Wind drift error

[Aenonar]

There has been an issue with the navigation system slowly drifting off for years now but recently I found both an error and a possible semi-fix while waiting for Heatblur to fix it.

The bug:
If a mission for example has the wind settings with a direction of 090° and 10 m/s (36km/h)the Viggen will report it as being 270330 in the CK37 on Vind/Ruta/Mål, aka direction of 270° and 33km/h (9,16 m/s). So the wind has both the wrong speed and 180° the wrong direction, thus the SI is also telling you to compensate in the wrong direction.

 

Aircraft is flying north with the wind coming from the right while the SI says you're drifting right as the CK37 thinks the wind is coming from the left.

This will slowly or quickly deteriorate both the nav system as well as BK90 bombing depending how long the mission and in which direction you're flying, making flying precision/BK90 missions with wind enabled basically impossible.

Semi-fix:
We tried to manually input the wind data to the altitude we would be flying on and even at 90km/h crosswinds the navigation system and BK90 drops seem to be holding up, although the BK90 parachutes of course just blew away once released but the navigation up until then went just fine and on target. If we kept the default wind values in the aircraft the BK90's were several km off target in just a minute of flying...

Note that inputting wind values manually only works if you fly at the altitude you took the values from as the wind can shift significantly in both speed and direction depending on altitude.

The Viggens system is supposed to be constantly reading current wind data for accurate navigation but it's simply not getting the correct data from the mission.

TL;DR - The CK37 doesn't obtain the correct wind values in both direction and speed - therefore the Viggens navigation system is constantly breaking down, please fix ASAP!

Edited December 6, 2022 by Aenonar
Crossed over erroneous findings...

[Spurts]

Actually, I think it's a ME thing where it sets the direction the wind is blowing to instead of from.

[TOViper]

55 minutes ago, Aenonar said:

There has been an issue with the navigation system slowly drifting off for years now but recently I found both an error and a possible semi-fix while waiting for Heatblur to fix it.

The bug:
If a mission for example has the wind settings with a direction of 090° and 10 m/s (36km/h)the Viggen will report it as being 270330 in the CK37 on Vind/Ruta/Mål, aka direction of 270° and 33km/h (9,16 m/s). So the wind has both the wrong speed and 180° the wrong direction, thus the SI is also telling you to compensate in the wrong direction.

 

1.
"BLOWS TO DIR" is the *sorryforblamingED* totally-OFF-any-world-standard-met-information regarding aviation. Its a ED thing.
In aviation, wind is always referenced to "where it comes from". At least, the ME says where the wind is blowing to.
The CK shows 270, which is correct. The 33 is not correct, it should read 36.

2.
Are you sure Doppler works at 2000 meters?
If not (which I am pretty sure it doesn't), a forecasted wind is put into the CK when the aircraft is spawning. Maybe at this stage something goes wrong with the 10% error.

3.
"Aircraft is flying north with the wind coming from the right while the SI says you're drifting right as the CK37 thinks the wind is coming from the left."
Considering 1. and having in mind that Doppler wind in the Viggen is in the standard-aviation-met system ("coming from"), this statement is wrong, the aircraft behaves correctly.

I can see that in the RC2 there is not even one hint that the Viggen uses another reference system for wind. As said, the "blowing-to" is an invention by ED, which makes absolutely NO SENSE. It caused and still causes headaches of people, and costs so much usesless time and efforts...

Edited December 6, 2022 by TOViper

[Machalot]

31 minutes ago, TOViper said:

As said, the "blowing-to" is an invention by ED, which makes absolutely NO SENSE.

I'll semi-defend this on the ground that the most straightforward version of the underlying math to create the wind in a sim uses the "blows to" direction, so a software dev writing the code would find it natural to do it this way, especially if unaware of the met convention. (Tbh I am an aerospace engineer and I find the met convention "backward" just like using negative charge for the electron or giving bodyweight in "kg" )

[TOViper]

"especially if unaware of the met convention" < That's the problem for this long term story.
 

