green river rc

My Feedback: (2)

I have a scratch built delta and need to find the CG. I did a search but some of the links didn't work anymore and the ones that did were confusing to me.

What exactly in MAC, and does anyone have a fresh link to a delta CG page.

Thanks,
David

xlr82v2

My Feedback: (1)

Try this calculator: http://www.flybyphotography.com/Craz...culator/CG.htm

MAC is Mean Aerodynamic Chord... Here's a website I found that explains it fairly well: http://www.bakehead.com/wing_CG.htm

For a Delta, I believe you want the CG located at 16-18%MAC for best results, as compared to 25-35%MAC for a conventional aircraft.

green river rc

My Feedback: (2)

Thanks, that is exactly what I need!

DanSavage

My Feedback: (5)

Actually, deltas balance at the same spot as any other planform. Where most people have problems with deltas, and jets in general, is they don't account for the wing area inside the fuselage. You can get away with this on a conventional wing but, not on a wing with extreme sweep, such as those found on a lot of jets. Here's a pic that shows what I mean.

I've built two models of the F-106 and first was balanced at 23% MAC and the second at 24%.

A quick and dirty way to locate the CG to use the landing gear location. In most instances the landing gear is located about 15 degrees or so from the CG. The second pic shows a side view of a drawing of the full-size F-106. The CG mark is shown in this drawing is located at 22% MAC. You have to be really careful with this on modern jet fighters, like the F-16, etc. because they're intentionally balanced tail-heavy to increase maneuverability. The LG is still located 15 degrees from the CG, but the CG is somewhere like 30-35% MAC. So, if you use this Q&D method on an F-16, it'll end up being massively tail-heavy.

xlr82v2

My Feedback: (1)

Dan, I have to respectfully disagree with your assessment of how CG could or should be determined on a tailless (no horizontal stabilizer) Delta planform wing, sorry. Let me explain:

First, please explain why you measure CG in "degrees"... I've been in model aviation since 1982, and full scale aviation since 1988, and fly Learjets for a living today, and I've never heard of or seen CG or LG location measured by "degrees" like that. One fault of measuring CG location with your quick and dirty method is that the distance of the landing gear from the CG will vary with landing gear height if you keep the angle at 17.7 degrees, or whatever arbitrary angle that you should choose... ie, if your landing gear were shorter than what is shown on your drawing, then they would be closer to the actual CG than what they are... if the landing gear were taller, they would be further away from the CG. It's just purely coincidence that the "distance" (actually, it's only the angle) from the CG to the landing gear tire contact patch center is 17.7 degrees. What happens to your angle when the struts compress or extend? I don't believe that this method really would be a good way to determine where the CG or the landing gear should be located.

Also, landing gear location in reference to CG is more of a design consideration for ground handling characteristics than for flight characteristics. Other than for rotation forces required at takeoff, LG location vs. CG will have very, very little, if any at all, bearing on flight characteristics except with retractable landing gear, and then there's a whole other set of variables to consider there, such as CG change with gear retraction, etc. I'd think it's a dangerous idea to base your CG placement on where you want to place the landing gear. A wise designer would always design the aircraft for proper flight characteristics first, then place the landing gear accordingly for proper ground handling characteristics, not vice versa. I'm not saying that your observations are incorrect, but that definitely is not the proper method to determine where the proper CG for flight should be located on a brand new, untested design like Green River RC is getting ready to fly.

Also, why are you considering wing area inside the fuselage, that is not fully exposed to airflow and is not truly an airfoiled lifting surface? By my way of thinking, using your method, then the entire length of the fuselage should be considered to some degree in the MAC calculation, should it not? If you calculate MAC based on chord at the wing roots (where the wing meets the fuselage, as is usually done) I think you'll find that 16-18%MAC will be very near to the same location that you're expressing as 22%MAC on your drawing which takes into account the "unwetted" area of the wing hidden inside the fuselage that doesn't see any airflow over either or BOTH sides. Now, I admit that I'm not an aeronautical engineer by any means, but I do have a pretty good grasp of what's going on, and I can't see why unwetted wing area should have any bearing on MAC calculation. If you can explain why it does, I'm all ears and willing to learn, and can be swayed if you can prove it.

***EDIT***

I've been playing with a drawing of an F106-ish planform, and I can see that taking into account unwetted wing area will increase the MAC and thus bring that line closer to the fuselage. However, 22%MAC taking into account unwetted area is on very nearly the same line as 18% using just the root chord against the fuselage side. This relationship may change somewhat with very wide fuselage widths, but for narrow fuselage widths such as an F106, it's just two different ways of saying very nearly the exact same thing, from what I can see. If you use the same CG@ %MAC location (such as 18%) then considering unwetted area in MAC computation actually moves the CG forward compared to using only wetted wing area. So, I guess I can see how it could possibly cause trouble on a delta with a very wide fuselage. But, with a pure flying wing delta, it's a moot point anyway.


