Are fairings useless?

[zarincos]

Here's an imgur album with my basic rover design: http://imgur.com/a/vc9ys. Since the rover itself is baking soda and a pat of butter away from being a pancake, I figured that it would save me some dv and look cool to put a fairing around it for better aerodynamics, and, curious as to how much it was saving me, I immediately started some tests (1.0.2 aero).

Abstract/introduction/procedure

As you may note, the launcher has a fair number of wings on it, this is because any design I tested had a strong tendency to flip in the atmosphere, including the fairing design. In my initial design, I used just one engine, and that could get my rover to around 8k without incident, but since that resulted in a TWR < 1 for the fairing, I use three engines for all tests to try to ensure that the tests are as fair as possible. I'm primarily testing for whether or not the extra mass of the fairing outweighs any aerodynamic benefit they give, and I submit that this may not be the case for all incidents, as I don't really want to try to design every reasonable rocket and put a fairing around it, I'm open to any screenshots by anyone else.

All tests are performed at 100% throttle using SAS to go straight up. All designs were also tested for gravity turns, and, while the fairing run did the turn the smoothest, none of them had serious issues performing up to a 45 degree turn. Obviously that's not efficient for an actual launch, but if they can do it at 3k they should be able to do it at 10+k.

Results:

As the album shows, the rover on top of a rocket with no special aerodynamic precautions reaches close to 4k maximum height, while the fairing'd rover reaches a mere 2380. I then recognized that maybe it's simply the additional weight of the fairing that causes the TWR to be very non-optimal, and so a smaller fairing would not have these issues. To account for this, I stick an extra ~10.5 tons of mass on my rover in the form of fuel tanks on top of it (that don't feed into the bottom engines) and repeat the test, interestingly getting 2823 for a maximum height, weighing .1t more than the fairing and getting an extra 500 meters in the air.

Conclusion:

Fairings are literally worse than dead weight. I can appreciate that the additional fuel tanks on top of the rover may have been, in some way, aerodynamic, but they should not provide a better aerodynamic surface than a fairing, and the completely stock weight should be accurate (I have precisely 0 mods installed in this save).

[Claw]

I heartily agree that fairings are too heavy. Indeed, I'm not completely surprised by your results.

[sal_vager]

The panels used to be massless, only the bases had mass, but people didn't like that and it was changed.

So now they are excessively heavy, I'm not surprised if this also affected their drag.

[Frybert]

The panels used to be massless, only the bases had mass, but people didn't like that and it was changed.

So now they are excessively heavy, I'm not surprised if this also affected their drag.

Actually the game used to CLAIM the fairings had mass (added to the base) when they didn't.

[himnhouw]

Fairings do work, but they are not magic. You still have to make them aerodynamic.

[Red Iron Crown]

I've found that fairings are less about reducing drag and more about shielding their delicate contents from atmospheric heating and drag forces. I suspect if you got the unfaired version up to more serious speeds it would start shedding parts.

[GusTurbo]

Well that doesn't seem right.

[Brotoro]

I tried fairings for the first time tonight. The payload was my 60 tonne Duna Base payload on a modified Big Advanced Reusable Rocket (you can follow the Developing Duna link in my sig to see this stuff if you care).

First I tried to launch the payload on the rocket with no fairing (for a control). Unfortunately, I wasn't able to get the payload boosted in a proper turn without flipping out of control. The payload stack was just a bit too flexible, I think, so the rocket bent enough to put it out of control.

Then I tried it with the fairing. The building of the fairing was difficult -- the interface wasn't responding well to my mouse clicks...but I'm not sure why. Maybe because the payload/rocket has 400+ parts, so there might have been lag in recognizing my clicks. Or something else...I dunno. Anyway, I finally got the fairing built and launched it. Arrrgh, it was terribly laggy. Quite a bit laggier that without the fairing. Painfully laggy. BUT...the fairing allowed me to get the rocket through the boost. But, oh, the lag.

Oddly enough, the lag continued even after I blew off the confetti at 40,000 meters. Weird. And painful.

