Sink rate with engine(s) idle

Hello you guys
Let me introduce myself as a hobby pilot, being with Microsoft flight simulators since 1992, flying “just for fun”. I never use the autopilot, also not in the latest -and great- MSFS2020. Recently I came to wonder about, whether the newest flight model was designed with the use of the AP in mind.
the reason:
I find the sink rate of the planes I fly (a320 neo, FBW 320, cirrus sr 22) to be extremely slow. It takes ages to get down to earth, with throttle cut and idle engine. Can this be realistic? In all earlier versions of the flight simulator I found the sink rate to be realistic, i,.e. normal aerodynamic behaviour.
I know that all you experienced pilots constantly fly with the AP turned on with ILS approaches and all the fancy stuff, and you only switch off the AP just before landing.
My modest request to some of you:
Will anybody be willing to give it a try and set throttle to idle at about 10 000 ft or so, cruise speed, and see if you can make it to the ground as expected?
ps: I never use spoilers, neither in ms2020 nor before.
thanks for any reply, as I no longer enjoy the game as before. I wish experienced simmers could confirm to me, that the aerodynamic behaviour in MSFS2020 actually is realistic

Have you tried any non-first party planes to compare? Default planes are not the best example of what is on offer.

Well, honestly: I want the planes I like to fly as expected. I have flown the Pilatus some times and had no complaints, and there might be many more. But the issue remains, as it should not be necessary to purchase third-party software to fix a problem - as long as I do not know-by now- if there exists a problem after all

Well with the FBW A320 I fly vnav managed descent and I don’t touch the throttles at all , they are left in the CL position until 100 feet from touchdown

That’s my understanding but I prefer the PMDG 737 and even in managed descent throttles are armed but at idle, so somewhat similar to what you do but I have ap on and use speed brakes if necessary and of course spoilers and sometimes reverse thrust on touchdown

Anyone else care to chip in on this , cos that my two cents worth :thinking:

@Deppi17 I think you misunderstand, unless you are referring to after market planes as third-party software.

In most sims, the default planes are functional, but not exactly the most accurate representations of their real world counterparts.

Third party developers create simulations that often far exceed the quality of default aircraft. There are also community mods that make improvements to default airfract. Notable ones would be the G36 project, and the forthcoming WB 172X mod.

Can you give us some numbers?

What are your actual sink rates?

thanks for the response, even if it was not the kind of response I had hoped for. This entry will be lengthy, as I have reached a conclusion and wish to terminate the thread with it.
As to numbers: it is tedious and laborious to provide reproducable data. I am reluctant to do so, mostly because I am afraid that I already know the result. As a matter of fact, I have been browsing the forum intesively the last couple of days, only to find out that my first suspicion seems to be confirmed:
few default aircraft in MSFS, if any, show performance comparable with data in the respective POHs of the aircraft. It is a well-know fact that Microsoft/Asobo focus on appearance of the entire world scenery, a task they did very succesfully and keep on improving upon. The aerodynamics of the aircraft fleet has only second priority, and they leave it to third party vendors to finish the work.
THE REASON WHY MICROSOFT can get away with it can be found in the autopilot, of course. Allmost every pilot, in real life as well as in simulation, does use the AP, even in smaller airplanes .
as a (retired) computer programmer I can imagine that in AP- programming one is able to, at least to some extent, to bypass some very tricky calculations with respect to balencing the 4 basic forces acting upon an airplane (plus considering wind and weather influences) - in order to achieve an aerodynamic result which satifies the vast majority of the audience.
people normally turn off the AP few hundred meters above ground, which simplifies gravity calculations.
the engine power of the SR22 is not as expected, others than me have discovered that, but these kind of things do not bother me.
However, when basic gravity calculations seem to be deficient, I just have to bail out. (I wonder if they entered the parachute gliding rate? - by the way, I did not activate the parachute,) Beig stuck i the air with idle engine in a time which feels like ages to me - is not acceptable.
so the final conclusion: I have to learn using the AP or give up being a flight simmer.
time will show.

