Jump to content

Uncommanded roll excursions on ILS to RW25R at LEBL


Recommended Posts

I have been practicing approaches to Barcelona LEBL RW25R and observed that on an ILS approach to this runway two uncommanded rolls to starboard followed immediately with a recovery to level flight occurred whilst on the ILS approach with autopilot and APPR selected. Nil wind is being used under a clear sky.

 

I am using the Simwings Barcelona airport scenery and Orbx EuropeLC and Orbx Barcelona city scenery.

 

Any thoughts about this behaviour? I would have thought that if the approach from 8nm had come under the authority of the ILS, then in the absence of buffeting etc from weather inputs, no such behaviour should have been observed. 

 

It might be interesting to remove the Simwings airport scenery and observe if the behaviour occurs with the default airport.

Link to comment
Share on other sites

I have just rerun the approach with initially the Simwings Barcelona airport removed then without the Orbx Barcelona city scenery installed and, in the first case the problem was still present and in the second case it did not occur.

 

It must be something to do with the magnetic attraction of the Orbx Barcelona scenery! I shall investigate further governing the use of the Orbx city scenery.

Link to comment
Share on other sites

  • Deputy Sheriffs
1 hour ago, LecLightning56 said:

I am using the A320 IAE variant in the Professional Series.

And which version exactly? 1.0, 1.1, 1.2.2.1 or 1.2.3.0?

Link to comment
Share on other sites

I am using Version 1.2.3.0. It is definitely the interaction of the Orbx Barcelona city scenery on this aircraft. If you deactivate the Orbx Barcelona city scenery and use the Orbx EuropeLC scenery, the observed behaviour does not occur.

Link to comment
Share on other sites

I have just tried another well-known product (a Boeing 777) on the same approach, with both the Simwings Barcelona airport and the Orbx Barcelona city scenery installed and there is no problem with uncommanded roll manoeuvres on an ILS approach to RW25R at Barcelona. It appears to be the case that the observed problem is aircraft-specific.

Link to comment
Share on other sites

This has been reported earlier, some called it 'wing wave on ils approach'... I'm trying to get more insight to this since I moved to 4.3 in October (im currently at 1.2.3.0 and P3D 4.4). My conclusions so far:

 

- it never happens on RNAV approaches, only ILS

- it only shows in 'calm' conditions, if crosswinds are too strong it will not show

- all models/variants will show it, also with min. and max. ZFW

- it is somehow related to scenery, for eaxmple i can reproduce it at LOWG (FSDG) with and without the addon, but not EDDF (Aerosoft)

- i believe it happens with Navigraph and NavData, but i'll recheck this (i have current navigraph cycle but NavData is the one coming with the bus)

 

I suspect it has something to do with ORBX, i have all of it (global, europeLC, etc...) and I'm somehow unwilling to remove all of that, took so long to install...  I have Simwings LEBL as you do so I'll try to reproduce this next.

 

Tom

Link to comment
Share on other sites

29 minutes ago, MajorTom said:

This has been reported earlier, some called it 'wing wave on ils approach'... I'm trying to get more insight to this since I moved to 4.3 in October (im currently at 1.2.3.0 and P3D 4.4). My conclusions so far:

 

- it never happens on RNAV approaches, only ILS

- it only shows in 'calm' conditions, if crosswinds are too strong it will not show

- all models/variants will show it, also with min. and max. ZFW

- it is somehow related to scenery, for eaxmple i can reproduce it at LOWG (FSDG) with and without the addon, but not EDDF (Aerosoft)

- i believe it happens with Navigraph and NavData, but i'll recheck this (i have current navigraph cycle but NavData is the one coming with the bus)

 

I suspect it has something to do with ORBX, i have all of it (global, europeLC, etc...) and I'm somehow unwilling to remove all of that, took so long to install...  I have Simwings LEBL as you do so I'll try to reproduce this next.

 

Tom

Thanks for bringing this known phenomenon to my attention. As you have pointed out if I inject a crosswind into the landing, the phenomenon does not show.

 

I hope that Aerosoft can find a fix for this soon.

Link to comment
Share on other sites

You probably see the same issue I also reported. Unfortunately Mathijs said, it is caused by a short frame drop that messes the frame-by-frame logic up, so the Bus leaves the ILS for 2-3s. There is no solution. 😕

 

 

Link to comment
Share on other sites

Well, I've read Mathijs answer in the old thread, but i don't buy it - yet ... There has to be proof ! At least I have to see the reason behind that.

 

The phenomenon is very consistent and reproducible at exatly the same points on final with a given ac/airport combination, even without addon scenery. My fps are well above 50, sometimes even 80, and there is no weather nor AI injection enabled which could cause a frame drop. I do know these frame drops very well when doing night time approaches to EDDF or EGLL, sometimes I have to kill AI completely in order to stop erratic behaviour of the bus, but the low fps is visible and can be logged. Here evereything is absolutely stable and out of nowhere the ac swings to the right. And on  RNAV or with crosswind it will not happen...

