Improved engine performance and flight dynamics

[derl30n]

Fixes #210
Fixes #359
Fixes #824
Fixes #850
Fixes #917
Fixes #974
Fixes #984
Fixes #1034
Fixes #1043
Fixes #1166
Fixes #1173
Fixes #1204
Fixes #1255
Fixes #1375
Fixes #1377
Fixes #1442 (VApp calculation)
(are all about climb rate/cruise attitude/not able to reach cruise/stalls/fuel consumption)

Summary of Changes

• Engine (redesigned performance table)
  • Taxiing thrust adjusted
    enables taxiing on idle thrust and also impacts aircraft performance in flight e.g. approach or decent
  • N1 on Approach will match real life on about 40 to 45%
  • Climb performance reworked
  • Cruise N1 adjusted
  • Fuel burn during cruise adjusted
• Aerodynamics
  • Flaps
    • Increased lift coef
      prevents early stalling of the aircraft on landing
      improves the overall feeling of the aircraft in flight
    • Decreased drag coef
      deceleration on approach now slower
    • Increased pitch momentum
      pitch now changes noticeable when flaps position changes
  • Speedbrakes / Spoiler
    • Decreased drag coef
  • Landing gear
    • Increased drag coef
  • Elevator
    • Increased momentum coef
      makes aircraft (more) responsive
• MCDU
  The following speeds are now calculated based on charts below
  • VLS
  • VApp
  • F (for approach)
  • S (for approach)
• PFD
  • F and S speeds for approach are shown based on charts below
  • The Alpha Max and Alpha Prot has been tweaked as best as possible but is not correct.
    This is an issue to be solved by the future flight model

This PR actually solves as a HOTFIX meanwhile development on the final engine model continues.

Screenshots (if necessary)

References

https://www.youtube.com/watch?v=cv41hm4JEB8&feature=youtu.be&t=100
https://www.youtube.com/watch?v=FODhSJucbK0
https://www.youtube.com/watch?v=rhI5XGwlnaI

Additional context

Discord username (if different from GitHub): DerL30N#3751

How to download the PR for QA

Every new commit to this PR will cause a new A32NX artifact to be created and uploaded.
The build.py will have already been run with the latest changes, so no need to rerun it once you download the zip.

1. Make sure you are signed in to GitHub
2. Click on the Checks tab on the PR
3. On the right side, slick on the Artifacts drop down and click the A32NX link

[donstim]

See my comments on an earlier version of the modification here: #1204 (comment)
Thank you for adding the change to the spoiler drag coefficient!

[joniatus]

This works beautifully as described. Thank you 1000x better

[derl30n]

I am currently working on an further improved engine parameter table.
I don't know how long this will take me tho.

[nobodyn0se]

I tested this PR, great work.
Note: I'm not with QA.

Date of testing: 13 Oct, 2020
Sim Version: 1.9.5

T/O weight: 67500kg
Departure: FACT
Arrival: FAPE

Low N1 performance is realistic. I taxied in idle thrust with no issues. Lift generated by flaps is much better.
Initial climb at 3500 fpm FLEX 65, acceleration was realistic. After climb thrust reached target speed of 250kts, the rate of climb was 2800 fpm to maintain 250 (the IRL video shows ~3200fpm), that continued till FL140 beyond which the climb rate slowed down to 1200-1500 fpm range.

Above FL280 though, the climb rate slowed down to 800fpm till FL310 where I initially leveled off to burn some fuel.
The aircraft struggled to climb above FL310, it did not go above 400fpm. I had to use TO/GA which sustained a 600fpm climb to FL330.
During cruise, the N1 thrust was at 91.5% for M0.78 with a 45kts tailwind. According to Airbus' cruise perf charts, at that weight it should be around 81% N1 for M0.78

Descent performance was stunning, had to use the spoilers to slow down sometimes. Approach was realistic with addition of flaps requiring pitch changes just like the IRL video.

Overview: Realistic acceleration/deceleration. Great low N1 performance. High N1 performance might have to be tweaked.

[derl30n]

please remove any changes you made to manifest.json and layout.json. The CI will automatically produce these

Will do.

