PDA

View Full Version : An AOA problem?



PRB
October 3rd, 2008, 16:35
Hi, aerodynamic gurus.

I’ve been trying to make a half-decent FS9 air file for the CFS2 Ki-45 by Akemi. I’ve got it flying ok. The excessive nose heavy CG is fixed, as well as most of the pitch instability. I think it’s about ready to be released on the unsuspecting public, except for one thing:

The reference speeds section says she stalls at 58 kts with full flaps, and indeed she stalls at just around that speed, but, in order to get the plane that slow requires an extreme nose high attitude. That means you can’t land the plane in the usual technique in which the ship is just on the edge of a stall as she touches down, because if you do that the ship will hit the ground tail wheel first, and you can’t see the runway anymore anyway. So you have to land hot.

My first thought was to tweak on the CG again, but I don’t think that’s the problem. Moving the CG forward will only result in more backwards force required on the yoke to keep the ship in the nose high attitude needed to fly that slow.

Having a hard time getting my head around this one...

Also, while doing stall tests at 5000 feet, I noticed that even when the power is pulled back to idle, the nose doesn’t fall, and in fact wants to come up, even while very close to stall speed. The ship is going down big time, as can be seen by checking the VSI and the altimeter, but the nose won’t drop. Even when she does stall, the nose takes a long time to fall off, and then quickly comes up again.

fliger747
October 3rd, 2008, 20:16
Paul:

What AOA is she stalling at? Is the CG above the aerodynamic center? You might try lowering the CG. Most twins are not full stall landed, the at rest angle must be about 7 deg or so and the stall at 15 deg or more? I do land the Supercub more or less tail wheel first, but with the tundra tires she is still a little bouncy. In service they probably wheel landed these.

Cheers: T.

sparks
October 3rd, 2008, 22:10
Tom, you're on the right track, but I don't think vertical movement of the CoG will change anything.
Paul, look at your lift curve. What's your critical angle of attack? How does that compare to the at rest angle on the ground? What tools are you using for the air file?

PRB
October 4th, 2008, 20:49
I figured it out: The plane had no flaps!! This is just too much fun. Ok, from the beginning.

At some point today I began observing that lowering the flaps didn’t seem to have much affect on the flight characteristics. Then I got to thinking that hey, the landing experience thus far, too fast, can’t slow down, etc., has been what I would expect in a flaps up landing. So I decided to do a flaps flight test. Took her up to 5000 feet, trimmed the plane for level flight and dumped the flaps. I heard sound of flaps moving, and jumped out to spot view to confirm that they were indeed down. But handling characteristics? Nothing changed!

Then I remembered something else. From the start of this effort, every time I went to the Advanced Flap Settings area of the Systems tab (AirWrench v1.01.64) all my flaps data would disappear as soon as I clicked the “Save current flaps configuration to Aircraft.cfg file” button. I didn’t know why this was, and simply stayed away from that button. I should have been paying closer attention to that page…

So I went back to the Advanced Flaps Settings page. First thing I noticed was that all three scalar values were set to zero. Then I noticed the System type field was set to 4, which means no flaps! When I set that value to 1 (hydraulic), the flaps data remained intact when I saved the settings to the aircraft.cfg file. Imagine that!

So I set the scalars to 1, and the type to 1, and took the beastie for a spin around the airport, and now the flaps work, and the plane lands like a plane should. And dropping the flaps results in that usual nose pitch up increase in lift thing.

Shane Olguin
October 5th, 2008, 08:36
Hey Broham,

If you would like to model in a clean stall break, try increasing pitch moment at a given AoA for your stall. Remember, 1 degree = .01747 radians on this table. It's table 473. Here's an example attached.

fliger747
October 5th, 2008, 09:45
Good info Shane!

Paul: Do you have Airfile Manager? It will give a graphic presentation such as Shane has shown above and allow you to enter base ten numbers, rather than having to directly use hex. However, any direct airfile mods must be done after your final efforts in Airwrench as it always goes through complete computations. However as an exception to this, if you do have some table mods that you like, you can save the copy of the airfile and copy the appropriate tables back into the active airfile using Aired.

The effect on the stall break of the CG height is not as pronounced as the airfoil charcteristics, however in My Supercub, the stall break is noticably different on floats as oppose to wheels or skis. With a low CG the CG also moves forward as the AOA increases, giving a more noticable increase in required elevator force for a given attitude and a more active stall break.

They only let us stall the 747 in the sim.... So I can't comment on it too much!

Cheers: T.

PRB
October 5th, 2008, 10:31
Thanks, Shane. Iíll have to try that.

Tom, I have Airfile Manager, but it wonít install on this machine (WinXP x64). I sort of stumbled on exactly the technique you described: Work first with AirWrench, get it as close as possible, then tweak things in the individual tables using AirEd. Sometimes I have to steal some of those tables from other planes, since when the air file comes out of AirWrench, some of the tables are no longer editable.

fliger747
October 5th, 2008, 16:14
Interesting. I have Airfile manager on both an XP-64 machine and a Vista machine. It does visually vibrate on the Vista machine and can be really pokey.

Cheers: T.

Shane Olguin
October 6th, 2008, 16:53
T,

If you manually adjust the view windows (the black windows) a bit you'll find that it stabilizes the flashing windows. It's an issue with Vista display drivers. If you need more detailed help on how to do this let me know. You can get it to stop. ;-)

fliger747
October 6th, 2008, 19:18
Shane: Thanks, it's been an irritation, not totally disabling the program, but certainly an obstacle. A question? Is it possible to add data points? Some tables would benefit from another waypoint or two.

Cheers: t.

sparks
October 6th, 2008, 20:39
Shane: Thanks, it's been an irritation, not totally disabling the program, but certainly an obstacle. A question? Is it possible to add data points? Some tables would benefit from another waypoint or two.

Cheers: t.

CL-AoA and CM-AoA can go up to 47 points each, but the stability coefficient tables are very limited. A good rule of thumb is do not exceed what the stock files are using, otherwise FS may crash or refuse to load the air file.

AirEd will generally edit any length table, it all depends on the definitions in it's ini file. Same for AAM, which uses the same ini format as AirEd.

Shane Olguin
October 9th, 2008, 17:20
If you want to add data points you'll probably need to use AirUpdate. What I do is use AirUpdate to dump the entire .air file, then I go in and transfer the table I want to add stuff to into an Excel spreadsheet and do my work there. When it's done I cut and paste the stuff back onto the dumped .air text file and update. As long as you adhere to the recommendations Sparks made about not adding too many data points to certain tables, you're good to go.

fliger747
October 9th, 2008, 17:33
Thanks Shane: Possibly adding at most two data points.

T.