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

[Jwmflying14]

Hey guys, just updated to .90 from .25. With FAR 0.14.7 installed both alone, and in conjunction with other mods, both with ATM and without (basically every scenario, including stand-alone), The FAR tab on the menu (top in game play, bottom left in VAB/SPH) is pure black.

Does anyone know why? Physics seem to be okay.

[cantab]

Not really. Prefectly realistic and intuitive. The wings inboard of other things absolutely have to be stronger (heavier) to be as structurally sound with more forces to deal with (it's own and anything attached to it) plus those forces further out are magnified by the lever arm that a wing is.

The thing is that the same arguments would apply to everything else in KSP. The fuel tanks in the first stage of a rocket "absolutely have to be stronger (heavier)" than the ones in the upper stages. But all that is ignored for simplicity.

Put it this way. This whole wing mass business is so contrary to what I expect from KSP that not only did I not know it despite having made a fair few successful FAR planes, but the idea that a part should change in mass depending on where it is on the craft would never even have occurred to me. If I had spotted it by myself I would have assumed it was a bug.

Would it not make things behave like the player expects to make the mass of a wing segment constant for a given strength rating, and then it's the player's responsibility to strengthen the wing roots? As indeed I frequently did on my own FAR planes because that to me was the obvious thing to do.

[Vegemeister]

The only problem with the wing mass thing is that the same wing made of smaller pieces weighs more. That is, a 2x1 wing made out of two 1x1 square parts weighs more than a 2x1 wing made out of a single 2x1 part. In fact, the inner 1x1 weighs the same as the 2x1 would, and the outer 1x1 adds its mass to that. I have no problem with the underlying idea. High aspect ratio wings should have more structural mass than stubby wings.

But, FAR wings are trapezoids, as I recall. I think if you say areal mass density within a wing part is a linear function of distance from the tip along the mid-chord line, with the y-intercept equal to max(root density of child wing parts) and integrate from tip to root, there should be a fairly simple closed-form polynomial solution, and summing over all the wing parts would give the same result no matter how many layers deep the wing is. If you wanted to be extra consistent, you could also integrate to find the center of mass of each wing part. I've never done center of mass integrals, so I'm less confident the solution would be simple.

It would also be neat to scale the result proportional to wing loading, so there would be more of a mass penalty to using stubby little wings like the X-15 or the F-104 instead of a delta. Maybe also make thinner wings heavier.

[DundraL]

Except that on one side almost all the forces have to be support by one or two panels rather than the 4 panels on the other side, so those 1 or 2 panels have to be stronger and so heavier.

Both wings are subject to the same aerodynamic forces. How I construct them is what decides how well they hold up.

It is FAR that is deciding that the front outer right piece needs more strength/weight than the back right inner.

And what do I get in exchange for this mass? Nothing really. This extra mass just makes that single part more resistant to aerodynamic failure. It doesnt actually make the wing stronger. Infact, in the shown example, the right wing, the much heavier one, comes apart far easier due to its unimproved tendency to flex and catch more air.

Using the amount of wing surface on the other side of a wing is just not the best way to calculate it. Do it according to distance from CoP or CoL.

My point is: why must I construct according to how much mass FAR will add rather than what is most structurally sound or convenient? I can distribute wing load myself. I can decide what is going to be subject to more forces myself. I can reinforce things myself. (in fact, I still have to anyway)

Also, Im drinking right now so please forgive any oddities =)

[KillerRaccoon]

If you don't like how the wing mass is done, don't install FAR. It's really not that big a deal, just build symmetrically. Honestly, I don't know how you guys/gals are getting this aggravated by something this trivial.

[futrtrubl]

The only problem with the wing mass thing is that the same wing made of smaller pieces weighs more. That is, a 2x1 wing made out of two 1x1 square parts weighs more than a 2x1 wing made out of a single 2x1 part. In fact, the inner 1x1 weighs the same as the 2x1 would, and the outer 1x1 adds its mass to that. I have no problem with the underlying idea. High aspect ratio wings should have more structural mass than stubby wings.