[derl30n]

Low N1 performance is realistic. I taxied in idle thrust with no issues. Lift generated by flaps is much better.
Initial climb at 3500 fpm FLEX 65, acceleration was realistic. After climb thrust reached target speed of 250kts, the rate of climb was 2800 fpm to maintain 250 (the IRL video shows ~3200fpm), that continued till FL140 beyond which the climb rate slowed down to 1200-1500 fpm range.

Do you know the N1 you had during climb?

Above FL280 though, the climb rate slowed down to 800fpm till FL310 where I initially leveled off to burn some fuel.
The aircraft struggled to climb above FL310, it did not go above 400fpm. I had to use TO/GA which sustained a 600fpm climb to FL330.
During cruise, the N1 thrust was at 91.5% for M0.78 with a 45kts tailwind. According to Airbus' cruise perf charts, at that weight it should be around 81% N1 for M0.78

Will look into that as well. Could you please send (or link) me the cruise perf charts, that would be nice.

Descent performance was stunning, had to use the spoilers to slow down sometimes. Approach was realistic with addition of flaps requiring pitch changes just like the IRL video.

Overview: Realistic acceleration/deceleration. Great low N1 performance. High N1 performance might have to be tweaked.
I will do some work for the high N1 performance.

[nobodyn0se]

Do you know the N1 you had during climb?

@MisterChocker Climb N1 was 88.6%

Here's the cruise performance chart. Credits to user: donbikes who sent it, I hope it's fine to upload it here.

[derl30n]

@nobodyn0se big thanks. Hell yeah I like the imperial system for weights... still not using metric ffs.

[sphectrum]

Discord: sphectrum#6484
Date of testing: 14/10/2020
Version of the sim: 1.9.5.0
PR Tested: #1295

Testing technique: Gate start/cold and dark, full flight
Steps you've made to test the Pull Request: Every procedure from cold and dark to shutdown after arrival

Tested comparing to latest master during a full flight with a manually programmed MCDU.

Overall Rating: Good, needs some more tweaks
Conclusions of the testing: Works as described, but FF at CRZ (FL360) is around 2000l/h per engine, N1 of 82.2% seems ok as well.
Lacks performance above FL270, climb rate is below 500fpm, after FL320 even 200-300fpm.
Low N1 performance is great, which can be seen during taxiing on idle thrust and during descent.
Modified coefficient of drag for flaps, speed brake etc is much more realistic, as well as lift and stall speed improvements visible on approach and final. Amazing work overall.

[donstim]

Some more charts that should be of some help. I wish I could give a proper attribution. All I remember is they were supplied by a pilot on the Discord channel. Sorry, these are in metric units, and the cruise altitude performance appears to differ a bit from the chart above.

For the maximum altitude chart, I'm not sure what criteria Airbus uses to determine the maximum altitude. Common criteria are either a 100 fpm or 300 fpm rate-of-climb capability remaining with climb thrust or 0 fpm rate-of-climb with cruise thrust.

[donstim]

Since the lower speed, lower altitude performance associated with takeoff and initial climb looks pretty good at this point with your thrust mod, I assume you are considering retaining that and only increasing the thrust for higher speeds/altitudes. Since the table has a dependency on Mach, you could probably fashion a curve similar to what you did for N1, where you keep your thrust below a certain Mach and then either blend it back to what it was at higher Machs or maybe even increase it if necessary.

The attached chart may help with determining what Mach number to "blend it back in" or increase it. Since it's probably good up to about 250 KIAS and 5-10,000 feet, maybe retain what you have up to about Mach 0.4? Then around Mach 0.6 it needs to be getting back to what it was or start to increase.

Also, for lower N1s like you would use for descent, I don't think you want to change what you have (at all Mach numbers). Only for higher N1 like would be used for cruise or climb.

[derl30n]

@donstim big thanks! btw metric system is better ^^

[derl30n]

Since the table has a dependency on Mach, you could probably fashion a curve similar to what you did for N1, where you keep your thrust below a certain Mach and then either blend it back to what it was at higher Machs or maybe even increase it if necessary.

Can you please explain to me what you mean? I don't understand. And yes it is tricky atm via the N1 to tweak high alt climb speeds etc without rocketing out of the airport.

