to Kerbin, usually with no fuel remaining, and using all but about 30 units of mono propellant from 80 to start (I use some fuel and mono on the descent final seconds also, after detaching the main stage. It works pretty well, and it's a design that's done the job for me thus far.

But I'm not convinced that this is the best method, and even a small weight difference on the lander can mean a significant fuel usage increase to get out of the atmosphere at the first step.

So, which is more efficient/easier/better as an option?

1. An all-in-one ship and lander, leaving behind no parts.

2. A lander leaving behind the landing fuel surplus, tanks, engines and legs, with enough power to return to Kerbin alone.

3. Lander docking with an orbital command module for separate Kerbin re-entry.

Eventually, I plan to build a Munbase Alpha (complete with flared trousers and implausible science fiction scenarios), and an orbital station for regular surface-orbit transfer activity, and also to serve as a design prototype for deep-space manned missions. For now at least, I wanted to know if anyone had an opinions on this, and whether anyone has actually tested the weight/fuel-consumption numbers to see what works best.

I'll upload a picture of my lander when I get back home from work. The best part in my opinion is that it also leaves zero debris in orbit and only a flag on the surface. That said, I kind of like the idea of leaving part of the spacecraft on the surface as with the Apollo missions, but that's really not an important feature, not crashing into empty stages is far more important to me...

P.S. I'm also playing this in career mode.

Edited November 28, 2013 by Osprey

[Elmun]

Once I learned that the space center can destroy any debris at the click of a button, I stopped worrying about it.

As for landing style, I've done both a whole lander liftoff and Apollo style lander. Both worked pretty well after some tinkering. The whole landers are simpler, where as my Apollo style lander had staging issues (mostly from how I launched the whole bit into orbit and docked the lander and command module after launch) and had to carry RCS fuel for docking. Here's the Apollo style CSM/Lander I did after docking them together:

One thing I have found is that I'm using laterally mounted nuclear engines more and more on landers simply because they are so efficient. From the highest point you can on you lander, throw a standard girder, then attach a small fuel tank, a fuel line from your main fuel tank to the small outboard tanks, then your nuclear engines. Like so:

That style of lander is very easy to work with and never runs out of fuel on Mun/Minmus missions. The added struts are from hard lessons learned after having parachutes rip it all apart. This type of lander will also leave very little debris depending on how you deal with your transfer stages (crashing them into the Mun etc)

[mhoram]

So, which is more efficient/easier/better as an option?

1. An all-in-one ship and lander, leaving behind no parts.

2. A lander leaving behind the landing fuel surplus, tanks, engines and legs, with enough power to return to Kerbin alone.

3. Lander docking with an orbital command module for separate Kerbin re-entry.

The most efficient is variant 3, since you don't need to haul the fuel for the returnjourney to the Mun surface and back to orbit.

Variant 3 is also the hardest, since you need to perform the additional step of docking.

For now at least, I wanted to know if anyone had an opinions on this, and whether anyone has actually tested the weight/fuel-consumption numbers to see what works best.

Head over to the BSC Kerbal X challenge: several of the contestants are able to land on the Mun and come back - it may give you some ideas and there is also an extensive review.

[Murph]

One thing I have found is that I'm using laterally mounted nuclear engines more and more on landers simply because they are so efficient.

You may well have tested sufficiently to know for sure, so I'm not saying that you're wrong… They are extremely efficient, but they are also HEAVY. Are you certain that 4.5t of a pair of LV-Ns is actually better than a 2.5t Poodle, 1t of a pair of LV-909s, or maybe even 0.5t for a single LV-909?

To me, there's a minimum dV requirement before a LV-N is actually a real benefit. If you're using them for landing, ascending, then return to Kerbin, the transfer burn back to Kerbin probably makes them worthwhile. If they are only for landing and re-ascent, with a different engine powering the long orbital transfer burns, I'm not sure they are the best.

A useful setup might be to use the same LV-Ns for orbital transfer away from low Kerbin orbit, landing, ascending, then transfer back home. Something like that might really allow the LV-N's efficiency to make a big difference, despite their very poor individual TWR.

Personally, I like the LV-Ns for long transfer burns, but prefer the LV-909 or Poodle for landers, and for Apollo-style returns home with just the CSM. As far as landing with the return stage or not, I'm not sure. I do like doing it Apollo-style, it's interesting and fun. Mun is probably about the heaviest gravity where it does make some sense to land with the return stage and fuel. My gut feeling is that it's really 50/50 which is better for Mun. Any heavier gravity, and the fuel for returning to Kerbin is probably better to stay in orbit, I think.

