[1.3.1] Ferram Aerospace Research: v0.15.9.1 "Liepmann" 4/2/18

[Gfurst]

Talking about engines, I was flying my humble attempt at a small plane, it reaches around 0.86 mach maximum, and I'm actually using MRS small basic engines, which is basically balanced to the regular jet engines.

With FAR I can get up to 15k altitude, which is much better than regular stock limits, however despise the very low thrust on the engines, I get very low drag too, allowing reasonable speed to maintain level flight, above this altitude the engine flames out, but still there more than enough airflow.

Question is? I'm aware of the new curve for engines and the change thrust due to speed, have there been any other modifcations to make air engines a bit more complex?

[ferram4]

@Kitspace: 1. Simple; build some deflection into the control surface in the editor.

2. Possibly, but at the same time this then raises the question of how to detect that. In the end, this means adding a lot more options to an already bloated tweakables menu, which I really, really, really don't want to do.

Also, simple reason why IAS > TAS: once you get above Mach 0.3, compressibility effects start becoming important. One of those effects is that the stagnation pressure increases faster than what would be expected using the simple "static pressure + dynamic pressure" measurement that a pitot tube uses to measure velocity.

@Neuntausend: Logs and reproduction steps. That's not supposed to happen, and I'm really confused as to how you managed to get lift with no drag.

@jrandom: Area ruling. You need more of it.

[MatterBeam]

Talking about engines, I was flying my humble attempt at a small plane, it reaches around 0.86 mach maximum, and I'm actually using MRS small basic engines, which is basically balanced to the regular jet engines.

With FAR I can get up to 15k altitude, which is much better than regular stock limits, however despise the very low thrust on the engines, I get very low drag too, allowing reasonable speed to maintain level flight, above this altitude the engine flames out, but still there more than enough airflow.

Question is? I'm aware of the new curve for engines and the change thrust due to speed, have there been any other modifcations to make air engines a bit more complex?

Jet engines have both an Atmosphere curve and a Velocity curve.

velCurve		{
			key = 0 1 0 -0.125804
			key = 0.2 0.98 0 0
			key = 0.6 1 0.2399533 0.2399533
			key = 1.05 1.167 0.5409369 0.5409369
			key = 1.67 1.46 0 0
			key = 2.1 1 -3.49991 -3.49991
			key = 2.278022 0.496117 -3.164169 -3.164169
			key = 2.5 0 0.004306508 0
		}
		atmCurve
		{
			key = 0 0 0 0
			key = 0.15 0.2 2.448394 2.448394
			key = 0.5 0.6 0.5020669 0.5020669
			key = 1 1 1.047895 0
		}

In essence, even if everything goes correctly, an atmospheric pressure of 0.15 drops your thrust to 20% of maximum. Going at Mach 2.27 drops you thrust to 49% of maximum. Going at Mach 2.27 at an altitude where pressure is 0.15 gives you about 10% of the thrust you had at ground level.

Area ruling. You need more of it.

Can I add this to my sig?

[jrandom]

I'm sorry! It's my first area ruling design! PLEASE DON'T HURT ME PHYSICS DOES THAT ENOUGH ALREADY!!

*curls up in corner sobbing*

Actually, I'm just gonna go build more planes! Whee!

[Pirke]

Since KSP 1.0 has come out, I updated today and gave it a spin. I am running KSP 1.0.2 x86_64 under gentoo linux and have the current FAR installed. Currently I am flying a Mk 1 Pod in a "stable" orbit with a periapsis at 22.5 and apoapsis at 25km. They go up a kilometer, then they go down one, but I don't lose any speed at all. The atmosphere is hardly slowing me down, despite the constant heating effect. Apparently, the lift is high enough to push the orbit before me up enough to keep me orbiting constantly. It's quite impossible to land a ship that way, because the parachutes come off if I return from a low Kerbin orbit and still have 2000m/s at 10km altitude.

Is that something that happens with the new FAR, and can I work around it somehow?

I don't think you have real lift, you're simply in a very low orbit as if there was no atmosphere.

I had a similar issue: 3 MK1 pods stacked on top of each other, during reentry they didn't slow down. As I had some tourist issues after that I made a bugreport with a link to logging here: http://forum.kerbalspaceprogram.com/threads/121042-Unable-to-resume-a-flight-with-tourists-from-space-center.

Note that the logging is made after restarting KSP again and being unable to resume the flight. After some save game hacks i managed to resume the flight but the issue didn't occur the second time.

