• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

incorrect pitch attitude during turns(737NG)

Messages
12
Hi everyone,

I'm working on the flight dynamics of a 737NG and have come across a problem,maybe one of you knows a fix for it.

At a given weight and speed the aircraft pitch attitude during turns
is completely off,it needs too much nose up attitude to maintain the present altitude.

Please no offense, but do not lecture me about the involved aerodynamics and the need for increased pitch during turns.I am fully aware of that and have taken that into account.I have quite a few hours in real aircraft and the values are off big time.

At a given weight and speed, on the real aircraft turns with 25degrees of bank require about 2degrees of pitch increase to maintain altitude during a turn.
The flighsim model needs almost 6 degrees pitch increase during 25degrees
bank to maintain altitude under the same conditions.Since this is nearly
triple the real value I would like to fix it.

However I don't know that much yet airfile tuning.Does anyone here
have an idea,which value could be modified in what direction to fix this?

The problem is only related to pitch attitude during turns, in level flight the pitch values of the flightdynamics are spot-on.

Thanks a lot already,

Oliver
 
Last edited:

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Oliver
This is not involved aerodynamics, just a few basics.
The key parameter in Angle of attack (AOA) which, for level flight at steady speed is the same as pitch angle but measured down from the fuselage or pitch angle to the horizon. AOA combined with the slope of Lift coefficient versus AOA gives Cl the lift coefficient. As you said you believe the pitch angles in level flight are spot on, but you also need to know the AOA and your stall margin.

In a 25° bank turn you require 10% more lift, which means more AOA. If this gives you a 6° increase in pitch/AOA you must be flying at a speed/weight where you have little stall margin. AOA increases rapidly as the stall is approached. In tests I'm doing at the moment I get a 10° increase in AOA for a 20 kt decrease in speed approaching the stall. A 20 kt increase in speed gives 2.5° decrease in AOA, so you can see how steep the increase can be at low speeds.

To be able to help you, I would need you to repeat your test at a variety of speeds and write down the data. I would expect that higher speeds would give less pitch increase in the turn.

FS simulates flight characteristics very accurately so it will be easy to solve your issue with some data on the flight conditions where it occurred. A pop-up gauge with read out of flight data would help you.

Roy
 
Messages
12
Hi Roy,

first of all thanks for the quick reply to my topic.I appreciate your help!

Without wanting to get into a deep discussion of aerodynamics,even in level flight and steady speed, AOA and pitch angle are not the same.The reason is, pitch angle is measured as angle between aircraft longtitudinal axis and the "horizon" so to say, while AOA is the angle between wing chordline and the relative airflow.Depending on the definition it can also be the angle between aircraft chordline/ long. axis and relative airflow.However that is not the same as pitch angle.
This document by Boeing explains the difference pretty well,with graphics too:
http://www.boeing.com/commercial/aeromagazine/aero_12/whatisaoa.pdf

However back to the topic.I agree with you,that because the lift vector is not pointing straight up anymore during banked turns , you need more lift than in level flight to compensate the angled lift vector (to act as an equivalent force to the unchanged aircraft weight).

I also agree that the behaviour is more pronounced at lower speeds than at higher speeds.That is because velocity has a larger influence on total lift than AOA.

In normal level flight FS simulates flight characteristics well,however during turns it s beginning show flaws. My problem is apparent in every
&large size jet transport I have tried in FS so far. The "loss of lift" is simply
too large and since all aircraft show it, I strongly suspect somewhere the FS internal flight calculation model is not perfect, concering bank influence.
Still I hope one can compensate for that by adjusting the individual flight dynamics of aircraft.

The problem becomes the most pronounced in the speed range between about 230kts to 120kts. In other words from slightly above clean speed to
full flaps approach speeds.
Ok sorry for writing such a long post, but I wanted to make sure we are talking about the same things before looking for a solution.

You said you need more data Roy .Please tell me exactly what you need and will try to supply it.And also please let me know what your suggestion is where to start fixing the issue.

I really appreciate that you're trying to help with this.Thanks a lot!