1 Like

Never noticed any problem like described here, why would the aircraft fly a realistic descent path with AP ON but not with AP off according to you? Idle descent on a jet would be around 3 nm for each 1000 ft. From 10.000 ft down would take you around 30 miles.

Wow, I’m sorry but you have not the slightest idea what you are talking about. This all does not make any sense, of course the default planes do not follow POH numbers 100% (fun fact, neither does the real aircraft) and their flight models are not the most accurate. But to say they are completely useless and built for autopilot is a strange conclusion. I did some glide ratio tests myself and it’s not as far off as you are describing.

I think you are comparing MSFS to other sims, you should compare MSFS to the real aircraft. Do you have any real world experience or what are you using to draw your conclusions, did you at the very least compare glide ratio to POH? And what do you mean by gravity calculation? Gravity is the same at a few hundred meters as it is at 30.000 ft, difference is negligible. And autopilot doesn’t affect the flight model, if you have problems getting down, it would be the same whether the AP is ON or OFF.

I’m sorry but it doesn’t sound to me like you know what you are talking about. It okay to have complaints and be a critic (so am I), but the problem must be real and you need to supply evidence to support your claim. This is a baseless rant.

When I launched this thread, I did it with the kick-off in a ball game in mind.I delivered the ball to be played and thrown back and forth forfinally being returned to me with constructive answers, either affirming my suspicion or - still constructive - rejecting it. Honestly, I had really been hoping for at least some entries telling me that the sink rate of the SR22 is OK, or at least as OK as it can be for the time being, when disengaging the AP at 10 000 ft or above. I received none of this kind, and this includes your response, too, sir.
As a matter af fact, your entry tells me that you did not play with the ball at all. Instead you smashed it right back into my face in a - which I cannot interprete otherwise than - rude and insulting manner.
So, when I finally decided to come up with an response (as the kind and gentle perosn I am) to continue this thread even after I finished it, I do this also with a broader audience in mind, i.e. friendly and non-insulting people, who might wish to know WHAT I AM TRYING TO TALK ABOUT (capitals are imnportant, as the subject is extremely difficult), which means that a definitive answer of course only can be found among Asobo programmers, and more about that later on.

For now I would like to come up with some examples, and please do not take them too literally:
the first example are weather forecasts. Their predictive power is worth a few days at best, most of the time. Why? when so much computing power is available today? I am neither a meteorologist nor a pilot, but it is clear enough that the point here is the fact that very small deviations (you might call them tiny) in mainly air pressure accumulated can lead to considerable deviations on a large scale. Before I return to flight sim matters, another example:-

When you plot any equation on a piece of paper with your pen (well, today you can let a printer do it…), you certainly do not draw an infinite amount of points on paper. Of course you chose some proper unit and do some interpolation afterwords to provide for a smooth and continuous curve.

Now comes the big question: what has all this to do with aerodynamic calculations in general and AP progamming in particulal? I will give it a try and please do not smash my face, as I do my best as a patient and gentle person;
Lets assume we have space ccordinates (x,y,z) at a given time. With the AP engaged and a given controls input and chosing some suitable time unit there will exist a target (which of course must be calculated) which means some guidance to future coordinates (x1,y1,z1), and so on in this manner. I cannot elaborate further on “some guidance”, because there are still equations to be solved, and of course they are not a breeze to perform. However, given some fixed numbers for important parameters (the AP input) I assume that the calculations are somewhat simplified, is not that a logic conclusion, as there will be less unknown quantities?

But let me revert to the met example: knowing their future position in space mets will be able follow air molecules from one point to another in time, almost precisely, and it would be possible to predict the weather for weeks. Please note: the meteorological equations will still have to be done, as well as the aerodynamic equations in MSFS.

