FAR, Shuttles, and What am I even looking at?

[G'th]

So I recently got the itch to do spaceplanes again, and because in my current game I wanted a high capacity crew vehicle, I decided to do a B9 based shuttle. Currently I'm at the stage where I'm working on making sure it will fly well and I've hit a bit of a road block, as the way FAR appears to work now is a lot different from the last time I played the game, and at this point I"m not sure what I'm looking at when I generate the information via FAR.

So this is what I have right now. Its no where near a finished state and is mostly just where I'm at right now. In order to go farther, I really need some help on what I'm looking at what I need to fix. I've already test flown this (For reentry) and while it stays stable it rolls for some reason. Then once I come out of reentry and start coming down below Mach 1 stability almost goes out the window.

Javascript is disabled. View full album

Some notes:

1. The engines are just there for mass atm. I haven't worked at aligning them yet. Same thing with the RCS pods.

2. The odd shape of the wings and CS' is just due to this being a WIP. My first stab at it didn't work at all and looked a lot more like actual wings.

3. I re-scaled the B9 parts so that they can be used to transport 2.5m parts.

[Empiro]

The area ruling tools are mostly for supersonic drag (as far as I can tell). It doesn't affect stability.

One immediate problem though is that the center of lift if way too far back, and that may cause your craft to nose dive. You might improve roll stability by adding a dihedral winglets at the tips of your wings.

[capi3101]

You'd probably have better luck on the Official FAR design craft thread, but since you asked...

What you're looking at there are area ruling graphs. The green curve shows you the amount of cross-sectional area over the length of your craft; for proper area ruling, you want no abrupt changes in the smoothness of the graph. The yellow line measures changes in cross-section ; you'd like to have this line as flat as you can possibly make it. Finally, the blue line measures pressure over the length of your craft; where the graph crests, your craft is creating the greatest amount of drag.All three lines help to identify where your craft will create drag and gve you an idea of where to make adjustments in the design.

Now, as far as your stability concerns go - you can check those in the dynamic stability simulation tab. Drain your plane of fuel and hit the simulator around the time all hell brakes loose in terms of altitude and speed. I believe the parameters you want to check are N_ß, L_ß, and M_w. Put a value of 1 in the beta and w parameters (not at the same time, necessarily) and then mash go - if you see a graph that oscillates but the oscillation damps down or stays the same, you're good. If you see anything else, your craft is a no-go, and will need to be redesigned to maintain stability under the listed conditions. Your static stability parameters can tell you this as well - any red numbers are pretty much an automatic wash on your design, especially if you then run the simulation and get the aforementioned "bad" graph shape (if you get a red number but then test it in the simulator and get a "good" shape, then it's something that's probably livable; I myself have never seen this particular happy state of affairs).

Suggestions - add dihedral to the wing, make the tail bigger and/or move the center of mass forward. That's if you want the thing to fly. If you're looking for aesthetics...well, there's not too much advice I can give there; it looks pretty nice as is.

Edited October 28, 2015 by capi3101

[Van Disaster]

During re-entry:

* Roll is likely because it's yawing & sideslipping, so just add some lateral stability. Dihedral and/or bigger vertical surfaces. However...

* Instability under Mach 1 is probably the CoL shifting forward of CoM. If you fix that you might find that the CoL being further rearward when you're supersonic has cured your lateral instability issues too.

What you want to do is ignore the CoL ball other than a gross marker and use the static derivative tab to make sure everything is green *at a range of speeds and altitudes and fuel loads*. FAR can't empty the craft for you unfortunately, you'll have to do that yourself. You might find if you empty the craft that at 0.95M one critical derivative has gone red, which will be the source of the problem. If it's Mw ( I think ) then it's a CoL-too-far-forwards issue.

[G'th]

Okay now that i know what im looking I was able to fix the wings very nicely and even managed to get along withot going dihedral.

In getting everything balanced and getting my CM movement minimized i even made the thing lighter which is nice.

