correct as is No adverse yaw, too much yaw stability and lack of departure qualities at high AOA

[Akiazusa]

This report details inconsistencies in the DCS F/A-18C Hornet's flight model behavior at high Angles of Attack (AOA), especially the stability and controllability derivatives, comparing public wind tunnel data by NASA, and other papers.

 

Issues:

1. Minimal Adverse Yaw: At high AOA (around 40-50 degrees), the aircraft exhibits minimal adverse yaw during aileron deflection rolling maneuvers with MSRM activated. This is unrealistic, as real Hornets experience significant adverse yaw at these AOA due to asymmetric induced-drag generated by both the aileron and the differiential stabs.

2. Excessive Roll Control Authority: At high AOA, the ailerons & differiential stabs appear to retain an unrealistic amount of roll control authority. Real F/A-18C Hornets experience a reduction by more than a half in roll control effectiveness at high AOA compared to those at 0 AOA, according the Cl-δA chart from NASA. Note that this also doesn't include the contribution by the differiential stabs.

3. Excessive Yaw Stability: Check post below.

4. Lack of departure quality: There's a lack of nose-slice and sideslip excursions that can result in a falling leaf OCF.

 

Supporting Evidence:

The report "AERODYNAMIC PARAMETERS OF HIGH-ANGLE-OF-ATTACK RESEARCH VEHICLE (HARV) ESTIMATED FROM FLIGHT DATA" (NASA TM 202692 with a link https://ntrs.nasa.gov/api/citations/19900019262/downloads/19900019262.pdf) provides real-world aerodynamic coefficient data relevant to the F/A-18C. Specifically, the following parameters are crucial:

Cl-δA: This coefficient represents the rolling moment generated by the aileron deflection. A decrease in this value at high AOA signifies reduced roll control effectiveness. Checked by ED.

Cl due to differential stabs: Not included in the NASA document. But worth checking.
Cn-δA: This coefficient represents the yawing moment generated by the aileron deflection. A positive value at high AOA signifies adverse yaw. Checked by ED.

Cn due to differential stabs: Not included in the NASA document, so the total amount of adverse yaw moement should be higher than a single Cn-δA.

Cn-β: Check post below.

The data from this report can be used to compare with the current flight model's behavior and identify discrepancies.

 

 

Testing Information:

MSRM (manual spin recovery mode) is used for separate control surface manipulation. Turn on the MSRM switch to the rightside of the right DDI before testing.
Thank you for your time and consideration.

 

Video link:

https://packaged-media.redd.it/uk23tvykr6uc1/pb/m2-res_480p.mp4?m=DASHPlaylist.mpd&v=1&e=1713337200&s=f9d5e02a332e95290ab60fe357a05fb884d866ab#t=0

F18 no adverse yaw at high AOA.trk

Edited April 18, 2024 by Akiazusa
update ED checking progress

[BIGNEWY]

Hi, 

we have checked out the data for aileron influence at rolling and yawing moments and they are very close to the evidence you presented. 

We are happy with the results and consider it correct as is. 

thank you 

[Akiazusa]

1 hour ago, BIGNEWY said:

Hi, 

we have checked out the data for aileron influence at rolling and yawing moments and they are very close to the evidence you presented. 

We are happy with the results and consider it correct as is. 

thank you 

Hello,

But I'm not seeing much adverse yaw in the track file. While I appreciate the confirmation that the underlying aerodynamic data is correct, there seems to be a disconnect between this data and how it is actually translated to roll rate and yaw rate in game.

In the track, when I'm testing with a full right stick at 40-50 AOA with MSRM activated, I'm expecting the nose to slice to the left, but all I got is a roll to the right, and there's not any perceivable yaw motion.

If you consider how the spin arrow works, it directs the pilot to move the stick into the spin direction. This is using adverse yaw to stop a yaw motion. But how can it stop a spin if adverse yaw is very weak like what is shown in my track?

 

From another reference document https://trace.tennessee.edu/utk_gradthes/2372, it says in the MSRM falling leaf entry procedure that: "Apply lateral stick to generate sideslip (some side-to-side cycling may be necessary to generate largest sideslip)." This is what I actually did in the track. But as you can see there's not much sideslip generated that can throw me into a falling leaf.

