PDA

View Full Version : Centre of lift and centre of gravity



hairyspin
March 8th, 2009, 10:03
Dumb newbie question here - it's my first model. The Hawker Tempest had its maximum wing thickness at about 37% back from the leading edge. Should the centre of lift coincide with this, and then the centre of gravity be near this? I'm building for CFS3 using AirWrench with ACM to set the geometry in aircraft.cfg first.

I'd pinch Jerry Beckwith's FM for the FS9 Tempest, but I think he'd mind.... :whistle:

sparks
March 8th, 2009, 12:10
When it comes to FS/CFS, don't get hung up on the wing. The flight dynamics are for the entire airframe, and when all is said and done, it's the tail that determines where the center of lift is. When an aircraft has been trimmed to fly straight and level, the center of lift and the center of gravity are in the same location. The CoG when empty is typically 25% aft of the leading edge of the wing.

I don't recall an FS9 Tempest. There's an FS9 flight model for the David Hanvey/Paul Barry Sea Fury and a CFS3 flight model for the stock Tempest on my web site. If you want to use any of these flight models 'as-is' for a CFS3 upload, just add a note to your readme file to say where they came from. Any one of the flight models on my web site that have been built with the workbook should also work well as a starting point for AirWrench.

hairyspin
March 8th, 2009, 13:07
Thank you Sparks, it's the David Hanvey/Paul Barry Tempest II (http://www.surclaro.com/Hawker_Tempest_MK_II_for_Fs2004_review/) for FS9 I'm thinking of - a personal favourite. So far with the geometry, AirWrench is reporting the empty CoG at 37% of MAC, so if I've got my inputs correct for aircraft.cfg then I'll keep going.

I'm not saying it's taking as long as learning gMax, but this is not a job for an end-of-a-day head.

sparks
March 8th, 2009, 18:10
What 3D model are you working with? Ideally, the CoG and model center should be in the same location. 3D models aren't always built that way, but the ones that are tend to be less trouble.

BTW, the quick and dirty way to set these centers is click the 'Repair Aerodynamic Balance' button on the balance tab.

hairyspin
March 8th, 2009, 22:57
It's my own Tempest II - a new build from a newbie, lol - so I don't have an accurate fix on the original aircraft's CoG.....

crossram
March 9th, 2009, 04:52
For a new build in gmax, locating the model 'in center', is the best way to have a good start for the FM.

In gmax, I'd recommend (with the grid turned on), moving the entire model, vertically onto the a/c's thrust line. In other words, in line with the prop (thrust).

The model's master node MUST be at 0,0,0!

Then, longitudely (fore/aft),for the visual model itself, move it until gmax's centerline it through 1/4 MAC on the wing.

Once you get into the FM, it will simply be a balancing act, to set up the COG, and have the a/c fly correctly.

hairyspin
March 9th, 2009, 11:32
Thanks crossram, I've done that with the top node and it was already built with the thrust line on zero. There's just so much to get right and it takes a while to get my head around it, plus I'm still finding out what's important and what can be assumed.

The lack of visual tools once the model is built is a small hindrance; I wouldn't care to say how long it took to get the contact points, static height and pitch right via Notepad. Learned a few things in the process, mind you. And then I discovered ACM - Aircraft Container Manager: it was available some time ago but has been withdrawn and is entirely unsupported. Nevertheless you can still find it if you look hard enough and I found it a revelation for setting the geometry in aircraft.cfg.

I am making progress, just not as fast as I'd like - plus ca change!

BTW Sparks, is there somewhere to set the prop rotation in AirWrench, or do I need to set that myself?

Ivan
March 23rd, 2009, 15:44
With CFS1, you really don't have a choice about fixing the CoG if it is located incorrectly. My preference given a choice is to put the center (0,0,0) of the model at about 1/3 chord of the wing and slightly below the centerline of the engine on a low wing aircraft because the wing structure and landing gear shift the whole CoG down a bit.

Hi Sparks,
I joined late in this thread, but here goes: My understanding was that the Center of Lift really describes the the lift of the wing. In level flight, it would be balanced out by trim effects mostly from the horizontal stabilizer. Ideally, for most efficient flight, the center of lift should be slightly ahead of the CoG so that the h-stabilizer is also providing a slight amount of lift, but still set up in such a manner that the wing stalls before the h-stabilizer. Some aircraft are set up so that the CoG is slightly ahead of the Center of Lift and the tail has to provide downforce but this is less efficient because the wing has to also counter the tail's downforce.

