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

[ferram4]

@jpinard: Well... that's not really possible. The only reason that SRBs are so limited in stock is because they waste so much thrust fighting drag; they've actually got a lot of dV, believe it or not. You can test that in stock by using the thrust limiter so that it doesn't waste as much fuel in the lower atmosphere and you'll be amazed.

Anyway, you can try the Kerbal Isp Difficulty scaler and see if that gives you something you like, but other than that, I don't know. These things aren't really linear, so trying to fix them makes little sense.

@Agathorn: How are you calculating VesselLift and VesselDrag? Are you grabbing the FARControlSys and using the reference area and Cl and Cd from there or are you iterating over every part? The former will allow you to calculate it quicker, using force = dynPressure * refArea * coefficient.

[Agathorn]

@Agathorn: How are you calculating VesselLift and VesselDrag? Are you grabbing the FARControlSys and using the reference area and Cl and Cd from there or are you iterating over every part? The former will allow you to calculate it quicker, using force = dynPressure * refArea * coefficient.

At the moment I am simply returning FARBaseAerodynamics.Cl and FARBaseAerodynamics.Cd...My numbers seemed to match what the flight data GUI shows.

[Agathorn]

BTW Ferram I sent you a PM about these changes. Just saying in case you didn't notice

[jpinard]

@jpinard: Well... that's not really possible. The only reason that SRBs are so limited in stock is because they waste so much thrust fighting drag; they've actually got a lot of dV, believe it or not. You can test that in stock by using the thrust limiter so that it doesn't waste as much fuel in the lower atmosphere and you'll be amazed.

Anyway, you can try the Kerbal Isp Difficulty scaler and see if that gives you something you like, but other than that, I don't know. These things aren't really linear, so trying to fix them makes little sense.

@Agathorn: How are you calculating VesselLift and VesselDrag? Are you grabbing the FARControlSys and using the reference area and Cl and Cd from there or are you iterating over every part? The former will allow you to calculate it quicker, using force = dynPressure * refArea * coefficient.

I am so dumb! That question was meant for the ISP Difficutly thread as those were the settings I needed! LOL

[p3asant]

@p3asant: Not FAR, launch clamps probably aren't straight. If you're playing with Real Solar System, it's a stock bug with launch clamps not being placed exactly being exacerbated by the size of the planet. There's really not a solution besides not using launch clamps.

I wasn't using launch clamps. The probable reason was that the rocket had a small mass (~25t). The issue seems to diminish as mass and size of rocket goes upwards.

[darcgekco]

I just downloaded the version in the OP and I'm still getting the error spam.

Made a clean install with just FAR, same issue.

Here's the output.log.

Edited February 1, 2014 by darcgekco

[Raptor831]

Ah, ok... I see that you've had some success in the later post.

:snip:

Thanks for the tips. I tend to enjoy KSP for the engineering/design aspects than the actual piloting. I'd like to think of myself as a good pilot all the same, but sometimes my results prove otherwise!

I haven't tried S-turns yet, but that's something on my "to-try" list. The B9 airbrakes are *quite* effective with FAR installed, so I've been relying on those to control speed. Those feel cheaty to me using FAR, so S-turns would be a nice alternative.

[Surefoot]

I haven't tried S-turns yet, but that's something on my "to-try" list.

If you know about piloting, it's a bit like this:

http://www.combataircraft.com/en/Tactics/Air-To-Air/Forward-Velocity-Vector/

or a kind of high-speed yoyo could do it (but watch that AoA !). Basically any move that costs energy / velocity.. And yeah i wonder about those B9 airbrakes too, they seem to get through a few mods including DRE (where they just burn off even when closed..).

[cardgame]

The new tweakables are nothing short of amazing. I've limited most of my planes' control surfaces to provide maximum maneuverability without ever putting me in danger of stalling from too high AOA or ripping apart from g-forces. It's beautiful

[prullenbak]

Has anyone here tried using FAR with mods like PlanetFactory and specifically Sido's Urania System?