[Aenonar]

1 hour ago, TOViper said:

 

1.
"BLOWS TO DIR" is the *sorryforblamingED* totally-OFF-any-world-standard-met-information regarding aviation. Its a ED thing.
In aviation, wind is always referenced to "where it comes from". At least, the ME says where the wind is blowing to.
The CK shows 270, which is correct. The 33 is not correct, it should read 36.

2.
Are you sure Doppler works at 2000 meters?
If not (which I am pretty sure it doesn't), a forecasted wind is put into the CK when the aircraft is spawning. Maybe at this stage something goes wrong with the 10% error.

3.
"Aircraft is flying north with the wind coming from the right while the SI says you're drifting right as the CK37 thinks the wind is coming from the left."
Considering 1. and having in mind that Doppler wind in the Viggen is in the standard-aviation-met system ("coming from"), this statement is wrong, the aircraft behaves correctly.

I can see that in the RC2 there is not even one hint that the Viggen uses another reference system for wind. As said, the "blowing-to" is an invention by ED, which makes absolutely NO SENSE. It caused and still causes headaches of people, and costs so much usesless time and efforts...

 

  1. Yep I missed the "blows to" as coming from is just the default in aviation... Another notch in ME weirdness. But that makes it pretty weird that it worked better when we manually input the wind direction shown in the ME, which then should be 180° wrong...

 

2. Got lazy when I took screenshots for this thread so I just picked the default ME FL of 2000m. Unsure if it works or not IRL however the error in game is even worse at 500m.

With 90km/h winds at 180° (0° blows to) at 500m altitude it displays 180700

 

 

However.... Things just got even weirder because earlier we tested this for several hours and the results were always the same. Wind active and no manual change would result in BK90's missing and the nav system failing in all wind speeds. Earlier today when taking the screenshots again the BK90's went off and the nav system failed. Now however when using 90km/h winds again... It's flawless without any input needed, except that the windspeed in the CK37 is displayed wrong...

Edit: Sum up from this latest test with 90km/ winds, 100km distance at M0.8: SI pointed me to the right place, BK90 dropped to the right place (within 50 meters ish, should have been way more), CK37 said speed was 70km/h and towards the end the CI was about 5 degrees off, so half the nav system slid away in the wind...

Hell I thought I finally had figured out what was wrong with the nav system drifting off with wind... So I guess it's an intermittent bug or something else...

Edited December 6, 2022 by Aenonar

[Aenonar]

Done some more testing and while I haven't been able to recreate the 10km drifts I had the first time I've found a few things...

Wind speed is displayed wrong in the CK37
CI points about 5° off from the SI with a 90km/h crosswind
Drift if more significant when starting from the ground (potentially due to the LS sync from increased/decreased takeoff due to headwind vs tailwind)
Drift is reduced when wind is input manually, albeit it should no longer update via doppler
Time standing on the ground prior to takeoff doesn't seem to have any effect of the drift. (We have previously theorized about the NAV system drifting along with the fuel going down over time while on the ground)

 

More testing required as sometimes the nav system drifts several km away while not doing anything "weird"...

Edited December 6, 2022 by Aenonar

[TOViper]

*LOL*, I already thought it would end in some complicate stuff.

@Aenonar, can you provide a solid track of the happenings. This should help the guys here to see what you describe ...

 

[Machalot]

When you report the CK37 estimates the wrong wind, how accurate should we expect it to be? Is 5° and 20% speed unreasonably bad for an airborne Doppler radar? 

In your test cases do you fly over water? If so, do you set the Doppler switch to SJÖ? This setting strongly influences nav drift. 

Are you generally aware how the TERNAV system works, and accounting for its function on your reports? For example, at 2000m it will not do anything, but below 500m you might get some nav fixes if there is some terrain variation below you.