Kind regards,

Oliver
 

mgh

Messages
1,413
Country
unitedkingdom
If the flght path vector is horizontal (level flight) the difference between pitch angle and angle of attack is constant. As a result a change in angle of attack equals the change in pitch angle.

At a steady speed, the change in lift coefficient required to maintain a level turn varies directly with an increase in lift. As Roy explained, this is 10% in the case of a 25 deg banked level turn. The corresponding change in angle of attack depends on that change in lift and on the straight and level lift coefficient. The higher that lift coefficent the greater the corresponding change in angle of attack. This means that any comparision must be made at the same initial lift coefficent.
 
Last edited:

mgh

Messages
1,413
Country
unitedkingdom
If the flght path vecror is horizontal (level flight) the difference between pitch angle and angle of attack is constant. As a result a change in angle of attack equals the change in pitch angle in level flight.

At steady speed, the change in lift coefficient required to maintain a level turn varies directly with the change in lift - 10% on the case of a 25 deg banked level turn. The corresponding change in angle of attack depends on the change in lift and the straight and level lift coefficient. The high that lift coefficent the greater the change in angle of attack. This means that any comparisions must be made at the same initial lift coefficent.
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Hi Oliver,
I was keeping things simple. Actually FSX measures AOA relative to the fore/aft fuselage axis. Hence, in the sim, pitch and AOA are the same in level flight, except that nose up pitch is negative. If level at -10.0° pitch, AOA is 10.0°. I know that is not the way it is done in real life, but this is what FS does.

In a level turn the ratio of Lift/Weight varies as 1/cos(bank) As I said, at 25° bank the ratio is 1.103. At a given speed CL needs to be 1.103 times what it was wings level. Since CL and AOA vary proportionally, AOA will need to be 1.103 times what it was in level flight. As I also said, especially at low speeds, AOA and speed do not vary proportionally.

While CL and AOA vary proportionally, the slope of the relationship will be different for different airfoils, as will CL max where the stall occurs. In FS this is specified in Table/Record 404 in the .air file. There are several other Records that affect it but 404 is the main place. Theoretically CL max should vary as the cosign of wing sweep angle, but that does not occur in the sim.

The problem becomes the most pronounced in the speed range between about 230kts to 120kts. In other words from slightly above clean speed to
full flaps approach speeds.

And that is the area where AOA is high.

The data I'm talking about is IAS, pitch angle, AOA, weight, altitude, flap settings. All of these are available for readout as A:Variables, so you need a gauge that will give you that data. Plus a copy of the air and config files.

I seriously doubt that there is a problem with the way FS handles bank, but I never fly any kind of airliner, all my jets have two seats max, so it may be something I have not experienced. There is an excellent paper on how FS handles the aerodynamics written by Zyskowski. Should be easy to find, if not I can post a copy.

Before you do anything to 404 (which is very easy to screw up), I would suggest adjusting this to more than 1.0
[flight_tuning]
cruise_lift_scalar=1.00000

It has no influence on the slope of the CL AOA curve, but it effectively increases CL, so the AOA at any speed will be less.

Always glad to help

Roy
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
@ MGH. You were correct in the post you deleted.

However, you are confirming that the required increase in AOA is higher at higher CL, which for level flight will occur at lower speeds. Absent an AOA indication, this will appear as a pitch increase which is the topic of this thread.

It is also a good explanation for the increased rate of AOA change as the stall is approached, plus how this is affected by the Cl Alpha slope.

Thanks
Roy
 
Last edited:
Messages
12
Hi Roy & MGH,

thanks for your further replies and explanations.

@Roy
I downloaded the paper by Zyskowski, which contains some interesting
information.Still it seems to be generic and does not specifically detail
how Flightsim for example processes the individual airfile tables or what
each table represents(I know some information on that is available with the
editors).
But still a good read,thanks for pointing it out.

@both:
Sorry if I sound a bit slow,but just to make sure,do I understand you correctly,that you both more or less agree that the issue lies somewhere
in the relationship between Coefficient of Life CL and AOA in the mentioned speed range?

As for fixing it,Roy is there a readout gauge somewhere available for download?Unfortunately I have zero XML gauge design knowledge to create
one myself to track the various factors.