[AndreyATGB]

Has anyone here tried using FAR with mods like PlanetFactory and specifically Sido's Urania System?

It works fine, why wouldn't it?

[celem]

Something weird is going on with my aircraft lately. Its no doubt a shortcoming in my understanding of aerodynamics as implemented by FAR, so maybe someone could suggest whats happening and what i might alter.

I built a load of sub-sonic light aircraft, and a few big slow beasties, all fine. I've got a handle on what my CoM does as fuel drains and am careful about where I pump it, my CoM drifts very little and sits just ahead of CoL, everything im designing manages to fly just fine in lower altitudes and sub-sonic.

However I hit a bit of a wall when I try and go supersonic or bust past say 10km. I rather suspect my CoL is moving in relation to my speed. My realworld piloting experience extends only to gliders, so im not sure whats happening physicaly as I start to mach. Does the CoL react when you increase speed? Which way is it drifting?

Some of my designs suffer loss of control authority up around the 12-15km line, the nose drops to prograde and nothing will persuade the nose to lift, the wings are not stalled as far as I can see from the mod's readouts, I can work out the CoM at those points to still be where I want it to be assuming the CoL stays as shown in SPH, if CoL is moving then I just dont know.

If I do manage to break through into upper atmosphere then its generally good from there, but I find that as I pass the 10km mark I have to be bang on my prograde. The window of angle-of-attack that doesnt produce a stall gets tiny. What is influencing this segment of the climb?

I do have a couple of designs that will climb to stable orbit reliably, I have managed to build the ssto's I wanted, im just not sure exactly why some of the designs work and others suffer these issues. Any advice on the physics behind these phenomenon is welcomed.

I can probably drum up a screenshot or two later on of designs that display these traits if needed. They all share the same underlying theme, in that they are fully rocket-powered heavy shuttles with TWR>1 taking off vertically along the runway. They each have customised top and bottom wing sections to provide the lift/balance. As a rule the wings are deltoid and hug the body (the lift surfaces below and above angle to hug the decreasing core diameter moving from cord to tip), They rotate very very slowly after popping off the end of the runway and then climb at 45 degrees before dropping towards the horizon. I tend to make the entire ascent as one burn rather than coast and circularise.

[lammatt]

Something weird is going on with my aircraft lately. Its no doubt a shortcoming in my understanding of aerodynamics as implemented by FAR, so maybe someone could suggest whats happening and what i might alter.

I built a load of sub-sonic light aircraft, and a few big slow beasties, all fine. I've got a handle on what my CoM does as fuel drains and am careful about where I pump it, my CoM drifts very little and sits just ahead of CoL, everything im designing manages to fly just fine in lower altitudes and sub-sonic.

However I hit a bit of a wall when I try and go supersonic or bust past say 10km. I rather suspect my CoL is moving in relation to my speed. My realworld piloting experience extends only to gliders, so im not sure whats happening physicaly as I start to mach. Does the CoL react when you increase speed? Which way is it drifting?

Some of my designs suffer loss of control authority up around the 12-15km line, the nose drops to prograde and nothing will persuade the nose to lift, the wings are not stalled as far as I can see from the mod's readouts, I can work out the CoM at those points to still be where I want it to be assuming the CoL stays as shown in SPH, if CoL is moving then I just dont know.

If I do manage to break through into upper atmosphere then its generally good from there, but I find that as I pass the 10km mark I have to be bang on my prograde. The window of angle-of-attack that doesnt produce a stall gets tiny. What is influencing this segment of the climb?

I do have a couple of designs that will climb to stable orbit reliably, I have managed to build the ssto's I wanted, im just not sure exactly why some of the designs work and others suffer these issues. Any advice on the physics behind these phenomenon is welcomed.