Reasonable?
- Ivan.

fliger747
March 23rd, 2009, 23:13
FS seems to work a bit differently than real planes, but we work with what we have. A real plane with a CG aft of the CL tends to be dynamically unstable in pitch. Yin fights yang to maintain speed stability. In this matter FS can be a bit more forgiving. Certainly minimizing the extra drag of tail downforce with a somewhat aftish (near the CL) optimizes cruise. We utilize this for instance in loading of the 747, however the autopilot is used to provide augmented stability for the preponderance of the flight.

Cheers: T.

Ivan
March 24th, 2009, 11:02
Hi Fliger747,
For every aircraft, there are established forward and aft CG limits. I am not suggesting loading an aircraft to push the CG past its aft limit, but *AT* the aft limit is where the aircraft will perform best as far as speed, range, etc. with hopefully enough stability to remain controllable (by someone's view).

I believe this aft CG limit is often behind the center of lift but am still looking for a documentable example.

- Ivan.

fliger747
March 24th, 2009, 19:43
The Forward and aft CG limits are established on a number of factors; The aircraft must be able to perform the flare manuver at the most forward CG and recover from a stall at the most aft CG and have satisfactory positive stability (for civil aircraft) charcteristics.

An aircraft with a CG forward of the CL utilizes a downforce of the tail to counterballance the nose down pich moment. As speed decreases, the downforce on the tail decreades, dropping the nose to increse the speed. Such a configuration is speed stable. A CG aft of the CL a drop in speed decreases the lift on the horiz stab and causes the nose to rise, further decreasing speed. Such a configuration is unstable in speed/pitch.

Cheers: T.

fliger747
March 24th, 2009, 19:45
P.S:

A book I reccomend as a general primer on aerodynamics is "Aerodynamics for Naval Aviators", available as a reprint at many pilot shops.

T.

Ivan
March 25th, 2009, 12:17
Hello Fliger747,

Actually the references I am using are mostly Federal Aviation Adminstration docs and NACA docs. The FAA docs agree with your statements that the CG must be kept ahead of CL and stability is reduced as the CG is moved aft. At some point, the aircraft will be neutrally stable and the CG limits should be set so that the aircraft remains positively stable. The point point behind the CL at which the aircraft becomes uncontrollable is much closer to the CL than I had realised.

Thanks.
- Ivan.

Henry
March 25th, 2009, 13:00
Yup Tom knows his stuff
he wouldnt take his dog with him if he didnt:gossip:
LOL
H

Ivan
March 25th, 2009, 17:02
I have no doubt Fliger747 knows his stuff.
:applause:

....But the neutral stability point is most likely a inch or so to several inches behind the center of lift. That was the most surprising thing from this discussion. For no reason in particular, I was thinking it should be more like 3-5 inches behind the CL on a single engine fighter. In looking at various references, there was also a comment that various aircraft have limits that are set to make them easier to fly rather than to optimise flight performance. The Boeing 747 seemed to come up a lot in examples as an aircraft that was flown with the CG fairly far forward!
:wave:

Consider the REALLY general case here: The forward control / lifting surface MUST stall first or the aircraft is not controllable. It makes no difference whether it is a canard, conventional, or something looking like the Langley Aerodrome. Just food for thought.

- Ivan.

fliger747
March 25th, 2009, 17:03
A possible exception to the above is the use of fly by wire technology in aircraft such as the F-16 in which control stability is artificially generated, but the advantages of manuverability so obtained are retained.

Some other aircraft, such as stealth models, are aerodyanmically unstable due to their configuration and require synthetic flight control to be flyable at all!

Cheers: T.

fliger747
March 25th, 2009, 17:15
Additional complication results from the relative verticle positions of the CL and CG. Their position with regards to each other can change with attitudinal change. Transonic aircraft will usually have some migration of the CL position at increasing Mach number. In many aircraft this causes a serious nose down moment (Mach Tuck).

Another complication, is the use of high lift devices on both leading and trailing edges of the wing and of course none on the tail.

