Posts by Jet-Pack (IPACS)

    The copilot is controlled via the overlay but I don't think you can see that in vr. We haven't had the time to add vr support for that.

    If you turn the copilot on it will maintain your present bank angle or if you are wings level it will keep that track. And it maintains the flight path angle, if you are almost maintaining altitude it will do that instead. And it holds the speed.

    If you are in an upset attitude the copilot will recover the aircraft for you.

    And with your duchess i am guessing you mean your installed just flight duchess and not a real world aircraft you could own?

    I think it's better to direct problems with add on aircraft to the developers of that aircraft. Here you probably won't see any responses from them and not from us either, since it's not our product.

    Hi Steve,

    We didn't change the radar altimeter, maybe you swapped the pivots from pilot and copilot?

    A button that is depressed and stays in would be called a switch in aerofly. Us an "input_switch" for the button position.

    To reset the switch you add an event_edge with a threshold of 0.3 and an input of your engine NH. Make it rising only and as Events add your Switch.Reset.

    I'm not sure it is possible to start the fms in the off position.

    As I said many times it is our intention to keep the default tmd file aircraft state as "running" because a lot of things depend on this. And if we add the option to load the aircraft cold and dark then it won't be via default tmd, there will most likely be a different configuration, like the clean, takeoff and landing configs.

    HI Jay, welcome back I guess :)

    Profiles are one possible solution but there could still be an even better solution that just nobody has thought of yet. That is why I keep asking what you want to assign specifically and what things are actually different between aircraft. Because as I see it only the throttle quadrants have been mentioned so far and a few custom home cockpits that have a lot of controls that could be used differently for different aircraft. But I'm not yet hearing anyone that says: no I want the landing light switch in aircraft A to be on my switch 1 but it in aircraft B it should be on switch 2. Nobody actually likes assigning all these buttons and switches and we all tend to assign the same stuff on the same switches to not get confused with what they do every time.

    Just some ideas that I want to throw in here which would mean less menu clicky thingy and more powerful and intelligent assignments.

    Maybe we can detect that you have a throttle quadrant connected and then prioritize which controls you have on there. Then all you need to do is connect it via usb and depending on the aircraft the assignments switch automatically - no configuration needed and you desired controls (e.g. left+right throttle for 737 and throttle and mixture for cessna) will be assigned for you. Just as an idea: we could also map the controls from left to right like in the virtual cockpit, so if you got 6 levers you will automatically have the exact same setup like in the virtual cockpit.

    I could also imagine a 3D menu for the assignments: click the thing in the virtual cockpit, move the hardware switch or lever to assign the control, click the next one and proceed until you customized your home cockpit. The downside is that that way still needs options to transfer a setup to multiple aircraft. Otherwise that would be a pretty neat solution.

    I'm pretty sure something in the 3D model can explain the strange behavior with the wheels, since all other 3d models seem to be fine to land on. Like a double sided face or a duplicated carrier deck geometry or sth. like that. It could also be an issue with the decals, which seem to have a collision as well, though they really shouldn't.

    Are you sure that the model was actually created by whitav8 ? Maybe he used a stock model from somewhere which you could do, too?

    The rotation speed of the power turbine is not actually defined within the engine, it is defined by the rotation speed of the propeller or rotor and the gear box ratio unless you actually disconnect the physical connection with a clutch or free wheel (drivetrain_freewheel) which you probably don't have yet.

    If your propeller is actually stopped then the power turbine is also stopped. Maybe you should add a way to visualize the propeller rotation, e.g. hook up the rotor geometry to the propeller rigidbody. If they don't spin then the shaft that the turbine sits on also doesn't spin.

    Check that you have an InitialVelocity for the joint that hold the propeller in place (PropellerJoint). And also check that there is no additional friction that would slow down the shaft.

    Hi Steve,

    the electric generators can and should be hooked up to the engine accessory gear like in reality (see C90 and Q400 for examples) but the hydraulic system component are not quite there yet, so I'd skip this for now.