I'd highly recommend to stay within the 16-18%MAC CG for the first flights, especially on an unknown, untested airframe, then one can experiment with CG location on subsequent flights, based on flight test performance, if desired. The 16-18% range is a good "safe" range for most any tailless delta planform. And, always set your landing gear location according to flight CG location, based on %MAC on a swept/tapered wing, or %chord on a rectangular planform wing. LG location is always a secondary consideration to flight CG location, and never the other way around, ideally.

DanSavage

My Feedback: (5)

First, let me apologize for the long-winded reply. Which model Lear do you fly?

FWIW, I've been building model airplanes since 1972 and have successfully designed and flown many scale models including a Pitts S1, Aerospacelines Super Guppy, 2 versions of F-106, Sukhoi Su-27, WB-57N and B-57B and Su-27IB Platypus. I also designed the flying surfaces for my brother's Airbus A300-600ST Beluga. In addition, I've calculated the CG for many, many other modelers for their scratch-built scale and non-scale models. While I only hold a PP certificate, I grew up in an aviation family. My dad and step-dad flew for major airlines.

As I said, it's a quick and dirty method. It's not scientific by any means and I, personally, would only use it as a last resort. Under normal circumstances, I prefer to calculate the CG using the wing planform.

This method assumes the original designer knew what he was doing. Generally speaking, most 3 views depict the aircraft at the normal ride height with the landing gear compressed as much as it would when sitting on level ground. All this does is work backwards from first locating the CG, then placing the landing gear according to the desired ground handling. (think: reverse engineering)

If you were to go through a few 3 views and look, you'll see that the LG location on most aircraft is around 15 degrees of the CG, as shown in the drawing I posted.

I gave the caveat about being careful which model you use it one because it doesn't work with modern jet fighters. As I noted, the original designers intentionally balanced the aircraft tail-heavy to make them more maneuverable. As you're probably aware, to make up for this lack of stability, they incorporated a flight control system.

The reason why many models of the F-16, such as BVM and Yellow Aircraft, sit on the ground with a nose-high attitude is that the designers didn't move the LG forward from the scale location when they relocated the CG. This places the LG beyond the 15 degree position which makes it harder for the tail plane to rotate the model because of the increased length of the moment arm between the LG and CG.

The usual result is that the model does not respond to elevator input at first when the pilot rotates. Then, finally, as the airspeed increases, the elevator is able to generate enough down-force to overcome the moment arm and the model leaps suddenly into the air.

The proper solution is to move the main LG location so it is within the 15 degrees of the CG location. I did this on my Su-27 and more recently on a foamie F-22 I put together. Both of these models sit level on the ground and rotate easily when commanded. In the case of my F-106, the original designers placed it there.

Because the wing thinks it continues all the way to the centerline. As far as the wing is concerned, it's a flying wing. The forces they generate don't stop at the outer skin of the fuse. They go all the way to the centerline and they follow the shape of the wing planform.

The reason why you don't include the entire fuselage is that the area behind the CG cancels out the area ahead of the CG. So, including the fuse area ends up being a wash. The exception to this rule is the modern fighter designs like the F-16, F-18, Su-27, MiG-29, etc. When calculating the CG for these designs, not only do you need to carry the wing all the way to the centerline, but you also need to include the strake area ahead of the wing. If you don't include the wing area inside the fuse and/or the strake area, the model will end up massively tail heavy. Here's pic of my Su-27 design showing the CG location difference between including and omitting the strake area.

I didn't account for either on my first jet, an F-16, and it was massively tail heavy and crashed. I'd only calculated the wing area outside the fuse and the CG ended up being something like 3" aft of where it should have been located.

Also, another reason for including the wing area inside the fuse is that it 'normalizes' the CG location across varying planforms. For example, if we were to follow what you suggest and only use the wing area outside the fuse of a design like the F-16 or Su-27, then the CG would be located somewhere like 10-15% of the MAC. The problem is that you never know where the starting point is located as you move from planform to planform. So, instead of using 'normalized' data, you end up taking a stab in the dark at it and crossing your fingers, hoping you're right.

Personally, I'm a little uncomfortable designing a model, building it, finishing it, then guessing at where the CG should be. Been there, done that and have got the crashed models to prove it.

The only designs that included the entire fuse area was the Guppy and the Beluga. Both of these have airfoil shaped fuselages. If you don't include the fuse area, they will end up being tail-heavy. The Guppy crashed on it's maiden voyage because I didn't include the fuse area. Once the fuse area was taken into account, it flew beautifully. When my brother built his Beluga, we took the fuse area into account from the beginning and it flew great. At the same time my brother was building his, another modeler was building one in Germany. The other modeler didn't take into account the fuse area and his model was massively tail-heavy. He built and crashed two different models due to improper CG. We calculated the CG for his third model and it flew perfectly.

As I wrote above, I've designed many models and have been using this method for many years. This method of CG calculation by carrying the wing planform all the way to the centerline works very well across many different planforms and eliminates all guesswork.

