TAS/GS weirdness
-
As far as I can tell, the new version doesn't change any of the reported behavior. The plane can't make book speeds, fuel flow looks to be artificially constrained, parasitic drag seems to be too high.
Summary: The plane is slower than real life by roughly 7 knots. I think it might be a combination of the way power is modeled and parasitic drag being too high. This is somewhat complicated by the fact that the Indicated airspeed (on the panel instrument) seems off (high by about 6 knots).
#1 : Airspeed indicator is indicating too high by 6 knots. Setup: 1356 ft msl, std pressure & temp, no wind. Power setting 25 squared mixture 100 ROP. From Little nav map, IAS is 119 kts. TAS=GS=121 kts according to EFB, GPS, & Little Nav Map. Panel airspeed indicator shows 125 knots. Conclusion: Airspeed indicator is indicating too high by 6 knots or so. This needs to be considered when doing further analysis on why the plane can’t make book speed numbers. Another image (next page) captured at 6,400 ft the panel air speed indicator showed 121 kts and little nav map showed 115 kts indicated implying that the error seems constant with respect to altitude.
#2 TAS Airspeed and fuel flow don’t jive with altitude throttle, prop RPM, and mixture controls. At 6500 ft, std temp & pressure no wind, Little Nav map reports airspeeds (knots); indicated: 115, ground 126, true 126. Fuel flow is too low (I varied mixture between where it is shown and full and it had no effect on Fuel flow or airspeed, 12.7 g/h is what the book say we want to get 75% power and I cant get that according to the fuel flow indicator). In the real world at 6500 feet std temp & pressure, a wide-open throttle, 2500 RPM, and 100 rich of peak EGT should give me 75% power and a TAS around 135. GS should be 135 as well as no wind condition & std temp/pressure.
#3 Parasitic drag seems too high. In an attempt to test drag independent of power I looked at change in speed vs descent rate. Level flight GS 127 kts. In real life, I start a decent from level flight 6500 ft at 75% power by trimming nose down for 500 ft/min. It takes 30 seconds or so to stabilize but this fairly soon results in +15-17 Knots/hr ground speed increase and indicated airspeeds just inside the yellow arch of the airspeed indicator. In the JF Arrow III 2.0 model, under these conditions I see a 10 knot GS increase about 6 knots less than expected.
If I continue to increase descent rate by trimming to 1000 ft/min descent I finally see +19kts, close to what I would expect with a 500 ft/min descent rate.
-
I agree, did very similar tests (your #1 and #2) and ended up with the same results as you did.
The issue is still not fixed, the aircraft is slower than the ODM states and in clear skies preset can't even reach the 75% power cruise setup at all, even with 50% fuel and 1 pax while the book states the figures for max GW.As for the IAS gauge, I think mine is rather 4kts too high, not 6, but that's a minor.
Also if you read the TAS from the EFB or LittleNavMap and dial the TAS in the IAS gauge (lower white scale) to see which setup for altitude and temp you should have had (in the white upper scale of the IAS gauge), it is WAY off, for me about 4000ft on that scale.
I opened a ticket (#50575) for this TAS scale misalignment as well since it is obviously wrong. -
I've found a work around if not the root cause of slower than expected TAS and lower than expected fuel flow. Basically the aircraft was unable to generate 75% power at altitude of 6-7000 ft due to inadequate fuel flow.
In engines.cfg, fuel_flow_scalar was set to .51. Through trial and error, I found this decreases max fuel rate obtainable by varying the mixture control as altitude increases. So much so that I could not get the 15 or 16 g/h I would expect in a full rich climb nor the 12.1 g/h required for 75% power at altitudes between 5 and 7,000 ft.
I set the fuel flow scalar to 1 and was see very high fuel flow rates (22+ gph) which is way to high. I suspect that's why the engine tuner set the scalar so low. Too low IMHO. I increased it to 0.73; the point were I could get 15ish g/h at 7000 feet full rich. leaning to 12 gph improved the TAS, but it was still too low 133ish.