[antbin]

Who says you need two nuclear engines?

Javascript is disabled. View full album

[Sternface]

Number 3, by far. Though the delta-V involved with the Mun is so small, it doesn't really matter.

Try it out with Real Solar System mod. Every single little kilogram counts. If that kind of not-optional optimization sounds fun for you, I highly recommend it.

[Dave Kerbin]

For the Mun with return to Kerbin I would say 2. The delta-v required to return from the Mun is very, very low. I've done a low mass Apollo style Mun landing, and even with everything stripped off (the command module was little more then a fuel tank and a docking port, and I ditched RCS to just fly it straight in) I still didn't really save mass over a direct land and return. When the cost of getting proper orbit and interception for docking are added I think it may have done worse. This is helped/compounded by the fixed selection of fuel tanks - you can't carry an arbitrary amount of fuel, so you're going to have a little left over in the lander anyway.

[Vanamonde]

The scale of the parts relative to the demands of a Mun-Kerbin return is such that it's more trouble than it's worth to discard parts on the way home. I mean, you can shave some weight off of something like this, but why bother?

[LethalDose]

I like the minimal lander that Vanamonde posted, but I've tried very similar designs and have had a LOT of trouble (read as "Stranded a lot of kerbals") using that much fuel to land on the Münar surface, return back to orbit, escape the Mün, and hit Kerbin's atmo.

Maybe I got greedy and had a science bay in there, too.

[Vanamonde]

I generally eject a stage just before the lander touches down, so that the lander is full for the return leg.

[WafflesToo]

A combination of #2 and #3 really. A separate CSM means not having to haul the return fuel down to the surface and back again. It doesn't have to be a manned CSM either; here's my favorite one.

A two-stage lander that leaves the bits not required for re-orbiting the crew means you need to bring less fuel for the ascent... which means you need less fuel for the descent since you're lighter. It's one of those great snowballing things

BTW- 1.5m/s? Well done, I mean that.

Beat this: 0.1 m/s

[LethalDose]

I generally eject a stage just before the lander touches down, so that the lander is full for the return leg.

Okay, so if you use another stage for 95% percent of your landing (orbit to surface), then you're basically leaving parts on the surface and using the option #2 of what the OP listed above.

[Kasuha]

Option 2 is quite sufficient for Mun/Minmus and you don't have to bother with docking. For destinations outside Kerbin's SOI, option 3 is optimal. Although there are people who can return with just lander from about anywhere.

[Vanamonde]

you're basically leaving parts on the surface and using the option #2 of what the OP listed above.

Huh? After it smashes into the surface and the dribble of remaining fuel explodes, the only "surplus" left is a small hardpoint or two. You're going to be ejecting stages along the way anyhow; what difference does it make if the last one hits the surface?

[Osprey]

Crud, I forgot that I upgraded my OS to 64-bit on Monday, and didn't transfer my screen shots before formatting.

As it happens Vanamonde, my design is pretty similar to yours, only with two radial orange engines (whatever they're called, I'm not in-game to look it up) on the outside, two lights for landing timing, RCS and a ladder, though the legs are somewhat higher up to lower the centre of mass also.

The most efficient is variant 3, since you don't need to haul the fuel for the return journey to the Mun surface and back to orbit.

Variant 3 is also the hardest, since you need to perform the additional step of docking.

Fortunately, I'm pretty handy when it comes to docking, so the difficulty isn't a problem, save for the requirement of extra RCS fuel for the tweaking and manoeuvring bit.

Edited November 28, 2013 by Osprey

[Murph]

One thing to consider, is that doing full Apollo style Mun missions is great for practicing orbital rendezvous and docking. Both are actually quite easy once you get used to doing them, despite seeming really very hard initially. Mun missions are fantastic for learning those skills, which may prove very useful for missions to more distant places, where it's simply not practical to carry the return fuel down and back up on the lander.