But, FAR wings are trapezoids, as I recall. I think if you say areal mass density within a wing part is a linear function of distance from the tip along the mid-chord line, with the y-intercept equal to max(root density of child wing parts) and integrate from tip to root, there should be a fairly simple closed-form polynomial solution, and summing over all the wing parts would give the same result no matter how many layers deep the wing is. If you wanted to be extra consistent, you could also integrate to find the center of mass of each wing part. I've never done center of mass integrals, so I'm less confident the solution would be simple.

It would also be neat to scale the result proportional to wing loading, so there would be more of a mass penalty to using stubby little wings like the X-15 or the F-104 instead of a delta. Maybe also make thinner wings heavier.

Ooooooh, I see. Didn't realize it was different parts. Thought it was the same parts just attached differently. Then yeah that may be a problem. Though building with larger one piece panels do give a weight saving in real life but probably not that much.

And what do I get in exchange for this mass? Nothing really.

While I understand your arguments and some of it is based on the limitations of KSP and its single parent policy the above I can't agree with. If you decide to hang all your wing mass off one point instead of having them distribute the load along the entire length of the wing then yeah you can expect increased mass to support the wing with no gain. I would say that is a sub-optimal design.

[DundraL]

If you don't like how the wing mass is done, don't install FAR. It's really not that big a deal, just build symmetrically. Honestly, I don't know how you guys/gals are getting this aggravated by something this trivial.

I hope to have a discussion in hopes of a better way for this to be implemented. If you dont like how the discussion is going, dont take part in the discussion. It's really not that big a deal, just skip the parts you dont want to read. Honestly I dont know how people get so aggravated by something so trivial... sorry I had to =) no offense intended

Anyway, back on topic, I DO spread the load out along my wings. I DO reinforce my wings as needed, adding wing spars ETC and strutting to them. I DO... err... USED to buff up and partially shield wing parts I expect to take an aerodynamic pounding.* These things must be done regardless of FARs participation. That mechanic is accounted for already without adding mass according to wing construction. If adding wing mass negated for the need for all of this, that would be great. But it doesnt.

I understand why it is there though, and I DEFINITELY like and appreciate the realism that it attempts to provide. However, I disagree with the methodology of how it chooses where to add the mass, and the scaling of the mass added. Thus the suggestion that the extra mass scale be distributed according to (distance from the CoL)*(Gobal multiplier). This IMO would allow much more freedom in construction methods while giving better and more predictable results.

*USED to... now I add more strength/mass as the great Ferram intended. That is, IMO how this feature shines most.

Edited May 6, 2015 by DundraL

[Pyromaniacal]

Good goddesses I need nuFAR. I just had a plane I'm designing flatspin because there was too much drag up front and the turbojets couldn't put out enough thrust.I'm not sure where the drag came from, but there you go. Maybe I should not design planes until this releases.

[jrandom]

So I'm having to wrap my head around aerodynamics again. Is it right that this design has a minimum flight speed of 200m/s? Any slower and it falls (not stalling, just not enough lift). Wings are set to 0.5 strength, and the only fuel is in the small side tanks. Weight is 9 tons.

Javascript is disabled. View full album

This is with the dev build of nuFAR.

[ferram4]

@DundraL: Your method fails horribly in one clear way: you will make tiny canards and fins much heavier than parts at the wing root because they will be far from the CoL. In addition, you will make wingtips much heavier than wing roots. You will make wing mass depend not on anything related to the construction and loading of the wing itself, but instead on the properties of some part of the aircraft far away from it that contributes little to the forces on the wing. Finally, you will make wing sections that are carrying comparatively little load heavier than wing sections that are supporting huge amounts of wing area heavier for arbitrary reasons. The inverse (having it vary with 1/(dist to CoL)) will result in even stranger behaviors and will cause NaN masses and encourage parts used a wing buffers to keep them away from the CoL.

This is not an exhaustive list though, considering I just got back from working an election and that I spent less than 30 seconds thinking about it.

I appreciate that you, like many other people, have this idea that how things are actually put together doesn't matter, only the way it looks matters. Unfortunately, In real life, if you have two unconnected metal plates sitting next to each other, the first plate won't do anything to help the second resist a shear force, and instead that will need to be accounted for elsewhere.

