CFD, broken!: Body not affecting airflow, coanda effect and prop wash

Questions & Community Discussions Aircraft
1

The body of the aircraft is not affecting airflow while using CFD. Air should
be flowing around the body and be affected by it, but air just goes through.
Air flow around a cylinder should look like this (perhaps not like pic 4 with
eddys) , but in the game it just goes through and is not affecting the air.

1280×720 760 KB
Here in FS20 there is no impact
on the air from the fuselage. The tail is performing correctly, but nothing is
happening around the fuselage. Side forces are calculated with the modern FM,
but the air is not affected. This is with a 25 knot crosswind.
1408×1121 903 KB
Also there is no Coanda effect
of the fuselage. The air pushed by the prop is just carried away by the wind,
making crosswind landings impossible, especially with a castering nosewheel.
1272×1249 1.04 MB
Airflow should look like this,
following the fuselage.
1234×1027 877 KB
This is
something Rush tried to get fixed in XP11, but Austin. You can see the clips
below.
https://www.twitch.tv/videos/251080674
https://www.twitch.tv/videos/251369670
https://www.twitch.tv/aerosimgaming/clip/FurryBadCocoaBigBrother Also the
math doesn’t add up, using the legacy prop model you can see the induced
velocity of the prop. 134 kph. (72.4knots). If look at the angle it looks like
the airflow is leaving the prop at 45°
1748×1317 1.28 MB
However simple math says the
airstream should come out at 19°, not 45° there is something wrong with the
calculations. So yeah. TDLR. - Body doesn’t effect CFD (CFD_ReinjectBody = 1)
- No coanda effect. - Prop wash with crosswind doesnt seem to be calculated
correctly.

2

This is Austin’s video on how XP11 deals with that: Testing Prop Wash on the
Ground for X-Plane Flight Model -
YouTube

3

Hello @MrTommymxr First, you have to understand CFD is as accurate as it
needs to be for what it’s used for. We don’t use CFD to compute Coanda effect
as the existing collision model perfectly does the job for this particular
effect. The precision/performance ratio of our CFD is adjusted to get proper
results for 2 effects: prop wash and deep stall. So yes, some lines will go
through the fuselage in the visualisation mode as the voxel grid is too broad
for this. But the influence on the computed forces is negligible. Do you have
any source to demonstrate that the airflow should move as you illustrated with
this picture?

1234×1027 877 KB
" However
simple math says the airstream should come out at 19°, not 45° there is
something wrong with the calculations." What “simple maths” are you talking
about? You seem to be just adding two vectors but you agree this is not how
fluid motion should be calculated, right? Regards, Sylvain

4

Coanda effect is too important to be left out. without it, planes with
castering wheels simply do not respond in the tiniest of crosswinds. my
sources are in my initial post

5

boop <https://devsupport.flightsimulator.com/questions/13095/why-does-cfd-
just-end.html>