Improving my Spaceplane

[DarthRiko]

I've been working on SSTO spaceplane for quite some time now, and I've hit a bit of a rut in my designs. So I would like some ideas on improving one of them.

This is my current design. Note that the COM and COL are show with tanks empty, not full.

As you can see, the main body is LF/O with two Aerospike engines, with two LF columns on each side fueling a total of four turbojet engines.

There are four shock cone intakes and fifteen structural intakes (from the Spaceplane Plus mod).

There exists four of every type of science module. This might be redundant, since some of them are re-usable, but the convenience of not needing to EVA to clean them is worth the very might weight/drag increase.

Right now, there is no RCS or docking port. I will add those as improvements are made.

Here's my usual ascent profile:

Air intakes automatically handled by mechjeb.

Activate turbojets.

35 degrees from liftoff to 15,000 m.

25 degrees from 15,000 m to 22,000 m.

15 degrees from 22,000 m until air depletion forces me back to 50% throttle. Usually I will be going around 1650 m/s at this point.

Activate aerospikes, incline to 35 degrees.

Deactivate turbojets when air depletion forces me to 25% throttle.

Once apoapsis is above 80,000 m, cut throttle and wait to circularize.

I can get into orbit and return pretty easily. However I don't usually have enough fuel to do anything more than that. I don't have the fuel to meet with my space station for refueling, so interstellar missions are a no-go.

So any ideas for improvements would be greatly appreciated.

Edited September 8, 2014 by DarthRiko

[Taki117]

Actually not a bad design, lacks some yaw authority (a vertical tail) but other than that you are fine. Seems like your biggest issue is mass, it's awesome that you are carrying science gear, but 4 is very redundant (also if possible put the radially mounted stuff in a cargo bay) Goo cans and science labs are heavy, so I wouldn't reccomend more than two.

[Red Iron Crown]

Four science modules has the benefit of gathering just about all the science from a biome in which they're deployed.

I'm still pretty green when it comes to spaceplanes, but I can offer some advice:

- Replace the LF tanks with LFO tanks and tweak the oxidizer out of them. The LFO tanks actually have better mass ratios than the LF-only tanks even when not carrying oxidizer. Plus, you can reuse that tank space for rocket fuel, which brings me to:

- Refuel in orbit. Getting to orbit is the big challenge and eats a lot of fuel. If you had a simple propellant depot in low Kerbin orbit (maybe an orange tank with some docking ports on it), adding the planned RCS and docking port and refueling once in orbit would give the craft enough delta-V for interplanetary work.

[Voculus]

Can you post a link to your CRAFT file? Also, are you using FAR, or just the stock aerodynamics?

[Wanderfound]

As Voculus says: FAR, NEAR or stock? Good advice for one set of aero is very bad advice for the others. Your SPH picture looks like you don't have FAR installed, though.

Also, if you can, give us screenshots in the SPH with indicators on (like the one you've already shown), but directly from the side and above. It's hard to see exactly what's going on from a 3/4 angle.

I'd also recommend that you grab the RCS Build Aid mod; as well as helping to balance your RCS, it also adds another indicator for dry (i.e. unfuelled) CoM, which is immensely useful when designing spaceplanes.

One thing missing from your ascent profile: after you shut down the last air-breathing engine, close the intakes. They create a lot of drag. For the same reason, once you've shut down the rockets to coast to apoapsis, point prograde to make your profile to the wind as narrow as possible.

[Red Iron Crown]

Definitely not FAR, there's no toolbar button.

[DarthRiko]

Actually not a bad design, lacks some yaw authority (a vertical tail) but other than that you are fine. Seems like your biggest issue is mass, it's awesome that you are carrying science gear, but 4 is very redundant (also if possible put the radially mounted stuff in a cargo bay) Goo cans and science labs are heavy, so I wouldn't reccomend more than two.

I decided against a vertical tail, as my planes all have dihedral wing angles (which you can see in a screenshot later down the post). I removed two of the science "clusters" and decreased my mass by 2.4% (which may seem small, but affects more than most realize).

