Failed Amicitas build files... · 7:42pm Feb 27th, 2020
Here:
https://www.dropbox.com/sh/38tpyyu3xsmrb4z/AAAaQh9Iz_olrWGT9b4mVa0xa?dl=0
Includes both the ship as designed in the spaceplane hangar and the Shuttle-type launch stack.
Center of mass and center of thrust issues bedevil me. I never yet have tried re-entering the ship into atmosphere, but I have no doubt it's a death-trap, tail-heavy as it is. Even in orbit without the stack, I have to activate RCS for SAS to keep it on target at 1/3 throttle; anything higher makes it backflip. However, with the main fuel tank still attached the thing FRONTflips in atmosphere beginning just before the SRBs burn out.
As for looks... well, this is as close as I could get with stock KSP parts to what I envisioned. In-story the wings are less obviously wings and more flattened-out curvy tailfins, sort of like making a cigar rocket out of clay, setting it on its side on a table, and very gently smooshing it down a bit so the table gives it a flat bottom. And, of course, it's supposed to be pink.
But you were curious, so here you are. Feel free to fix it if you can; I have my doubts.
The main problem with shuttles is that you will always be off center of mass and thrust. That’s why all shuttles at ISS is done by capsule.
To be perfectly clear: my primary goal was to get as close as I could to what I described in the stories. Flyability was a SECONDARY concern.
So the ESA has been fudging aerodynamics with healthy doses of the magic of friendship? I can't say I'm surprised.
5210651 No, worse. There is one magic more powerful than friendship, and that is the Rule of Cool.
But even semi-realistic physics don't like it when you design by Rule of Cool.
Physics has a wedgie coming to it.
Is that why the Shuttles engines point Outwards so far, so within their Gimbal range, they can be kept pointing towards the centre of mass of Tank and Orbiter?
5210749 It's possible. I should probably look at the tail a bit more closely to compare the big three main engines with the much smaller three OMS (Orbital Maneuvering System) engines to see how they're angled.
5210764
The front flip at end of burn is due to the backwards movement of the fuel surface, especially the lower liquid hydrogen tank? so towards the end, the flight dynamics are that the external tank is almost T to flight path?
5210772 Not sure what you're asking.
Towards the end of the SRB burn the ship's tendency to tilt backwards (the fuel tank leading the orbiter) overpowers the orbiter's engine gimbal and reaction wheels. When the SRBs burn out, that becomes totally uncontrollable, with further burning of the main engines only accelerating the speed. Burning the main engines on the orbiter in orbit, without the fuel tank, produces a similar backflip, but of lesser intensity.
5210789
Looking at pad side Shuttle lauches, that thing pushes forwards a fair way even in the first seconds, as well as the expected upwards, which is very confusing as the rotation centre with the SRB is high, and empty tank it low, means the Main Engine gimbals would have to have a heck of a range to compensate?
As a proud owner of KSP I have download the files and will begin testing with stock parts to see if I can make it flyable with minimal part loss or add on.
Edit 1: First Major Issue I'm seeing...What are those engines? Swivels? Your gonna need bigger and stronger ones I feel, commencing Testing.
Edit 2: Stronger Engines have produced greatly improved results and the usage of fuel lines to move the centre of mass around has proved useful. Looking at ways to improve SAS strength without going too far, continuing testing.
Edit 3: Sneaking a reaction wheel set into the engine compartment has improved performance yet the flipping still perplexes me...new tests are in order...
Edit 4 : Partial Orbit and it managed to get back! I think I'm getting closer! (Its looks are largely the same however its components less so)
Edit 5: ORBIT! Will be tweaking fuel loads now
Edit 6: Alrite I've done what I can, sending the file in a new comment, hope this helps!
5210810 The tricoupler in stock parts only accommodates 1.25-tier connections.
here ya go this is my take on what you described and linked flies to. it can take off and with the launch vehicle achieve orbit and refuel something else it can be landed but i have not tested reentry to landing just yet.
https://www.dropbox.com/sh/45prvq2vuidd6po/AABOu8DDxPPE-c6TBMY02ffSa?dl=0
steamuserimages-a.akamaihd.net/ugc/781875933868129593/073E2606551B4DF49424BC75475AACDA13BDCA97/
5210647
Well I've done my best to make it look the same and now be able to achieve orbit! Here ya go https://www.dropbox.com/s/tlxf8woxihhyl3e/Amicitas%20Launch%20Stack.craft?dl=0
5210847 I need the loadmeta file too to test it.
5210845 It's good work, but remember my goal was to recreate the ship designed in the stories. Yours is likely more flyable, but it's a different ship.
