Thinking about making the switch to FAR.

[Van Disaster]

Congrats on up & back.

Don't descend to 20km at 150m/s down, for starters on my trip down I just held the nose up slightly ( that 2 deg pitch in the pics ) until my descent rate got to around 50m/s, and then changed to actively piloting. TWR of 0.75 is excessive for a spaceplane, there's a reason you have wings try more like 0.5. As for landing, there's a reason I turn to finals over the island runway, gives all that time to get lined up properly... and if you're not lined up properly, well, you're in a plane: go around & do it again.

FAR removes drag from air intakes completely, doesn't matter if they're open or not so don't worry about that. Stock's idea of air intakes is a bit odd...

[Hodo]

I can report a successful flight to orbit last night in the Screech Owl:

{snip}

Anyways, I got into the pattern, relatively lined up on the Runway, slowed my rate of descent to 5 m/s, then fouled the approach and crashed. I think it was because I wasn't level at touchdown; that sort of thing used to screw me up with stock aero on occasion. It's been a few weeks since I've done a landing of any kind - and my first attempt at one in FAR. I did quick-save the game so I can try again; any pointers on things I should do differently this time (like, not put myself in a spin to lose speed)?

Ok from what I can tell on the craft and from your description you have more than enough power to get to space, which is how you are doing it, brute force. You basically have a manned missile. The wings are really not that affective at their thickness, they should be wider at the root to generate more lift at lower speeds. This is why you are running into so many problems on re-entry and landing. Your wings are not generating enough lift to keep your craft in the air at those points.

Take this craft for example, it is a fair bit smaller but it can go to an orbit of 100km and back and land at the KSC.

Notice this is a delta wing design this gives it a GREAT deal of lift at low speeds and at supersonic speeds.

But this light duty SSTO that I use currently is also a delta wing design but laid out a bit differently.

Javascript is disabled. View full album

But my most successful SSTO and heavy lifter for a good while, almost a solid year was a traditional wing setup.

Javascript is disabled. View full album

[capi3101]

Let me ask this question real quick - I've been largely considering the aspect ratio of the wings as a function of span-over-chord (you know, if it's a 2:1 ratio, then the span length is twice the chord length). I haven't been giving much thought to going the other direction of chord-over-span (i.e. a 2:1 ratio having the chord twice the length of the span). Would there be any point in me trying a design like that?

[Van Disaster]

Concentrate on wing area first, you'll naturally arrive at a suitable shape given the rest of the design. But I said all this already

[Hodo]

Let me ask this question real quick - I've been largely considering the aspect ratio of the wings as a function of span-over-chord (you know, if it's a 2:1 ratio, then the span length is twice the chord length). I haven't been giving much thought to going the other direction of chord-over-span (i.e. a 2:1 ratio having the chord twice the length of the span). Would there be any point in me trying a design like that?

You are over thinking the design process. Van Disaster is right this has all been said before.

With any aircraft if it looks right it will fly right. Now there have been some odd designs that have flown. But at first stick to the basics of getting an aircraft to fly. I started small, I built my SSTO program to where it is today by starting with simple planes then building supersonic aircraft then craft that could achieve near orbital velocities then finally aircraft that could achieve space interface then last a true SSTO space plane. Now almost 2 years later I can build some pretty decent space plane SSTOs and planes in general that work at almost any situation, from VTOLs, to gliders.

It just takes patience and practice.

[eddiew]

You are over thinking the design process. Van Disaster is right this has all been said before.

With any aircraft if it looks right it will fly right.

Have to echo this Capi You've got a habit of trying to find magic mathematics that will make everything work; it's probably great if you're in the engineering trade, but KSP doesn't need you to be that formulaic. KSP just needs you to make a plane that looks like a plane ^^

Delta wings or conventional, you can take your pick really, but the thing that really shouts out about the Screech Owl is that the canards are bigger than the tail, and either have no control surfaces, or are entirely control surfaces. I think both those points make it unique and therefore something to be suspicious of (It's ok to have small canards that are control surfaces, but not such a large proportion of your lifting surface! If I'm right about your design, then you end up with more of your wing being off-axis than on-axis when you try to pitch, which just doesn't sound safe...)

