Space Planes and Drag and Drop Tanks

[MrOsterman]

So here's what I've got:

I finally decided to try out a space plane.  I'm in Sandbox mode but I don't like the concept of the Rapier (one engine for both L+O and airbreathing) so I committed to not using it.  

I designed a plane using the MK2 body, a pair of Whiplash Jet engines and a pair of Spike Rocket engines.  My main lift is provided by the Big Delta wings because for some reason when I try to build a wing by hand, I get a lot of flopping up and down so I'm trying to find something that will give enough lift on its own.  I was able to consistently limp this up into orbit with about 350 m/s to spare for things like rendezvous with my space station, and deorbiting.  Sadly I rarely had the fuel to deorbit after a rendezvous unless I refueled at the station.

I had a mission I wanted to fly that I knew I needed some extra Detla V to use once I got up there so I had the idea to add drop tanks with liquid fuel.  I used Decouplers to stick two Mk1 tanks with nose cones under the wings and then fuel lines to connect them to the engines.  This was supposed to get me enough extra fuel to get up and start my acceleration run without using any of my onboard fuel.  The plan was to take off, get up to 4km in altitude, drop the tanks and then ramp into space.

What happened was this:

I got up to about 270 m/s and then the plane stopped accelerating.  I dropped my tanks and was able to get up to 310 m/s or so  and then I hit an acceleration wall.  If I dove, I could maybe push it up to 350 m/s and again, a wall.  

Now I sort of fixed this by moving the drop tanks to under the main body and changing to Mk0 tanks (plus I'm tight on space down there on the run way).  This seemed to half cure the problem because now I could get up to around 320 m/s around 4km of altitude and my acceleration drops off.  I drop the tanks, dive a bit and eventually ramp back put to an AoA of 7 degrees or so and make my run at space.  But it also doesn't seem to work well because I'm not getting to space with that much spare dV either.  The tanks don't seem to be helping me.

Question 1: Do the radial decouplers create extra drag even once you've detached the tanks?

From there I'm still figuring out my whole profile  to space.  Before I started to play with the tanks I could set AoA to around 5 degrees and just fly up, changing from Whiplashes to Spikes around 18km in altitude when the acceleration started to drop off and the plane slowed.  Now that I'm using the tanks it's harder.  It feels like once I start to slow my acceleration (not my speed, that caps or constantly increases even slowly), I lose my ability to punch into the 400 m/s range.  Is the extra drag from the tanks at speeds under 300 m/s making it too hard to get past the sound barrier?

Question 2:  Is constant acceleration crucial for the Whiplash engines?

I'm thinking but frankly don't have the energy to experiment that I need to drop the tanks before I hit 300 m/s and the drag really becomes, well, a drag.  So my profile would be to take off, position my flight to my desired orbital inclination, point up, and hit the gas.  Once I see the thrust on the engines start to go down, ditch the tanks and minimize my drag profile.

This whole thing came too because rather than design a new space plane (which I'm trying to use more and more rather than rockets because they're re-usable), I just kept tweeking an imperfect design.  Moving wings, adding drop tanks, etc.  On the one hand, my frustration came because I'd been using this plane to go up to the station consistently, and then swapping out the extra passenger cabin for more cargo space, or fuel.  

Question 3:  Am I the only person who refuses to give up on a design just because it fails and instead keep trying to tweek it better?

 

[bewing]

1 hour ago, MrOsterman said:

Question 1: Do the radial decouplers create extra drag even once you've detached the tanks?

They're not supposed to. What version of the game are you running, and are you running mods?

But yes, you are definitely running into a drag problem. That's exactly what happens with a spaceplane when you add just a little too much drag ... suddenly you can't get past mach 1. Did you maybe used to have incidence on the wings, and then perhaps remove the incidence (by accident) when you attached the droptanks?

(BTW, you don't need fuel ducts with the drop tanks -- just go into the decoupler's action menu and click the "allow crossfeed" button.)

Keep in mind that MK2 may be pretty, but it's the absolute worst formfactor for a spaceplane to try to get into orbit, because it has much higher drag than MK1, MK3, or 2.5m.

1 hour ago, MrOsterman said:

Question 2:  Is constant acceleration crucial for the Whiplash engines?

No. I've seen it make a difference of 5 m/s or so in top speed, but "crucial" isn't the proper description.

1 hour ago, MrOsterman said:

Question 3:  Am I the only person who refuses to give up on a design just because it fails and instead keep trying to tweek it better?

Oh god no. Stubbornness is pretty much a required trait for spaceplane design. You need to expect to put in several hundred hours of design time, if you even start to optimize your craft.

