Aerodynamics thread

[thunderchief]

Re: Chinese Engine Development

... Do canards not generate lift? Eh?

Basically, there are two types. First type generates lift during whole flight (just like wings) but it also generates drag. Seldom used in modern airplanes.

Second type doesn't generate lift until deflected and it is used as control surface. Unstable fighters like J-20, Rafale etc ... must use them all the time to stabilize flight and they are controlled by FBW . When fully deflected, this type of canard generates lots of lift , increases turn rate but it also bleeds speed. In normal flight deflection is very small and very rapid (dozens of time each second) so loss of speed is negligible.

 

[Inst]

Re: Chinese Engine Development

Yeah, I did satellite images as well. The dimensions are roughly 13.4 by 20.6, which is actually the figure we get from big shrimps on Chinese boards.

Please, Log in or Register to view URLs content!

Draw two lines alongside the wings to determine the length of the wing area for body lift adjusted wing area. Find the pixel length of the square section of the J-20's wings, then add it to the full length of the wing area. This gives you a length that would be approximate if the J-20 had traditional delta wings that ended in a straight line, instead of tapering off towards the rear. Then find the pixel length of the aircraft, divide the delta-equivalent wing length by the pixel length of the aircraft, then multiply by .5 (for triangular calculation), 20.6 (for length), and 13.4 (for width). The approximate figure I got was 77 m^2.

 

[latenlazy]

Re: Chinese Engine Development

Yeah, I did satellite images as well. The dimensions are roughly 13.4 by 20.6, which is actually the figure we get from big shrimps on Chinese boards.

Please, Log in or Register to view URLs content!

Draw two lines alongside the wings to determine the length of the wing area for body lift adjusted wing area. Find the pixel length of the square section of the J-20's wings, then add it to the full length of the wing area. This gives you a length that would be approximate if the J-20 had traditional delta wings that ended in a straight line, instead of tapering off towards the rear. Then find the pixel length of the aircraft, divide the delta-equivalent wing length by the pixel length of the aircraft, then multiply by .5 (for triangular calculation), 20.6 (for length), and 13.4 (for width). The approximate figure I got was 77 m^2.

Not sure what you're doing with your calculations. There are two methods that can be employed. One requires trigonometry. The other treats half of the wing as a trapezoid. Using the trapezoid area equation the area of half of the wing would be h((b1+b2)/2), with height being half of wingspan, b1 being length of the wing with its leading and trailing edge lines extended to the center of the body and b2 being the length of the straight part of the wing tip. Because that's half the wing area, wing area would be 2h((b1+b2)/2). Since h in this case is half the wingspan (2h=w), the equation would be w((b1+b2)/2), or, alternatively, h(b1+b2).

There's a reason I wouldn't use that image you posted. Tilt angle and perspective means a high rate of error where it's hard to get a range. With satellite pictures of course there's error based on pixel resolutions, but it's a range with known reasonable upper and lower bounds. As I found when using an actual satellite picture differences of .1 meters had a huge impact on wing area (hence the range from 80-90 m^2.

 

[latenlazy]

Re: Chinese Engine Development

How about this from wiki :

That drawing is from paralay, and clearly not an accurate drawing of the shape and dimension of the J-20.

I believe PLAAF didn't want dogfighter for its first 5th gen fighter. Considering China lags behind other powers in engine development , that dogfighter would be underpowered and that is a great disadvantage in close combat. It is better to field interceptor/strike fighter for now , until engine development picks up.

Except the J-20 program was designed specifically with the WS-15 in mind, with the requirements including supermaneuverability. The J-20 and the original heavy fighter version of the J-31 were designed with the same requirements in mind. The PLAAF was asking for a fighter that reflected future technological capability, not its capability at the start of the program. Keep in mind that it's not that the engine got delayed, but that the rest of the program was accelerated. We expected the J-20 to be first revealed 3-5 years later than it was precisely because of the engine.

 

[Inst]

Re: Chinese Engine Development

The problem with satellite pictures is that with such poor resolution it's adequate enough to get length + wingspan, but it's a lot harder to directly get area for measurements. It is hard to tell where shadow ends and wing begins, so you're going to have a much higher error rate.

With regards to what I'm doing, however, is that if you see the wing area calculation for the F-22, it's based on calculating the triangle formed by the forward sweep of the wing, then using that to form a hexagon including the rearward sweep of the wing.

Please, Log in or Register to view URLs content!

Doing something equivalent for the J-20, the shape formed is roughly a hexagon and you can solve it trigonometrically.

Split the shape into roughly 3 sections, upper triangle, lower triangle, and middle rectangle. The area of the upper triangle is wingspan * length of upper triangle * .5, the area of the lower triangle is wingspan * length of lower triangle * .5, while the area of the middle rectangle is length of the middle rectangle * wingspan.

