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

[Korn Sarum]

I am having a proble, similar to the one Dreadp1r4te is having. It happens every time I re-enter after loading a quicksave in orbit: the craft either enters a perfect flatspin, or I lose all control and it drifts to the ground at less than 50 m/s, even though it should be moving at at least 100 m/s. The problem seems to occur only after loading a quicksave. If I re-start the game and enter the craft from the space centre I encounter no problems at all.

[Dreadp1r4te]

I am having a proble, similar to the one Dreadp1r4te is having. It happens every time I re-enter after loading a quicksave in orbit: the craft either enters a perfect flatspin, or I lose all control and it drifts to the ground at less than 50 m/s, even though it should be moving at at least 100 m/s. The problem seems to occur only after loading a quicksave. If I re-start the game and enter the craft from the space centre I encounter no problems at all.

This could be another problem entirely, and one I've heard of before; since your quicksave is already in orbit and (hopefully, if its orbital) out of the atmosphere, no aerodynamic calculations are made for the craft. I'm not sure if this was fixed or ever addressed, as I only today started using FAR, but I've heard mention of it before in my quiet forum lurking. Make sure you're not encountering the same issue I was before you panic, though; it might help to post some pics of the craft for Ferram (or anyone more aerodynamically adept than I) to take a look at. I know enough to get a plane in the air, but not enough to bring it safely back down. No wonder I can't find any test pilots...

[Virindi]

I have an interesting 'problem' and I am not sure if it is caused by FAR, B9 intakes, or the game itself. When I switched from 0.22 to 0.23, the flameout point for jets (vanilla turbojet) moved WAY up. A spaceplane that in the past flamed out at 25km can now run at full throttle at 32km, so almost zero deltav is needed to circularize. My spaceplane has one turbojet and two B9 "divertless supersonic inlets". It is now easily able to top at 2400m/s in the atmosphere, where before top speed was 1500m/s with the exact same plane on KSP 0.22 and FAR 0.11. Also possibly noteworthy, Mechjeb's flameout protection utility still attempts to throttle back at 25km as if the old flameout point was still there.

Any idea what is going on?

EDIT: Here are a couple screenshots of before and after flameout. Note the 'air req met': 10.2% and the engine is still running.

Edited January 7, 2014 by Virindi

[DresCroffgrin]

As much as I hate to be the bearer of bad news, I don't think 0.12.5.1 completely fixed the reentry issue from craft loaded in space. Another one of my crafts just suffered from instantaneous deceleration at the edge of the atmosphere. However, closing the game with a panic-induced Alt + F4 and repeating the process with the same craft on the same trajectory resulted in a perfectly normal reentry.

[Surefoot]

By spinning it up so everything is in tension you'll cause centrifugal forces to act as a sort of restoring force to counter any aerodynamic forces that might cause it to flex out of control.

You mean the gyroscopic effect is accounted for in the game physics engine ? Or is it a by-product of other physics rules ?

[illectro]

You mean the gyroscopic effect is accounted for in the game physics engine ? Or is it a by-product of other physics rules ?

It's a by product of simple physics in a cartesian coordinate system.

[theSpeare]

Has someone written a tutorial somewhere on how to use your stabilization simulation graph thingy?

[illectro]

I was actually starting to write a script for a video tutorial, but I've run into a few things that aren't documented (like, what unit is 'Density' in the Data & Stability Derivatives page).

[Konnor]

Greetings. I'm here with a bug report. A problem I mentioned in the Deadly Reentry thread is, I believe, actually FAR-related. I've made a test case with nothing but FAR (v0.12.5.1) and Kerbal Engineer (v0.6.2.2, but no parts were used during the test). To reproduce it, I simply set a lander just near the runway, took off, made a circle and came back; when I get close to that lander, it always explodes (with structural failures and collisions in the log). But this does not happen with FAR v0.12.3.

output_log.txt: http://ge.tt/1lFELxC1/v/0

The fun begins after the line 9259 (but I'm not sure this line itself has something to do with the explosions):

[Lander]: ground contact! - error: -0.008m

[ferram4]

@Korn Sarum: You aren't using version 0.12.5, right? There was a fix in v0.12.5.1 that was supposed to address that.