- Replace the LF tanks with LFO tanks and tweak the oxidizer out of them. The LFO tanks actually have better mass ratios than the LF-only tanks even when not carrying oxidizer. Plus, you can reuse that tank space for rocket fuel, which brings me to:

I'm going to have to disagree with you on the wight efficiency of LFO vs LF tanks. At 150 LF, a MK-1 Fuselage adds 0.9t to the vehicle mass. A FL-T400 Fuel Tank with 144 LF and 0 Oxidizer adds 0.97t. I'm not sure if I'm missing something, if you could elaborate, that would be quite helpful.

Can you post a link to your CRAFT file? Also, are you using FAR, or just the stock aerodynamics?

I am using stock aerodynamics.

Also, if you can, give us screenshots in the SPH with indicators on (like the one you've already shown), but directly from the side and above. It's hard to see exactly what's going on from a 3/4 angle....One thing missing from your ascent profile: after you shut down the last air-breathing engine, close the intakes. They create a lot of drag. For the same reason, once you've shut down the rockets to coast to apoapsis, point prograde to make your profile to the wind as narrow as possible.

Mechjeb handles my air intakes automatically, which includes closing all of them when the Turbojets are deactivated.Here's a couple more screenshots with some updates made (with COM and COL reflecting empty fuel tanks).Two science "clusters" removed, two remain.Docking port added.TopSideFrontBack

[Red Iron Crown]

I'm going to have to disagree with you on the wight efficiency of LFO vs LF tanks. At 150 LF, a MK-1 Fuselage adds 0.9t to the vehicle mass. A FL-T400 Fuel Tank with 144 LF and 0 Oxidizer adds 0.97t. I'm not sure if I'm missing something, if you could elaborate, that would be quite helpful

You're absolutely right, sorry for the misinformation. The dry mass of the Mk-1 fuselage was reduced by 200kg in 0.24, before that my statement was accurate (honestly I hadn't noticed the change). My point about being able to reuse it for liquid fuel + oxidizer if you decide to refuel in orbit might make it worth the slight penalty though.

[Heagar]

There are Three issues with your design on witch you can improve:

• From the picture i guess you have no vertical stabilizers? When the jets loose in about 22-26k altitude thrust - mostly in one of it at first - they come in handy allowing you to set a higher thrust. Also at higher altitudes with less thrust in a flameout senario. I suggest a pair of "Delta-Deluxe Winglet".
  
• The CoL should be at any time behind the CoW, otherwise you may experiance an flipover during reentry.
  
• Now to your main problem, getting into orbit without using too mutch fuel: I have to admit, that i play mostly stock-KSP. Because of that i don't know the structural intakes of the mod you are using (weight, intake area, drag e.g.).
  
  • In my experiance 2 RAM air intakes per jet are enough to go up to 32k at slower airspeeds.
    
  • But you should reach at least >1800 m/s before leaving 26k or reaching 30k. Sounds weard?
    That's right but: The less intakes you are carrying around you have less weight and drag witch slows you down, so higher airspeeds are possible. And with 1800 m/s + you hit a mark in witch the stock enviroment allows you to use your jets up to 45k at least, even when the air-intake readout is zero in the recource-tab.
    If you slow down your ascend at 33-34k to 10 m/s - you may have to pitch down! - you can reach airspeeds up to 2300- 2400 m/s. Then orient you plane horizontally and "let it go". Your speed will slowly drop with the altitude since your jet engines will be running at ever less thrust.
    But you should have an AP in space and an PE (within the atmosphere) as well at the moment you have to shut down your jets. So it's only a tiny thrust at the AP with your rocket engines for a full orbit.
    The backside of this is obvisiously the long time you stay in the atmosphere running on jets (nearly an quarter of an circumnavigation). But because they use mutch less fuel than any rocket engine it is worth that.
    

[Daveroski]

Just a suggestion, but one which may mean changing your design as little as possible.

You have a space station.

Have a fuel delivery vehicle in permanent residency at your station. Take the fuel to your Space plane. Job Done.

I use this method. My vehicle is an unmanned, low powered, high isp, and carries life support to replenish any losses and gather any waste during the wait for refuelling. I keep enough fuel back to return to the station where any waste can be recycled as possible.

[O-Doc]

If you're not air-hogging then more wings are going to be the key to getting out of the atmosphere efficiently. I think you really need more lift considering your current ascent relies mostly on a high TWR. Also, your are going to waste fuel without good horizontal stabilization because your control surfaces are going to be working(losing lift) on keeping your craft pointing in the right direction.