But fairings DID allow me to fly a payload that I was no longer able to boost in 1.0.2.

[Farex]

Oh Brotoro tried fairings, maybe something in the making for the Duna-Mission?^^

@Topic:

In 1.0 the fairings did their Job perfectly. But in 1.02 all my rockets keep flipping over (the same that worked so well in 1.0).

Just let that sink in: If you put your payload into a fairing in 1.02, it makes your rocket flip over. Don't do the fairing and you are going well. This is not right.

[zarincos]

Yeah, I'm noticing a lot of flippage in 1.0.2. I've been trying to put this rover into orbit for nearly an hour, and, with or without fairing, it gets nigh-uncontrollable at 15-25kish. I think that's the area where the drag is enough to play havoc with my heading but the air isn't dense enough for the wings to do their job. Adding the fairing does make it slightly more stable, but not just terribly more so.

As far as the fundamental state of fairings go, I'd suspect sal has it right, with them being balanced for massless panels and them weighing something now had a negative effect on drag. In general right now, after a fairly large amount of testing on, admittedly, a fairly specific situation, I'd say they have a use, just a very niche use, plus being nice heatshields.

[hawk_za]

rather launch your rover facing vertically it would save you quite allot

yet another example of trying do do things the old way we need to try and keep our rockets slimmer

[AllOfIt]

I haven't updated from 1.0 yet, so my fairings still work great.

[Superfluous J]

I've only started to fiddle with fairings (because I just got to the point where I'm not sending stuff that's already aerodynamic up) but every ship that I've put them on had 3 things in common:

1) They were very unaerodynamic without the fairings.

2) Adding fairings kept the ship controllable.

3) dV to space went down - sometimes significantly - with the fairings added.

So useless? No. Far from it.

[michaelsteele3]

Maybe, but so is having 30 Billion threads on this.

[Brotoro]

Ah ha! Somebody pointed out in a different thread that there is apparently a color change (from orange to green) of the text that tells you whether or not you can click with the left mouse button to stick down a fairing spot while building. But I certainly can't see that color change, since I'm colorblind. So probably my problem with building fairings was because of that.

I hope squad improves that interface.

[Rocketfeline]

Yeah, I'm noticing a lot of flippage in 1.0.2. I've been trying to put this rover into orbit for nearly an hour, and, with or without fairing, it gets nigh-uncontrollable at 15-25kish. I think that's the area where the drag is enough to play havoc with my heading but the air isn't dense enough for the wings to do their job. Adding the fairing does make it slightly more stable, but not just terribly more so.

As far as the fundamental state of fairings go, I'd suspect sal has it right, with them being balanced for massless panels and them weighing something now had a negative effect on drag. In general right now, after a fairly large amount of testing on, admittedly, a fairly specific situation, I'd say they have a use, just a very niche use, plus being nice heatshields.

Ive noticed this with the old Procedural fairings mod that got updated to 1.0.2 It seems to flip at a certain altitude. Another thing that bugs me is that fairings can act like heatshields. So when your making a skycrane-rover-to-duna mission all you gotta do is add fairings.

[stibbons]

I hope squad improves that interface.

Please, pretty please file a bug about that. In the support forum at least. You're right, it's really something that should be improved.

[Brotoro]

Please, pretty please file a bug about that. In the support forum at least. You're right, it's really something that should be improved.

That's a bug? The program is working as they intended it to work.

[stibbons]

It's unusable by a significant number of people and should be changed. A bug report is how you bring it to the attention of people who can change it.

[panzer1b]

Try lifting a 30+m radius capital ship using the 3.5m fairings....... THATS HEAVY!

i literally have over 500t of fairing for a 100t capital ship right now. Funny YES, but not quite practical.

That said, what do u expect from a 30m fairing (i had to edit the .cfgs to increase the max radius limitation, so annoying for me). It looks amazing though, the largest and fattest thing ive ever gotten into orbit.

that said, fairing weight is too high, it should be heavy, but not absurd levels where your fairing is over twice as heavy as the payload.....