Here's a link videos of [link=http://SavageLight.com/video/su-27_full_flight.mpg]my brother's Su-27[/link] (my design) and [link=http://SavageLight.com/Dart/jetday-f-106.mpg]my first F-106[/link], the [link=http://SavageLight.com/video/platypus_test-flight.wmv]maiden voyage of my Su-27IB[/link] and finally, the [link=http://SavageLight.com/video/beluga_test_flight2.wmv]maiden voyage of my brother's Beluga[/link]. The Su-27 was balanced at 23% MAC, the F-106 at 22%, the Su-27IB at 25% and the Beluga at 25%.

See the pattern? This is what I meant by 'normalizing' the data. All of these widely varying designs used CGs calculated this way. All these designs CG ended up between 22% and 25% MAC and all are rock-steady in flight. And, for those that that have landing gear, the main LG is located right around 15 degrees of the CG.

xlr82v2

My Feedback: (1)

Dan,

OK, I see where you're coming from now. A long winded thorough explanation is much better than a short one IMHO, so thanks for that, no need to apologize at all.

So, how do you factor the strake area into your MAC calculation? That's kind of a complex shape on most aircraft. Like I said, I'm not an AE by any stretch...

DanSavage

My Feedback: (5)

Just for grins and giggles, I calculated the CG for the Su-27 without including either the fuselage or the strake area. Just the wing alone.

This is defined by the green outline. Also in green is the MAC and the CG at 25% MAC.

In the case of the Flanker, when you use just the wing area (shown in green), the true CG location, shown in white and proven under actual flight conditions, ends up being a negative number as it's ahead of the leading edge of the wing. And if you attempt to fly the model with it balanced at the 25% MAC location, certain disaster awaits. This is exactly what happened with my first jet. (F-16)

This is an extreme example, but it pretty clearly illustrates why you need to include the portion of the wing inside the fuselage when calculating the CG.

Dan

DanSavage

My Feedback: (5)

No problem. I'm not an AE, either, nor do I play one on the internet.

I use a function in available in most CAD programs that finds the geographic center of a geometric plane, otherwise known as the centroid. As it turns out, the centroid also happens to fall directly on the MAC. So, first, I trace what I think the wing planform should be, then find the centroid. Once I know where the centroid is, I simply draw a line that intersects the leading and trailing edges of the wing and that is the MAC. Even though my Su-27 drawing shows both sides, you really only use one-half of the planform. If you use both sides, the centroid will fall on the centerline, which is pretty useless because the MAC would run the entire length of the planform.

To find the centroid in the pre-CAD era, most model designers would cut out the wing planform out of cardboard, then found where it balanced so the cardboard planform would sit perfectly level. Once they found this point (centroid), they simply drew a line from the leading to trailing edges to define the MAC. Then, to confirm the CG, they would make a balsa chuck glider.

You can still use this method today, though it is a bit slower.

xlr82v2

My Feedback: (1)

Dan,

I fly the Lear 55 and Lear 60, and Lear 35 every once in a while, but I try to avoid the 35 if I can

Here's an interesting piece of Learjet CG trivia: Allowable CG envelope: -4 to +29%MAC (Learjet 55)

[8D]


So what CAD program do you use? I bought TurboCAD 14 and all the training materials with it, but I haven't sat down and tried to learn it yet... need to do that someday...

DanSavage

My Feedback: (5)

That's very interesting. I never knew that.

Here's Q&D CG for a Lear (not sure which model). The entire wing all the way to the CG is in blue and the wing area outside the fuse is in green. The lines which extend all the way to the centerline are at 25% MAC. The blue and green boxes are the CG range -4 to +29% MAC.

I'll bet it flies really bad at -4% MAC.

I use DesignCAD 3D MAXX. It's a little clunky compared to the newer programs. I'm going to use Rhino 3D to design my next model.

I've never used TurboCAD, but I have read a lot of nice things about it. Yup, that's what it takes. Eyeballs in front of the screen. It helps if you start off with small, but real projects to get your feet wet. Things like servo 3-views, retract 3-views, etc. If you try to learn a CAD program by designing an entire airplane, you'll get pretty frustrated pretty quick.

Dan

xlr82v2

My Feedback: (1)

Dan,

That's a Lear 35 (or it could possibly be a 36... there's no real way to tell externally between the 2 of them)... I'm not sure what the CG envelope is for the 35, as my type rating was done in the 55, and I'm much more knowledgeable on the 55...

As far as flying characteristics, It likely flies similarly to any other CG location, just will require more nose up trim for takeoff. I think the furthest forward that I've seen it in day to day operations is about 2-4% MAC... as far as the way the aircraft flies, you can't feel it... it just requires a different trim setting.

bps

My Feedback: (3)

I have built many delta origional designs over the years. I use the balsa scale glider approach to deturmining the CG. Have not lost a delta yet and they all flew off the drawing board including a twin engine tractor.