A better Mun landing approach

[FleshJeb]

3 hours ago, MPDerksen said:

Okay, let's give this a go then.  I don't believe I used "Toggle Snap" to a hexagon so my six boosters might not be aligned perfectly.  I figured that doing them as sets of 2 was sufficient.  Feedback most welcome.

That's a very respectable rocket.   I built a reasonable facsimile of it, and flew it. The very abrupt changes in diameter make the 2.5m tank cause a LOT of drag and difficulty with control. You could fix that by streamlining it, or putting the lander it in a fairing.

However, I just used the power of engine gimbal. I put an LV-T45 Swivel on the center column, and did the same with the two last asparagus stages. Those two help with roll control as well as pitch and yaw.

I got it to orbit and checked the remaining fuel. You have plenty. Flew to the Mun, landed, came back with a perikee of 40km. Held Surface Retro,  Turned the engine on for about 30 sec near Pe to lower Ap, and help maintain orientation. No problem. Popped up out of atmo a little, and fell back in. Turned off SAS and it flipped nosedown because of the drag from the big tank. Activated the single chute, and it righted itself when it fully opened. Splashed down at about 11 m/s and broke the Science Jr and everything below it.

I might suggest a couple of additional radial parachutes on the capsule, if you want to bring the whole lander back. You could ditch the heatshield, and still have a lighter craft. I flew it that way because I wanted to see if the tank could take the heat, and if the craft would remain stable. The solar panels blinked a little, but the tank and engine were fine.

All in all, a good little ride. You could add a total of seven T-400 tanks to the top of each launch stack, and still have plenty of thrust for liftoff.

[GoSlash27]

I concur with FleshJeb about the streamlining. I would add that that's an awful lot of rocket for just a Mun trip. I understand "better safe than sorry" and you're probably better off doing it that way for now...

Best,
-Slashy

[MPDerksen]

 

Here is my current unlocked Tech Tree.  I don't see any shrouding in my options, yet.  Also, while you guys are more efficient, I've made 6 runs to the Mun, and back, and nearly run out of fuel each.  That's a primary reason I'm asking for some help, to find where it's being wasted.

Note that my decoupler is right below the MK1.  I don't try to bring home the Science Jr or Service Bay.  As soon as I'm lined up for reentry, (so i don't need any more electrical), I dump them.  This is why I'm using the OKTO and an Scientist, instead of a pilot.  I need to capture all the science before I come home.  I don't want to beam it over either, since I don't get as much that way.

Would a "reaction wheel" help?  How about moving the batteries into the service bay?

Edited March 27, 2018 by MPDerksen

[Leopard]

designed a pretty standard Mun lander on console a while back, its not very efficient but it works.

booster is two of the 2.5m tanks (one an orange if you have one), skipper under it - six fins and six thumpers - with nose caps and struts.

trans stage (which also finalises LKO) is the mid sizes 2.5m tank, with a poodle and a streamlining nose.

thats the booster

the lander it two stage, descent stage (which handles the final stages only, the trans stage does most of the work really as I learn how to fly it), is built around a materials bay, three of the fuel tanks the same size, each with a terrier under it, some legs and various science bits on the outside. Optional probe core and antenna and it stays on mun as a probe.

on top of that is a Mk1 pod riding a half sized 1.25m tank with another terrier, parachutes etc.

 

mission profile is get to orbit, burns the booster and about a quarter of the trans stage fuel.

burn to mun, obtain orbit at around 20k, usually about 25% fuel in the trans stage remains

burn to sub orbital, aiming for only just clipping the ground.

burn to zero horizontal speed at about 5k up (estimated as no real way to judge it), ditch the trans stage even if it has fuel it it at this point - watch its impact, this tells you how high up you are.

descent engines get you to the ground, with enough fuel reserve to move around if desired.

the ascent stage can get you back home with fuel to spare.

 

also works to Minmus.

its inefficient but it works.

 

key points:

1. land during the day

2. use the trans stage as a sounding line, watch its distance below you when it goes bang, thats your altitude above the ground, can get a rough idea of the ground altitude from this.