And Roy, if I increase the cruise_lift_scalar value in the aircraft.cfg to increase CL, won't that in turn mess up the correct pitch attitudes during wings level flight?With an increased value, these pitch attitudes will
then be too small ,won't they?

Last issue, the aircraft in question is a payware product, so I can't simply copy the .air and cfg file and upload them here.

Sorry if its a lot of trouble,thats the problem if someone is new to something and doesn't have the necessary knowledge yet.

Thanks again and looking forward to your reply,

Oliver
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
This links you to the best description of what FS does in the air file.
http://www.fsdeveloper.com/forum/showthread.php?t=332348

It will take you some time to read. But it has most of the necessary equations as they apply to the air file records.


I have several pop-up flight test gauges and I'll see which is most suitable and post it here.

It will give you accurate pitch info, much better than you would get from the panel.

Best way to get accurate pitch/AOA data is to start at some high IAS and very slowly decrease speed in autopilot altitude hold, pause when you hit one of your target speeds write down the data and continue down to the stall.

Since you rightly can not post payware documents, but can post test results, the ball is in your court!

Roy
 

mgh

Messages
1,413
Country
unitedkingdom
I find that AFSD is very helpful. It's a self-contained application that reads a fairly comprehensive amount of FS data direct from FS using DSUIPC. For this case it can display: speeds, AOA, pitch, CL (and its components). plus lots more.

You can download free from:

http://www.aero.sors.fr/afsd.html
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Oliver
Attached are the zipped background and XML code for a Flight test debug gauge I put together at the request of the author of AFSD. It is designed to be used as a pop-up.
In your panel.cfg you will need an entry like:
[Window Titles]
Window07=Flight Test

You call up the gauge with Shift 8

You also need
[Window07]
Background_color=0,0,0
size_mm=250,310
window_size_ratio=1.000
visible=0
ident=Flight_Test
window_size= 0.250, 0.310
window_pos= 0.70, 0.20
gauge00=xxxx!Herve_Flight_Test_Debug

where xxxx is the folder in the Gauges directory in which you saved the background and XML code.

You may need to edit these depending on what pop-ups already are in place

Most of the outputs are obvious, the right side is mostly engine stuff. The left side gives pitch and AOA plus three outputs to do with Record 404.

STALL gives the AOA used by FS to illuminate the stall warning and it is pretty much CL Max. Units are degrees
ZERO LIFT is the AOA where CL is zero. Units are degrees
CL ALPHA gives the slope of the CL vs ALPHA straight part of the 404 curve. Units are per radian

The above three are static as is the CG PCT.

Now you can start getting data.

Roy
 

Attachments

  • Herve.zip
    15.1 KB · Views: 395
Messages
12
Hi Roy and MGH,

thank you both for your links to the data tracking tools.Glad someone
already programmed such a gauge.
@Roy:
That explanation of the airfile-readout is excellent!Great link!

Roy, I installed your gauge and did the testing already, however before
I start posting lots of numbers, I wanted to ask one thing.
On your gauge, the readouts for Stall, Zero lift and CL ALPHA all showed
"0.0" all the time, during all my test flights.
Is that correct or something wrong with the readout/installation of the gauge?

All other parameters were tracked and varied during the testing.

Regards,
Oliver
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Oliver,
I get the correct readouts for the one's you had as zero. If you get any readouts, the gauge is installed correctly


I know FS calculates them from table 404. If you have zeros it kind of suggests there is something wrong with your 404. If that is the case, you found where your problem is coming from.

I would suggest installing the gauge in the stock 737 and see what happens.

My next suggestion is to save your current 404 and substitute the one from the stock 737.

Sometimes FS appears to use default values when the air file is corrupt, although the thing I usually see is idiotic antics, back flips etc.

If it was using default values it is possible that the A:vars concerned are not addressed.

When an air file record is corrupt, the record concerned usually does not show up in an editor like AAM. So, that would be the first thing to check.

