Aircraft does not balloon in ground effect

[onox]

Some months ago I was trying to fly and land without a yoke (so only trim wheel, rudder, and throttle) using this video. The aircraft in FG does not really strongly balloon in ground effect like in the video (see the end of the video when they land).

@Juanvvc @dany93 @tigert Don't know how much it depends on the atmospheric conditions, but should the effect be increased a bit?

[gilbertohasnofb]

@onox I had seen this video before and I was planning to try the same :smile:

[onox]

Flying the approach is very nice, but I just always end up with 1 or 2 broken gears 😅

[dany93]

I tried it. Amusing. With no training, there is a large risk of touching with the nose wheel first, which is of course very bad. And if you succeed at first touch, the rebound gladly does the rest. Nevertheless, the pitch can be controlled easily enough with the elevator trim. I had a very nice soft touch like that. Hitting 1 and 7 on my keyboard to avoid touching my JS :smile:

should the effect be increased a bit?

Easy to change for testing. I presume you know that the effect is already in the FDM (lines about 1047 - 1070). Current value: 1.20 at max (+20%).

The problem is which value to take. I found very few publications giving tables or values. e.g. Ground Effect Aerodynamics by Erjie Cui and Xin Zhang (Fig.4). Also (NASA) FLIGHT EVALUATION OF GROUND EFFECT ON SEVERAL LOW-ASPECT-RATIO AIRPLANES It obviously depends on the aircraft configuration, but for our c172 which has no wing endplates, +20% seems not so bad. Increasing it makes soft landing easier, more pleasant, but I have no personal feeling scale to judge it.

If you want to try, I have added a factor. It makes it easy and quick to change the value for doing tests.

        <function name="aero/function/kCLge">
            <description>Change_in_lift_due_to_ground_effect</description>
            <product>
                <table>
                    <independentVar>aero/h_b-mac-ft</independentVar>
                    <tableData>
                          0.0000	1.2030
                          0.1000	1.1270
                          0.1500	1.0900
                          0.2000	1.0730
                          0.3000	1.0460
                          0.4000	1.0550
                          0.5000	1.0190
                          0.6000	1.0130
                          0.7000	1.0080
                          0.8000	1.0060
                          0.9000	1.0030
                          1.0000	1.0020
                          1.1000	1.0000
                      </tableData>
                </table>
                <value>1.10</value> <!-- 1.10 dany for test (1 gives the initial FDM) -->
            </product>
        </function>

[onox]

Shouldn't the last value (second column):

1.1000	1.0000

remain 1.0 (after multiplication)? Now it becomes 1.10 (due to the factor).

[dany93]

Yes, you are right. It was too easy. We need to recalculate all the table.

[wlbragg]

Do we have a set of keyboard commands for trim, may be nice to have to practice the training of landing without use of yoke.

[wlbragg]

When starting on final using the Internal Launcher the aircraft starts with engine off. Have we missed something? I used start with engine running and at least it starts the engine but with no forward momentum, thus you start to stall. I tried this yoke-less landing several times and I get the same results, broken gear. But I am not convinced that with trim applied at the right time you could successfully land without breaking the gear. The closest I came was similar to the described effect of bouncing 3 times, except my first bounce broke the front gear.

[dg-505]

Do we have a set of keyboard commands for trim

There's Home/Numpad 7 for nose down, and End/Numpad 1 for nose up.

Anyhow, for the Twin Otter I added F5 and F6 for aileron trim left/right, and F7 resp. F8 for rudder trim left/right

[wkitty42]

On 12/01/2016 02:40 PM, wlbragg wrote:

When starting on final using the Internal Launcher the aircraft starts with engine off. Have we missed something? I used start with engine running and at least it starts the engine but with no forward momentum, thus you start to stall.

this is what the new startup states the sim devs are working on are for... in our case, if one choses to start in air, the startup state xml or section in the xml would init everything as needed and have JSBSim up and operational so that there is proper forward motion with the desired altitude, heading and slope angles needed... this is something that has just been barely touched on a time or two in the past but "we", the c172p team, have not undertaken as it is so new and not fully understood yet...

-- NOTE: No off-list assistance is given without prior approval. Please keep mailing list traffic on the list unless private contact is specifically requested and granted.

[wlbragg]

OK, I guess I didn't realize the "on approach" startup was broken. See further comments on State support in the C172P #856.

[dany93]

I tried again several times the "without the yoke" landing (and turns with rudder only). I succeeded about 1 out of 2 or 2 out of 3 times. :grimacing: Full flaps (or 20 deg like in the video). Throttle at idle preferably, but the throttle can be used to control the descent rate if we are short.

Throttle at idle, 1000 to 1500 ft height, starting from elevator trim = 0 gives about 85 kt descent velocity. Using 1 and 7 numeric pad keys (rather 1 for trim up), I progressively decrease my velocity to 70 ... 60 knots during the flare. Keeping in mind that at too low speed, the elevator will has no efficiency left. By hitting, rather holding the 1 key (elevator trim -0.4 ... -0.7), I control my velocity and flare. Avoid too much or too lately trim up with still too much velocity because the aircraft will go up and stall. The important (and the most difficult) is to anticipate and do it progressively. At very low speed and height, the trim up also makes the nose wheel go up.