3. brake to about 20ms about 500m above the ground, then ride a low thrust to the ground, landing about 3ms to 5ms

 

practice helps

[GoSlash27]

I thought the 1.25m shroud unlocked pretty early. The wiki says it's in Advanced Construction, which you have unlocked. The wiki may be out of date...

My first career mun lander had the same payload as yours, but needed a tonne less fuel and corresponding tank mass. Probably  landing technique made it easier for me. The lighter lander and improved aero of the shroud allowed for a much cheaper lifter (about $26k at launch)

 I use a technique called the "reverse gravity turn" for my landings. I wrote a tutorial on it, but Photobucket borked it last year. There's a video demonstrating it's use, but it wasn't executed quite right. 

 

The Pe isn't supposed to be lowered and the burn is done continuously rather than in spurts. It can't quite  match the efficiency of a perfect zero descent rate landing, but it's much easier to do well and is pinpoint accurate.

Best,
-Slashy

Edited March 27, 2018 by GoSlash27

[suicidejunkie]

When your rocket tips North or south and you just can't figure out why, rotate the whole ship 90 degrees in the VAB so that it now tips East automatically!

 

Also, if you're using the landing lights, try grabbing the BULB mod, and set them to different colors.  You can do boat style with red and green for port/starboard to help you remember which way is which. 

Or what I did, which is 3-way symmetry and red/green/blue.  As I approach the surface, the light beams start to overlap; when I see the cyan/yellow/magenta arcs, it is time to slow down, and when they merge to white in the middle, it is time to slow to 1m/s and touch down like a feather.

[Johnny Wishbone]

3 hours ago, suicidejunkie said:

Also, if you're using the landing lights, try grabbing the BULB mod, and set them to different colors.  You can do boat style with red and green for port/starboard to help you remember which way is which. 

Or what I did, which is 3-way symmetry and red/green/blue.  As I approach the surface, the light beams start to overlap; when I see the cyan/yellow/magenta arcs, it is time to slow down, and when they merge to white in the middle, it is time to slow to 1m/s and touch down like a feather.

You can change the color of stock lights. No mod needed.

With version 1.0.5, all dedicated lights, and the LY-10 Small Landing Gear can have their colors changed in the VAB and SPH using tweakables. It is possible to change all three RGB color channels independently to produce nearly any color of light.

[GoSlash27]

I just checked on my Caveman career install. The 1.25 shroud *is* still in Advanced construction and you have already unlocked it.

If you add 1 going up from the stack to the bottom of the lander tank and 1 going down from the capsule to the top of the lander's fuel tank, it should be much more clean and stable.

If you do that, you can also get rid of the cargo bay, saving some weight and lowering the CoG for a more stable lander.

Finally, I'd recommend reducing the number of fins to 4 and placing them orthogonally opposed (one pair n/s the other pair e/w) so your control response won't be wonky. Or better yet, do as @FleshJeb did; replace the core and late asparagus motors with Swivels. You won't need fins then.

*EDIT* Oh... while I'm thinking about it:
I hope this comes across well using ascii art...
Instead of feeding your asparagus like this   o -> o -> o -> o <- o <- o <- o
Try it like this: 8 > o -> o <- o < 8
Take the first 2 pairs and reroute them so they feed the last booster pair in parallel, then drop all 4 at once. That way you won't have that underpowered phase in the sequence and early burnout of the first stage.

Best,
-Slashy

 

 

 

Edited March 27, 2018 by GoSlash27

[Vanamonde]

This question about how to play the game has been moved to Gameplay Questions. 

[FleshJeb]

7 hours ago, MPDerksen said:

Here is my current unlocked Tech Tree.  I don't see any shrouding in my options, yet.  Also, while you guys are more efficient, I've made 6 runs to the Mun, and back, and nearly run out of fuel each.  That's a primary reason I'm asking for some help, to find where it's being wasted.

Note that my decoupler is right below the MK1.  I don't try to bring home the Science Jr or Service Bay.  As soon as I'm lined up for reentry, (so i don't need any more electrical), I dump them.  This is why I'm using the OKTO and an Scientist, instead of a pilot.  I need to capture all the science before I come home.  I don't want to beam it over either, since I don't get as much that way.