I can probably drum up a screenshot or two later on of designs that display these traits if needed. They all share the same underlying theme, in that they are fully rocket-powered heavy shuttles with TWR>1 taking off vertically along the runway. They each have customised top and bottom wing sections to provide the lift/balance. As a rule the wings are deltoid and hug the body (the lift surfaces below and above angle to hug the decreasing core diameter moving from cord to tip), They rotate very very slowly after popping off the end of the runway and then climb at 45 degrees before dropping towards the horizon. I tend to make the entire ascent as one burn rather than coast and circularise.

i aint trained in physics; i am a chemistry teacher, and i know very little about aerodynamics, but from what i've read from WIKI, (if i didnt misundersand)

when you are in the trans-sonic regime, aka ~mach 0.7, vortexes will be form at the tip of the lifting surface and that will increase the drag and hence the loss of control.

[Surefoot]

Screenshots will help a lot. Loss of control authority at transonic speeds is called Mach Tuck, you can do a quick research about it as it's well known, and simulated properly by FAR. Ways to counter it depend on your design, usually all-moving control surfaces (or elevons on a tailless delta) that are shielded from the compression wave will keep some level of control. Then there are cases where your design is TOO stable and will not move away from prograde, thats where you introduce dihedral /anhedral and other unstable elements of design..

[Raptor831]

Something weird is going on with my aircraft lately. Its no doubt a shortcoming in my understanding of aerodynamics as implemented by FAR, so maybe someone could suggest whats happening and what i might alter.

...snip...

The loss of control seems to be Mach Tuck. I think the only real way to fix it is to have large pitch control surfaces (or more of them in KSP). Maybe pump the fuel around to see if that helps. I know the CoL moves back in supersonic flight towards the rear of the plane. Keep that in mind when you're balancing your craft. Also, the higher alt you go, the less your control surfaces work anyway. No air means no response. Large control surfaces help here too.

[TheHengeProphet]

I'm having an interesting "problem" in that jet craft I have constructed (even using stock parts) find themselves rocketing past their supposed maximum 2400 m/s.

This jet is built entirely stock and only uses two radial intakes for air consumption. Just exceeding 4100 m/s before running out of air at 35.1km...

Javascript is disabled. View full album

So, my question is: Is this a bug?

[ferram4]

Well known issue with FAR's changes to the turbojets. The fix is to go into the config and look for the last "key" in the [turboFanEngine] block. Get rid of the "@" and ",4" on it and it should work properly.

[Surefoot]

So i just finished this delta-winged beauty:

It flies perfectly at any speed from 50m/s to mach 6+ (yay for delta wings). I can almost hover around 40m/s with the leading edge flaps down... and manoeuver around.

Only trouble is, it wont take off by itself !! It's not my first plane that does this, i use the B9 landing gear and KJR (just in case). I am not sure what is going on, but it's like it's totally stuck to the ground, my tests at north pole confirm that i can max speed on ground without being able to leave it at all. From my flying tests, i should be able to take off from 50m/s ...

If i take a small bump or the ski jump at the end of the runway, it goes up immediately without any trouble and good authority. I tried to use a taller landing gear on front to have it point upwards significantly, but no change. To be clear it has excellent pitch authority, i can make full inverted loops at 150m/s and such.. but no take off O_o Any idea what is going on ?

Edited February 3, 2014 by Surefoot

[ferram4]

Perhaps the rear landing gear are too far back, and it can't rotate on takeoff? That's the only issue I can think of and if those are the gear clipping through the wings right behind the smaller inboard intakes, then those are really, really far back for landing gear.

[Surefoot]

Oh i see, but should the nose lift by itself by virtue of ground effect or similar ? Or maybe that's not implemented yet ?

I'll try and move the rear wheels forward, that will be trouble to avoid tail hits upon landing / take off though.. ahh difficult design choices here. Interesting.

[ferram4]

The nose will lift itself a little, but since all of your pitch surfaces are at the back that means that the only way to make the craft pitch upwards is by pushing the back end down. If the control surface doesn't have a large enough lever arm to help it counteract the mass of the craft times it's lever arm then the plane won't be able to pitch up.