[jrandom]

Transonic curve question: I just slapped together this fuselage and the yellow line is all squiggly -- much squigglier than I'd expect. What am I looking at here?

Javascript is disabled. View full album

[funkcanna]

Hi there, is there an updated tutorial or documentation on FAR? Im having trouble with the interface and getting rockets to be stable (somebody mentioned to be stable I need a "negative slope along the yellow line", which makes zero sense to me!), and a brief scan through the 9000+ posts was a bit difficult!

[Wanderfound]

Transonic curve question: I just slapped together this fuselage and the yellow line is all squiggly -- much squigglier than I'd expect. What am I looking at here?

http://imgur.com/a/3LFff

The graphs auto-scale in relation to the maximum amplitude of the curves. So, if you have a design that has a very consistent (but not perfectly consistent) cross section, your lines are going to look wiggly because the graph is "zoomed in" a great deal. In contrast, a ship with massive divergences in cross section is going to have a relatively smooth looking yellow line due to the compression of the Y axis.

The black lines provide scale.

- - - Updated - - -

Hi there, is there an updated tutorial or documentation on FAR? Im having trouble with the interface and getting rockets to be stable (somebody mentioned to be stable I need a "negative slope along the yellow line", which makes zero sense to me!), and a brief scan through the 9000+ posts was a bit difficult!

See http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1353891&viewfull=1#post1353891

That post is a spaceplane design guide, but it includes an explanation of how to use the FAR analysis tools (apart from the new area rule stuff, which you don't need to worry about yet).

The yellow line being referred to is the Coefficient of Manoeuvrability in an AoA sweep, as seen here:

Edited May 13, 2015 by Wanderfound

[jrandom]

The black lines provide scale.

So that's what those are.

[blowfish]

Transonic curve question: I just slapped together this fuselage and the yellow line is all squiggly -- much squigglier than I'd expect. What am I looking at here?

http://imgur.com/a/3LFff

The yellow line represents the curvature (second derivative) of the green line (which is cross section). Look at the green line - notice that there are "kinks" in it where the slope of the cross section (green line) changes very quickly. Those will lead to very large deviations in the curvature (yellow line).

[FaterW]

Hello!

I've recently started experiencing some serious lag in VAB and SPH. I have several mods installed but by switching on and off the few that make modifications to ship editing in different combinations I've managed to determine that I only have the lag if I have FAR and TweakableEverything installed at the same time. I looked at the log and if both mods are active it is spammed with records like this every 1-2 seconds:

[LOG 02:50:16.902] Updating FAR Shuttle 'Montauk'

[LOG 02:50:17.547] Voxelization Time (ms): 643

With FAR on and TE off it only updates when I change something on the ship, but if both mods are active it seems something is constantly making FAR update.

I've tried installing and uninstalling them with CKAN and manually too, nothing changed. Later I will try the game with only these two mods installed.

[pandastratton]

TY for the new update!

[Amram]

So that's what those are.

yep, and the take away to be gained from that is that you will likely never achieve a perfectly smooth line, because even when your design borders on perfect, the parts themselves may have deviations that rob you of that perfect line. The closer you get, the more far will zoom in, and the more squiggly the remaining line will look.

[werk62]

I am having a minor problem where the FAR debug settings do not save between game sessions so every time I launch the game I have to set them again. Not game breaking, but moderately annoying. Unfortunately I am running a crapload of mods so it is pretty much non-diagnosable. Just wondering what types of mods might interfere with that so I can narrow it down.

[Wanderfound]

There's a bit of an issue with the landing gear setup on the Firehound...

[Hektos]

Getting this thing off of the runway is a feat of aerobatics, and my only "landing" recovered only the cockpit and a wheel.

But it goes fast! And on a single, standard jet engine!

http://i.imgur.com/5E6P3fu.jpg

http://i.imgur.com/XiZrUUJ.jpg

My engine and intakes look funny because of Ven's part revamp. They're the stock basic engine and intake.

Dang! Crazy design! I guess for speed it's all about stretching the plane then?

I just can't seem to smooth out the area curve any more with the parts available to me (only at the starter science building tech wise).

Not using any parts mods either (I had adjustable landing gear but it actually messed everything up).