Your center of lift is too high and far forward. I personally like the center of the CoM ball to align with the front edge of the CoL ball. The more unstable your aircraft the more "out of shape" you tend to get on ascent Which, is lost dV.

Your ascent for this design can use some work too. Once you cease to accelerate and a decent rate with only turbos, punch your rockets and level out. When that acceleration starts to slow down, you should be losing altitude at that point. So, pull back hard and get as steep an AoA as your can up to 60 degrees and throttle down your power as your turbos start to cut out. But, keep the power on for as long as your can. Let your wings pull you out of the atmo. Once your turbos are throttled so low as to be damaging your ascent, switch them off, shut your intakes and punch out with your aerospikes just like you would with a VAB launched rocket.

Pro tip: you're currently trying to do an air-hogging ascent with a non-air-hogging design.

EDIT: Here's an imgur album of that maneuver from a long time ago with someone else's design, who complained of not having enough fuel to get into orbit:

Javascript is disabled. View full album

Edited September 7, 2014 by O-Doc

[DarthRiko]

Well this started off well with several good suggestions that really helped. Then I get replies from people that I honestly don't know if they even read the thread.

I've said twice now that the CoM and CoL indicators are with empty fuel tanks, specifically to demonstrate that I've accounted for CoM-drift from fuel use. I have had zero stability issues. And the last guy states "you're currently trying to do an air-hogging ascent with a non-air-hogging design" while ignoring the NINETEEN AIR INTAKES that I detail in the first post. That's 4.75 air intakes per turbojet.

[O-Doc]

Well this started off well with several good suggestions that really helped. Then I get replies from people that I honestly don't know if they even read the thread.

I've said twice now that the CoM and CoL indicators are with empty fuel tanks, specifically to demonstrate that I've accounted for CoM-drift from fuel use. I have had zero stability issues. And the last guy states "you're currently trying to do an air-hogging ascent with a non-air-hogging design" while ignoring the NINETEEN AIR INTAKES that I detail in the first post. That's 4.75 air intakes per turbojet.

SP+ air intakes are broken. Sorry for trying to help.

[DrMonte]

Here is StevenVLawyer's incredible tutorial on (space)plane building

Ignore VTOL in the name.

[Wanderfound]

Well this started off well with several good suggestions that really helped. Then I get replies from people that I honestly don't know if they even read the thread.

I've said twice now that the CoM and CoL indicators are with empty fuel tanks, specifically to demonstrate that I've accounted for CoM-drift from fuel use. I have had zero stability issues. And the last guy states "you're currently trying to do an air-hogging ascent with a non-air-hogging design" while ignoring the NINETEEN AIR INTAKES that I detail in the first post. That's 4.75 air intakes per turbojet.

I know that you emptied the tanks; my point in mentioning the dCoM indicators provided by RCS Build Aid was that it removes the need to do that, and allows you to simultaneously view dry and wet CoM's.

But, whatever. The spaceplane community here are, in general, exceptionally helpful. And O-Doc builds very nice stock aero planes.

You've just substantially reduced your chances of getting help the next time you ask for it.

[DarthRiko]

SP+ air intakes are broken. Sorry for trying to help.

That's actually helpful. I'll look into that and do some testing.

I know that you emptied the tanks; my point in mentioning the dCoM indicators provided by RCS Build Aid was that it removes the need to do that, and allows you to simultaneously view dry and wet CoM's.

But, whatever. The spaceplane community here are, in general, exceptionally helpful. And O-Doc builds very nice stock aero planes.

You've just substantially reduced your chances of getting help the next time you ask for it.

I wasn't referring to your help. Your post was actually helpful and made a lot of sense and improved my ascent profile. I have not looked into the mods you suggested yet, but I intend to.

[Red Iron Crown]

SP+ air intakes are broken.

How so? They seem to work well enough, though I haven't tried to push any boundaries with them.

[DarthRiko]

How so? They seem to work well enough, though I haven't tried to push any boundaries with them.

Well I did some testing and he is right and wrong. I tested the same plane with three different air intake configurations.

The first, being 4 Ram Air Intakes, at a static 30 degree climb from launch, went to 19.5 km before throttling back became necessary.

The second was 4 SP+ Shock Cone Intakes, at the exact same ascent profile, went to 21 km. This demonstrates the Shock Cones do work, and are slightly more effective than stock Ram Air Intakes.