[EDIT] My best at this time (twice in a row): Descent: flaps 20 deg, engine at idle, elevator trim -0.2 ==> 65 kts. Flare and slow down: Elevator trim progressively for nose upwards, to -0.4 or -0.45 ==> touching at about 52 kts, nose wheel slightly high. Main wheels touch first.

[dany93]

I have made a spreadsheet to calculate the Change_in_lift_due_to_ground_effect table. Any values (by +10%, +15% increase from the current table at 0 ft height) can be quickly applied. If someone is curious to try...

I can upload the spreadsheet and give a link, or give one or two calculated tables in this thread. Putting the numbers in a good shape is longer than calculating the values. Although not mandatory for tests.

[onox]

You could put the new tables in a comment here.

[dany93]

Initial (current) values: (slightly corrected for 0.4000)

        <function name="aero/function/kCLge">
            <description>Change_in_lift_due_to_ground_effect</description>
            <product>
                <table>
                    <independentVar>aero/h_b-mac-ft</independentVar>
                    <tableData>
                        0.0000    1.2030
                        0.1000    1.1270
                        0.1500    1.0900
                        0.2000    1.0730
                        0.3000    1.0460
                        0.4000    1.0250
                        0.5000    1.0190
                        0.6000    1.0130
                        0.7000    1.0080
                        0.8000    1.0060
                        0.9000    1.0030
                        1.0000    1.0020
                        1.1000    1.0000
                    </tableData>
                </table>
            </product>
        </function>

Max factor (for h_b-mac-ft = 0): 1.1

                    <tableData>
                        0.0000    1.3233
                        0.1000    1.2295
                        0.1500    1.1841
                        0.2000    1.1608
                        0.3000    1.1221
                        0.4000    1.0902
                        0.5000    1.0746
                        0.6000    1.0590
                        0.7000    1.0447
                        0.8000    1.0334
                        0.9000    1.0212
                        1.0000    1.0111
                        1.1000    1.0000
                    </tableData>

Max factor (for h_b-mac-ft = 0): 1.15

                    <tableData>
                        0.0000    1.3835
                        0.1000    1.2807
                        0.1500    1.2312
                        0.2000    1.2047
                        0.3000    1.1601
                        0.4000    1.1228
                        0.5000    1.1024
                        0.6000    1.0821
                        0.7000    1.0630
                        0.8000    1.0472
                        0.9000    1.0304
                        1.0000    1.0157
                        1.1000    1.0000
                    </tableData>

Max factor (for h_b-mac-ft = 0): 1.2

                    <tableData>
                        0.0000    1.4436
                        0.1000    1.3319
                        0.1500    1.2783
                        0.2000    1.2486
                        0.3000    1.1981
                        0.4000    1.1555
                        0.5000    1.1302
                        0.6000    1.1051
                        0.7000    1.0813
                        0.8000    1.0609
                        0.9000    1.0395
                        1.0000    1.0202
                        1.1000    1.0000
                    </tableData>

Max factor (for h_b-mac-ft = 0): 1.3

                    <tableData>
                        0.0000    1.5639
                        0.1000    1.4344
                        0.1500    1.3724
                        0.2000    1.3364
                        0.3000    1.2742
                        0.4000    1.2207
                        0.5000    1.1857
                        0.6000    1.1511
                        0.7000    1.1180
                        0.8000    1.0883
                        0.9000    1.0577
                        1.0000    1.0293
                        1.1000    1.0000
                    </tableData>

[EDIT] NB: there is some discrepancy between

• the parameters used in JSBSim Aero input and outputs Altitude of MAC (Mean aerodynamic chord) divided by wing span: /fdm/jsbsim/aero/h_b-mac-ft
• and Ground Effect Aerodynamics by Erjie Cui and Xin Zhang where in h/c c is wing chord).

I don't understand, but I prefer the JSBSim point of view.

[onox]

Cool, I'll try some of those tables.

Throttle at idle, 1000 to 1500 ft height, starting from elevator trim = 0 gives about 85 kt descent velocity.

I think the video mentioned 65 kts as the approach speed?

[dany93]

~~I don't know for the video.~~ Right, 65 kts. (difficult for me to understand what is said) POH says Airspeed -- 60-70 KIAS (flaps DOWN). This value of 85 kts that I gave was just an indication as a start at descent with default elevator trim.

[tigert]

Note that they came in fast, and thats why it floated excessively. Try approaching at 70 ot 75 kts and see how it goes. This (with yoke operational) is the most usual way to break a Skyhawk in landing whrn you bounce and instinctively push yoke when it bounces, resulting in a collapsed gear. So it is not totally unrealistic. I can try to fly it tonight and see how it feels with current fdm.

Also note that the weight plays a big factor on the 172, if you fly solo with half tanks vs crossweight..

[tigert]

65 is quite ideal flare speed, then it floats a lot less.

[tigert]

Also check the short field technique on the POH for the no float procedure