In the 747-400 a typical CG position for efficent cruise is around 24% MAC. Allowance must always be made for CG shifts as fuel is consumed in flight. This can be a significant factor, especially in swept wing aircraft.

The whole point of CG limits is to retain good controllability junder a wide vareity of conditions. This can include icing, turbulence, incorrect loading etc.

Cheers: T.

Ivan
March 27th, 2009, 15:35
Regarding shifting CoL depending on use of flaps, slats, etc. What I found interesting in various flight manuals was that the net effect was typically a strong nose-down pitching moment when flaps were deployed. That suggests that although the wing has a higher coefficient of lift, the center of lift has shifted aft.

Another issue would be that the CoL might change depending on the angle of attack and thus change its relative position to the CoG. Hopefully it just shifts aft and increases rather than decreases stability.

I guess in wartime, the "rules" sometimes don't apply. A case in point would be the fuselage fuel tank on a P-51D or late P-51B. Another case in point is the Yakovlev piston-engine fighters carrying an internal bomb load aft of the wing.

I also saw a note in the Spitfire IX / XVI manual to NEVER use the aft fuselage fuel tank on a plane with a bubble canopy. Makes you wonder why the tank was installed in the first place.

With any of these aircraft, it is obvious that using any of these features means the CG must be shifted VERY far aft and make the aircraft dangerous to fly, but I guess that in war, the operational considerations outweigh safety.

- Ivan.

Ivan
March 27th, 2009, 15:51
....swept wing aircraft. I take it that a lot of the fuel on a 747 is carried in the wings well aft of the CG?

- Ivan.

fliger747
March 27th, 2009, 19:00
An empty airplane first fueled from the reserve (tip) and outboard main tanks will fall on it's butt... The weight and ballance form we are presented shows the change in CG throught the flight, including numerical enumeration (redundency....) of the tko and lndg CG.

An airplane can be flown with the CG and CL quite close together, even with an aft CG, but it becomes dynamically unstable. Continuous flight control imputs will be necessary, either manual or augmented. It however increases the posibilities for manuvering to some extent. In such a system artificial stick force augmentation (eg. bungees or springs) may be necessary to achieve a suitable stick force per G (for example 10# per G was considered good for a WWII fighter).

Cheers: T.

sparks
March 28th, 2009, 07:31
With CFS1, you really don't have a choice about fixing the CoG if it is located incorrectly. My preference given a choice is to put the center (0,0,0) of the model at about 1/3 chord of the wing and slightly below the centerline of the engine on a low wing aircraft because the wing structure and landing gear shift the whole CoG down a bit.

Hi Sparks,
I joined late in this thread, but here goes: My understanding was that the Center of Lift really describes the the lift of the wing. In level flight, it would be balanced out by trim effects mostly from the horizontal stabilizer. Ideally, for most efficient flight, the center of lift should be slightly ahead of the CoG so that the h-stabilizer is also providing a slight amount of lift, but still set up in such a manner that the wing stalls before the h-stabilizer. Some aircraft are set up so that the CoG is slightly ahead of the Center of Lift and the tail has to provide downforce but this is less efficient because the wing has to also counter the tail's downforce.

Reasonable?
- Ivan.

Correct on CFS1, the FM center is the MDL center.

The wing has a center of lift and the airframe has a center of lift, which may or may not be in the same position.

In FS, we only have one CoL, which is the sum of the lift of the wing, horizontal stab and elevator.

The purpose of the elevator is to change the moment coefficient (CM), but changing CM can also be thought of as changing the postion of the CoL.

In order for an aircraft to fly level, the CoL and the CoG have to be in the same location. If they are not, you will have pitch rotation - i.e. the nose will pitch up or down. That's why the CoL and CoG are usually located close together with the plane loaded and fueled, and the elevator and trim neutral.

fliger747
March 28th, 2009, 09:42
This way of computing the pitching moment of the wing can have some interesting results. An example is a taildragger that sits on the ground at a lower angle than the stall angle. Firstly the tail is reluctant to come up till the wing is about to fly, as concentrating on the moment of the wing tends to undervalue the greater effect on the tail at low speeds, especially with prop blast. Secondly, zipping along in a three point attitude, pulling back on the stick with the tail on the ground will cause the mains to lift off the ground first! In reality with a conventional tail position this cannot happen. It is posssible to fly off in a three point with enough speed, but at a lower speed, the elevator cannot increase the AOA by dropping the tail further (it's already on the ground) we have a departure due to the way FS calculates things.