Would a "reaction wheel" help?  How about moving the batteries into the service bay?

I think you've gotten some very good tips on better piloting. They should help a lot. However, there's a certain pleasure in overbuilding, because it frees you up to try stupid things, and stupid things are the best learning experiences. You've got enough fuel to land on Duna and come back. Or you could land on the Mun, and have enough to transfer to Minmus, land and return to Kerbin. (Minmus is a good early teacher of the effectiveness mid-course inclination corrections.)

DOH! I've been a Sandbox player so long, I forgot you could move science.

A reaction wheel would help, but the aero drag is so severe in this case, that it wouldn't fix the problem. They can lead to "over-control" problems where the SAS will swing wildly trying to stay on a maneuver node. I usually err on the side of too little control. When I need more, I typically have a gimballed engine.

I think the batteries may be dragless (I'd have to do an experiment), but I think it's good practice to use service bays, even if they're not strictly needed. Service bays are fantastic heat insulators, and if you open the doors, they make decent airbrakes.

@GoSlash27 Last I checked, his fins were entirely passive.

I had another thought, and that was 4 * T200 tanks with the hemispherical noses on both ends, attached radially to the service bay. Put 4 legs on them, and he's good to go. It will be wider, lower, and probably less draggy than the adapters.

 

[GoSlash27]

1 hour ago, FleshJeb said:

I think the batteries may be dragless (I'd have to do an experiment), but I think it's good practice to use service bays, even if they're not strictly needed. Service bays are fantastic heat insulators, and if you open the doors, they make decent airbrakes.

FleshJeb,

 Batteries aren't dragless, but they do exhibit their drag in the CoM of the parent part. Having them out in the slipstream will slow you down, but it won't pull you off- course.
 I'd normally agree with you on the service bay, except in this case he doesn't need it. The shroud will protect the experiments during ascent and the experiments aren't coming home. All the bay is doing is adding mass and raising the CoG in this case.

1 hour ago, FleshJeb said:

Last I checked, his fins were entirely passive.

 Good catch. Yeah... I'd clean up the payload, add the Swivels,  and ditch them completely. The drag of the asparagus boosters is enough to keep him on course.

Best,
-Slashy

[Capt. Hunt]

one thing that might help with tipping on landing is to try and put it down somewhere flat if possible.  the center of a crater is usually the flattest spot around.  If you were landing on minmus, aim for the flats.

Edited March 28, 2018 by Capt. Hunt

[Kryxal]

I usually use service bays because they make for a nice standard subassembly ... probe core of choice, 1-2 batteries, and MJ/KER nicely protected and ready to go on any rocket you make.

[90VG]

I used to have a heck of a time landing on the Mun.   I've got it down now where I can pretty much land where I want without tools (MechJeb) on the Mun.

A couple things I didn't see listed here:

1.  Use the move tool to push your engine up or tank down so that the lander engine doesn't stick out so much.  You can hit the ground harder and not bust the engine.

2.  If you are using a PC/MAC, get a joystick with a throttle.  I only hooked up the throttle to the game, could care less about the stick.   I've done quite a few landings without gear that way, as you can touchdown at  0.2-1 m/s.  Gives very accurate throttle control.  I have a Logitech Extreme 3D Pro, $40 USD.

I'd say practice landing without gear, land on the engine.

I just made a video for you.  Landing on the Mun with an OKTO, FL400 tank, OKTO, Battery, Reaction Wheel (important for control speed), Solar Panel, Antennas, 60KN engine, no gear.

I had audio recording, so you can hear me clicking away on the keyboard, but you can kinda see how I approached it.  I hid Kerbal Engineer but still had the altitude displayed top right.   If there was a flag  around on the lit side of Mun, I would have landed on it, or close, but you'll get the idea.

 

 

Depending on when viewed, it may have my mistake of throttling the wrong way, then I did a quick load.  I edited that out, but it's still processing.

 

 

Edited March 28, 2018 by 90VG

[MPDerksen]

Lots to cover this morning!  Skipping to the results, I brought my Scientist home with 445 science from a successful landing on Mun.  I actually targeted, and landed directly in the big crater like I wanted.  However, I used almost all my fuel, in spite of the comments that this design is over-kill.  So I need some idea where to improve.

