4500+ dV not enough to get to low Laythe orbit?

[problem]

So it's my first mission to Laythe and I got my lander down on Laythe's peaks, at 4000 m altitude. Kerbal Engineer says I have 4595 m/s of delta V (thats just in the atmosphere). Every time I try to launch this ship back to orbit, it runs out of fuel before I manage to circularize, around 200 m/s short. I don't understand what's wrong here, most of the reddit and forum posts I read say you need about 3000 m/s to get to Laythe orbit from SEA LEVEL. I thought I overcompensated with this lander but looks like that's not the case. Any advice on why this ship isn't working/what kind of ascent profile I should use? Here's a screenshot with the lander stats for reference.

Edited May 3, 2018 by problem

[Snark]

Hello, and welcome to the forums! 

16 minutes ago, problem said:

Any advice on why this ship isn't working/what kind of ascent profile I should use? Here's a screenshot with the lander stats for reference.

Well, from looking at the ship, seems to me that you should absolutely have no problem getting to orbit.  You've got plenty of dV, you've got enough thrust to have a reasonably high TWR, I see no reason why it shouldn't work just fine.  I don't see anytihng wrong with your design, looks like a great little Laythe lander you've got there. 

I see that you're relying on fuel ducts for the asparagus staging, rather than just enabling crossfeed on the radial decouplers.  Therefore you're going to be dependent on getting the fuel lines hooked up correctly.  Just to verify, you've got the fuel lines going from the radial tanks to the central core, and not the other way around, correct?

As for ascent path, there's nothing special-- it's just a simple plain-vanilla ascent like you'd do for any vertical rocket launch.  Standard gravity turn.  You should easily make it to orbit with lots of dV to spare.

Unless there's something pathologically wrong with the design that's not visible in the screenshot (e.g. some oddball clipped piece that's totally horking your aero, or something), about the only thing I could think of is if you're doing some extremely sub-optimal ascent path.  Could you describe the ascent path you're using now?  What angle from the vertical are you when you reach 300 m/s, for example?

Your local TWR is pretty high, 2.74, so you'll want a somewhat aggressive gravity turn.

[problem]

39 minutes ago, Snark said:

Hello, and welcome to the forums! 

Well, from looking at the ship, seems to me that you should absolutely have no problem getting to orbit.  You've got plenty of dV, you've got enough thrust to have a reasonably high TWR, I see no reason why it shouldn't work just fine.  I don't see anytihng wrong with your design, looks like a great little Laythe lander you've got there. 

I see that you're relying on fuel ducts for the asparagus staging, rather than just enabling crossfeed on the radial decouplers.  Therefore you're going to be dependent on getting the fuel lines hooked up correctly.  Just to verify, you've got the fuel lines going from the radial tanks to the central core, and not the other way around, correct?

As for ascent path, there's nothing special-- it's just a simple plain-vanilla ascent like you'd do for any vertical rocket launch.  Standard gravity turn.  You should easily make it to orbit with lots of dV to spare.

Unless there's something pathologically wrong with the design that's not visible in the screenshot (e.g. some oddball clipped piece that's totally horking your aero, or something), about the only thing I could think of is if you're doing some extremely sub-optimal ascent path.  Could you describe the ascent path you're using now?  What angle from the vertical are you when you reach 300 m/s, for example?

Your local TWR is pretty high, 2.74, so you'll want a somewhat aggressive gravity turn.

Yeah I was really proud of this ship up until this point, usually I'm pretty good at planning ahead but I feel stumped here.

As far as the lander design, I did actually stick all the science modules on top of the main pod, which I guess could be a major source of drag, but I can't see it being enough to completely kill all my dV like this.

Fuel lines are actually attached to the Rockomax adapter between the engine and the main fuel tank. I thought that would be alright but it seems like they aren't actually transporting any fuel. Maybe that's a bit less efficient than a working asparagus but definitely better than making them go the wrong way.