[Eric S]

Personally, for Munar landings, the Apollo style isn't worth the extra work unless you want the feeling of accomplishment. Just getting back into orbit from the Munar surface is 75% of the delta-v required for the entire trip home. Having a separate descent/ascent stage is even worse. I've had to make custom tanks just so that my landing stage didn't have so much fuel it could land, take off, and go back to kerbin all by itself. The reason this worked well for the LEM is because the LEM was designed to cut every corner possible as far as mass went (I seem to remember it didn't even have seats), and there's no way it could have survived reentry. In KSP, we don't have a manned pod that can't survive reentry. We also don't have the quantity of equipment that we can leave down there and still get the science benefit.

As for leaving things on the Munar surface, sometimes I do, but only drop tanks and maybe the landing struts attached to them. Leaving an engine down there is a waste, since you're almost never leaving enough mass down there to allow for a smaller engine. Every LEM-style lander I've used on the Mun has had two identical engines, so I bailed on that design and went with drop tanks instead. The only exception to that is when I do a lander that's got enough delta-v to hit several biomes on the same trip, in which case I might throw a pair of 48-7S's on the bottoms of two of the drop tanks.

Vanamonde, if you want to improve that design, ditch the 909. The 48-7S may only be half the thrust, but it will still be more than enough, and at a fifth the mass, will easily get you more delta-v despite the lower ISP with tanks that small.

[Vanamonde]

Thanks Eric S, but that's actually one of my ships I built for newbies, so I always like to give them a little margin for error in thrust and fuel.

[Jananton]

Because in career mode one doesn't have all the parts available to start with, I'm quite fond of this contraption, if I may say so.

The initial retro burn is done with the stage that brings the lander to Mun or Minimus so it crashes into the planet and the final descent is done with the lander with enough juce to spare to get back to Kerbin with all the science data gathered.

Greets,

Jan

Edited November 28, 2013 by Jananton

[Scarecrow88]

This one is a little over-engineered, but it was designed for a contest on another forum, and had to be easy to fly for other pilots. In that regard the descent stage has plenty of fuel to get to the surface, as well as return and dock with the orbiting command ship. However, it was designed as a 2-stage lander and flies really nicely in that mode too.

[deadelitezealot]

I have a two staged lander with two small lander cans and rover being very fuel efficient espescially the ascend stage is full of fuel its easy to land its reliable and great for docking has lots of fuel and rcs to dock if you know how to dock its relatively easy and a two stage lander that I create is always more fuel efficient there are also small mun landers with a minimum of fuel two staged to land and dock.

here are some pics of my own lander

[Kerolyov]

You can calculate which option is best in Kulebron's chapter on whether separate lander is better than all-in-one

http://flyonbudget.onegiantleap.info/chapter3.html

Most Mun landing are better all in one, unless particularly heavy.

Edited November 28, 2013 by Kerolyov

[sjwt]

My current muner ship runs on the following stages.

1200DV Boosters for speed!

3100DV Reach orbital height and 90% circulation

2500DV Circularize - Burn to Mun and land

1200DV Takeoff and return to Kerbin

Its a long burn time on a 48-7s to goto the mun and return, but its very fuel efficient @ 905 units of fule+ox to get from less then LKO to the Mun and back, and thats with all that extra weight of Pretties and science.

The finale stage is fuel linked to the Muner intercept and landing stage, so that if you need a little more DV for landing it is their.

Your most efficient is an Apolo type Orbital command module, followed by a lander that leaves as much mass on the surface when taking off.

Easyset would most likely be a lander leaving parts behind on take off as a all in one needs a lot of DV and the complexity of docking adds a bit of danger for a newbie.

http://ksp.sjwt.org/2013-11-18/Ezh-01a.craft

Edited November 28, 2013 by sjwt

[Eric S]

Thanks Eric S, but that's actually one of my ships I built for newbies, so I always like to give them a little margin for error in thrust and fuel.

The 48-7S version of that craft would have 100 more delta-v, not less. As for the TWR, to be honest, when I was learning to land on the Mun, a TWR of almost 10 would have been an issue, far to easy too overcorrect and wind up going back up. Even the 6.9 TWR of the 48-7S might have been a little much.

Not trying to say you need to change it for the sake of the newbies, I just think you're underestimating the OPnes of the 48-7S, especially in the case where the 909 would be a significant fraction of the mass of the resulting craft. I honestly expect that engine to get nerfed before final release. Half the thrust in a fifth the mass more than makes up for the lower ISP on lighter craft.

[Dun Kirk]

until costing comes in and the science rehash its hard to tell......

I used a reusable lander with a fuel depo in orbit. I think building minimal fuel profile is ignoring the cost of launch....