Yes, I found the shroud.  It looked like a standard nosecone.  Still ended up with 5 degrees of inclination.  Piloting skills needed.  This time, I made a low, 20-30K orbit around Mun before my main fuel tanks were empty, so that's progress.  I wasted a bunch on landing and return.

Other than to return to a few of the Missions, to make money, I still have a stranded Engineer in a highly inclined orbit around Mun, and I'm sure I could repeat what I did last night, find a completely different Biome and get another batch of science.  Yes?

A last question:  Other than fuel cost, is there anything really different about a Minmus trip?  Should I wait until it shows up in the Missions to get the cash for doing it?  Note that I'm playing Career mode.

[suicidejunkie]

21 hours ago, Johnny Wishbone said:

You can change the color of stock lights. No mod needed.

With version 1.0.5, all dedicated lights, and the LY-10 Small Landing Gear can have their colors changed in the VAB and SPH using tweakables. It is possible to change all three RGB color channels independently to produce nearly any color of light.

Oh, nifty!  I didn't even notice that it wasn't needing the mod anymore.

Edited March 28, 2018 by suicidejunkie

[Zhetaan]

@MPDerksen:

Minmus is a different beast, but not a horrendously different beast.  Travel time is longer (take solar panels), landing is a bit easier (you can use smaller engines), the fuel requirement is roughly the same (you spend more to get there but less to land there), the science haul is roughly equivalent also (Minmus has a larger multiplier but fewer biomes), but getting the encounter is a bit more difficult because Minmus is a small target.

Actually, if you time the launch correctly, your persistent five-degree inclination will help to get you to Minmus, because Minmus is inclined six degrees with respect to Kerbin.

Since you've landed on the Mun, you should start to see Explore contracts for Minmus soon.

Anyway, your fuel wastage at this point is probably mostly in your piloting, but there's undoubtedly room to improve your rocket design in that, as well.  Consider continuing to put the science experiments below the heat shield:  I didn't see anyone else mention this, so I will tell you that you can EVA your pilot and collect the science from the experiment modules; any Kerbal can do this, but only the scientists can restore the one-time use experiments (Mystery Goo and Science Jr.).  If you're on EVA, get close to the module and right-click it to get a menu that has an option for 'Collect Observations' or 'Take Data'.  Then you can elect to 'Store Data' if you right-click the command pod while you're near it.  If you put the science experiments on a detachable pod, then you don't need to take them home, and that saves fuel.  Of course, science experiments are expensive, so you may want the recovery value, but so long as you take good contracts and don't blow up too many of your vessels, you don't need to worry about the cost.

When you said that you try to encounter the Mun from a Pe on the opposite side of Kerbin from the Mun, this is a waste of fuel.  The reason is because the Mun is a moving target:  where the Mun is now is not where it will be when you get there.  You have to consider that it will move while you make the trip.  Someone else mentioned waiting until the Mun just appears from behind Kerbin (that's a great rough rule), but another way to look at it is to go to the map view and, if you look down at Kerbin's north pole, then rather than put the manoeuvre node exactly opposite the Mun from Kerbin, you should move the node about forty-five degrees later in the orbit.  In other words, if you put the node at the exact bottom middle of the screen (the six o'clock position) you want the Mun to be at roughly forty-five degrees to the upper right (the one-thirty position).

For landing, try to contact the ground at between four to six m/s.  The landing legs can take more, but rough landings are the stuff of tipping nightmares.

On the subject of differently-coloured lines, it depends on what they are.  The solid lines indicate your current trajectory.  They change colour as you move through different spheres of influence.  This is important because it tells you when you have an encounter (this is a lot more useful for interplanetary trips, where you are zoomed out too far to see the planets).  Blue is your path in your current sphere; orange/brown appears with the first change of sphere, purple is the next after that, and then I believe it continues to green, red, and back to blue.  You will only see those as solid lines if you change a setting (raise the patched conics limit):  they do not appear by default because it taxes performance to make predictions through so many sphere changes.  If it's a dotted line, then it's a prediction of your path after a manoeuvre node.  It uses the same sequence of colours, however, so your new orbit after setting a burn will be shown with a dotted orange/brown line, and if you pass through a sphere change (or have additional nodes placed after that) then it will change the colour again, and again, and so on.