You know what really irks me? Vegemeister's criticism (that for any wing shape, as number of wing parts N -> infinity, mass -> infinity) is actually a legitimate and serious downside of the current wing model, but instead of taking that tack, you're arguing that wing mass should be divorced from how it's constructed.

@Surefoot: Old issue from a very old build. Keep up-to-date with the builds before reporting issues.

@jrandom: How much wing area do you have? Based on the math, you need a combined Cl * Area = 3.6 m^2 to keep it in the air at that speed. Frankly, given the way that it's designed, you probably just don't have the wing area for all the heavy junk on it.

Edited May 6, 2015 by ferram4

[xXIndestructibleEVAXx]

So, how stable/accurate is nuFAR? At the moment, this and DRE are the only mods I'm waiting to update before I can begin playing.

[jrandom]

How much wing area do you have? Based on the math, you need a combined Cl * Area = 3.6 m^2 to keep it in the air at that speed. Frankly, given the way that it's designed, you probably just don't have the wing area for all the heavy junk on it.

Heh. There was some debate on reddit so I figured I'd come here and ask the people who actually know what they're talking about.

Max cross-section area is 3.9m^2, so it looks like I need more wings after all. Thanks! (And I think I'm going to bury the monopropellant inside a module or something -- in retrospect, tacking them on the exterior wasn't the greatest idea I've ever had.)

[blowfish]

Heh. There was some debate on reddit so I figured I'd come here and ask the people who actually know what they're talking about.

Max cross-section area is 3.9m^2, so it looks like I need more wings after all. Thanks! (And I think I'm going to bury the monopropellant inside a module or something -- in retrospect, tacking them on the exterior wasn't the greatest idea I've ever had.)

An alternate explanation is that the CoM is so far back that you don't have enough pitch authority to raise the nose and actually generate lift. If you're falling but not stalling that means that you're unable to maintain the required AoA for level flight. You could try adding some canards at the front which would both reduce the pitch moment required to lift the nose and increase your available pitch authority. What you have now is basically a lawn dart.

[Kurtvw]

So, how stable/accurate is nuFAR? At the moment, this and DRE are the only mods I'm waiting to update before I can begin playing.

It runs pretty nicely, but some of the UI is still a little flakey (as of yesterday's build) Not much point of installing it if you're waiting for DRE anyhow since its pretty clearly one will not release without the other one. But Installing the current DEV of FAR is useful for getting your head around some of the changes, its quite a bit more sophisticated than it was in KSP .90

[jrandom]

An alternate explanation is that the CoM is so far back that you don't have enough pitch authority to raise the nose and actually generate lift. If you're falling but not stalling that means that you're unable to maintain the required AoA for level flight. You could try adding some canards at the front which would both reduce the pitch moment required to lift the nose and increase your available pitch authority. What you have now is basically a lawn dart.

Oddly enough, it actually has decent pitch authority. The CoM is farther forward than you'd initially think, being placed just behind the cockpit canopy. The CoL is right behind it. I can pitch up just fine, but below 200m/s it always heads downwards regardless of pitch (even with no stalling).

- - - Updated - - -

Say.. for testing out voxelFAR, should I be using the voxelAero or voxelAeroPort branch? Edit: Ah, looks like voxelAeroPort has the most recent updates, so I'm on the right branch. Good. Yay.

Edited May 6, 2015 by jrandom

[Vegemeister]

voxelAeroPort is for 1.0x. You probably want that one.

[DundraL]

Yea, I did a bad job of explaining it. I DID mean the inverse, and failing to explain that was an oversight. Though as you said, it would encourage keeping wing surfaces away from CoM to reduce mass, the same can be said about adding parts between wing sections in the current model to prevent them from counting the wing surfaces attached via those other parts. Adding a fixed amount to distance or mass in that formula could prevent those NaNs EDIT: another oversight. This would punish small planes. Perhaps some other way or a way to grasp the scale of the aircraft its working with.