Edited April 16, 2024 by Akiazusa

[Akiazusa]

If anyone succeeded by using the MSRM falling leaf entry procedure please tell me:

 

[maxTRX]

2 hours ago, Akiazusa said:

If anyone succeeded by using the MSRM falling leaf entry procedure please tell me:

 

No...

Some 'historian' here should probably explain the timeline and the progress from the the early days of the Hornet through early 2000's.

[Akiazusa]

In addition to the control surface effectiveness discussed earlier, the yaw static stability, also known as the Cnβ parameter, likely plays a significant role in the insufficient sideslip excursion experienced in the track file.

As Cnβ is expected to become negative above 30 degrees AOA, the aircraft would naturally tend to diverge from its original zero-sideslip state once sideslip is initiated. This characteristic should be contributing to the difficulty in establishing and maintaining sideslip during maneuvers at high AOA when MSRM is engaged, that all the feedbacks are removed from control loop.

 

Still from NASA:

 

Another track of attempting to generate large sideslip, but any sideslip excursions will soon reduce to zero, as if it's statically stable:

F18 MSRM sideslip generation prolonged.trk

[Figaro9]

3 hours ago, Akiazusa said:

If anyone succeeded by using the MSRM falling leaf entry procedure please tell me:

 

yes, think it works.

f18, centerline bag, 50% initial fuel

1) stabalize at 150kts, 35kft
2) select spin recovery switch. since the leading edge flaps drives to 34° and the trailing edge to 0°, nose drops...
3) power idle, slowly full after stick
4) smoothly apply full rudder
5) slowly increase opposite power lever to mil , other engine idle
when yaw rate is achieved
6) both engine idle
7) spin recovery swith to norm

Recovery need some adjustments to the natops. You need mil power on the engine against  spin direction.

msrm spin recovery.trk

[Akiazusa]

11 minutes ago, Figaro9 said:

yes, think it works.

f18, centerline bag, 50% initial fuel

1) stabalize at 150kts, 35kft
2) select spin recovery switch. since the leading edge flaps drives to 34° and the trailing edge to 0°, nose drops...
3) power idle, slowly full after stick
4) smoothly apply full rudder
5) slowly increase opposite power lever to mil , other engine idle
when yaw rate is achieved
6) both engine idle
7) spin recovery swith to norm

Recovery need some adjustments to the natops. You need mil power on the engine against  spin direction.

msrm spin recovery.trkFetching info...

Hello,

I was talking about intentional falling leaf entry. A falling leaf is not the same as a spin. The entry procedure is described in the tennessee paper:

MSRM Falling Leaf Entry:
1. Stabilize at wings level, 0.5 IMN /40K (145 KCAS).
2. Select spin recovery switch to RCVY to enter MSRM and reduce power to IDLE.
3. Establish dive angle of approximately 20 degrees.
4. Smoothly apply full aft stick and hold.
5. Apply lateral stick to generate sideslip (some side-to-side cycling may be necessary to generate largest sideslip).
6. Neutralize controls. Falling leaf motion is characterized by large sustained oscillatory yaw rate motion with approximate 5 second period. Allow motion to persist for two cycles or until 25,000 feet altitude.
7. Recover by selecting spin recovery switch to NORM and neutralizing controls. Falling leaf motion should be damped promptly after CAS is enabled due to the sideslip rate feedback and other control law features incorporated in the upgraded flight control system.

See if you can generate enough sideslip to initiate an in-phase yaw and roll oscillation, just like the one in the video.

Edited April 16, 2024 by Akiazusa

[Akiazusa]

44 minutes ago, Figaro9 said:

Recovery need some adjustments to the natops. You need mil power on the engine against  spin direction.

If talking about spin recovery, this would be another issue. Acccording to yet another reference document https://trace.tennessee.edu/utk_gradthes/2312/ , the upgraded flight computer (which we have by the hornet version in DCS), is capable of killing yaw motion, even a slow rate spin. Since the computer is actually using the above mentioned adverse yaw, to actively drive the aileron and the differential stabs to oppose the yaw rate. If there weren't any adverse yaw available, I can't imagine how this mechanic can work.