@Dreadp1r4te: That sounds like a bunch of people who don't know what the code does talking out their asses and trying to convince everyone how smart they are. All of the aerodynamic properties are determined when the craft loads; they just aren't applied until there's an atmosphere to apply the forces.

@Virindi: That was a fix to unintended behavior in the stock game. Engines originally did not respect their "ignition threshold" value and would flameout the second they tried to pull more resources than were available. Now, so long as there are more than 10% of the requested resources available the engine will throttle down, rather than flaming out. This was behavior that was supposed to be there all along, but was not.

@DresCoffgrin: Uploaded a new version with all the code that could cause that removed. Hopefully that fixes it.

@Surefoot: Physics engines are great. On one occasion I put a satellite into orbit, spun it up (like real satellites do to maintain a particular orientation) and then used a coning maneuver to change its axis of rotation relative to the world. I'm not sure what the practical use of it in KSP would be, but it was interesting.

@theSpeare: Help functions are in the GUI, which should answer most of your questions. Without anything more specific, I can't answer them.

@illectro: Density is in kg/m³. Any questions you have, I can answer.

@Konnor: Yeah, just checked that in the new v0.12.5.2 and I couldn't reproduce it, so it looks like removing the other weird area-code got rid of that issue.

So the v0.12.5.2 hotfix is out, to fix the errors found by DresCoffgrin, Konnor, and possibly Korn Sarum. Quick testing wasn't able to reproduce the issues, but let me know if they come back.

[Virindi]

@Virindi: That was a fix to unintended behavior in the stock game. Engines originally did not respect their "ignition threshold" value and would flameout the second they tried to pull more resources than were available. Now, so long as there are more than 10% of the requested resources available the engine will throttle down, rather than flaming out. This was behavior that was supposed to be there all along, but was not.

Okay, but a turbojet is supposed to have 200kn of thrust max, and yet at 10% air it is able to provide 110kn of thrust (shown in screenshot)? That certainly isn't linear, and sounds more like the thrust isn't being reduced at all. I would think that if you had 10% air, the maximum thrust for the jet would be closer to 20kn (10% of the maximum).

[taniwha]

I won't claim that this applies in this case, but 1/10 the air at 5x the velocity gives 1/2 the thrust. My point being that non-linearity is to be expected. The question is how much.

[Virindi]

I won't claim that this applies in this case, but 1/10 the air at 5x the velocity gives 1/2 the thrust. My point being that non-linearity is to be expected. The question is how much.

Sure, but it is giving vastly more thrust than it did under the old setup when you kept the engine running by throttling down (manually or letting mechjeb do it). It really seems like thrust is not decreasing at all.

I tried to fly the same plane under 0.22 and I couldn't get anywhere NEAR the speed with MJ automatic thrust reduction to prevent flameout.

Regardless, if getting [nearly] maximum thrust at 32km altitude is expected, then it presents a serious and negative change in game balance. Spaceplanes went from 'challenging' to 'you can get in orbit by accident'. I actually discovered it by accident with a regular plane just trying to fly around the planet, I lost control and when I looked I found that it was because I was in space with an apoapsis of 500km!

Edited January 8, 2014 by Virindi

[junkie_business]

Hey Ferram, first of all, great mod!

(Also I can only use a phone for this at the moment, so apologies if this has been covered) I've been playing 12.5.1 (just noticed you've updated but haven't had a chance to try it yet) on a modded .23, and if I set a Pe to ~20k to reenter, once the ship gets to around 68k the ship starts to 'vibrate' around an axis, so much so the G meter maxes out and the ship slows to around 60m/s (if it hasn't disintegrated). If I manage to get a parachute out, it still vibrates, but then some strange interaction between the parachute and the vibration can end up making the ship climb back up at a rate of up to 100m/s, then back down, then back up again. This happens with stock or mod parts. If I plan a 40km Pe, this doesn't happen. Sometimes this vibration will fire the ship out on a kabillion m/s route out of the system. Luckily Jeb had stayed behind on board Salyut.

