Low-tech plane CoM vs. CoL

[AlexinTokyo]

I'm trying to build a very low-tech (initial aviation tier) plane for puttering around on Kerbin, but I'm having trouble believing the so-called design helpers.

I build the basic airframe with appropriate wings, control surfaces, etc., but it ends up with the CoL way behind CoM. In order to bring the CoL forward, I have to mount the wings so they're basically hanging off the cockpit, which looks ... well, see for yourself in the spoilered image below, but 'not great' is not an exaggeration.

So is the CoL really that far backward with those winglets as a tail? What part should I be using as a tail? Is there some other way I can bring the CoL and CoM into their appropriate places?

Does anyone have any viable, not-stupid-looking, basic-tech-only plane designs?

Edited July 22, 2015 by AlexinTokyo
Answered

[Starhawk]

You don't need those huge horizontal stabilizers as part of the tail. If you feel you must have some horizontal control surface near the rear, just add a couple of the smallest control surfaces directly to the fuselage.

Of course, this may fall into your definition of 'stupid-looking'.

Happy landings!

Edited July 21, 2015 by Starhawk

[Sharpy]

- don't hang the tail fins so far back. It's their lift that moves the CoL so far back.

- use Liquid Fuel Fuselages instead of rocket tanks. All that oxidizer space is a waste.

- maybe add a pair of canards (just tail fins but on the cockpit)

- Elevons don't need to be attached to any wing surface to work. You can use them directly instead of tail fins or as alierons as well, and they don't generate lift unless out of "null" position normally.

...and in the end just try how your plane handles with CoL at different positions. I can assure you with it so far back you'll have a lawn dart.

Edited July 21, 2015 by Sharpy

[AlexinTokyo]

You don't need those huge horizontal stabilizers as part of the tail. If you feel you must have some horizontal control surface near the rear, just add a couple of the smallest control surfaces directly to the fuselage.

Of course, this may fall into your definition of 'stupid-looking'.

Happy landings!

I hadn't thought of having only control surfaces with no fixed horizontal stabilizer. I'll give it a burl and see how it goes

- don't hang the tail fins so far back. It's their lift that moves the CoL so far back.

- use Liquid Fuel Fuselages instead of rocket tanks. All that oxidizer space is a waste.

- maybe add a pair of canards (just tail fins but on the cockpit)

- Elevons don't need to be attached to any wing surface to work. You can use them directly instead of tail fins or as alierons as well, and they don't generate lift unless out of "null" position normally.

...and in the end just try how your plane handles with CoL at different positions. I can assure you with it so far back you'll have a lawn dart.

Re: LF fuselages: I wish! Unfortunately, the basic aviation tech doesn't include any LF-only tanks. I need 90 science (ninety, gorram it!!) to unlock those, and I will probably have other priorities at that point. I'm playing on hard mode (entirely my own fault), and the science isn't so easy to come by. But I will certainly try ditching the stabilizers for control surfaces only.

[T-Bouw]

In my humble opinion, I blame the overly forward mass of the airbreathing engines.

It was meant to make balancing planes easier, but in practice, this is only true for medium-sized planes (mk 2 plane-parts).

Anything bigger is virtually unaffected by this mass-shift forwards and small planes (like yours) come out looking "off" with their main wings needing to be placed waaay forward.

[Sharpy]

oh, yep, sorry, didn't notice that they are "a level above".

As for the control surfaces - the farther they are from CoM (in the plane they are acting, e.g. pitch:front-back) the stronger they act. Alierons on wings at a level flush with CoM are almost inert. So you might add a pair to the cockpit to act as canards if the plane is not very responsive.

[capi3101]

Re: LF fuselages: I wish! Unfortunately, the basic aviation tech doesn't include any LF-only tanks.

In that case, be sure you're draining out the oxidizer in the SPH. While you're at it, do the same for the monoprop in the cockpit; this plane isn't going to space, it needs neither.

If you've happened to unlock the AV-R8 winglet (available with Flight Control, a 45-point tier tech), those make for great horizontal stabiliators in low-tech. They make for good canards even after you've got some better parts unlocked. The Delta-Deluxe is a pretty reasonable piece for a low-tech fin/rudder, though.

How are you for gears? I'd turn the brakes on the rear gears up if you can.

EDIT: You might not want your CoM and CoL to be right next to one another. If you're into mods, I'd suggest one called RCS Build Aid. Among its many useful features is a "dry center of mass" marker that shows you where your CoM will be when the craft runs out of gas. For planes, it's a handy way of being able to tell if your CoM is going to go aft of the CoL at some point during the flight. In stock, you could just drain all the tanks completely and use that position to set your CoL - but that gets tedious once your planes start getting bigger, and you have to remember to put the fuel back if you want the plane to run at all...