Best advice (and how I got to grips with FAR); mimic someone else's plane that does roughly the same thing. You'll be surprised at how it doesn't fly... Then you'll look twice and realise all the ratios are wrong, and it'll be glorious after fixing it

[capi3101]

(It's ok to have small canards that are control surfaces, but not such a large proportion of your lifting surface! If I'm right about your design, then you end up with more of your wing being off-axis than on-axis when you try to pitch, which just doesn't sound safe...)

Do you think that's why the plane is behaving the way it is once I'm up in space?

[Wanderfound]

Misbehaviour in space is probably due to offset CoM/CoT. Aircraft very rarely have perfectly vertically centred weight distribution.

Just eyeballing it isn't enough; a tiny misalignment can cause a lot of torque. Use RCS Build Aid to check it.

[eddiew]

Do you think that's why the plane is behaving the way it is once I'm up in space?

To be fair, probably not; as Wanderfound said, that's most likely thrust misalignment. Happens easily in SSTOs since you tend to have vertical stabilisers only on the top. Sometimes using the offset tool to nudge an engine up or down a 'pixel' or two can help this enormously, but the RCS Build Aid will tell you exactly the kind of torque you're getting

[UnusualAttitude]

I use the rotate tool (angle snap off) to try and get my main engine(s) to fire through the craft's CoM. You won't get it perfectly aligned, but you can get it close enough so that full throttle can be held straight by SAS on all but the largest ships.

A quick and dirty fix, if your orbital TWR is high enough, is to simply limit the thrust of your engines.

[capi3101]

I do have RCS Build Aid - that's that torque parameter in Engine mode I should be looking at then, right? Okay - I can do that.

Didn't have a chance to play last night on account of busted air conditioner and crummy weather, so nothing new to ask y'all today.

[Wanderfound]

When you look at it, make sure only the relevant engines are in the first stage.

[Hodo]

When you look at it, make sure only the relevant engines are in the first stage.

Or you can just set the thrust limit to 0 on every engine but the ones that provide power for that mode.

[Van Disaster]

If it's a little off then it's a little off, turn RCS on. If it's a lot off then moving the thrust vector is worth doing.

I keep banging on about wing area because it's all that really matters - FAR isn't sophisticated enough to pay all that much attention to shape. I've paid full attention to principles of supersonic design, tried to keep as much of the craft inside the shock cone as I could ( although mach angle at mach 5 is 11 degrees and that's nearly impossible with what we have ) ... and then built gigantic forward sweep aircraft that look more like subsonic transports and had them work just as well. You don't have to worry about vortex issues because there aren't any, and so on. What you have to do is give the craft enough wing, which means a low enough wing loading. The next thing you have to do is give it enough control leverage, which is just control surface area * distance from fulcrums, and anything else is just dressing or tuning.

You're building a spaceplane though, and it has to be a viable if limited envelope aircraft or you should just dump the wings and make it a rocket. First thing you need to do when you think you've built something is to take off, fly until just over mach 1, slow down, turn round and land it. Don't stick a payload in it, don't try going to space, just a simple 15 min flight. A lot of potential issues will show up just doing that.

And one last thing - what AoA is the static derivative tab giving for your designs at 25km alt/Mach 4?

[capi3101]

Haven't been checking 25k/Mach 4 - usually I check 0k/M0.35, 5k/M0.85, 10k/M1, 15k/M2, 20k/M3, 30k/M4. I general I haven't been getting red numbers until 30k/4, and those have been in the N_ß parameter (i.e. "needs moar tail"). Occasionally I'll get M_w and X_u instabilities at that level too. Level AoA at that level is up around 20 degrees as a rule.

The X_u baffles me - "X-force changes with forward speed"; does that mean I need to throttle back or something? Usually I do whatever I need to do to fix the pitch-up tendency and the X_u goes away in the process. That usually involves me having forgotten to not put my tailplane level with the wings instead of moving them up towards the top of the fuselage and giving them dihedral; it goes away when I do that. Seems I still need to un-learn some of my old stock aero habits...