I do realize that I am not even in the vicinity of being able to perform aerodynamic calculations in MSFS, but I do have some experience programming in the C language.Based upon this experience, and all the above mentioned I can imnagine the worst scenario for me as a - fictive - aerodynamic programmer:
an aircraft floating in space at great height with the AP disengaged and no input from the controls! this situation will require utmost precise calculations to provide just an acceptable simulation. Utmost precise.

finally: someone might come up with arguments and postulate that programming with the AP switched on is at least just as difficult as otherwise, or perhaps even more complicated. Of course noboby can know, except Asobo programmers, as mentioned in the beginning of my entry. So in a certain sense it is right that I do not know what I am talking about. My answer to this will my and others experience with the simulator:

For the first I never had these annoying gravity problems in earlier versions of the FS, so Bruce Artwick and successors maybe did an even better job than they are recognized for. Secondly it has become clear to me, after browsing this and other fora for some time, that the risc of things going bananas is significantly greater when the AP is disengaged than otherwise. There are countless examples of all kinds of aircraft deviating either to the right og left shortly after take-off, which cannot be accounted for by propellor effects. I could go on, but I will not, because it has never been my intention to criticize neither MS nor Asobo.

I will finish here, and wish the audience lots of playing pleasure, with or without the AP engaged! And yes, in some way I still do not know what I am talking about. But at least I have tried to explain my thoughts on the gravity question, which very well might be wrong, but which I nevertheless feel having the right to express publicly - or am I even wrong here? Please do not smash the ball in my face one more time - I am a fragile person.

While I agree with you that the aircraft in MSFS are quite ‘floaty’ compared to previous simulators, especially in ground effect before touchdown, I do not believe the AP has anything at all to do with this. The AP ‘simply’ provides control inputs to the aircraft in place of the pilot. It has to overcome the exact same aerodynamic forces as the pilot would.

It sounds to me like you are suggesting the whole flight model changes when the AP is engaged. That’s what @Nijntje91 is rejecting above. I also reject it.

AP or no AP, the aircraft should fly using the same rules of physics. Same equations. Everything the same.

Where MSFS seems to fall short (compared to my own experience of simulators over the last 30 years) is the slow speed regime. That includes both power off and power on. Aircraft float excessively, and aircraft in full landing configuration can be yanked and banked with no real risk of an accelerated stall occurring or the inner (slower) wing stalling. Speaking mainly of GA aircraft. To me it feels odd, but I have little real life experience to call upon.

So I agree with you about the power off flight modelling feeling a bit wrong, but it may be more correct than the old simulators. But it has no relation to the AP. That part is where you are ‘barking up the wrong tree’ as the saying goes :upside_down_face:


The “gliding angle” (best angle depending on speed) of an A320 is at around 15… roughly! This means sailing with engines idle from a hight of 1000ft leads to a flying distance of 15000ft. So if you are at 2.5nm till the touch down point on the runway you need to be at 1000ft above the runway to reach the touch down point with idling engines. Interesting question if this works in fs2020 and I’ll try tonight. :sweat_smile:

ps: just found some data for the SR22… it says gliding angle 11 at 87KIAS… so might give it a try too…

1 Like

You could check with the SR22 Handbook if there is some data in there.
Page 58 has some info for example.
Glide Ratio is 9.6:1

1 Like

Thanks for the data, I only had the numbers for the SR20 G3.

1 Like

I wasn’t rude or insulting, just direct :joy:.

I don’t get the point even after reading your long post twice, and I still don’t get what you mean with “gravity problem”, if you mean that glide ratio is unrealistic (which has absolutely nothing to do with the use of autopilot), that’s not a gravity problem, rather a lack of drag. Once more the question, what do you base your claim on? Did you compare glide ratio with POH? Or are you comparing simulators?

Ground effect is overdone and there are a few other things, like the accelerated stall issue you are talking about is because Asobo reduced maximum control surface deflection with speed, you can only see this in dev mode.

1 Like