Edited July 21, 2015 by capi3101

[Foxster]

I'm not great at aircraft but this flies real nice and the wings are in the proper place...

[pincushionman]

Grab those horizontal tails with the rotation tool and point them down a click (non-angle-snap mode). This will have a surprisingly big effect, and will allow you to keep the tailplanes far back and keep the long moment arm you're gonna want with the tiny control surfaces.

- - - Updated - - -

By the way, while we say "keep your CoL behind your CoM," that's not exactly how real planes are made. What happens is the lift center of the main wing is placed behind the CoM, and the resulting moment is canceled out by downforce (on tailplanes) or upforce (on canards). The result is the overall lift vector is exacly over the CoM with no net moment, or "trim." It's just REALLY hard to zero the moment using the tools we have in KSP. And it's why the main wings of rear-engine aircraft (think T-tails like the DC-9/MD-90/B717 or the various CRJ's) are so far aft.

Edited July 21, 2015 by pincushionman

[Sharpy]

EDIT: You might not want your CoM and CoL to be right next to one another. If you're into mods, I'd suggest one called RCS Build Aid. Among its many useful features is a "dry center of mass" marker that shows you where your CoM will be when the craft runs out of gas.

If you're not into mods, just drain your tanks dry and see where your CoM goes.

This is especially important in SSTO, where you depart the atmosphere with tanks half-full and reenter almost empty (and without the payload!) - I was in for a naaasty surprise the first time I did it. Especially that rocket engines drain fuel starting from the front tanks, picking the rear tanks last, so all my leftover fuel was in the rear meaning my CoM was waaaay back.

[GDJ]

I'm not sure if this would help, but here is just an example of what people have already stated:

Rear view: If you see the "spheres" that represent the Centre of Lift, Centre of Gravity, and Centre of Thrust (CL,CG,CT) you'll notice that in this photo that they are all aligned on the same plane ( or at least really close to the same plane. You can be off a tiny bit without any detriment to performance).

Side view: In this view, the CG (yellow) and CL (blue) are very close together, with the CL slightly behind the CG. This promotes stability since the CL is behind the CG, but close enough that the plane will lift off easily and fly fairly neutral. If the CL is ahead of the CG, the plane will pitch up uncontrollably since the CL want to try to flip behind the CG.

Hope this helps. For the most part, as long as your CL is slightly behind your CG you'll be okay. The CL can be slightly higher and behind the CG to promote stability, or slightly below and behind the CG to enhance cornering and maneuverability.

If the CL and CG are exactly in the same space, your plane will fly fine, but any over correction in your piloting will cause the plane to lose control. Any imbalances in thrust from multiple engines will cause the plane to lose control very quickly. Also if the plane's CG shifts due to fuel being burnt it could become unstable in flight. I would compromise and keep the CL slightly behind the CG. Your plane will be a little slower to change direction, but much easier to handle in the long run.

Also keep the rear landing gear slightly behind the CG and CL. The easier the plane can pitch up on the runway for takeoffs, the easier the plane will take off at a lower speed. Be cautious since the plane can pitch up easier, it could smack the engines down on the runway and you'll lose them. Adjust the position for flying style and taste.

Edited July 21, 2015 by GDJ

[Sharpy]

CoL can be quite a bit behind CoM - if you drain your fuel you get a different position of CoM so aim for CoL at a location where it always stays at least a little bit behind CoM. If it's far, the plane will be lousy to fly, constantly trying to pull downwards, but that's still something you can live with. If it's in front, your plane will be flip-happy, loving to turn tail forward, and that is NOT something you can just live with. It must be way back (in your example screenshots that did count as "way back") for the plane to pull downwards so hard you won't be able to compensate by pulling up.

[Alshain]

There is a lot of strange advice here but if your wings are on the CoM, you will need horizontal stabilizers for pitch. It's really not that hard to build a simple plane. In your last picture you are using an LFO tank. I sure hope you drained the Oxidizer out of it. This plane has only 1 LF tank, the middle section is just structural and the front of course is intake.

Edited July 21, 2015 by Alshain

[GDJ]

CoL can be quite a bit behind CoM - if you drain your fuel you get a different position of CoM so aim for CoL at a location where it always stays at least a little bit behind CoM. If it's far, the plane will be lousy to fly, constantly trying to pull downwards, but that's still something you can live with. If it's in front, your plane will be flip-happy, loving to turn tail forward, and that is NOT something you can just live with. It must be way back (in your example screenshots that did count as "way back") for the plane to pull downwards so hard you won't be able to compensate by pulling up.