Didn't get a chance to play again last night - same reasons (the upstairs of my apartment, where my KSP box is located, has a temperature currently being measured in Scoville units/percentage of Hells). Spent the day looking at RL supersonic designs and Wanderfound's designs. I do have a question there - I noticed a fair number of the "engine unit" designs included two FL-T800 tanks but a few utilized one, and a few utilized three. What kind of criteria do y'all use when determining how much fuel to load? I think I might have asked this question before (and probably didn't understand the answer). I did note that the SR-71, X-15 and Concorde all had dry weights that were about 45% of their normal takeoff weight; is that a good metric to shoot for in KSP?

Edited April 2, 2015 by capi3101

[Hodo]

Haven't been checking 25k/Mach 4 - usually I check 0k/M0.35, 5k/M0.85, 10k/M1, 15k/M2, 20k/M3, 30k/M4. I general I haven't been getting red numbers until 30k/4, and those have been in the N_ß parameter (i.e. "needs moar tail"). Occasionally I'll get M_w and X_u instabilities at that level too. Level AoA at that level is up around 20 degrees as a rule.

The X_u baffles me - "X-force changes with forward speed"; does that mean I need to throttle back or something? Usually I do whatever I need to do to fix the pitch-up tendency and the X_u goes away in the process. That usually involves me having forgotten to not put my tailplane level with the wings instead of moving them up towards the top of the fuselage and giving them dihedral; it goes away when I do that. Seems I still need to un-learn some of my old stock aero habits...

Didn't get a chance to play again last night - same reasons (the upstairs of my apartment, where my KSP box is located, has a temperature currently being measured in Scoville units/percentage of Hells). Spent the day looking at RL supersonic designs and Wanderfound's designs. I do have a question there - I noticed a fair number of the "engine unit" designs included two FL-T800 tanks but a few utilized one, and a few utilized three. What kind of criteria do y'all use when determining how much fuel to load? I think I might have asked this question before (and probably didn't understand the answer). I did note that the SR-71, X-15 and Concorde all had dry weights that were about 45% of their normal takeoff weight; is that a good metric to shoot for in KSP?

KSP is a fair bit smaller scale than real life where the SR-71 and Concorde are concerned. That and jet engines work a bit differently here than in real life. In KSP they are a LOT more efficient than the real life jets. I have built a mach 4+ aircraft using Realism Overhaul with AJE and FAR.... it was roughly 60% fuel just to get to those speeds.

You can look at some of my designs if you want to see some alternate designs to Wanderfound.

Imagur Album

[Wanderfound]

Spent the day looking at RL supersonic designs and Wanderfound's designs. I do have a question there - I noticed a fair number of the "engine unit" designs included two FL-T800 tanks but a few utilized one, and a few utilized three. What kind of criteria do y'all use when determining how much fuel to load?

The only criteria I use is intended mission. LKO stuff gets smaller tanks, interplanetary stuff gets larger tanks. Most of the longer-ranged things are designed to land and return on Duna or similar.

To LKO, one FL-T800 is usually plenty for a Mk1 ship, two do just fine for a Mk2. And if you push the limits, you really don't need much oxidiser at all...

Edited April 2, 2015 by Wanderfound

[Van Disaster]

I did note that the SR-71, X-15 and Concorde all had dry weights that were about 45% of their normal takeoff weight; is that a good metric to shoot for in KSP?

Probably not - KSP parts usually have considerably higher dry mass than they would in real life ( unless you're using RO or some other mod that edits everything ) and we don't have the facility to custom build fuselages unless you're an actual modeller. ( I am a modeller and I still won't custom build fuselages anyway ). The X-15 is a bit wonky given what it is - basically just a rocket with control surfaces - although it's useful to note the hypersonic design features of it.

My fuel criteria is "will it get to orbit", and that's it. If it's heavily loaded I might use LF droptanks because it's going to burn a fair amount just getting to altitude, but on the odd chance I do get to orbit with excess fuel I'll either leave the fuel in orbit in a tank somewhere, or just it back.

[Hodo]

Probably not - KSP parts usually have considerably higher dry mass than they would in real life ( unless you're using RO or some other mod that edits everything ) and we don't have the facility to custom build fuselages unless you're an actual modeller. ( I am a modeller and I still won't custom build fuselages anyway ). The X-15 is a bit wonky given what it is - basically just a rocket with control surfaces - although it's useful to note the hypersonic design features of it.