Next through trial and error, I settled on parasite_drag_scalar set to 0.825. This allowed me to get book speeds at 6 or 7000 feet.
The airspeed indicator is still off, but this IMHO goes a long way towards allowing the modeled Arrow III to perform closer to real life.
Summary of changes:
In flight_model.cfg
parasite_drag_scalar =0.825
In engines.cfg
fuel_flow_scalar =0.73
Inflight stats with the new settings
And yes, I did update the related support tickets as well :)
-
@BernieV said in TAS/GS weirdness:
This allowed me to get book speeds at 6 or 7000 feet.
Wow thanks for the efforts!
Do these changes impact the climb performance? did you compare that to the book numbers?
Also when getting the book speeds at 7000ft, did you lean it also according to the book? 100°C rich of peak? Or do these changes also modify the position of the mixture to gain this fuel flow?Will give it a try later as well :) It's not the fastest aircraft out there, but having it getting at least the book speeds would be so nice.
Unfortunately I have no idea about most of the flight model configs (opposed to lights) so whatever I'd change here, it would just cause something else to break for sure. :DAs for the IAS gauge, I am not sure where this offset could be fixed. For the time being one could simply rotate the texture of the gauge, but it'd look weird.
The wrong TAS vs. altitude scale can be fixed by turning the altitude texture 4000ft to the left (leaving the TAS scale at the old position) at least.
Whoever does that, remember to also rotate the luminance texture since with panel lights on, they get overlayed. -
Do these changes impact the climb performance? did you compare that to the book numbers?
Climb performance continues to be better than real life. At 25 squared full rich in real life I see 700 ft/m climb initially (from sea level) dampening to around 500 ft/m between 5,000 and 7500 ft. The model happily climbs at 800-1000 ft/min with 25 squared throttle/rpm and full rich. I don't think I made that worse, but I would like to figure out how to slow the climb without impacting cruise speeds. I'll do some reading. Others may find this setting fine. I suspect in real life this is very aircraft dependent and varies with engine health, prop condition, rigging, paint, wax, etc.
Also when getting the book speeds at 7000ft, did you lean it also according to the book? 100°C rich of peak? Or do these changes also modify the position of the mixture to gain this fuel flow?
The book gives climb performance at 2700 rpm. Thats nuts. Its redline speed rpm. Perhaps mechanics and marketeers will point out that in theory you can run that way all day, my instructors and all the Arrow owners I know set climb power at 25 squared MP/RPM full rich (below 5,000 ft).
So I set it like I do in real life. 25 inch MP 2500 rpm full rich in the climb to 6,500. Fuel flows looked right with these settings (15-16 gph prior to leaning at altitude).
I initially set fuel flow in cruise by looking at fuel flow reported by Little nav map (I set it to 12 gph based on book 75% power chart. The EGT read WAY low at 12 gph. I did try adjusting the fuel flow using mixture control and looking at egt, but at 100F ROP, I was seeing north of 15 gph which is way too high. So, not sure if the EGT gauge is being modeled correctly. in the end it did not seem to matter much if I used fuel flow as reported by the FF gauge or using the EGT and setting 100F ROP, the cruise speed was still in the neighborhood of 137 kts at 6500 ft 2500 rpm.
As an aside, I tend to cruise at 2400 rpm or as low as 2100 rpm if not in a hurry. 2500 rpm is loud and produces a lot of vibration that over a 1-2 hour trip wears on you.
Will give it a try later as well :) It's not the fastest aircraft out there, but having it getting at least the book speeds would be so nice.
Agreed! You don't buy an Arrow to go fast :) Its a wonderfully stable IFR platform and with the gear up trips of 200 miles or so are not too painful. The one thing that was missing and motivated me to get the speed up a tad is that if you don't plan descents well, you will still end up too high or fast for precision instrument approaches. I think Its to the point now that it behaves pretty close to the real deal. I still want to slow the climb a bit without effecting cruise speeds. I'll look at that over the next few days.