The lander is actually a bit unstable so I have to stay a bit slow when ascending, I'll actually start turning slowly at 12000m and break mach 1 near 18000m. By 30000m I'm at a 45 degree angle and I'll thrust to an apoapsis near 60 km to meet the mother ship. I'll actually just share a video of me launching it for even more reference. 

https://streamable.com/ngewe (this is one of my better attempts but still not good enough)

Edited May 3, 2018 by problem

[Snark]

1 hour ago, problem said:

As far as the lander design, I did actually stick all the science modules on top of the main pod, which I guess could be a major source of drag

Hm?  I don't see anything that looks like it should be a problem.  I see a Mk1-3 command pod (fine), with a Science Jr. on top of it (fine), and a few little bitty science instruments attached radially (absolutely fine, any drag they induce is completely trivial).   I see nothing visible that would cause this craft to have hideous aerodynamics-- it looks perfectly reasonable to me.  The only thing I could imagine that would make the craft's aero a lot worse than it looks would be if you did something weird with clipping (e.g. taking big un-aerodynamic things and clipping them inside the vehicle under the completely mistaken impression that that will shield them from drag).  I'm guessing this is not your issue; I only mention it just in case, because that's a not uncommon mistake that people sometimes make.

You might want to close the doors on that Science Jr. on ascent-- I think that may help its drag profile a bit.  But honestly, that's only a fairly minor thing-- with well over 1000 m/s of extra dV to spare, it really shouldn't make the difference between getting to orbit or not.  I think your problem likely lies elsewhere.

1 hour ago, problem said:

Fuel lines are actually attached to the Rockomax adapter between the engine and the main fuel tank. I thought that would be alright but it seems like they aren't actually transporting any fuel.

I think that shouldn't matter-- as long as they're connected to a part that has fuel crossfeed capability (which is most parts), it ought to flow just fine.

If the problem is somehow something to do with your fuel ducts and they're not transporting fuel, though, then the good news is that that's easily solved on the spot with your existing ship that's already sitting there-- no need to go back to the VAB to redesign.  Just turn on crossfeed on the radial decouplers.  You'll get exactly the behavior that you want:  the radial tanks will drain completely before it starts draining the central tank.

The one thing to watch out for in that case (i.e. the one thing that behaves differently than it would if you had properly functioning fuel ducts) is that the radial engines won't automatically stop when the radial tanks are drained.  That's because, unlike the fuel ducts (which are one-way), enabling the crossfeed is bidirectional.  So if you just took off and burned, what would happen would be that first the radial tanks would get drained, but after they're empty, all three engines would keep burning while the central tank is drained.

So, if you're using crossfeed-enabled decouplers as your asparagus mechanism, it still works just fine, but you can't just "wait until the radial engines die and then stage" as you may be used to doing.  Instead, you need to watch the fuel content of the radial tanks instead, and stage when they reach zero.

Other than that one little difference, should work exactly the way you want.  (It's why I basically never use fuel ducts anymore-- crossfeed-enabled decouplers give me the behavior I want, with a simpler build process, and less mass and drag.)

1 hour ago, problem said:

The lander is actually a bit unstable so I have to stay a bit slow when ascending

Just to verify:  You've set the fuel flow priority on those radial tanks so that the bottom 2-ton tank drains before the top 0.5-ton tank, yes?  If you haven't... do that.  It'll help your stability.

1 hour ago, problem said:

The lander is actually a bit unstable so I have to stay a bit slow when ascending, I'll actually start turning slowly at 12000m and break mach 1 near 18000m. By 30000m I'm at a 45 degree angle and I'll thrust to an apoapsis near 60 km to meet the mother ship.

Oh.  Wow.  Yeah, that's a seriously inefficient ascent curve.  You should be starting your turn immediately upon takeoff (i.e. immediately point the nose about 10 degrees or more east of the zenith) and following all the way up-- by the time you hit 300 m/s, you should already be at 45 degrees, long before you get to 12000m.  You should be at 45 degrees by 6 km or thereabouts, I'd guess.

If you're climbing slowly straight up to 12000m... you're going to get absolutely clobbered by gravity losses.  This is a major part of your problem right there.