My fuel criteria is "will it get to orbit", and that's it. If it's heavily loaded I might use LF droptanks because it's going to burn a fair amount just getting to altitude, but on the odd chance I do get to orbit with excess fuel I'll either leave the fuel in orbit in a tank somewhere, or just it back.

I know when I am designing a SSTO I put a minimum of 2.5km/s delta/V on it for fuel, but I have been known to go up to 6km/s delta/V when designing. But it depends greatly on the designs job and mission. Currently I am in the process of redesigning my whole spaceplane SSTO cargo hauling fleet. But not as actively as I once was. I only recently got around to finishing three SSTO cargo planes that I plan on using for missions. And I cant say I have messed with the new larger 2.5m stock aircraft parts. I just dont like the look of most of them, they are a bit clunky when used with other stock parts and dont help with the parts count at all.

But I digress, it really does come up to the basic mission goals for that craft what it has for fuel and engines. If it is a cargo craft it will have a higher TWR when empty then it will when it is fully loaded. My combat craft all have higher TWR across the board but have lower fuel range and no real cargo ability they are also a fair bit more nimble. My science/recon SSTOs are easy to fly and have good range and a VERY small cargo/utility bay for science experiments.

[capi3101]

Okay - so general rule for a payload to orbit craft around 2500 m/s? That's pretty much the only kind of plane I flew while I was still in stock.

[NathanKell]

To orbit alone? You shouldn't need much more than 1200m/s tops, since you're transitioning at over 1km/sec anyway, and that already leaves a good cushion for deorbiting. 2500m/s is only if you want to, like, go see the Mun first.

[capi3101]

To orbit alone? You shouldn't need much more than 1200m/s tops, since you're transitioning at over 1km/sec anyway, and that already leaves a good cushion for deorbiting. 2500m/s is only if you want to, like, go see the Mun first.

Well now - there's getting the plane up to transition speed as well. Granted, that probably doesn't involve all that much fuel, but it ain't exactly zero either...

Good related question then - what is the Isp of Turbojets and RAPIERs in FAR? Are they any different than what they are in stock?

[Hodo]

Well now - there's getting the plane up to transition speed as well. Granted, that probably doesn't involve all that much fuel, but it ain't exactly zero either...

Good related question then - what is the Isp of Turbojets and RAPIERs in FAR? Are they any different than what they are in stock?

The ISPs dont change much it is the power levels that change in FAR. They are pretty much half of stock when sitting still so getting up to speed is a bit more of a challenge and thus gaining altitude is a bit harder than stock.

As for my fuel amounts I like having enough fuel for any situation including orbital transitions and returns. I also find that your required fuel depends on the engines and again its role.

Lets take this craft for example, SP-410.... It is my newest SSTO cargo hauler.

It hauled 70 tons into an orbit of 100km x 100km and still had over 700d/v left with the cargo still inside, and over 1000m/s when the cargo was removed. This was without even touching the cargo fuel amount.

But this design had a LOT more d/v when sitting on the ground because it was designed as a VTOL that used rockets to get it off the ground.

And this small fighter had in rocket mode less than 1200m/s delta/V. And it was still quite capable of reaching my test orbit of 100km x 100km.

Again it comes down to the design.

I personally find no point in ever going to anything less than 100km orbit my lowest space stations sit at 100km. But I also am a bit odd in some of my thinking. I tend to over-engineer most of my designs.

That was a sustained turn of 13.5Gs!

[Crzyrndm]

Good related question then - what is the Isp of Turbojets and RAPIERs in FAR? Are they any different than what they are in stock?

I normally budget around 60-65 units of LF per turbojet per 10 minutes of flight time (fuel usage per engine at full throttle sits between 0.08 and 0.11 units per second IIRC, with the most efficient altitude being 5200). Rockets to orbit/deorbit I normally pack around 1000-1200 m/s of dv (typically transitioning at 1400-1450m/s and 25km) depending on rocket TWR and planned ascent profile (some craft can't pitch up very hard so you get hit with a bit more drag).

[Van Disaster]

Up where you spend most of the jet part of the flight the stock engines have 1230ish ISP for me, although I run KIDS in ISP fix mode. I don't think that affects jets though.

Note TWR, reference area, AoA. Not the best mach 5 wing shape but it works fine, so hey whatever. I can't give any fuel fractions because it's delivering ships to Mun & has an extra tank.