High Altitude Airport Operations
-
@BernieV said in High Altitude Airport Operations:
how broken the power modeling is in the JF Arrow. In the real world, power varies directly with RPM and TAS varies directly with RPM.
I'm not sure how deeply you have looked into the actual engines within the flight model, or within the sim in general, but we are seriously limited by what the sim provides us with. The engines are now (finally, after battling against the sim's internal editor messing up and even re-writing on occasion) laid out with the correct figures as far as possible, and then adjustments have been made to the areas which are 'flexible'. However, it is always going to be a 'best fit' to some extent.
You mention in another post about power being a result of the combination of RPM, manifold pressure, fuel flow and density altitude - the last of these is ENTIRELY worked out by the sim. There is nothing I can adjust which will affect this. Fuel flow is almost as bad - we have 'fuel_flow_scalar =1' and 'BestPowerSpecificFuelConsumption=0.49', nothing which will allow us to set fuel consumption at various points to specific levels. For all intents and purposes, you can forget these when trying to create an engine in the sim until you are almost done! The fuel flow is set as a result of the rest of the engine build, once that's all done you can play with those two scalars to try and hit as many points as possible but that literally is it. It gives you readings on a gauge but doesn't do any more.
We've just lost two of the four items which control power. That's down to the base sim's modelling of aircraft. If you take away two sides of a square, is it still a square? Can you make a square with what is left? This is a piece of core simulator code wich we are trying to manipulate so that it gives you a reasonable facsimile of what the real aircraft does. Even external engine builds are not perfect and look at the amount of time they take - you certainly would not have this aircraft in-sim yet if we were waiting for one of those.
I guess it's going to have to be a case of managing expectations - this is a new sim, there is reliance on it to a large extent and also a big lack of knowledge about it (the whole engine files are different to what went before, yet there is no supporting documentation. We're just left to guess). If you're flying the real thing every day, you're going to be able to pick holes in ANY sim aircraft. I don't think it's fair, though, to point the finger at the JF Arrow specifically when we're fighting against the simulator's own limitations and doing our best with limited resources to overcome them.
-
@Delta558 Understood. Don't read my comments as bitching about the model. I was more motivated to correct what I thought was inaccurate advice being given about how things work in the real world. After doing a few tests, I understand at least a bit of what the original poster was getting at (although most of it was very confusing) and its related to the power modeling breaking down at certain edge conditions. I've built a lot of models for computer capacity planning, enough to know this; Every model is wrong. Some are useful. Your Arrow is definitely useful and fun as well.
I think I also said recently that they way fuel flow works (and power modeling in general for the JF Arrow), it might be a compromise (in the way its modeled) so you can get what you need, but not necessarily what you want. And to restate my opinion of the model, its damn good. Good enough for me to practice instrument procedures without worrying about practicing unrealistic behavior due to working around model issues.
-
Thanks for the comments and clarification, Bernie - your input throughout has been very useful and a great help in refining the model.
My main concern, when I mention 'managing expectations', is for those who may be new to the sim and think we just dial in some figures and it works perfectly. Again, there may be less experienced / in-training pilots who decide to use this to practise and as long as they remain aware that we cannot guarantee absolute fidelity (much as we would like to!!), that there will always be compromises, then it's all good
-
@BernieV said in High Altitude Airport Operations:
Best power and best economy are mutually exclusive, so I am a little confused by your instructions in the last two sentences.
My use of the terms "Best Power" and "Best Economy" may have been confusing since they have specific meanings in the POH. I didn't mean to imply the same definitions, but only that the max amount of power being transferred to the prop can be seen by max fuel flow, and that peak EGT represents the most efficient fuel/air ratio and results in the least fuel usage.
-
I have a line in the engine.cfg that I've been messing with for awhile. It started pre v0.5 along with probably everything that has been discussed in other threads. Looks like all of which were included in the new flight model of v0.5
;engine_mechanical_efficiency_table=0.000000:0.770000, 700.000000:0.770000, 2000.000000:0.670000, 2200.000000:0.540000, 2700.000000:0.540000 engine_mechanical_efficiency_table=0.000000:0.770000, 700.000000:0.770000, 2000.000000:0.520000, 2200.000000:0.490000, 2500:0.52000, 2700.000000:0.540000By adding in an extra RPM point and tweaking values, I've gotten pretty close to POH values. And thanks to @BernieV chart up above, greatly reduced to power at 2000 RPM. I've never noticed that before lol Now what I'm not sure about on that band, is where to adjust it to as I have no reference. I'm thinking it needs to come down a bit more as I can match or exceed the speeds of 2200 RPM even though the MP is the same at max throttle.
I've also adjusted fuel_flow_scalar =1.1 ; 1 Now as @Delta558 has pointed out, theres not much you can do with this. Right now its about in a middle ground, little high here, little low there. If i increase a touch to better match the Power cruise settings, it would then be further off from the econ settings.
Here's I chart I made showing the performance. All testing was done at 5000ft, 10 OAT, clear skys, max weight.
-
@weptburrito said in High Altitude Airport Operations:
engine_mechanical_efficiency_table=0.000000:0.770000, 700.000000:0.770000, 2000.000000:0.520000, 2200.000000:0.490000, 2500:0.52000, 2700.000000:0.540000
Nice find. I assume the algo digesting that data set will do a ( linear?) interpolation between data points. Given that the lowest RPM setting the modeled prop governor will restrict to is 1700 RPM (and I have no data from real life to suggest it should do otherwise), perhaps adding values explicitly for 1700 rpm as an anchor to prevent TAS rising as RPM is reduced (via the prop lever) would be helpful. With your new values, the algo is likely interpolating between 2000 and 700 which might result in wonky behavior between 2000 and 1700 RPM (e.g. where 1700 RPM might give you better TAS than 2000). Just a thought after a few glasses of wine last night :)
-
So at the moment I have reduced the 2000 band down to .50. After I let the speed stabilize, reducing the RPM to 2000 I'm still holding the same speed. I'm wondering though if it should reduce some as I did not increase the MP? I then reduced the RPM more, down to as low as it would go, just a hair over 1600 and actually got a 1-2 kt increase? With a small increase in fuel.
I'm assuming that cant be right. I'm also assuming if I drop the 700 band down some, I would then see a drop in speed and FF as like you I'm assuming its linear from point to point. I'll probably do that in a minute just to see what happens. Problem is I'm making a lot of assumptions and just don't know what it SHOULD do. I don't have a chart for reference lol
-
Been tweaking some. This is where I'm at for now. Theory is as long as an increase in MP can be had depending on alt with a decrease in RPM then power/speed should be steady. Problem is I don't know how true this holds IRL at lower RPMs. Theoretical numbers where calculated by seeing around a -100:1 RPM to MP ratio in the 2500/2200 POH values.
engine_mechanical_efficiency_table=0.000000:0.770000, 700.000000:0.660000, 2000.000000:0.460000, 2200.000000:0.490000, 2500:0.52000, 2700.000000:0.540000
-
@BernieV Thanks and please do, would love to hear some feedback. Small warning, I've noticed that it has affected idle rpm, it dropped down around 500. But his can easily be corrected by adjusting as I did
idle_rpm_friction_scalar= 0.6 ; 1It can also be adjusted with idle_rpm_mechanical_efficiency_scalar, but unfortunately does anyone know what the difference between the 2 are lol
-
Be careful with trying to exactly match the POH Power Setting tables...
https://community.justflight.com/topic/2217/poh-power-setting-table-errors