[tater]

"Capital ship."

Can't you just build them in space? Make a large station, then launch the same mass of "parts" to the station (whatever adds up to the right mass), then hyperedit it to station rendezvous and assume it was built there---or build it there using EL.

Ksp really needs a build in space functionality built in for stuff that would never be launched to orbit.

[Murph]

Fairings seem just fine to me in 1.0.2, used for stuff that's "in scope" for the game. I've been using them to launch small satellites for contracts. The weight isn't excessive (around 0.25t, if memory serves), no aerodynamic issues, rocket perfectly controllable, no flipping. Frankly people experiencing control issues with fairings have either a design issue (e.g. lack of control authority), or are are not following a valid flight profile (45 at 10 is ***BAD***, wrong, incorrect, etc; real gravity turns are not a single sharp pitch change), or are showing a lack of pilot skill by letting the nose get too far from prograde.

Other complaints here seem invalid to me, based on trying to use the fairings in ways that is far out of scope for the game, and not something they are intended for, e.g. launching a capital ship. They work fine for the normal use case; totally different use cases need something different, not a change to the existing fairings. "One size fits all" is rarely a good thing, so it's a very bad idea to try to make a single set of fairings work for all use cases; the current ones work for the standard cases, and should remain tuned for them; other cases need different fairings (probably from the mod community, if the use case is out of scope), or some alternative solution.

I'm not saying they are perfect, as there's still scope for fine tuning just about everything in the game, but they are perfectly reasonable for their normal purpose. They are useful, they work, they don't have a huge problem in normal use cases.

[Farex]

Well wait a moment, I will start up KSP right now and take 2 Screenshots for example and load the craft file up so everyone can try it out.

€dit: Here you go

The craft for you to test: http://www./download/bskd6hlr3gyplih/Raumstation+mk2.craft

And for the ones that think all complains are because of incompetence: my ascent profile looks like this, turn over from start at about 5° than turn of sas and let the aero do it's work or turn slowly until the craft hits 45° at about 8-10km.

Edited May 5, 2015 by Farex

[Murph]

And for the ones that think all complains are because of incompetence: my ascent profile looks like this, turn over from start at about 5° than turn of sas and let the aero do it's work or turn slowly until the craft hits 45° at about 8-10km.

Ok, so it's not items #2 or #3 that I mentioned. Without actually spending time flying it and fixing it, I'm 99% certain that it's item #1: a design issue, namely lack of control authority. I don't see very much control on that rocket, for its height and mass, just the thrust vectoring and a couple of 1.25m torque wheels in the payload. I reckon if you add some 2.5m torque wheels to the main lifter, the problem will be solved, either that or a mixture of torque wheels and some tail fins.

Adding the fairing both increases the weight, and increases the side surface area (i.e. more surface for air to react against when it's anything other than directly prograde). Smoothing the air flow does nothing to eliminate the large surface area itself, it just minimises the drag for the surface area, and eliminates friction heating of the protected parts. Both of those factors require more control authority, and I'm guessing that it's just enough between them to push it over the limit of what you have available. I've yet to put much effort into big rocketry in 1.0, but past experience (pre-1.0) with things very like the rocket in your picture tells me that the thrust vectoring alone is frequently not enough for a 400+ tonne rocket. (Your boosters do seem well strutted, which rules out a different issue where the flexing of the mounting negates the thrust vectoring.) With really big payloads, I've used a stack of 4x 2.5m torque wheels (more typically on top of a 3.75m core + boosters, rather than 2.5m), and tail fins in the past.

[thereaverofdarkness]

And for the ones that think all complains are because of incompetence: my ascent profile looks like this, turn over from start at about 5° than turn of sas and let the aero do it's work or turn slowly until the craft hits 45° at about 8-10km.

It is your incompetence, but not as a pilot. There are struts going through your fairing wall.

You can do it with struts and without the fairing, or you can do it with the fairing and without struts. But don't do it with both, at least not with struts going through the fairing wall. Or maybe it's something else, I don't know, but it looks like it's the struts causing you trouble.