I also removed FAR and it seemed to be fine, then returned when I reinstalled it. I don't use any other mods that do anything like FAR, as far as I know. I'll try the update when I get a chance, but thought I'd post so you're aware of it. Thanks for the great mod.

David

[Selentic]

I've got a little quibble with FAR, that i'm not sure if it has been noticed or not or if i'm doing something wrong - if you reentry the atmosphere with a pod or small ship while spinning you keep spinning all the way down (even after the chutes (RealChute) open). I'm pretty sure that's not supposed to happen!

Anything I can do about that - or should I just stop spinning? >_>

[Virindi]

I am having a proble, similar to the one Dreadp1r4te is having. It happens every time I re-enter after loading a quicksave in orbit: the craft either enters a perfect flatspin, or I lose all control and it drifts to the ground at less than 50 m/s, even though it should be moving at at least 100 m/s. The problem seems to occur only after loading a quicksave. If I re-start the game and enter the craft from the space centre I encounter no problems at all.

This just happened to me, too. My spaceplane exploded as soon as it hit the atmosphere on reentry, and now it is drifting at 10m/s at 40km up. Flight data lists Cd as 133743...that seems a bit high to me

This is with a plane that I designed and launched using the same version of FAR.

[egreSS]

I'm sure there is a guide that breaks down the stability derivatives visually, would anyone mind throwing me a link?

I'm skeptical on my understanding of the longitudinal "Xw" Down Velocity derivative. From reading the help I'm translating this as: the movement of the craft (forwards or backwards) with the respect to the movement of the craft downwards - but again, I just don't trust my understanding.

[ferram4]

@junkie_business: Try v0.12.5.2. I removed everything that should cause that issue from the code.

@Selentic: What would stop it from spinning? There's only a very small amount of skin friction that would act to stop it from spinning.

@Virindi: Get version 0.12.5.2. I update the mod for a reason.

@egreSS: Yep, that's correct. I believe positive means that as the craft goes down more (which would correspond to higher angle of attack) that there are more forces pushing it forward. This is in the plane-relative coordinate system, mind you.

[Virindi]

Sorry, hadn't seen that there was a new update.

[Selentic]

@Selentic: What would stop it from spinning? There's only a very small amount of skin friction that would act to stop it from spinning.

True enough. It just seemed really strange the couple of times it happened. Not something really worth worrying about. (PS: FAR doesn't try to take into account the mess that is parachutes, does it?)

[taniwha]

Regardless, if getting [nearly] maximum thrust at 32km altitude is expected, then it presents a serious and negative change in game balance. Spaceplanes went from 'challenging' to 'you can get in orbit by accident'. I actually discovered it by accident with a regular plane just trying to fly around the planet, I lost control and when I looked I found that it was because I was in space with an apoapsis of 500km!

Getting to space with a space plane isn't really all that challenging, even with TVp&a's nerfed engines. If you want challenge getting to space, use RSS .

[AbeS]

Is it even possible to make a working SSTO in RSS?

[NathanKell]

Yes. People have posted examples. Although it's even *more* possible to make an SSTO period, not an SSTO spaceplane. Once again, KSP forum teaches us the wrong terminology (c.f. NERVA).

Man, I really should try one one of these days. It'd make a nice challenge.

[AbeS]

Now that I think about it you are right, I have made it to orbit with a single stage with rockets I meant SSTO spaceplane! It is a real challenge indeed.

[ferram4]

Sure it's easy to make a SSTO in RSS. All you need to do is take an old Atlas rocket, replace the original engines with an NK-33 or some kind of equivalent and make sure you're only using a Mercury capsule. Bang, SSTO.