However, my last issue is that im getting some major SAS type wobble at subsonic speeds. It bobs up and down like crazy. Turning off SAS and going manual solves it, but keyboard flying is less than stellar and it makes landing harder because i cant hold a specific AOA.

Any tips for fixing this? Im going to try reducing torque but i also dont want the torque to be so weak that on orbit becomes problematic.

Another option may be to reduce the CS' deflection, but im unsure how much i could reduce it before its too weak. Presuming everything is at 100% and the plane flies super well at all speeds (aside from the bobbing) how much do you guys think i could reduce?

Id post some pics but my internets down atm so im posting from my phone.

[Van Disaster]

I'd suggest mods for the last issue:

* Pilot Assistant - better PIDs for surfaces and numerical holds for heading/altitude/speed, and a better SAS along with it

* Dynamic deflection - lets you set up control deflection range based on dynamic pressure ( kinda like the FAR assist but considerably more flexible ).

You could possibly fix the sensitivity by shifting CoM forwards a bit too, but I'd probably recommend those mods anyway.

[G'th]

^ ill definitely check those mods out, however my initial ideas proved to be right on point. I made my first landing from orbit just a few minutes ago. Ended up on the side other side of the mountain by KSC, but i still landed safely despite the rough terrain.

Almost have the deorbit position nailed.

[Empiro]

I usually turn off SAS at low speeds -- your plane should be stable enough without SAS if it's relatively well designed. The FAR assists for Roll and Yaw help a lot, and I use trim settings to get a steady glide slope for landing.

[Wanderfound]

OP: most of what you're asking is covered here: http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1353891&viewfull=1#post1353891

But, as stated above, the short version is: the DD or PA mods (or reduce control authority) to fix the SAS wobbles, a bit of dihedral to fix the roll issue. Focus on the static analysis graphs; get it all green at 5,000m/Mach 1 and 20,000m/Mach 4 and most of your troubles should be sorted.

[G'th]

^ Ooh, definitely gonna bookmark that one. Kudos!

Right now I'm working on integrating this thing into a STS like system. Been having trouble though figuring out how to get the thrust offset zeroed out without affecting vehicle performance and without a payload. And of course the other option is to have the COT fire through the CM, but I'm having trouble nailing that down. Its really hard to tell what the proper angle is and with the engines I have not being strong enough to really affect the thrust direction its even harder.

[Van Disaster]

If you're launching off Kerbin rather than RSS Earth, then honestly a STS system is daft ( actually it's pretty daft for Earth but that's another matter ). One of the biggest problem is most engines don't have the gimbal range of the SME. A long time ago I tried something like Shuttle-Saturn to see how it'd go:

As you can see massively overfuelled and the orbiter didn't even need engines although you can see what angle I had to put them at to balance everything. The main thing is orbital velocity for Kerbin is so low that making a HOTOL spaceplane is quite trivial and I've made ... maybe not hundreds, but certainly lots of successful and pretty heavy ones.

[Wanderfound]

^ Ooh' date=' definitely gonna bookmark that one. Kudos!

Right now I'm working on integrating this thing into a STS like system. Been having trouble though figuring out how to get the thrust offset zeroed out without affecting vehicle performance and without a payload. And of course the other option is to have the COT fire through the CM, but I'm having trouble nailing that down. Its really hard to tell what the proper angle is and with the engines I have not being strong enough to really affect the thrust direction its even harder.[/quote']

Both RCS Build Aid and Kerbal Engineer provide thrust torque figures. Angle the engines until the torque approaches zero, and you'll have 'em firing through CoM.

Keeping the torque zeroed while the fuel drains may require some thought, however.

[G'th]

If you're launching off Kerbin rather than RSS Earth, then honestly a STS system is daft ( actually it's pretty daft for Earth but that's another matter ). One of the biggest problem is most engines don't have the gimbal range of the SME.

Well, that's why I have the Space Shuttle Engines mod. And besides, if I was worried about what was efficient or even simple then I wouldn't have bothered with any of this lol.

Both RCS Build Aid and Kerbal Engineer provide thrust torque figures. Angle the engines until the torque approaches zero, and you'll have 'em firing through CoM.