You might get some benefit out of Aerogav's MK2 spaceplane tutorial:

https://forum.kerbalspaceprogram.com/index.php?/topic/165435-basic-mk2-spaceplane-guide
 

[MrOsterman]

23 minutes ago, bewing said:

They're not supposed to. What version of the game are you running, and are you running mods?

But yes, you are definitely running into a drag problem. That's exactly what happens with a spaceplane when you add just a little too much drag ... suddenly you can't get past mach 1. Did you maybe used to have incidence on the wings, and then perhaps remove the incidence (by accident) when you attached the droptanks?

(BTW, you don't need fuel ducts with the drop tanks -- just go into the decoupler's action menu and click the "allow crossfeed" button.)

I'm running the current version as far as I know.  I just installed it on my laptop about 6 weeks ago.

No mods either except MechJeb because I don't want to do ANOTHER manual docking, or hand calculate ideal launch windows.

I did miss the cross feed with the couplers and I've been building asparagus rockets for a while now.  Need to start doing that.

As for my space plane, the only changes that took me from "able to get to space" to "not able" was the addition of the drop tanks and the addition of two radial parachutes on the back of the body to help with landing.  I know they're not strictly needed but I tend to really suck at landing and need to be able to stop NOW to keep from running off the back of the runway.  I could use the MechJeb to land but I find my planes don't have the maneuverability to let autopilot land them without crashing on the way in.  

The other problem is that I seem to keep getting stuck at the sound barrier with the 'new' design.  I can't shake that the problem is my thrust dropping off as my acceleration goes down.  In theory, as long as I'm going faster I should be seeing my thrust keep going up unless the problem is with altitude and I'm just getting into air that's too thin to fast.

 

[Laie]

32 minutes ago, bewing said:

Keep in mind that MK2 may be pretty, but it's the absolute worst formfactor for a spaceplane to try to get into orbit, because it has much higher drag than MK1, MK3, or 2.5m.

Everybody just knows that Mk2 parts are THE WORST. No spaceplane thread that doesn't point this out. I just wondered, how bad is it really?

I've done some drop tests. It's pretty hard to come up with comparable Mk1 and Mk2 vessels, but I put together two barebone planes of about similar size and mass, dropped them from high altitude on Eve, and took note of the highest airspeed attained.

Guess what? The difference was 298 vs. 305m/s.

My planes were very barebone, only fuselage and wings. I guess adding landing gear would make them even more equal. The decision to add a retractable ladder, or not, would probably have a larger drag effect than the choice of fuselage.

 

[bewing]

1 hour ago, MrOsterman said:

I should be seeing my thrust keep going up unless the problem is with altitude and I'm just getting into air that's too thin to fast.

Well, no. It's not nearly that simple. The thrust does increase with speed, but it's a complex curve. The thrust does decrease with altitude, but that's a complex curve, too. What ends up happening is that each engine and design has a sweet spot altitude where you get the maximum thrust vs. drag. And that's the best altitude for going supersonic. It may very well be in the 8km to 10km range for your plane.

53 minutes ago, Laie said:

Guess what? The difference was 298 vs. 305m/s.

But your drop tests were exactly prograde. And that's not what happens with a real spaceplane, and that's exactly the problem with MK2. As soon as they get the tiniest bit off prograde, the drag triples.

[TheTripleAce3]

If you can post some screenshots of the craft and drop rank arrangement I'll try to get back to you later today on any tips I have (aside from whiplases liking 8km more than 4km on MK2 bodies).

Best of luck.

Edited April 8, 2019 by TheTripleAce3

[Kryxal]

1 hour ago, bewing said:

Well, no. It's not nearly that simple. The thrust does increase with speed, but it's a complex curve. The thrust does decrease with altitude, but that's a complex curve, too. What ends up happening is that each engine and design has a sweet spot altitude where you get the maximum thrust vs. drag. And that's the best altitude for going supersonic. It may very well be in the 8km to 10km range for your plane.

But your drop tests were exactly prograde. And that's not what happens with a real spaceplane, and that's exactly the problem with MK2. As soon as they get the tiniest bit off prograde, the drag triples.

I never thought to ask, but that drag, is it parasitic drag?  It sounds too high to be induced drag at a high speed...

[OHara]

27 minutes ago, Kryxal said:

that drag, is it parasitic drag?

Yes, in the sense that it not proportional to the square of the lift of the part, like induced drag is, and it is included by KSP in the parasitic drag sums that you can see with alt-F12: Aero GUI.

But, it does depend on the angle of attack of the Mk2 parts, as their flat surfaces present them selves to the airflow. 