The reason that you see a wild explosion of coloured lines when you set up your Mun encounter is because, when you first get the encounter, the time you spend in the Mun's sphere is very brief.  That leaves you with an orbit that is almost completely about Kerbin--but it is also in phase with the Mun.  If the short time you spend in the Mun's sphere of influence doesn't change your orbit much, then it means that the same orbit that gets the Mun encounter can encounter the Mun again at some point even farther into the future.  KSP sees this and tries to draw the predictions, some of which encounter the Mun several times, thus giving you the wild lines and colours.  However, since (I assume) you plan to orbit and land on the Mun in the first encounter, you can ignore the others.  If need be, there are tricks to adjust the node to get the encounter you want, but first, I think it best to ensure that you know how to read the display--it's a lot easier to fly if you can read the gauges!

 

[GoSlash27]

1 hour ago, MPDerksen said:

Lots to cover this morning!  Skipping to the results, I brought my Scientist home with 445 science from a successful landing on Mun.  I actually targeted, and landed directly in the big crater like I wanted.  However, I used almost all my fuel, in spite of the comments that this design is over-kill.  So I need some idea where to improve.

MPDirksen,

 Just guesstimating by appearance, you should have about 5,400 m/sec DV in the lifter (vacuum), which should get you comfortably into LMO with fuel left over. Likewise, your lander should have about 3,200 m/sec DV ; about 3 times what would be required for a "perfect" landing/ takeoff cycle.

Again, these are just guesstimates. If you can post a craft file, I'd like to play with a copy and see what performance I get with it.

What techniques are you using for launch to LKO and munar descent to landing?

Best,
-Slashy

 

[MPDerksen]

So much help it's overwhelming!!!  LOL

@90VG that video was great, but brought up just a few questions.  How do you know when your horizontal speed is gone?  A few times I was drifting down very gently, but skimming over the surface, so I tip over.  Second:  The joystick with throttle is genius!  I have one around and never even thought of using it.  Last, Reaction Wheel?  Huh?  LOL.  I haven't read up on that .

@Zhetaan your explanation about orbits was VERY helpful.  Next time I'm baffled, I'll screenshot and post, but I think I generally understand better now.  Could you please post a picture of what a GOOD orbit to intersect Mun from Kerbin looks like?  Assuming the inclination is 0, where do I want my Pe to appear?

Slashy, more than happy to share my craft file, if you can explain how I would do that.  I got to a stable Mun orbit at about 20K before I had to use my lander fuel last time, so I think I've improved there.  To come in for a landing, I just try to keep the Retrograde marker centered and feather the throttle until touchdown.  I get what you said about the suicide burns, so I'm trying to get closer to that style, but I know NOTHING about mods, and so I don't know my relative altitude etc.  I'm doing all this from the stock software interfaces.

[GoSlash27]

MPDerksen,

No worries. Just go to http://wikisend.com/

You'd follow this path: KSP_Win64>Saves> (career name) > Ships > VAB and find the .craft file. Post the download link here and I'll give it a try.

Best,
-Slashy

 

[90VG]

I put another rocket down, right next to the first one, no landing gear:

 

[MPDerksen]

Now you’re just showing off....   I’m off for vacation, but I’ll try the throttle trick when i get back.  I do seem to fair better each time I venture out.

[90VG]

@MPDerksen  I can tell when people actually watch the video, especially when I troll like I did in that one.   

If you watch the first video I posted, you will see the vector in navball hit the horizon when I zero horizontal speed.  That means horizontal speed is basically zero.  Then if you burn retrograde till zero (and stop in space), you will be totally zeroed.  I've done this a few times, so I can tell if I'm about zero just by looking at the terrain.  You have to watch those navball indicators.

This is the rule of those markers:

   (I'll skip prograde and retrograde)

   Take your hand, close it like you are going to punch someone.  Stick up your thumb.  Rotate hand so it goes in same direction of orbit (from palm to finger tips) (rotate view also).  Your thumb points to Normal,  opposite is anti-normal.  That's how you kinda turn to adjust where you will land.