 

I'll do some more testing with logging enabled, i might even try to run with fixed fps in order to see the difference, and I'll report back.

Link to comment
Share on other sites

Hello,

same problem, with an approach for LFPO (Paris Orly) RW 24 under ILS . When Loc is alive and G/S also, the plane roll right/left several times before a good alignment.

I shall try again tomorrow with ananther flight.

Thanks for yours replies. I don't believe that is a problem with a scenery.

Michel.

Link to comment
Share on other sites

4 hours ago, MajorTom said:

Well, I've read Mathijs answer in the old thread, but i don't buy it - yet ... There has to be proof ! At least I have to see the reason behind that.

 

The phenomenon is very consistent and reproducible at exatly the same points on final with a given ac/airport combination, even without addon scenery. My fps are well above 50, sometimes even 80, and there is no weather nor AI injection enabled which could cause a frame drop. I do know these frame drops very well when doing night time approaches to EDDF or EGLL, sometimes I have to kill AI completely in order to stop erratic behaviour of the bus, but the low fps is visible and can be logged. Here evereything is absolutely stable and out of nowhere the ac swings to the right. And on  RNAV or with crosswind it will not happen...

 

I'll do some more testing with logging enabled, i might even try to run with fixed fps in order to see the difference, and I'll report back.

 

The fact that it happens only at certain airports is a very good proof indeed. Your basic framerate might be well above 50, but when you reach a certain position in the approach the frames drop momentarily to a value below 18. You probably don't even see it because it happens just momentarily. Unfortunately that is enough to cause the issue. I often see P3D hanging for a short time when it loads new scenery, especially when getting close to an airport.
A second later it recovers because your frames go back up again and so the Airbus gets sufficient data again.

 

The following might also be a good indicator that it is scenery causing fps issues and that it is totally unrelated to the approach:

 

On 2/3/2019 at 2:55 PM, LecLightning56 said:

I am using Version 1.2.3.0. It is definitely the interaction of the Orbx Barcelona city scenery on this aircraft. If you deactivate the Orbx Barcelona city scenery and use the Orbx EuropeLC scenery, the observed behaviour does not occur.

 

Probably Barcelona City is too heavy on your computers performance and thus the frames drop too low.
When deactivated your fps are probably a bit higher and thus the issue does not appear.

Such an fps drop is totally expected when you fly to a very detailed scenery such as ORBX's city scenes. Keep in mind each and every object needs to be calculated by your computer so if you have a very detailled city this means thousands of objects need to be loaded even outside the airport area.

Link to comment
Share on other sites

vor 4 Stunden , Emanuel Hagen sagte:

 

The fact that it happens only at certain airports is a very good proof indeed. Your basic framerate might be well above 50, but when you reach a certain position in the approach the frames drop momentarily to a value below 18. You probably don't even see it because it happens just momentarily. Unfortunately that is enough to cause the issue. I often see P3D hanging for a short time when it loads new scenery, especially when getting close to an airport.
A second later it recovers because your frames go back up again and so the Airbus gets sufficient data again.

 

The following might also be a good indicator that it is scenery causing fps issues and that it is totally unrelated to the approach:

 

 

Probably Barcelona City is too heavy on your computers performance and thus the frames drop too low.
When deactivated your fps are probably a bit higher and thus the issue does not appear.

Such an fps drop is totally expected when you fly to a very detailed scenery such as ORBX's city scenes. Keep in mind each and every object needs to be calculated by your computer so if you have a very detailled city this means thousands of objects need to be loaded even outside the airport area.

 

I don't know the code of the Airbus but couldn't this be programmed as "if frames too low, don't do anything (fly straight ahead)" imstead of turn sharply to either side?

Link to comment
Share on other sites

Hello Emanuel, that sounds like a stringent explanation, but you're missing several facts (sorry...) :

 

- it shows with default scenery

- it goes away with crosswind

- its not there with rnav approaches: same airport, date, conditions etc. so your passing the exact same scenery points...

 

Default scenery in my case is ORBX Global, Vector and openLC Europe, no airport addon enabled - would you recommend to test without any ORBX product ?

 

If this is related to something like scenery reload causing a sharp fps drop this would be 'visible' in performance montitoring for several hundred milliseconds, so I'll go and try to show this.

 

Tom

Link to comment
Share on other sites

  • Aerosoft
22 hours ago, MajorTom said:

Well, I've read Mathijs answer in the old thread, but i don't buy it - yet ... There has to be proof ! At least I have to see the reason behind that.

2

 

Well if you don't believe me, there is not a lot of use for giving you support is there now?