Unfortunately, if you have a really unstable lander, you might be outa luck, here.  The fact that you're using aerospikes is working against you, since they don't have any gimbal at all and therefore can't help with stability during ascent-- you're completely relying on your command pod's reaction wheels.  There are various ways that this design could be tweaked in the VAB to make it more stable and controllable, but of course that doesn't help you if you're already sitting on the surface of Laythe.

About the only suggestions I could offer to perhaps help your aero stability:

• Close the doors on your Science Jr. before ascent.  It'll help lower drag a bit in front, I think.
• Definitely make sure that the bottom radial tanks drain before the top radial tanks.
• Perhaps leave your landing legs extended during ascent?  It'll hurt you a bit by adding some extra drag... but on the other hand, that drag might help stabilize you some, like feathers on a badminton birdie.  Not actually sure whether this will help or not, but it's worth a try, I think.

[problem]

Thanks for the advice. I know my ascent would be pretty bad but I really prioritized getting out of the thick part of the atmosphere too much. I did test my lander on Kerbin and it seemed to do well, but even though Laythe's atmosphere is thinner it seems to me like it wants to keep the same thickness even up to 20-30 km. 

I'm away from my computer now but I'll try your suggested ascent profile later. The lander isnt extremely unstable, but it will try to flip over early in the flight. I have RCS thrusters to fight it, but it gets a lot more stable when I drop the radial engines. Perhaps I could use them to boost me past mach 1 and immediately drop them before I start to flip.

Also another thing I recently noticed is that the delta V values that kerbal engineer displays seems to be more than it actually is... When im going 1000 m/s horizontally near apoapsis it says I have about 1200 dV left, yet I ran out of fuel when going 1700 m/s. Maybe I'm missing something or KER is broken right now.

[bewing]

57 minutes ago, Snark said:

Unfortunately, if you have a really unstable lander, you might be outa luck, here.  The fact that you're using aerospikes is working against you, since they don't have any gimbal at all and therefore can't help with stability during ascent-- you're completely relying on your command pod's reaction wheels.

There might be one more chance. There's RCS on this thing -- and RCS can do a pretty powerful job of control authority if you get lucky with your CoM and thruster placement.

So ... as snark says, to get to orbit efficiently you need to get going a heck of a lot faster right off the ground, and you need to get quite tilted to the east pretty quick, and SAS locked prograde. So besides snark's suggestions about the landing legs and locking your forward fuel tanks -- turn on RCS before you launch!

 

[sh1pman]

(Possibly related question) Is there no decoupler between the heat shield and the fuel tank?

(Another possibly related question) Did you, by any chance, build this lander as a subassembly, with the engine being the root part? Or, did you place the central engine first, and then build the rest of the lander?

I'm asking because KER gets reeeally confused when the rocket is build bottom-up instead of top-down, and gives completely incorrect delta-v numbers.

Edited May 4, 2018 by sh1pman

[problem]

33 minutes ago, sh1pman said:

(Possibly related question) Is there no decoupler between the heat shield and the fuel tank?

(Another possibly related question) Did you, by any chance, build this lander as a subassembly, with the engine being the root part?

1. That's a silly design flaw I have a habit of doing, I hate having a fairing over a heat shield so I thought I could just use the heat shield as a decoupler itself. Doesnt take much more than common sense to see the issue with that though... I always manage to re enter kerbin perfectly fine without a heat shield anyway so it's not a huge issue. (Besides weight)

2. That's actually exactly what I did. I tend to use the same rocket for every Joolian moon, with different landers optimized for their specific mission. I guess this means something bad??

Okay I see what you said about KER. Well now I'm really curious about what my actual dV is.

I guess if one things for sure it's that Milfurt, Halberry, and Bobney are gonna have an unscheduled 2-3 year stay on Laythe... At least the view is nice. Definitely bringing a spaceplane with rapiers for the rescue mission.

Edited May 4, 2018 by problem

[Snark]

32 minutes ago, problem said:

That's a silly design flaw I have a habit of doing, I hate having a fairing over a heat shield so I thought I could just use the heat shield as a decoupler itself.