   With radial out and radial in, that's just away from the center of orbit, or to it.  In the case of the 1st video, you'll see the the radial (cyan colored) hit the horizon line on navball.  Now you know that your orbit is heading straight down, hence to radial out would be horizontal.  Get that?  You orbit includes the center of the Mun, and you are close to it, therefore to orbit "out" from the center of your orbit is horizontal.  Once you are close to that crossing, just aim retrograde and put it down.

   Also, if you can get close to zero on "ORBIT" mode, then switch to surface and burn retrograde until velocity is zero.  Then you will fall straight down.  Keeping aiming at retro, paying attention to altitude and shadow, you'll be on the ground no problem.  When you are falling to the ground, and you have your landing spot targeted, when you burn, it will push the retrograde marker away from where you are pointing to.  Once they line up on top, just adjust and keep moving that retrograde mark over the anti-target.  

   That is really easy way up high, but it gets a little high paced as you get about 500m out.  The video was a joke, but in reality, I landed 117m from the craft.  I don't know the exact numbers, but I think for most operations you only need to be within 2.5km.  That's very easily done.  5m is done via computers, it's just too fast paced for me to do that manually with a keyboard (for me).  But honestly, I have a few beers while playing, except for this morning when I made the first video, I was sucking down coffee before work.

[GoSlash27]

http://wikisend.com/download/212054/MPDMunGus.craft

A slightly- tweaked version of your design. I was able to do a mission with it with plenty of excess DV. I made 8km LMO with 325 m/sec DV to spare in the core asparagus stage. The lander did the mission and achieved Kerbin interface with a (scandalous) 1,131 m/sec DV remaining in the tanks. And that was with a really bad mission. I actually tumbled the lander and broke half the solar panels (too busy taking pics). This required a re-hop and lots of burning to generate electrical charge and directional control.

If you can figure out which part of the mission seems to be taking too much DV, let me know and I can give you some tips on technique.

Best,
-Slashy

 

 

Edited March 29, 2018 by GoSlash27

[MPDerksen]

1 hour ago, 90VG said:

@MPDerksen  I can tell when people actually watch the video, especially when I troll like I did in that one.   

If you watch the first video I posted, you will see the vector in navball hit the horizon when I zero horizontal speed.  That means horizontal speed is basically zero.  Then if you burn retrograde till zero (and stop in space), you will be totally zeroed.  I've done this a few times, so I can tell if I'm about zero just by looking at the terrain.  You have to watch those navball indicators.

I just don't know how to add WWF to my landings, but that's a separate post.  Yeah, it will take practice, but I have a current mission to pick up someone stranded on Mun, so I'll get my chance soon.  I had suspected that you just lower your Pe to just beyond your target, and burn retro until you fall on it.  No?

46 minutes ago, GoSlash27 said:

http://wikisend.com/download/212054/MPDMunGus.craft

A slightly- tweaked version of your design. I was able to do a mission with it with plenty of excess DV. I made 8km LMO with 325 m/sec DV to spare in the core asparagus stage. The lander did the mission and achieved Kerbin interface with a (scandalous) 1,131 m/sec DV remaining in the tanks. And that was with a really bad mission. I actually tumbled the lander and broke half the solar panels (too busy taking pics). This required a re-hop and lots of burning to generate electrical charge and directional control.

If you can figure out which part of the mission seems to be taking too much DV, let me know and I can give you some tips on technique.

Best,
-Slashy

 

 

I tried to upload my file on wikisend, but it kept stalling (even though I created an account).  I was able to get your version, and it's very close, just better

I'm going try to the mission with your ship.  But I'm not yet speaking dV very fluently.  You say you had 325 m/sec left when you were in LMO.  I have your design on the launchpad right now.  I see starting levels of 5,400 Liquid Fuel.  (what is Oxidizer anyway???)  But since dV, as I understand it, is a measure of ability to change the velocity of a mass, we need to know that mass, don't we?  For now, I'll simply post my amount of remaining fuel until I understand it better.