After some more investigations[1] I think there are a number of different issues all coming together.
The main manifestation of the problem is that it can't perform engine out taxiing; there are other manifestations but these are less definitively wrong (as I'm fairly sure that an aircraft that can't perform single engine taxiing would never make it off the drawing board, or pass flight testing, or gain approval).
The way I'm testing is to load the aircraft, use the config dialog for "ready for taxi", then shutoff an engine, gradually increase power and use the keyboard rudder controls to stay in a straight line. The result is a very excessive yaw moment about a point that seems to be aft of the CoG.
So to identify the problem I've:
1. Checked the geometry using
http://www.aircraft.airbus.com/fileadmi ... 20_WTF.zip [2] - maybe the nose gear should be a bit further back, but basically the geometry seems ok.
2. Checked the nose and main gear configurations. Apart from the nose wheel having no brake group it looks ok.
3. Checked the moments of inertia, against p44 of
https://www.diva-portal.org/smash/get/d ... TEXT01.pdf - revised based on these figures.
4. Added a flight control system for nose wheel steering with rudder
5. Removed all contact points to ensure that it's not an errant ground strike (it isn't)
6. Remove all of the pointmasses (by setting to minimum values).
The main problem is (4) and (6). If I set the masses to the minimum values (especially those aft of the CG), and the revised NWS I can control in a reasonably straight line up until 120kts (when I run out of tarmac).
The default steering of +/- 70 degrees means 3.5 degrees per keypress; this isn't enough to control with any precision, and more than 15 degrees tends to lead to skidding at anything more than 10kts.
Inspecting the nosewheel compression (with the default mass configuration) the compression is lower than I'd expect[3] - I now think this is the main problem - but the NWS is aggravating the situation, and possibly there is a pendulum affect of the rear masses, however this isn't as easy to determine.
I can either set PAX C to the minimum, or configure as in the picture and (with the NWS change) accelerate contrallably.
The NWS is pretty much out of the F-14, but it's near enough for test purposes.
- Code: Select all
<flight_control name="FCS">
<channel name="Nose Wheel Steering">
<switch name="position/aircraft-on-ground">
<description>Aircraft on ground</description>
<default value="0"/>
<test value="1" logic="OR">
gear/unit[0]/WOW ne 0
gear/unit[1]/WOW ne 0
gear/unit[2]/WOW ne 0
</test>
</switch>
<switch name="systems/NWS/engaged">
<default value="0"/>
<test value="1">
gear/unit[0]/wheel-speed-fps le 135
<!--systems/hydraulics/combined-system-pressure ne 0-->
position/aircraft-on-ground ne 0
<!--systems/holdback/launchbar-engaged eq 0-->
</test>
<output>systems/NWS/engaged</output>
</switch>
<scheduled_gain name="fcs/scheduled-steer-pos-deg">
<input>fcs/steer-cmd-norm</input>
<table>
<independentVar lookup="row">gear/unit[0]/wheel-speed-fps</independentVar>
<independentVar lookup="column">systems/NWS/engaged</independentVar>
<tableData>
0 1
0 0 87
80 0 15
135 0 0
</tableData>
</table>
<output>fcs/steer-pos-deg</output>
</scheduled_gain>
</channel>
</flight_control>
-------------
[1] already been trying to help Josh with this.
[2] Airbus Drawings:
http://www.aircraft.airbus.com/support- ... -drawings/[3] Josh mentioned earlier that it might be nose light.