 

But we have been using this same code for 7 years now and we have seen this same issue hundreds of times and almost always been able to solve it by making sure the FPS stays stable. But if you want proof, just put the bus in cruise and start a CPU/GPU benchmark in the background that will stress the system fully. The bus will go out of control.

 

You are looking for something inside your sim, but it could just as well be in the OS or somewhere else. Recently a user having the same issues found out that he forgot to disable the RAID options in his BIOS (or how that is called these days). every few seconds that system would try to find the non-existing drives and that would drop FSP by 80% for a very short moment. If that happens at the wrong time you got problems.  Crosswind makes the code a little bit more stable because it forces more calculations (and is thus less susceptible to a loss of some calculations so that is also explainable.

 

If you do not believe this, that's fine by me, but I think we better close this topic in that case.

Link to comment
Share on other sites

  • Aerosoft
12 hours ago, Meyerflyer said:

 

I don't know the code of the Airbus but couldn't this be programmed as "if frames too low, don't do anything (fly straight ahead)" imstead of turn sharply to either side?

 

Not really, because this would just different problems. Besides testing for actual fractional FPS drops inside the sim is damned hard, if not impossible. You would need code that slows the whole sim to a crawl.

Link to comment
Share on other sites

vor 5 Stunden , Mathijs Kok sagte:

 

Well if you don't believe me, there is not a lot of use for giving you support is there now?

 

But we have been using this same code for 7 years now and we have seen this same issue hundreds of times and almost always been able to solve it by making sure the FPS stays stable. But if you want proof, just put the bus in cruise and start a CPU/GPU benchmark in the background that will stress the system fully. The bus will go out of control.

 

You are looking for something inside your sim, but it could just as well be in the OS or somewhere else. Recently a user having the same issues found out that he forgot to disable the RAID options in his BIOS (or how that is called these days). every few seconds that system would try to find the non-existing drives and that would drop FSP by 80% for a very short moment. If that happens at the wrong time you got problems.  Crosswind makes the code a little bit more stable because it forces more calculations (and is thus less susceptible to a loss of some calculations so that is also explainable.

 

If you do not believe this, that's fine by me, but I think we better close this topic in that case.

 

Hello Mathijs,

thanks for your answer and shedding some light to the 'crosswinds' phenomenon - first of all I want to stress that I have flown all your busses (and other ac) from the beginning, I do like them very much and believe they form an outstanding product ! And you and your team do an outstanding job given the constantly moving P3D environment (a nightmare for every project/program manager I believe...)

 

Now why I'm so insisting on 'answers' : I'm in Hard/Soft-Ware Development now for 40+ years and had to learn the hard way  to 'dig deep' when software showed 'unusual beaviour'. What puzzles me most is the fact that this 'excessive roll' or 'wing wave' will always happen  (given bus model and airport) at the exact same points in the final path, up to four times. So this makes it hard to believe something outside the sim is the cause, this being a very predictible 'coincidence'...

 

I do understand that these questions are not really a support issue, given the fact that only a small number of users seem to experience this. So what about moving this thread to the 'general' forums where we could continue to search for answers ?

 

Best regards,

Tom

Link to comment
Share on other sites

  • Aerosoft

Feel free to discuss it here. 

Just keep in mind this is not something we can change easily. It would mean months of work to redo the code to works in a different (and in hindsight, probably better) way.  But as a 40-year veteran, you know every project has things that in hindsight should have been done differently. The way we choose to do it was the most efficient one and suitable for the hardware of the day. It's still very efficient, that why so many people commented on how smooth and easy on FPS the bus is. But with modern 6 core CPU's it's probably better you do not work in FPS clock cycles.

Link to comment
Share on other sites

16 hours ago, MajorTom said:

What puzzles me most is the fact that this 'excessive roll' or 'wing wave' will always happen  (given bus model and airport) at the exact same points in the final path, up to four times.

 

For what it is worth, the FD commands the momentary roll on approach also with AP off.

 

Link to comment
Share on other sites

What is interesting to me is that it does not take much crosswind to eliminate the phenomenon altogether. I have just tried a 8nm ILS approach to LEBL Barcelona RW25R, on a course heading of 245 degrees and injected a negligible crosswind into the sim of 180 degrees at 2 knots (constant, non-varying), and the momentary rolls did not occur. Note that I used the full monty of scenery here with Orbx EuropeLC and Orbx Barcelona city scene and also the Sim-wings Barcelona airport scenery.

 

How the provision of some motion (albeit very modest and indiscernible) in the atmosphere can prevent the anomaly, perhaps needs to be discussed and understood. I guess for those affected this is at least a workaround. Without it the roll is very uncomfortable at a few hundred feet AGL on an approach.

 

Any comments from Aerosoft about this?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. Privacy Policy & Terms of Use