Third was 4 SP+ Shock Cone Intakes and 7 SP+ Structural Intakes. This test went to 22.5 km before throttling back, demonstrating that Structural Intakes do add to air intake.

Fourth test was of 8 SP+ Shock Cone Intakes in a more traditional hogging arrangement. This test went to 24 km before throttling back. This demonstrates that the Structural Intakes are not efficient.

I am doing further tests.

[DarthRiko]

I've done more tests, and replacing the 15 structural intakes with 4 additional Shock Cones made the difference. I can now establish orbit using only 20% of my Oxidizer, that is plenty to rendezvous with my space station for a refueling.

I do apologize for my harshness, O-Doc. While your help did lead me to this vast improvement, it was done in an indirect way that I mistook for carelessness.

[O-Doc]

How so? They seem to work well enough, though I haven't tried to push any boundaries with them.

I think there's some x64 issues with Firespitter which sets your intake air at less than a quarter. They still supply air but, not even close to their rating. I doubt this is intended.

I've done more tests, and replacing the 15 structural intakes with 4 additional Shock Cones made the difference. I can now establish orbit using only 20% of my Oxidizer, that is plenty to rendezvous with my space station for a refueling.

I do apologize for my harshness, O-Doc. While your help did lead me to this vast improvement, it was done in an indirect way that I mistook for carelessness.

I admit, I do skim alot of forum posts because, you know, time etc. But, the major issues I see coming out of SP+ builds in stock are related to the aforementioned intake level problem. I don't think this intake stuff exits with FAR installed if that's worth anything. I could be wrong about the whole intake stuff but, that's been my experience and I've done alot of testing with those structural intakes.

[juanml82]

I've done more tests, and replacing the 15 structural intakes with 4 additional Shock Cones made the difference. I can now establish orbit using only 20% of my Oxidizer, that is plenty to rendezvous with my space station for a refueling.

I do apologize for my harshness, O-Doc. While your help did lead me to this vast improvement, it was done in an indirect way that I mistook for carelessness.

So in 32 bit KSP one shock cone is roughly equal to 4 structural intakes?

[Red Iron Crown]

I wish I knew how intake air was calculated. The two relevant stats in the part.cfg seem to be "area" and "intakeSpeed". The stats for the SP+ intakes:

Shock Cone

area = 0.012

intakeSpeed = 12

Structural Intake

area = 0.0025

intakeSpeed = 10

Stock Ram Air Intake (For comparison)

area = 0.010

intakeSpeed = 10

Area-wise, it would seem that the Shock Cone is equivalent to just under 5 Structural Intakes, and 4 Structural Intakes is equivalent to a Ram Air Intake. Not sure how intakeSpeed affects the Shock Cone though.

[Wanderfound]

I wish I knew how intake air was calculated. The two relevant stats in the part.cfg seem to be "area" and "intakeSpeed". The stats for the SP+ intakes:

Shock Cone

area = 0.012

intakeSpeed = 12

Structural Intake

area = 0.0025

intakeSpeed = 10

Stock Ram Air Intake (For comparison)

area = 0.010

intakeSpeed = 10

Area-wise, it would seem that the Shock Cone is equivalent to just under 5 Structural Intakes, and 4 Structural Intakes is equivalent to a Ram Air Intake. Not sure how intakeSpeed affects the Shock Cone though.

I'm purely guessing, but I expect that it's a value that represents the speed at which the intake reaches peak efficiency. Intake air is a function of speed and AoA as well as altitude; have a look at the required/available air percentage figures reported by Mechjeb/FAR/Kerbal Flight Data and you'll see the available air rise as you increase speed while holding altitude.

This is partly why a lot of my ascents tend to have a bit of an S-curve pattern; climb until the air runs out, dive a bit to get to maximum speed, then climb again. The air runs out a lot later on the faster second run.

[O-Doc]

I've had a think about it and the reason the structural intakes are wack might be due to the direction of flow being the same as their attachment node direction. I've stopped investigating it because I'll be waiting for those parts to become stock anyway.

I'm glad the plane is now getting to orbit "properly".

[DarthRiko]

For anyone interested, here is the plane after revisions.

Edit: Crap, I didn't notice one of the intakes didn't mirror. You get the idea though.