Added together, it's roughly (wingspan * length of upper triangle * .5) + (wingspan * length of lower triangle * .5) + (length of the middle rectangle * wingspan), factor out and you get ( (length of upper triangle + length of lower triangle) * .5 + length of middle rectangle) * wingspan. Multiply and divide each term by 2 and you get roughly (length of upper triangle + length of lower triangle + 2*length of middle rectangle) * wingspan * .5. If you consider that length of upper triangle + length of lower triangle + length of middle rectangle = length of wing, then you can simply the equation to (length of wing + length of middle rectangle) * wingspan * .5.

Now, to get length of wing + length of middle rectangle, you can simplify by finding these two figures as a proportion of total aircraft length. In my case, using

Please, Log in or Register to view URLs content!

(if you insist on using satellite pictures), I get 56% as the percent of total length formed by the length of the wing + the length of the rectangular section.

Finally, multiply total length by .56, then multiply it by wingspan, then divide by 2, and you get 77m^2, which is less than that of the F-22.

===

Also, with regards to using the J-35 as a maneuverable front-line fighter, it makes sense, considering that larger aircraft tend to be more optimized for BVR than their smaller siblings. The trick is that the effectiveness of your radar scales linearly with a one-dimensional scaling factor, while the detection range of enemy radars against you only scales as the square root of such a scaling factor.

As an example, let's say you have an aircraft with an RCS of 1 m^2, width of 5, and a radar aperture of 1 meter. Bear with me. You scale the aircraft up by 10%, so its wingspan increases by 10%, its height increases by 10%, and its length increases by 10%. That increases the RCS by 21%, so now your RCS is 1.21 m^2. Your radar's size also increases by 21%, so now you transmit 21% more energy per radar pulse, as well as having 21% more surface area with which to receive incoming radar signals, so the total signal received on return has gone up by 46.41%. That translates to having 10% more detection range versus enemy targets.

On the other hand, your RCS has also gone up, but the way RCS works is that having a 21% larger RCS only results in an increase of 4.88% detection range against your aircraft.

So put together, your aircraft is now more capable as a BVR platfom, because your increased size isn't wholly balanced by increased detection range versus your plane.

===

The other part is that in WVR, it is a lot harder to have a disproportionate technological advantage over your opponent, because it is a lot easier to make a missile ultra-maneuverable, so kill-death ratios are going to be steadily approaching 1. In this sort of case, if you are aiming to specialize in WVR warfare, smaller becomes better, simply because smaller means cheaper and that you'll win in a war of attrition.

 

[latenlazy]

Re: Chinese Engine Development

The problem with satellite pictures is that with such poor resolution it's adequate enough to get length + wingspan, but it's a lot harder to directly get area for measurements. It is hard to tell where shadow ends and wing begins, so you're going to have a much higher error rate.

As I mentioned, that's preferable to using a picture with perspective and perhaps geometric distortion issues. Pictures where perspective is an issue can yield as small a wingspan as 12 m or as large as 15 m.

With regards to what I'm doing, however, is that if you see the wing area calculation for the F-22, it's based on calculating the triangle formed by the forward sweep of the wing, then using that to form a hexagon including the rearward sweep of the wing.

Please, Log in or Register to view URLs content!

Doing something equivalent for the J-20, the shape formed is roughly a hexagon and you can solve it trigonometrically.

Split the shape into roughly 3 sections, upper triangle, lower triangle, and middle rectangle. The area of the upper triangle is wingspan * length of upper triangle * .5, the area of the lower triangle is wingspan * length of lower triangle * .5, while the area of the middle rectangle is length of the middle rectangle * wingspan.

Added together, it's roughly (wingspan * length of upper triangle * .5) + (wingspan * length of lower triangle * .5) + (length of the middle rectangle * wingspan), factor out and you get ( (length of upper triangle + length of lower triangle) * .5 + length of middle rectangle) * wingspan. Multiply and divide each term by 2 and you get roughly (length of upper triangle + length of lower triangle + 2*length of middle rectangle) * wingspan * .5. If you consider that length of upper triangle + length of lower triangle + length of middle rectangle = length of wing, then you can simply the equation to (length of wing + length of middle rectangle) * wingspan * .5.

Now, to get length of wing + length of middle rectangle, you can simplify by finding these two figures as a proportion of total aircraft length. In my case, using

Please, Log in or Register to view URLs content!

(if you insist on using satellite pictures), I get 56% as the percent of total length formed by the length of the wing + the length of the rectangular section.

Finally, multiply total length by .56, then multiply it by wingspan, then divide by 2, and you get 77m^2, which is less than that of the F-22.

 

[Inst]

Re: Chinese Engine Development

As I mentioned, that's preferable to using a picture with perspective and perhaps geometric distortion issues. Pictures where perspective is an issue can yield as small a wingspan as 12 m or as large as 15 m.

Which is why you don't take dimensional measurements from flight pictures. On the other hand, if you're aiming to take proportional figures, where you just measure the ratio of features within the same axis you can get away with it, and it's better than working with satellite pictures where you can't tell a shadow from a wing.