5210887
Ah oops let me go procure it!
5210887
Here you go, https://www.dropbox.com/s/gjrfq4h40j7thnw/Amicitas%20Launch%20Stack.loadmeta?dl=0
5210808
Yep, you are right. The engines are pointed outwards to ensure the thrust vector is pointing through the center of mass, and this imparts some horizontal velocity on liftoff and during flight. You can sort of see this on the topmost engine (Engine 3, I think), as its visibly pointed towards the belly of the Shuttle.
You are also right that the Shuttle engines have a large gimbal range — 10.5 degrees according to Wikipedia. Real-life engines tend to have much larger gimbal ranges and TWRs than KSP engines in general IIRC (or at least that was the case when I played more frequently; not sure if that has changed since then). The Vector has the same gimbal range at least, so that may work better for flying, although they may not be as good for story accuracy.
5210910 Um...could you walk me through the changes you made? Because this looks exactly like what I had, including the two Swivels I tried sticking on the fuel tank in an attempt to balance thrust / gain control authority.
5210645
That is not really true; shuttles can be perfectly flyable with the right design. In fact, the vast majority of the ISS modules were brought up by the Shuttle.
5210932
Damn maybe the file didn't save correctly? I'll try re exporting it later sorry man
Alright, after playing around a little bit, here are my (incomplete) notes:
- Those Swivels probably are probably underpowered both in terms of thrust and gimbal range. The CoM moves quite a bit during launch, so you need a decent gimbal range to compensate. I replaced them with Vectors.
- Angle the shuttle engines inwards so your thrust vector is closer to pointing through your CoM. I rotated them in enough to get to orbit semi-reliably; they can probably stand to be fine-tuned some more. At this point, the engines on the bottom of the external tank can be removed.
- RCS Build Aid is your friend.
- I feel SAS has gotten really squirrely since I last played. Test launches tended to go well until right before solid booster burnout, where SAS would freak out for no discernible reason. MechJeb's autopiloting may work significantly better, assuming it works as well as I remember.
- Depending on how closely you want to emulate the real-life Space Shuttle, the Thuds on the Amicitas should be disabled during launch and only used for orbital maneuvers. The real-life Shuttle only used its main engines to get to orbit; once it was there, those engines were not used. If you end up going this route, the Thuds will need to be angled downwards to thrust through the Amicitas's CoM.
">/embed.js" charset="utf-8">Craft files (sorry for the pastebin links; I don't have Dropbox installed yet):
Amicitas Launch Stack.craft
Amicitas Launch Stack.loadmeta
With this I can reliably get into orbit, albeit in a really ugly manner. I turn the throttle to ~2/3, enable SAS, then launch. I manually move the nose down to about 85 degrees, wait until I hit about 200 m/s, then switch to prograde SAS. The throttle is increased to 100 by Mach ~2. By the time the solid boosters burn out, my apoapsis is well over 100km, so even though SAS has put me into a tumble I can just coast and regain control once I'm out of the atmosphere. From there, circularizing is fairly straightforward, although you may end up in a spin right as the fuel runs out.
5210954
Well, 2 out of 6 space shuttles will disagree
5211031
Except those disasters had nothing to do with CoM/thrust vectoring issues. The flight characteristics of the Shuttle on launch/return are quite sound.
5210971 I'm thinking I'll go in and cant the tricoupler holding the Vectors as you did, but add three Thud engines around them similar to the OMS engines at points that add up to the center of thrust passing through center of mass of the orbiter. That gives me something that can still make orbital transfer burns.
I didn't think of Vectors because I've never gone so far up the tech tree as to unlock them in career mode.
5211403
I can't remember when during my messing around I took that screenshot, but I think in the final product the outermost Vector engine is also angled outwards relative to the tricoupler, similar to the topmost engine of the Space Shuttle IIRC.
I don't remember the flight characteristics after jettisoning the external tank, but there's a chance that it won't be possible to angle the Vectors in such a way that you can fire through the CoM both with the ET attached and after it has been jettisoned. If you're aiming for a Space Shuttle-like flight experience then that's OK, since the Shuttle only used its main engines to get its apoapsis high enough and used the OMS engines to take care of circularization and subsequent orbital maneuvers.
Vectors are wonderful. Same TWR as Mainsails, large gimbal range, great sea-level and vacuum efficiency, all in a nice 1.25m package. Main problem is that they're super expensive, but that seems like that could be fitting for the Amicitas.
You might be able to get the wings to curve to better match what you had in mind with liberal
abuse of the offset tool and clipping, too. I probably won't have time today, but I might be able to find some time later.