Yeah, I'm really not a fan of the heat shield shrouds myself.  That's why I use this little ModuleManager snippet to change their behavior so they automatically do exactly what I want by default:  the fairings are turned off, and the connector node is place such that when I attach a decoupler, it's nice and snug right against the heat shield. 

If you'd like the same, just download that file and stick it anywhere in your GameData folder.

32 minutes ago, problem said:

Okay I see what you said about KER. Well now I'm really curious about what my actual dV is.

Ah, there you go.  I didn't bother checking your numbers.  Looks like you've got them drastically overstated.  (The issue never comes up for me, because I never use KER; I just manually calculate the dV as needed.)

For example, you've got 23.49 tons, with 5 tons of fuel in your radial tanks.  Given the aerospike's Isp of 290-340 depending on pressure, that should give you in the range 680 to 797 m/s for that stage-- not 1481 as it's saying, there.

And when you stage those away-- I'm guessing that'll reduce your mass to something around 15.8 tons.  With 8 tons of fuel to burn, that would give your second stage something around 2350 m/s of dV-- again, nowhere near the 3113 it's saying.

So, your actual dV for this craft is somewhere in the range 3030 - 3150 m/s.  Not 4500 m/s.  That ought to be enough to get to orbit... but not by a huge safety margin, meaning that you'll absolutely have to follow an efficient ascent path to make it.  You're certainly not going to be able to do a slow under-mach-1 vertical ascent to 12 km before starting the gravity turn.

So, yeah.  Problem diagnosed:  You've got only a fairly narrow safety margin to get to orbit, and you're following an inefficient ascent path that uses up that margin and then some.  So either you figure out a way to keep it stable so that you can climb efficiently to orbit, or else you will not go to space today.

[Thiagobs]

Yup, Sh1pman and Snark nailed it..

I replicated your craft on my stock-test install, with only KER installed, based on your screenshot..

With the capsule as root, KER reports roughly 3100m/s Delta V, but rerooting the craft and having the main engine as root, KER then reported almost 5000m/s delta-v

By cheating, I put the craft on Laythe's surface, about 4300 meters high and couldn't get to orbit in any way, missing just a bit of delta-v for full circularization... My version of your craft also was quite unstable, so I definitely lost some DV fighting the controls

[bewing]

Just for fun I also sort-of replicated your craft. I didn't try to figure how much ablative you have on your heatshield. I reduced mine to 160 units to save on mass. From a 3660m starting altitude, with the landing legs extended and RCS on, I made it to orbit (50x58km) fairly easily with 69 units of LF remaining, and almost all my monoprop left (for a little bit of last-ditch deltaV) -- and a not-particularly-efficient ascent. So, depending on the details, I'm pretty sure you can do it. (BTW, with RCS on, I didn't see the slightest bit of instability.)

Edited May 4, 2018 by bewing

[OHara]

2 hours ago, Thiagobs said:

With the capsule as root, KER reports roughly 3100m/s Delta V, but rerooting the craft and having the main engine as root, KER then reported almost 5000m/s delta-v

Usually, KER will give a clue of the misunderstanding, by associating a large delta-V to stage S0 or whichever stage is running after the capsule is separated.  That is, KER indicates that after dumping the dead weight of the capsule, the main rocket and its fuel tank can go pretty far. 

The screenshot in the top post, though, does not show the heat-shield-jettison in staging.   With no capsule-separation staged, KER for KSP v1.3.1 calculates correctly for me as if the capsule stays attached for all engine burns.  

Based on the paint schemes, the top post is from KSP v1.4.x, so maybe the experimental KER for v1.4.x (recently adopted by new modder) has a new bug.

 

[Laie]

6 hours ago, problem said:

I hate having a fairing over a heat shield so I thought I could just use the heat shield as a decoupler itself.

Heatshields have two attachment nodes: one very close to the shield, and one a little further away. Depending on which one you use, you will not get the ugly heat shield fairing.

[Rauko]

Do you use KIS/KAS? If you do, perhaps you can ditch the science stuf, fuel lines and parachutes, so you minimize drag and gain a bit of dv (have you seen "the martian"?)

You then get to the mother ship and make for a lko. And a quick rescue mission on lko.

It would make for a nice heroic aar.