But anyways, this thread is about pure aerodynamics, so I chose to activate MSRM to bypass the computer.

[Akiazusa]

Hi, I've updated OP to reflect the data checking done for Cl-δA and Cn-δA, and listed the remaining issues that could contribute to each other.

[DummyCatz]

To add to OP's argument, the primary yaw device at high AOA should be the differential stabilators, instead of rudders.

Ref page 17 of https://trace.tennessee.edu/utk_gradthes/2372/

Quote

Differential Stabilator (Diff-stab) was used as a primary yaw device instead of a rolling device at elevated AOA (above 30°): This is because a significant amount of adverse yaw is generated by differential deflection of stabilators at greater than 20º AOA.
The stabilator that is deflected trailing edge down (TED) creates very high induced drag on that side, producing yaw opposite the commanded roll. The Legacy Hornet’s old v10.5.1 FCC software used considerable rudder (up to full 30º deflection) to coordinate even small lateral stick inputs due to this diff-stab induced adverse yaw. The old software therefore drastically limited roll command gains (diff-stab + aileron) when the rudders were saturated, resulting in sluggish roll performance when greater than 30° AOA.

I've already checked the trackfile and didn't notice the nose move in the yaw direction either, so I suggest to do another checking for the differential stabilators. @BIGNEWY @cofcorpse

 

Regarding how effective the anti-spin control is that utilizes adverse yaw created by both the aileron and the diff-stabs, here's a quote from page 22 of https://trace.tennessee.edu/utk_gradthes/2312/

Quote

While the pirouette inputs were held, the nose tended to oscillate between 15 to 50 degrees nose low while angle of attack fluctuated around 60 degrees ± 10 degrees. Once the inputs were removed after 1.5 turns (540 degrees), spin arrows appeared a half turn (180 degrees) later. Spin arrows were displayed long enough to accomplish the training objectives of executing spin recovery procedures. When anti-spin control input was applied into the direction of the spin arrows, yaw rate seemed to cease almost immediately - within 30 degrees of turn. Removal of the spin arrows seemed to coincide with the ceasing of the yaw rate. The test pilot commented that the removal of the spin arrows seemed much quicker than with v10.5.1.

I'm yet to see the anti-spin controls being effective in the track provided by @Figaro9, despite it being held for a total of 3 consecutive spin cycles, and finally has to resolve to splitting throttles. This may explain the lack of adverse yaw.

Edited April 28, 2024 by DummyCatz

[DummyCatz]

I'm tracking the progress of this issue that is related to the effectiveness of anti-spin controls. Seeing that in the changelog of DCS 2.9.6.57650:

• Fixed: Spin Recovery Mode is not working.

I retested all the above tracks. Unfortunately nothing has really changed w.r.t aerodynamics. The anti-spin effectiveness is still not close to what is described as 'When anti-spin control input was applied into the direction of the spin arrows, yaw rate seemed to cease almost immediately - within 30 degrees of turn'.

[DummyCatz]

Beside the above proof of lacking adverse yaw implementation, I'd like to point out other missing points as seen in the provided tracks:

 

1. Lack of wing rock at around 38-42 deg AOA, from page 16 of https://trace.tennessee.edu/utk_gradthes/2312/

This is commonly due to asymmetric flow separation and asymmetric vortex breakdown at those AOA ranges. Parameters such as Cn_beta-dyn should reflect this. Can be calculated from NASA wind tunnel data. FCC OFP v10.7 is able to eliminate the wing rock completely, but here in DCS we don't have it implemented. (See my other threads about FCS OFP v10.7)

2. Lack of falling leaf mode according to page 9 of https://trace.tennessee.edu/utk_gradthes/2372/

Both are about uncommanded oscillatory roll-yaw motions at high AOA, which are currently lacking in DCS. Again the v10.7 flight control is able to mitigate this, but not implemented in DCS.

I hope the currently ongoing FM review and rework take the above evidence into consideration.

Edited December 16, 2024 by DummyCatz