Roy
 
Messages
12
Hello again,

not sure what to do now as I am a bit confused.
@Roy:
I did install your test gauge into the stock 737-400.
Strangely enough the result is the same, all variables are displayed
and vary during flight, but the 3 mentioned above still show 0.0
the whole time.
I installed the gauge into another 3rd party aircraft with the same
result.Not sure what is the reason for that,or why you see results
for those 3 factors and I don't

@MGH/Roy
I followed your advice and cross-checked with AFSD and interestingly
it seemed to be able to readout the airfile including the 3 factors
missing from Roy's gauge, at least I think so, they are named slightly
different in AFDS

So whats the next step now?I'm not sure where the problem is located.

The 3 values are :
Stall AOA : 12,04
AOA CL=0 : -1,75
Lin CL alpha (is that the same as CL Alpha in your tool Roy?): 5,5753

Oliver
 
Last edited:

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Oliver
You never mentioned what version of FS you are using.The three variables are not in the "New for FSX" section, but I know of some others that were new, but were not mentioned, so you must be on a version earlier than FSX.

I just checked again and the test gauge works fine in FSX, including in the stock 737-800 note -800 not -400
STALL 15.986
ZERO LIFT -1.2564
CL ALPHA 5.954, This is the same as AFDS reads I just abbreviated it to fit the space it should read linear CL ALPHA

The -800 has a zero lift AOA to stall AOA range of 17.24° and a CL max of 1.79.
Yours has a range of 13.79°, which with the slope gives a CL Max of 1.34

For the same weight and speed yours is going to be operating at a higher AOA and closer to the stall than the -800.

I know nothing about the NG version other than the fact that the -800 is one

Any how, now you have the 3 variables through AFDS and so 400 is OK though a bit shy on lift, take some speed/AOA/pitch data and see how the AOA relates to 12.04 when in a turn

Roy
 
Last edited:
Messages
12
Hi Roy,

I'm sorry, I completely forgot to mention the very basic data.I'm still
FS2004 here.
Thats also why I was talking about the stock 737-400, FS2004
still features the -400,unlike FSX which includes the -800

The Flight dynamics I want to modify are for a 737-700 & -800 addon.
My issue is present on the -800 variant and is extremely obvious in the -700
variant,where its even worse.

Ok I will do some testing and report back.
Roy thanks a lot for bearing with me and your continueous advice.Thats very
kind of you.

Oliver
 

Roy Holmes

Resource contributor
Messages
1,803
Country
us-virginia
Hi Oliver,
No problem, that is why the three variables do not show.
From what little I know, the NG series had a different wing and maybe your model does not have that, both in dimensions and the correct CL alpha characteristics. AFAIK the wing area is now 1344 sq ft, span is 117 plus and the root chord is 21. If you have those dimensions, then my guess is the problem lies in 404, Low CL Max, low AOA range.

Roy
 
Messages
12
Hi Roy,

hmm various things have come up.Not sure what conclusions to draw
from them.There are slightly varying figures on the internet regarding
the wingdata of the 737NG, however they differ not by much.
The models wings according AFSD matches those roughly( root chord seems
to differ by 2, 19 compared to 21).

Then I first crossed-checked again, level flight pitch valus at different weights
against a pitch&power table from the real QRH.These values fit pretty well for a flightsim adddon.
So I guess one could say, the cruise_lift_scalar value is set correct?

Then I did some testing at different weights&speeds using your gauge.
There seemed to be a small difference in pitch attitude readout between your
gauge and what the primary flight display showed, your gauge seems to
be more accurate.The PFD shows a slightly (just barely 1 degree) higher pitch,
than what your gauge says.


The AOA value change between level flight and 25degrees bank varies not so much. The increase from level AOA to AOA at 25 degrees of bank was between 0.3-1.1 units , depending on the speed.

The maximum AOA was 6.4 for the -800 (at 160kts Flaps15 maneuvering speed &gear down)
The maximum AOA was 7.9 for the -700( again at 160 , but for -700 this is flaps 5 maneuvering speed with gear up)

At speeds above 160 and below the maximum AOA was less.

Roy do you need the indivial figures for each weight,speed,pitch and AOA?
If so let me know.

Regards,
Oliver
 
Top