In my posts, the whole reason I mentioned anything is that construction DOES matter... in an odd way relating to mass. I'm not trying to say that the 2 wings pictured should perform the same structurally. They dont, and shouldnt. Im saying that, once *I* strut them all together symmetrically and add wing spars, the biggest thing that differs in terms of performance is the mass distribution. Im not trying to dispute the idea that wingroots should be larger/heavier/stronger than wingtips etc. Thats reality. I'm saying that the way FAR detects what pieces are wingroots and wingtips results in great changes according to the order you build wings.

Edited May 6, 2015 by DundraL

[Jovus]

Would it be possible to use the voxelized shape to calculate mass based on total supported distance from center-line in a cross-sectional manner? That might allow two wings with the exact same geometry but different construction methods to mass the same. It would also still preserve low-mass tiny canards.

At first glance it seems like a huge headache and potentially a slow-down during assembly, especially if you have to recalculate and then apply the voxel after every part-change.

[ferram4]

@DundraL: So the effects of parts very, very far from the lifting surfaces under consideration will still have a large effect on the mass of the parts connected. Come up with a method that doesn't result in strange interactions between parts and then maybe I'll consider it.

Also, as much as you might insist that a single bar strut is identical to the flat braced area at the bottom of a wing, it isn't. It never will be. They are structurally different, flex differently, and if constructed that way in reality would behave differently. So they will behave differently in FAR. Build your wings sensibly; when you insist on supporting a large mass of wing not through a solid connection to fuselage, but through a crazy daisy chain through other wing structure plus a single strut, it will require additional mass.

@Jovus: Maybe? First I need an algorithm to pull wing shapes out of it to begin with. No, I don't want, "maybe you could do X..." unless it's been backed up by at least pseudo code. Otherwise, I'm busy looking at things like DundraL's idea where there's strange behavior all over the place due to not thinking things through.

[Lord_Potato]

@ferram4: Are you going to reiterate on Better buoyancy? It seems that some of nuFar's ideas might work there.

[smunisto]

Anyone else have occasional slowdown(as in a heavy FPS drop) when tweaking a part menu? For example it happens when I try to adjust the color of some projector/light part. A quick look at the Alt+F2 screen shows FAR is recalculating something and from the (I guess) normal 100ms it peaks to about 450ms. And repeats the process many times, until it finally settles back to 100ms.

[blowfish]

Anyone else have occasional slowdown(as in a heavy FPS drop) when tweaking a part menu? For example it happens when I try to adjust the color of some projector/light part. A quick look at the Alt+F2 screen shows FAR is recalculating something and from the (I guess) normal 100ms it peaks to about 450ms. And repeats the process many times, until it finally settles back to 100ms.

There is a recent commit which addresses this issue. Do you have the latest version from Github?

[smunisto]

There is a recent commit which addresses this issue. Do you have the latest version from Github?

Probably not and I made the newbie mistake of not checking up with GitHub before posting. Apologies.

EDIT: Unfortunately the newest GitHub dev version did not fix the issue for me. I managed to reproduce it with 100% success rate using the latest version(AeroPort, updated 8 hours ago), when right clicking a Procedural fairing(the mod), which is free-floating in the VAB - i.e. not attached to a node. I guess FAR tries to recalculate its parameters each time it changes and I guess something in the code of PF tells a free floating fairing to constantly update in order to search for craft dimensions.

Edited May 6, 2015 by smunisto

[John FX]

Probably not and I made the newbie mistake of not checking up with GitHub before posting. Apologies.

EDIT: Unfortunately the newest GitHub dev version did not fix the issue for me. I managed to reproduce it with 100% success rate using the latest version(AeroPort, updated 8 hours ago), when right clicking a Procedural fairing(the mod), which is free-floating in the VAB - i.e. not attached to a node. I guess FAR tries to recalculate its parameters each time it changes and I guess something in the code of PF tells a free floating fairing to constantly update in order to search for craft dimensions.

Myself I found the slowdown started when I installed proc fairings. YMMV

[cantab]

Come up with a method that doesn't result in strange interactions between parts and then maybe I'll consider it.

Make wing mass depend only on the part and the strength rating, and leave it up to the player to set the wing roots stronger than the wingtips. What, if any, problems would that cause to FAR?