The obvious solution is to use the tall landing gear for the back of the plane so that you have more clearance to rotate, but put them on the wings and the nose gear on the fuselage so you have some built-in AoA on the runway.

[softweir]

Alternatively, see if you can find shorter rear gear or longer nose gear so the craft has a nose-up attitude on the runway. It might then just lift itself up, with minimal control input until the rear gear is rolling on air.

[TheHengeProphet]

Well known issue with FAR's changes to the turbojets. The fix is to go into the config and look for the last "key" in the [turboFanEngine] block. Get rid of the "@" and ",4" on it and it should work properly.

Thanks! I just swapped in the default keys and it works just fine now.

[celem]

Screenshots will help a lot. Loss of control authority at transonic speeds is called Mach Tuck......

Seems to be this that is likely my issue. It might also explain why I find it so very hard to rotate off the end of the runway, since my craft are generally supersonic by 500m (or more accurately take so long to rotate that they are several klicks downrange while still being below 500m yet breaking 300ms).

This album shows one of the designs im having issues with. It does fly to orbit, yet needs constant TLC to keep it airborne. Is entirely stock other than some nosecones. It is almost impossible to launch, leaving the runway at 200ms without having achieved enough lift it gets catapaulted by the drop at the end of the runway and then rotates like a brick. (as said above its often supersonic before its finished rotating for a 45deg climb.)

This one doesnt fall along the prograde in thin air, but it's window of acceptable AoA's shrinks drastically above 9km, it loves to stall out right around that transition. In addition, it doesnt like to hold pitch. I gave it lots of elevators and can get the nose up, but it has a tendancy to yoyo.

The final image shows CoM and CoL. Bear in mind that the first tank to drain is the 1.25m nosetank, which shifts the CoM back noticeably whilst rolling down the runway, the SRB's are functioning as Jaydos here and get dropped on the runway.

Still could be mach tuck? I've definately seen that on some of my planes so i'll keep an eye open for it. Looking back through these pictures makes me think my wing isnt Delta enough, so I might try and extend it forward toward th enose a bit more. I was trying to avoid pushing the CoL forward, but if its sliding back as I mach nyway then theres some tweaks can be done.

Javascript is disabled. View full album

Tanks draining cord to tip. Thanks >

edit: hmm controls; the forward winglets are roll only, the surfaces eitherside of the centre-tank, mounted on wing boards in the tail are elevators, the surfaces mounted to the 2 delta-pieces on each side are combo pitch/roll

edit2: might experiment with some flaps to assist the launch

edit3: more reading on mach tuck was a bit of a lightbulb. You'd be amazed how much that didnt come up as a glider pilot, it'd probably have torn my wings off. My designs are likely all encountering this immediately on launch then i'd guess since it triggers while still sub-sonic. Infact i now realise i've seen this in a couple of designs even as they accelerated to the runway end, where they 'tucked' right before I went to pull off. I'd guess my CoM is a little too forward in several stages of flight, will play around later this evening and see if I can stabilise this.

Edited February 4, 2014 by celem

[taniwha]

celem: I'm sorry, but that does not look like airplane (nor does it look like a space-plane). That looks like a stock soup-o-dynamics monster.

That has to be at least 150t (I see 3 Jumbo-64s, 2 X200-32s plus even more tanks), with relatively little wing area. Yaw stability is probably nonexistent. Even with FAR, if you put enough power behind a gold brick, it will fly, though the question of how well remains. Actually, the gold brick fairs better in FAR than it does in stock .

For planes (air or space), start small (< 10t if you can (easy with this mod, some fantastic tiny engines, and nice one-piece wings in it)), get familiar with how planes handle, then slowly work your way up through speed, altitude and size. It's surprisingly fun buzzing around in a small plane that can't break 300m/s in level flight, yet can glide at 60m/s .