[donstim]

Since the table has a dependency on Mach, you could probably fashion a curve similar to what you did for N1, where you keep your thrust below a certain Mach and then either blend it back to what it was at higher Machs or maybe even increase it if necessary.

Can you please explain to me what you mean? I don't understand. And yes it is tricky atm via the N1 to tweak high alt climb speeds etc without rocketing out of the airport.

Sure, do you understand how the table is formatted? I assume you do, but just in case...there are sets of 11 numbers separated by commas. The first 11 numbers are the Mach numbers in the table. Actually, the very first number is meaningless. The next 10 are the Mach numbers - 0 through 0.9 by 0.1. For each subsequent group of 11 numbers, the first number is the N1 for that group of Mach numbers. The next ten numbers are the thrust factors for the 10 Mach numbers (0 Mach through 0.9 Mach by 0.1).

So, my thought is to leave the thrust factors for the low N1s (maybe up though 60 or 70?) for all Mach numbers the way you have them. For the higher N1s (above 60 or 70), at Mach numbers of about 0.4 or lower, also no change. But for the higher N1s at higher Mach numbers, increase the thrust factor such that for Machs of 0.7 and above maybe increase the thrust factor above what was in the table before you started. It would probably be best to plot it against Mach number for each of these higher N1s so that this is done smoothly.

Does that make sense?

[derl30n]

@donstim but you do talk about this table right? n1_and_mach_on_thrust_table.
Yeah it makes sense, now I got a bit confused there xD
And by increasing the numbers further back (.6, .7, .8, .9) I would assume that it ups the engine's performance at the given mach.
The increment atm is linear. (its by default)

[donstim]

Yes, exactly. I have just taken an initial cut at it by doing the following: At N1s of 70 and above, I made the following changes:
At Mach 0.5 -- Took the average of the thrust factor from your file and the current master
At Mach 0.6 - Used the current master thrust factor
At Mach 0.7 - Used the current master thrust factor increased by 3% (i.e., multiplied it by 1.03)
At Mach 0.8 - Used the current master thrust factor increased by 5% (i.e., multiplied it by 1.05)
At Mach 0.9 - Used the current master thrust factor increased by 5% (i.e., multiplied it by 1.05)

This is just an initial cut - I haven't flown it yet.
I assume these changes to the Thrust-N1 will have an effect on fuel flow, which will have to be dealt with later.
engines.zip

[derl30n]

Sounds great! I will try this approach and lets see how crazy we need to crank these numbers up xD

[donstim]

Flying now. My first impression is that I definitely overdid it with the higher Mach/higher N1 values. Going to revert back to the original values (before either of our mods) for Mach 0.7 to 0.9 - no 3% or 5% boost and see how that looks.

Edit: After those changes, thrust for N1 still seems too high. Climb rate is a little high, but mainly at cruise, N1 is way too low. Also noticed that the climb N1s given on the ECAM gauges goes above the rating value given in the upper right corner of the ECAM. Wish I knew how what controls those values in MSFS. Seems like if the N1 shown on the gauges was limited to no greater than the rating N1 when in open climb, things would match up better.

[derl30n]

I have done some more tweaking to the engine performance table.
I still need work on the climb rate.

[derl30n]

@donstim I noticed that ECAM shenaniganz too.
I will work on the N1 at cruise and thrust / climb rate tomorrow.

[donstim]

Another chart of interest...But I don't know yet where the climb N1 is set in MSFS (as I noted previously)

[St54Kevin]

@St54Kevin Look at the VAPP speed from MCDU. That will not change if I am in selected speed??

Oh I thought you meant the speed tape. No that will of course not change. Why should it?

[derl30n]

I think you missed my point....indicated VLS on speed tape is 111 knots and so with that logic, VAPP should be 116 knots. but MCDU says 125 (and VLS 120)

Again, the speed tape does not reflect the real world, since this flight model is not accurate.

[derl30n]

@St54Kevin

• Managed speed for climb after takeoff was S speed again. Should be 250 or a speed constraint with CLB detent. My SID didnt have a constraint

Don't know why, did not touched it.