-
@BernieV Another data point that the drag in the climb is too low is that I was able to climb to 20,000 ft in the 0.2.0 model with the tweaks I made to parasitic drag and fuel flow (albeit at less than gross weight). I barely made it to 13,000 on a hot day between Napa CA and Yellowstone in real life and book shows 15K as the service ceiling. So now I am looking for a way to slow the climb and reign in the service ceiling (hopefully with a single parameter!)
-
I appreciate all the research and testing you're doing, Bernie. I've noticed the Archer is running slower than I thought it should be -- a friend suggested maybe it was because they modeled the 3-blade prop instead of the 2-blade prop. I don't have the real world experience on the Archer to really add to or comment on it beyond that.
-
@piedmonitor Whatever they modeled, we all agree that the numbers should match to the ODM that is included in the addon, don't we?
I also doubt that the 3-blade version isn't capable of a 75% power cruise no matter what the TAS is that results for that version. So there is definitely some issue with the flight model and I hope that someone from JF Staff is reading these findings and the devs are looking into it. -
Thanks for these tweaks. I noticed similar discrepancies with Little Navmap.
On the 25/25 rule, not everyone agrees. Came across this recently. See in particular John Collins’ comment and subsequent discussion.
https://www.pilotsofamerica.com/community/threads/the-25-25-myth.74115/
-
@j225 said in TAS/GS weirdness:
Thanks for these tweaks. I noticed similar discrepancies with Little Navmap.
On the 25/25 rule, not everyone agrees.
Agreed. I think that 25X25 is specified in my POH (for a '67 Arrow, not a III like this model, but pretty close in most aspects and both have 4 cylinder IO-360s engines) for climbs, but I get your point. Get a bunch of pilots together on a non-flying day, add a few adult beverages, and mention either lean-of-peak operations or he "right" combination of MP & RPM for a given engine, and watch the fur fly ;) I think the conversation you referenced managed to work both topics into the conversation!
I've put 1850 hours on my engine since rebuild and compressions are still good, so I'll stick with what got me this far. Fingers crossed and hoping to get to 2000 hours on this engine. The nice thing with a sim is you can do whatever you like and you wont have to pull out your wallet and pay for a top end overhaul or rebuild.
-
I've done one more round of testing and have have achieved a good balance of 75% power cruise TAS , climb performance, airspeed buildup in the descent, and service ceiling. They are as close to book at I can get for now.
In flight_model.config
parasite_drag_scalar =0.8
induced_drag_scalar =1.88In engines.config
fuel_flow_scalar =0.73
-
@BernieV , Thanks for doing this research, and the recommended settings for achieving more realistic performance. For reference, the Parasitic Drag of any aircraft (parasite_drag_scaler) affects both the cruise speed and climb performance, while the Induced Drag (induced_drag_scaler) affects mainly the climb performance. The speed and climb performance are both determined by the ratio of Total Thrust vs. Total Drag of the aircraft...
-
@RetiredMan93231 You are welcome RetiredMan. In practice, the induced drag scalar was the parameter that tuned rate of climb, the "slipperiness in the descent" (not a book value, but one I know from experience), and service ceiling. The final values posted gave me 137 kts level cruise at 75% power at 6500 ft (FT, 12 g/h, 2500 rpm as per book), an increase in TAS of 10 kts in a 500 ft descent at 75% power out of 6500 ft (from experience I expect 15 kts pickup, but I didn't want to detune ceiling or rate of climb for this subjective performance metric), and a service ceiling of 15000 ft (book).
-
@BernieV , In the real world Arrow III what is the expected RPM / power level on final at 75 knots with full flaps? It seems to me that the landing gear drag and the flaps drag are both currently a little too high, requiring too much power (RPM) to maintain the glideslope on final. What are your thoughts?
-
@RetiredMan93231 Between 15 and 17 inches of MP, prop level full forward, mixture full rich (unless operating at high elevations)
-
@set3times No problem. The last (3rd) notch of flaps is almost all drag. In real life, I delay using it until I am sure I'd make the airport if I lost power or unless I was high/fast on the approach and need to get down or slow down.