Jerry's above points are a good illustration of how FS departs from reality in some minor ways, with the stability factors being assigned seperatly. I expect that some of the FS ways of doing things are a result of a legacy of much slower computers in the past. For the designer the general slow evolution is generally beneficial in retaining our understanding of how things work.

Cheers: T.

Ivan
April 4th, 2009, 20:41
Hello Sparks, Fliger747,

Sparks, From your diagram (using FDE notation):
lt = 1205,8,Horizontal Stabilizer Position
l't minus lt = 1204,22,Main Wing Center of Lift

So unless I am misunderstanding, 1204,22 represents the Center of Lift of the wing alone. Of course there are other contributing factors such as angle of incidence of wing, angle of incidence of hstab, Non-dimensional factors (1101,134,Elevator Trim Center) and perhaps a bunch more less obvious factors which contribute to the CoL of the aircraft as a whole. How much of this interpretation is incorrect?

I hesitated a bit before responding to this trail because I was wondering if anyone else would notice an interesting thing in your diagram: The Center of Gravity of the Aircraft is behind the Center of Lift of the Wing (if I am interpreting the diagram correctly). It seems somewhat odd to illustrate something which is never supposed to happen.

Fliger747, FS/CFS does have its strangeness such as using the elevator to increase AoA with a taildragger while on the ground, but some things like prop pitch effect on the tail surfaces are represented:

1101,154,Pitch Prop Effect on Elevator
1101,212,Yaw Prop Effect on Rudder

I have no idea what the proper values are here, but these variables DO seem to produce the effect that is claimed.

- Ivan.

fliger747
April 4th, 2009, 21:01
Having flown a lot of real taildraggers (and owning one) I still find the FS effect quite anemic (other than P-factor etc). In fact many FS airplanes will fly quite well at aft CG location because the pitch stability is assigned a value rather than a deriveed one. Generally however we are able to do a reaonable job of modeling most of the virtues and vices of a typical aircraft.

One of the real issues (no pun intended) with realism in FS is trying to replicate the visual and dynamic realm of flight in a system that lacks many of the peripheral visual, aural and tactile clues that make flying IRL often easier than the same in FS. We all do the best we can....

Cheers: Tom

Ivan
April 5th, 2009, 10:11
Hi Fliger747,
Your hands-on experience is what I am missing when tweaking FS/CFS aircraft. Questions I can't answer are the ones like: "How fast does the tail come around under full rudder deflection in the propeller slipstream?" "At what airspeed does the propeller effect on the tail surfaces become unnoticeable?" "What is the minimum speed for directional control with the rudder under full power conditions?"

No argument regarding differences between real flying and the 1G Flying. Lack of "outside" view is a pain and joysticks with a lot of buttons don't even come close to addressing the problem. I figure that with the creativity of folks like Sparks, and others from back in the days of the 714th, we can try to get a little closer.

- Ivan.

hairyspin
April 5th, 2009, 11:21
Lack of "outside" view is a pain and joysticks with a lot of buttons don't even come close to addressing the problem.

TrackIR. And I wish I had the spare cash to get it!

fliger747
April 5th, 2009, 22:30
Progress is made by community interest and contribution! Many real advancements have been made by the gauge wizzards which can work around some of the "flat spots" in a sim engine, which as good as it is, does not fully apply directly to some situations. FSX accel has gotten a little closer.

HS: I have a track IR, but being of the age of needing progressive lenses which require much head turning for distance adjustment, it makes Track IR a bit jumpy. But it is a cool effect, making the cockpits seem almost 3D.

hairyspin
April 13th, 2009, 11:54
Ah. My eyes are going the same way - thanks for the heads up.

fliger747
April 13th, 2009, 13:22
HS:

I went to contacts and readers, I might try the Track IR again. Some brave and dedicated souls have made multiple monitor setups, though I think a really big HDR TV that you can connect to your computer might be a decent solution.

Cheers: T.