• Vapp calculation was wrong initally. I had 60T GW and landing on runway 23 with wind 230/16 selected. Vapp in MCDU was 140, Vls 126. Later on final it corrected itself to 131, which was correct. But it should be there from the beginning.

I have seen that as well as on approach when FLAPS 3 are selected that managed speed will increase by a few knots and VLS and VApp in mcdu are different. (This is only when on FLAPS 3).

[St54Kevin]

@St54Kevin Look at the VAPP speed from MCDU. That will not change if I am in selected speed??

I think you missed my point....indicated VLS on speed tape is 111 knots and so with that logic, VAPP should be 116 knots. but MCDU says 125 (and VLS 120)

Oh you referring to the yellow VLS band. Now I get it :D yes that should match you are right. And I don't see the point why the MCDU Vapp should change in selected speed mode.

@MisterChocker Isnt the VLS in the speedtape linked to the calculated VLS in the MCDU?

[derl30n]

Isnt the VLS in the speedtape linked to the calculated VLS in the MCDU?

@St54Kevin currently it's not. The VLS in the speedtape is calculated with simvars, the MCDU uses real tables.

outdated

[Benjozork]

LGTM, minor issues that will be fixed with flight model and my personal optimization of PFD elements

[dc1ps2]

After I updated the A32NX#1295 version, the entire throttle response is very strange. When idling (F1), it will push forward? Unless I reverse the throttle, I can stop. I feel that the throttle position is confused.

[lve0200]

I think, I remember from old FS9/Phoenix times, that one had to give a little bit throttle to overcome the initial inertia, but then it kept rolling @ idle.

[derl30n]

@dc1ps2 0% throttle does not mean 0% acceleration. It does not work like a car. Actually when driving automatic transmission, the car will engage the clutch and start rolling, when the pressure on the brake paddle is lowered.

[dc1ps2]

@dc1ps2 0% throttle does not mean 0% acceleration. It does not work like a car. Actually when driving automatic transmission, the car will engage the clutch and start rolling, when the pressure on the brake paddle is lowered.

So now if I want to stop completely, what should I do?

[Lessar277]

Hit the brakes, then engage the parking brake

[St54Kevin]

@dc1ps2 0% throttle does not mean 0% acceleration. It does not work like a car. Actually when driving automatic transmission, the car will engage the clutch and start rolling, when the pressure on the brake paddle is lowered.

So now if I want to stop completely, what should I do?

It's the same IRL. The plane accelerates with idle. If you have a light aircraft you have to brake a lot while taxiing

[Lessar277]

@MisterChocker Not sure if I should log a bug for this or not (not a real pilot, no clue how the a/c should behave in this situation): at high temperatures, TO/GA thrust oscillates wildly. See: https://streamable.com/uga58q

The above clip was taken at 37 celsius OAT. For that temperature, FLEX solved the issue for me and stopped the oscillations. However, going up to 45 degrees OAT, FLEX behaves the same as TO/GA. From what I noticed, thrust is reduced in order to keep EGT out of the red. FLEX seemed to have a higher tolerance to red EGT than TO/GA for some reason.

This happened to me regardless of wether I spawned on the runway or cold&dark. Above location is Paro with live weather (I initially thought altitude was the reason), but it's reproducible everywhere, as long as you set temperature from the weather menu.

[derl30n]

@Lessar277 we need a FADEC in order to sort that and we are working on that as well.

[ElSrJuez]

I think fuel consumption is still rather high, considering the supposed ~3500nm range for the Neo.
Today I am seeing between 44 and even 80 KG/min @ FL380 on v.0.4.1, more often than not its around 60KG/min.
This is even higher than v.0.4.0?

[derl30n]

Well then the new MFS update broke something. Wouldn't actually surprise me...

[Benjozork]

@ElSrJuez this change is not present in 0.4.1.

[ElSrJuez]

@ElSrJuez this change is not present in 0.4.0.

Roger that, I asked on Discord and was left under the impression that all commits to master were included in 0.4.1.

Anyway, thanks. I kind of figured that later... this is planned for 0.5, no?
The reason I ask is, it would be great that the fixes but specially the fuel consumption would be there before VATSIM CTP.
: D

[ElSrJuez]

Fuel consumption is roughly halved (~30 KG/min) on the current Dev (8-Nov).