    You're welcome :)

    Hi Steve,

    Fuel Controller

    here are all properties that are available currently:

    The FuelFlowLowIdle, Maximum control the fuel flow range (in kg/s) which are basically the hard limits for your low ground idle (where the engine can run without assistance and your full power (which melts your engine).

    PFuelFlowNH is a constant that defines how much fuel is to be added per target NH increase, that is roughly: (fuel_max - fuel_idle) / (nh_max - nh_idle) - more means the engine will add a lot of fuel to accelerate up to target nh and less means not burning up in flames.

    PFuelFlowThrottle is a feed forward control from the InputThrottle, that's like how much fuel goes in - can be tuned via throttle map

    FuelFlowLightUpFlat controls how much fuel is injected for the first part of the engine start - that fuel should be enough to stabilize the flames but shouldn't flood the engine at such low speeds - NHLightUpFlatEnd is defines the NH at which the controller switches to NH governing mode. The NH start is when you manually introduce fuel but you could just use a logic_greater NH > some value to change your condition lever input.

    NHTarget is the target rotation speed at the start of the simulation, needed because internally the FADEC ramps the target up and down slowly.

    NHRating is the maximum takeoff power core rotation speed, which could be 1.07 or 107% or 1.04 (104% NH)

    ThrottleMap is a mapping from your throttle input to a fraction between NHIdle and NHRating, for input 0.5 that mapping above returns the half way point between idle and rating. So basically nh_target = nh_idle * ( 1.0 - mapping_value ) + nh_rating * mapping_value. This is very useful because the actual thrust or power curve of the engine tends to be non-linear, this would be a way to counter that and could artificially increase the target NH early so that you have 50% thrust or power for 50% throttle input. Normally this would not be the case because the engine components get more efficient for higher power settings (towards their design power setting).

    The condition lever has a cut off range from 0.0 to about 0.1 I think and above that it goes from low idle target to high idle target (which you can adjust if you need) - you can think of that as your engine start/stop switch if your helicopter has that, if not - it probably has a cut off switch, then use that.

    ITT

    Yes the ITT rises nearly instantly, this is realistic in a sense that the actual ITT follows very close to the fuel flow, if you burn more fuel you heat up the gases very quickly. What you have experienced in the real world is the time delay caused by the heat transfer within the ITT probe. You can't directly measure the temperatures, your probes would melt, so you use other methods with metal parts that don't melt, but those need to get hot first before you can see an ITT increase.

    If you want to see your ITT increase slower you can use a first order delay:

    Code
                <[first_order_low_pass][EngineTOT][]
                    <[string8][Input][EngineStation45.OutputTotalTemperature]>
                    <[float64][TimeConstant][3.0]>
                    <[string8][Value][755.8]>
                >

    This simulates the probe heating up and your indicated turbine temperature becomes slower.


    Generally speaking this is how our engine stimulation works:

    Each turbo component simulates the thermodynamic process of that part with compressor maps, turbine maps, efficiency coefficients, etc. etc. These parts together create what we call an engine. Currently the only place we can adjust is the fuel flow, but we might add stuff like reverser flaps, mach inlets, variable stator vanes, bleed air extraction, etc. And we have connections to the outside using gear boxes to the core shaft and also the power turbine shaft. Those connections are necessary for the generators and starters.

    The fuel flow that we add to the combustion chamber is in kg/s and it can either be controlled fully manually if you create some form of mapping from the throttle lever or it can be fed through the simple fuel_controller, which is pretty stupid on its own, or it can be generated by the FADEC which can have more complex, fully digital control loops.

    So far we have only implemented a few FADECs specifically for that particular engine (see q400) but I think when we add more FADECs for the turbofans for example, we might create a more generic FADEC which can handle a certain engine category (e.g. twin spool high bypass turbofans or an afterburning jet fighter engine or maybe even a helicopter one with rotor speed governing, etc.)

    What I love about this

    The problems that you will encounter with controlling the engine are actually really similar to real world problems. It helps me better understand the physics behind real world engines.