True, but for KSP where the CL and CG are, the plane does handle fairly neutral.

Also in this plane the CG did shift slightly closer to the CL as fuel was being burnt, but stayed ahead of the CL. It's a proven, stable design that completed the Circumnavigation of Kerbin in 52 minutes, with a top speed of just over 1400 m/s at 22,000 metres altitude.

Edited July 21, 2015 by GDJ

[Coga19000]

Its a matter of relative size. Those horizontal Stabilisers, as you can notice, are around half the size of the main wing, and for such a small plane you don't need a much pitch control. Now, if you want to keep them, try to balance them with a similar pair on the nose. They are called canards in this configuration (you will get some specialised Canard parts later) and in my opinion, they look GREAT

[cantab]

Well, this is the cause of the problem!

Squad, why did you think think that was a good idea?

As far as the solution goes, three options spring to mind.

1: Shift the centre of mass back. You might add an extra empty fuel tank or a structural fuselage, putting the engine out somewhat behind the tailplane.

2: Reduce the wing sweep, which will bring the lift forward while the wing root stays in place. You can angle the swept wings, use the straight wing connector or even flip the swept wings round so they sweep forwards.

3: Shrink the tailplane. Unfortunately that's hard to do right now, the smallest elevons and wing pieces are in a 160 point node, but you could use a bare elevon.

I think the second option will make the best looking plane.

[JetJaguar]

Well, this is the cause of the problem!

https://farm1.staticflickr.com/258/19706928210_7a07b4ab0a_b.jpg

Squad, why did you think think that was a good idea?

It makes sense if the model represents just the external portion of the tailpipe. It's probably about where a jet engine's c.g. should be since the bulk of the engine is buried in the fuselage:

Anyway, your plane looks almost the same as my usual early game observation plane:

I use the AV-R8 winglet for the horizontal control surfaces, which is under the "flight control" node of the tech tree (at the same level of "aviation"). I do not bother with ailerons as the winglets have enough control authority to handle roll as well as pitch. I moved the center of lift forward by rotating the winglets and giving them a very slight downward angle - so small as to be barely noticeable. Try turning on the CL and CG display and play with the rotation yourself, you'll see how angle influences the position of the CL. This is with RCS build aid turned on, the CG moves forward as the tanks drain:

[Wanderfound]

One of the flaws of stock aero vs FAR is that stock tends to demand lift a bit further forwards than it should, resulting in hunch-shouldered planes or an overabundance of canards.

However, there's also the point that (in stock or FAR) your CoL does not need to be that close to your CoM. It can be if you want it to be, but it doesn't have to be. The design guides insisting that the markers should always overlap are based on old stock aero; they're outdated.

[AlexinTokyo]

Thanks for all the advice everyone.

Turns out the best way to achieve what I wanted was to rotate the wings to reduce sweep, which pulled the CoL far enough forwards.

Was able to complete the survey contract, including a rough field landing and takeoff. I did have to use the emergency parachutes to brake on the runway, as I had come in too high too fast, but that was more piloting error than design woe, so I'm calling it a win

[Nich]

They really need an AERO mod for SAS or just give us trim tabs we can control. It drives me crazy when I try yawing and my plane starts diving. I am just about to install MJ so I can adjust the PID filter parameters :\

[capi3101]

I've heard that there is a PID tuner mod for FAR; haven't tried it out myself but I've been meaning to check it out. I'm the same - try to make a simple little adjustment to line up on the runway, suddenly have to adjust my pitch to maintain my glide slope, things like that.

[pincushionman]

They really need an AERO mod for SAS or just give us trim tabs we can control. It drives me crazy when I try yawing and my plane starts diving. I am just about to install MJ so I can adjust the PID filter parameters :\

Trim is applied with mod+WASDQE. I know. I huffed and puffed about that myself for some time before another forum user pointed it out to me.

Won't help any coupling you're getting, though.

[Mic_n]

I'd also suggest (though it wouldn't be a HUGE change) moving your ailerons inboard. It's surprising how much they influence your CoL.

Because they're attached to the trailing edge of your swept wings there, moving them inboard will also bring them forward. It's highly unlikely you actually need (or indeed want) all the authority they'd provide out there on the wingtips like that, especially when your weight is all quite centred along your roll axis like that. You could also use the option of attaching them on the leading edge and clipped back within the wing..

You could also (if you wanted) switch out your 'conventional' tail section for a two-piece 'V' tail, which can still produce the same directional forces but generates less lift (and drag, for that matter) in the process.