Keeping the torque zeroed while the fuel drains may require some thought, however.

I'll give Build Aid a try and see if it can help me narrow it down. Problem I've been having is that it was impossible to really tell if my engines were angled properly as the COT is affected too much by the SRB's. I wish there was a way where the COT snaps up close to the COM so you can actually tell where its pointed.

[G'th]

Just to give an update on the nights progress and to get even more advice. I eventually gave up on the shuttle bit as the SRB/ET combos I had available just wouldn't mesh with the orbiter as I had it. I'd need super fine tweakables in order to make it work. (as the closest I could get to zeroing the torque was 2ish, and even with that flight was still a nightmare even with heavy SAS)

So, I decided to step back a bit and approach it from a different way. I decided to start off with a Dreamchaser inspired vehicle that covers the large crew capicity of the previous vehicle, and then once I get a solid design with that I'll work on a larger cargo version.

And so far I've done pretty well and have a good design that I'm just working out the low speed kinks on (mostly just finetuning the controls so a tap of a key doesn't send me over my max AOA). But something I'm curious about even though it doesn't seem to be affecting anything is this rather large unexpected dip in drag just before the expected drop when you hit the engines. I'm wondering if you guys see anything that might be causing it. I had a feeling it might be the Tail (as I made it rather thick) but removing it doesn't kill the first dip and its presence is necessary anyway to keep green numbers across the board.

Javascript is disabled. View full album

[Hodo]

Just to give an update on the nights progress and to get even more advice. I eventually gave up on the shuttle bit as the SRB/ET combos I had available just wouldn't mesh with the orbiter as I had it. I'd need super fine tweakables in order to make it work. (as the closest I could get to zeroing the torque was 2ish' date=' and even with that flight was still a nightmare even with heavy SAS)

So, I decided to step back a bit and approach it from a different way. I decided to start off with a Dreamchaser inspired vehicle that covers the large crew capicity of the previous vehicle, and then once I get a solid design with that I'll work on a larger cargo version.

And so far I've done pretty well and have a good design that I'm just working out the low speed kinks on (mostly just finetuning the controls so a tap of a key doesn't send me over my max AOA). But something I'm curious about even though it doesn't seem to be affecting anything is this rather large unexpected dip in drag just before the expected drop when you hit the engines. I'm wondering if you guys see anything that might be causing it. I had a feeling it might be the Tail (as I made it rather thick) but removing it doesn't kill the first dip and its presence is necessary anyway to keep green numbers across the board.

[url']http://imgur.com/a/cnmN0

While those graphs are useful they dont help that much for some of the more stability issues you may run into.

I would rely more on the information found in this screen.

This will tell you if your craft will have yaw stability issues or roll issues, or even pitch issues. It will even tell you at what speed, you will most likely stall at.

[Van Disaster]

That's not drag as such, that's the second derivative of cross-section - which I must admit is quite hard to read off the shape sometimes. I may not have got my head round it properly but it's the magnitude of the peaks that causes wave drag rather than the direction of it. What is it telling you about cross-section area?

[G'th]

Hm, interesting. The stability derivatives screen doesn't give me any red flags at pretty much any situation I could think of (outside of impossible ones, like M6.9 at .5km up) so that does explain why my plane flies so well despite that yellow line.

The cross-section is listed at 9.65m/s2. Not sure if thats good or bad lol.

[capi3101]

What's your critical mach and wave drag area? Really, that's all the yellow squiggly line is telling you - where the wave drag is being generated and how much. In this case, your surface area ramps up near the tail, hence the huge spikes in that graph. It's not the end-all be-all of FAR flight, though; in my experience, stability is far more important, and you can get a craft up into orbit even with ungodly amounts of wave drag (it just might not have as much delta-V available to it once it's up there).

I'd still run the dynamic stability simulation for 5,000m/Mach 1 and 20,000m/Mach 4 just to be on the safe side. Checking the w and ß parameters should be sufficient - just type in a value of 1 and mash go, and watch what the graphs do. As I've said before, you can be green across the static stability board and still have dynamic stability issues.