I have argued that there is a design flaw in that KSP figures drag from the bottom surface area times sin(AoA), rather than sin²(AoA), which makes us notice the drag at smaller AoA than would seem natural.  

If you rotate your wings to an angle of incidence just enough that the fuselage flies straight along prograde as you cross the sound barrier, all designs have an easier time, especially those with Mk2 fuslages.

5 hours ago, MrOsterman said:

Do the radial decouplers create extra drag even once you've detached the tanks?

Well, they do on the subassembly to which they remain connected.  Usually the bulk of the decoupler falls off with the tank; the remaining studs on the fuselage are only decoration, do not affect flight, and disappear on the next save/load.  But if you use the decouplers the other way around they and their drag could remain on the main craft.

[MrOsterman]

Here's a pic of it with the overlays turned on.  The drop tanks were under the wings out beyond the tailfins so that they didn't hit the fins when they dropped and fell back.  This changed when I moved them up under the fuselage and moved them down to MK0's.

 

https://drive.google.com/file/d/1gzGgyEYaveUR9mJjxYOIJWVz2XhANp_R/view?usp=sharing

 

 

[TheTripleAce3]

In short it looks like you should be fine, just that testing at 4km is often too low for a accurate measure of performance (for me at least).

Maybe next time see what you get at 12km, after re-reaching that 440m/s mark.

Edited April 9, 2019 by TheTripleAce3

[swjr-swis]

6 hours ago, MrOsterman said:

Here's a pic of it with the overlays turned on.

Mainly, it looks like you're carrying way too much fuel for what that plane is meant to do. Even if you carry dense overweight payload in that cargo bay (like full ore tanks), it should be able to make orbit on a lot less fuel. Aside from that: control surfaces add drag too, so cutting down to the bare essentials will help as well.

Try emptying your tanks completely and add only just enough LF to keep your jets running until they flame out on ascent (add a notch or two extra for contingency and for powered landing on reentry). Then add just enough LF+O for the aerospikes to take it to orbit, a midge more for orbital maneuvers, and another drop or two for the deorbit burn. At this point you may need to add a little extra LF, as the added weight will make the jet run longer.

You will probably find the plane can reach orbit, deliver payload, and return safely without needing to fill all those tanks. That'll allow you to remove some of those tanks (and dead weight, and drag) and further optimize your design.

Oh and add some (more) angle of incidence to your wings. The prograde marker being below the flight direction is a dead giveaway that you lack lift and/or incidence on the wings. This is the typical situation where the Mk2 fuselage suffers most from drag.

Edited April 9, 2019 by swjr-swis
AoI

[MrOsterman]

So one of my hopes was to eventually push this to be able to do more interesting things than just be a single plane to go up, do some orbital stuff and come back.  I've been loading more and more fuel onto it with the hopes that I can eventually get to space with enough gas to do a round trip of Mun.

I'm getting the impression that with just two Whiplashes I'm not going to do that because they just can't push the fuel needed for all that space travel.  So if I want to do the Whiplash/ Spike combo I'm going to need to replan how to build the engines and wingspace into the design to double my engines and have the UMPF to push it out into orbit with enough reserves to do the round trip of Mun, etc.

 

 

[Kryxal]

Be VERY careful, this sounds like the start of a loop that will make your plane explode (in size, though literal explosions aren't uncommon in KSP design phase).  You think you need more of something, so you add it, then you need more of something else, and so on, and so on...

[TheTripleAce3]

5 hours ago, MrOsterman said:

So one of my hopes was to eventually push this to be able to do more interesting things than just be a single plane to go up, do some orbital stuff and come back.  I've been loading more and more fuel onto it with the hopes that I can eventually get to space with enough gas to do a round trip of Mun.

I'm getting the impression that with just two Whiplashes I'm not going to do that because they just can't push the fuel needed for all that space travel.  So if I want to do the Whiplash/ Spike combo I'm going to need to replan how to build the engines and wingspace into the design to double my engines and have the UMPF to push it out into orbit with enough reserves to do the round trip of Mun, etc.

 

 

If you want to do Mun flyby's then you may want to switch out the spikes for NERVAs and the LFO tanks to LF only.

 

2 whiplashes not being able to push the thing is very unlikely if you're dropping the tanks, and as pointed out above, you're about to meet the Rocket Equation head-on in a situation where it will win.

I have a few whiplash/spike planes and what might end up happening is you reserving a quarter of your drop tank mass for Oxidizer to run the spikes for a few seconds at mid-high alt WITH the jets still pushing as well.

Best of luck.