EDIT: @Wanderfound That "front" gear looks like its WAY too far back xD

Edited May 13, 2015 by Hektos

[futrtrubl]

Dang! Crazy design! I guess for speed it's all about stretching the plane then?

http://i.imgur.com/ETCNgvX.pnghttp://i.imgur.com/9h1WIIT.png

Stick an antenna on the front of the plane, that will help with that initial spike in cross section. Also don't be afraid to add low weight parts just for the smoothing they provide. You look like you have ~20 parts so you have a little leeway part count wise even for the lvl1 SPH.

Edited May 13, 2015 by futrtrubl

[ferram4]

@FaterW: That indicates that TweakableEverything is updating the scale of parts way, way, way too often. There is nothing I can really change on my end, because I have to update the shape every time the part is rescaled; if TweakableEverything is rescaling parts when it doesn't have to, that's on that mod to be more efficient.

My only other solution is to drop any kind of TweakableEverything / Tweakscale support. Which, given the frequency of bugs with it, might be the best option at this point.

@werk62: I thought it was saving... or do ScenarioModules not automatically save when they have an OnSave() block defined?

In any case, I added some code to ensure that it saves.

@Wanderfound: Cannot reproduce, mine ends up sitting quite nicely. Hell, I balanced it accounting for that. Are you using a mod that changes the mass of parts?

[Wanderfound]

@Wanderfound: Cannot reproduce, mine ends up sitting quite nicely. Hell, I balanced it accounting for that. Are you using a mod that changes the mass of parts?

Not that I'm aware of; the only parts mod I use is TAC-LS, and if anything that would add mass to the cockpit. It's still possible to launch if you use the jet thrust to force the nose down before releasing the brakes. Maybe it's close enough to the tipping point for a random wiggle to make the difference?

[blowfish]

ferram4, Wanderfound: neither of you are using AJE, are you? AJE definitely both changes the masses of engines and moves their CoM around.

[Wanderfound]

ferram4, Wanderfound: neither of you are using AJE, are you? AJE definitely both changes the masses of engines and moves their CoM around.

I'm not. Ferram?

[funkcanna]

See http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1353891&viewfull=1#post1353891

That post is a spaceplane design guide, but it includes an explanation of how to use the FAR analysis tools (apart from the new area rule stuff, which you don't need to worry about yet).

Thank you, this is massively helpful, however I'm still confused as to how to CHANGE any of those lines on the graph. For example, I posted a thread about this (thinking it was the CoL that was causing instability - http://forum.kerbalspaceprogram.com/threads/120928-Weird-Center-of-Lift). I used the FAR graphs for this rocket design and my Coefficient of Manoeuvrability is positive. Based on your post, this means its unstable (which it is!) - but I don't know what to do to make the Coefficient of Manoeuvrability negative in terms of design. Sorry if theses are rudimentary issues, I just don't know the answer yet Also very confused by why Cd is so high and how to change it!

Also when inputting the Lower and Upper along with Mach numbers, for rocket building I assume I should put lower at 0, upper at 5 (as I don't really want to pass 5º AoA I believe? and have Mach at 5. Is this correct?)

[Crzyrndm]

Not that I'm aware of; the only parts mod I use is TAC-LS, and if anything that would add mass to the cockpit. It's still possible to launch if you use the jet thrust to force the nose down before releasing the brakes. Maybe it's close enough to the tipping point for a random wiggle to make the difference?

It's balanced if I fill up empty fuel tanks (Mk2 to 1.25m adapter and the bicoupler both have no LF in my version).

Also when inputting the Lower and Upper along with Mach numbers, for rocket building I assume I should put lower at 0, upper at 5 (as I don't really want to pass 5º AoA I believe? and have Mach at 5. Is this correct?)

For rockets, I find using the Mach sweep is often more helpful. You can see at which what speeds a certain AoA will be stable/unstable (instead of entering a speed, you enter an AoA, everything else is the same.). Increasing stability requires decreasing drag forward of the CoM and increasing it aft of the CoM (in this case, if you make your center stage extend further above your outer stages, it'll shift the drag they create back and increase vessel stability)

Edited May 13, 2015 by Crzyrndm

[ferram4]

@Wanderfound: I don't use AJE when designing stock example craft. Random wiggle isn't going to do it, considering that I've launched it a ton without that happening. You didn't angle it back down from its 5-degree-up position before launching, did you?

@funkcanna: You'd want the rocket to be taller, not as wide, and have more fins. But a more practical solution would be to replace all of those engines with ones that gimbal and use the control authority from that instead.

[EnDSchultz]

Hey Ferram,

In the latest build it's looking like the "Allow Aero Failures" option is getting re-activated any time I revert or return to the space center.