That said, how do you manage to get 90 m^2 as a J-20 wing area? There is no way you can claim it's around 90 m^2, unless you're going to assume the J-20 is Flanker-sized, which it is not.

 

[latenlazy]

Re: Chinese Engine Development

Which is why you don't take dimensional measurements from flight pictures. On the other hand, if you're aiming to take proportional figures, where you just measure the ratio of features within the same axis you can get away with it, and it's better than working with satellite pictures where you can't tell a shadow from a wing.

That said, how do you manage to get 90 m^2 as a J-20 wing area? There is no way you can claim it's around 90 m^2, unless you're going to assume the J-20 is Flanker-sized, which it is not.

My measurements suggested that 13.5 m was the lower bound wingspan. The ratio of wingspan to length I got was above .66. When I first measured it out I actually got a ratio of .7, which I thought was an unbelievable figure. My gut is that 13.5-13.7 is the closer estimate, but it shows how a .1 difference in ratio has a dramatic effect on wingspan and wing area. A small difference in wingspan can yield large differences in wing area because everything gets multiplied by the other dimensional factors. A 14 m wingspan would easily yield 90 sq meters, so it's not unusual for an 80 sq meter estimate for a 13.5 m wingspan. There are also variations for how long the length of the wing is based on sweep angle (I tried 42 and 43 degrees) and the length of the wingtips.

You can't use flight pictures to determine ratios either because proportions undergo a scaling factor based on distance from the camera and perspective angle. The proper ratios are not preserved

Btw we should move this discussion and stop derailing this thread. I'd be down for a thread dedicated to discussing measurements where this conversation could be moved to, if the mods are willing to accommodate.

Also, I forgot to ask this earlier but which big shrimp mentioned 20.6 and 13.4?

 

[Engineer]

Re: Chinese Engine Development

About wing loading - dimensions are probably wrong, but I believe proportions are right. As you can see, J-20 has proportionally smallest wings.

Wing loading is a meaningless metric for comparing modern fighters. J-20's configuration can generate 1.8 times the lift of an equivalent sized aircraft that purely relies on lift from the wing.

Don't get distracted by canards, I believe they do not generate lift, they are used just as control surfaces.

They do generate lift. They stick out in the airstream and it is impossible for them not to generate lift. Not only do they generate lift by exposing to the airstream, they also form votices creating additional lift on the wing.

My guess is that when PLAAF decides to get 5th gen dogfighter, they will go for something like J-31.

The J-20 is PLAAF's 5th generation dog fighter. J-31 is a much downgraded version of an inferior design.

 

[Engineer]

Re: Chinese Engine Development

The problem with satellite pictures is that with such poor resolution it's adequate enough to get length + wingspan, but it's a lot harder to directly get area for measurements. It is hard to tell where shadow ends and wing begins, so you're going to have a much higher error rate.

People run into that problem because they are using satellite pictures the wrong way, and that wrong way is to try to get absolute measurements rather than relative measurements. Satellite pictures have shown J-20 to be shorter than a Flanker, even when taking blurring into account.

Also, with regards to using the J-35 as a maneuverable front-line fighter, it makes sense, considering that larger aircraft tend to be more optimized for BVR than their smaller siblings. The trick is that the effectiveness of your radar scales linearly with a one-dimensional scaling factor, while the detection range of enemy radars against you only scales as the square root of such a scaling factor.

As an example, let's say you have an aircraft with an RCS of 1 m^2, width of 5, and a radar aperture of 1 meter. Bear with me. You scale the aircraft up by 10%, so its wingspan increases by 10%, its height increases by 10%, and its length increases by 10%. That increases the RCS by 21%, so now your RCS is 1.21 m^2. Your radar's size also increases by 21%, so now you transmit 21% more energy per radar pulse, as well as having 21% more surface area with which to receive incoming radar signals, so the total signal received on return has gone up by 46.41%. That translates to having 10% more detection range versus enemy targets.

On the other hand, your RCS has also gone up, but the way RCS works is that having a 21% larger RCS only results in an increase of 4.88% detection range against your aircraft.

So put together, your aircraft is now more capable as a BVR platfom, because your increased size isn't wholly balanced by increased detection range versus your plane.

That analysis doesn't work. RCS doesn't work in that way because of shaping.

The other part is that in WVR, it is a lot harder to have a disproportionate technological advantage over your opponent, because it is a lot easier to make a missile ultra-maneuverable, so kill-death ratios are going to be steadily approaching 1. In this sort of case, if you are aiming to specialize in WVR warfare, smaller becomes better, simply because smaller means cheaper and that you'll win in a war of attrition.

Major cost of an aircraft lies in the engines and avionics. A smaller airframe does not mean cheaper, especially when that smaller airframe uses the same level of engines and avionics. Moreover, there are examples of dog fights that show bigger aircraft being superior in WVR engagements. For example, in Su-27 vs. MiG-29 combats, Su-27 won every time.