Jump to content

Recommended Posts

On 5/31/2021 at 9:23 AM, quasiguru said:

I just wanted to add in, I am still experiencing zig zagging too, I had this happen on a flight last night that I streamed. TNCC/11 SCAP1K SCAPA DANDE AVAKI TNCM/10 This happened between SCAPA to DANDE.

image.thumb.png.a952927aa626c9c69d496d803a2fb45d.png
@Mathijs Kokstill having the same behavior on this same route same two waypoints.. This is on 1.0.5... Only way to stop it is go heading mode to bring it closer to the nav line, then reengage nav mode.
 

 

Link to comment
Share on other sites

23 hours ago, Mathijs Kok said:

Make sure you are all updated to version 1.0.5

that's the point, i already reinstalled the plane to make sure it wasn't any update install issue... but the problem, at least for me, still :( maybe its a problem with the sim... or the version downloaded from the MP... idk, i just gave up :( 

Link to comment
Share on other sites

On 6/4/2021 at 3:27 PM, Mathijs Kok said:

Make sure you are all updated to version 1.0.5

Here is the first flight I've done after 1.0.5: LSZH/28 DEGE2W DEGES Z1 ETAGO Z3 ABUMO N850 EBANA T842 PELUN PELU3H EDLP/PAD.I06. Out of all the flights I've done in the CRJ so far, some of them had a bit of zig-zagging (nothing too serious) but this one was particularly bad. Especially on the SID, but throughout the entire flight as well.

 

To add to this issue, the AP got lost on the final left turn towards final on the ILS 06 approach and just went to roll mode in the middle of the turn, I had to fly in heading mode in order to capture the LOC.

Link to comment
Share on other sites

10 hours ago, FHSim said:

that's the point, i already reinstalled the plane to make sure it wasn't any update install issue... but the problem, at least for me, still :( maybe its a problem with the sim... or the version downloaded from the MP... idk, i just gave up :( 

i just checked the files in "C:\Users\GUILHE~1\AppData\Local\Packages\Microsoft.FlightSimulator_8wekyb3d8bbwe\LocalCache\Packages\Official\OneStore\aerosoft-crj\Data" into the Readme, it does not show the 1.0.5, it shows as least update the 1.0.4.0, ai think that the MP did not really updated the plane, i'll reinstall the plane (again) '-'

Link to comment
Share on other sites

  • CRJ Forum Moderator
On 6/5/2021 at 6:00 PM, Lessar277 said:

Here is the first flight I've done after 1.0.5: LSZH/28 DEGE2W DEGES Z1 ETAGO Z3 ABUMO N850 EBANA T842 PELUN PELU3H EDLP/PAD.I06. Out of all the flights I've done in the CRJ so far, some of them had a bit of zig-zagging (nothing too serious) but this one was particularly bad. Especially on the SID, but throughout the entire flight as well.

 

To add to this issue, the AP got lost on the final left turn towards final on the ILS 06 approach and just went to roll mode in the middle of the turn, I had to fly in heading mode in order to capture the LOC.

I flew this route and saw no major problems. On the DEGES2W departure from runway 28 at LSZH it is essential to keep your speed below 210 knots (as annotated on the chart) when making the course reversal turn at ZH552 back to KLO in order to complete the turn before arriving at the VOR. This would require keeping the slats deployed until on course to KLO. Also, be aware that DEGES is a mandatory overfly waypoint, (the waypoint on the chart has a circle around it), meaning that legally, the aircraft has to fly directly over and past DEGES before it can begin the turn to your first enroute waypoint ROMGA (on Z1), so it will loop back to ROMGA from outside the course line drawn on the MFD. The real aircraft would do the same thing.

 

I had no problem intercepting the ILS06 at EDLP.

  • Upvote 1
Link to comment
Share on other sites

  • Root Admin
2 hours ago, JRBarrett said:

I flew this route and saw no major problems. On the DEGES2W departure from runway 28 at LSZH it is essential to keep your speed below 210 knots (as annotated on the chart) when making the course reversal turn at ZH552 back to KLO in order to complete the turn before arriving at the VOR. This would require keeping the slats deployed until on course to KLO. Also, be aware that DEGES is a mandatory overfly waypoint, (the waypoint on the chart has a circle around it), meaning that legally, the aircraft has to fly directly over and past DEGES before it can begin the turn to your first enroute waypoint ROMGA (on Z1), so it will loop back to ROMGA from outside the course line drawn on the MFD. The real aircraft would do the same thing.

 

I had no problem intercepting the ILS06 at EDLP.

 

I just tried this one myself and as JR did not see any issue.  It is hard to debug these issues that we are unable to recreate, but I assure you will will keep on trying.

Link to comment
Share on other sites

  • CRJ Forum Moderator
22 hours ago, quasiguru said:

@JRBarrett can you attempt mine as well? I am tempted to just drop more money for the Aerosoft version to avoid the issues with the FS Marketplace version. 

I flew your example flight plan, and I agree that the aircraft is taking an excessively long time to get established on the leg from SCAPA to DANDE. I did the flight with the clear sky weather theme, to insure winds aloft was not a factor.

 

I do not know the details of how the autopilot is coded, but am somewhat familiar with the mathematical algorithms used for doing great circle navigation. Two factors that might make this particular leg problematical are the length of the leg from SCAPA to DANDE (242 miles) and the relatively southern latitude of the two waypoints (approx 15 and 17 degrees).

 

I am using the equivalent of the Aerosoft version for testing, so I do not think this is an Aerosoft vs. Marketplace issue.

 

I am in the process of putting together some information from the real Proline 4 autopilot systems manual which I will send to Hans this week, with some suggestions on possible ways to improve the capture of an outbound course leg when executing a turn at an enroute waypoint. 
 

Many users have remarked that the aircraft turns “too soon” when turning at high altitude, but this is actually a characteristic of the real CRJ, which can only use a single fixed bank angle of 12.5 degrees when turning under autopilot control above 31,600 feet. The real aircraft can begin a course change turn as much as 11 nautical miles before arriving at a waypoint, and the turn anticipation distance the sim version is using appears to be very close to the value the real aircraft  would use - the distance is calculated by a specific algorithm that uses current airspeed, expected bank angle, and the number of degrees of the required heading change as inputs.

 

Because the CRJ can only turn using a single fixed bank angle, the long turn anticipation distance (in the real CRJ) is used to insure that the aircraft will not reach the new course line prematurely, causing an overshoot, which could easily happen if the aircraft did not start the turn until closer to the waypoint. 
 

This initial part of the turn sequence in the sim appears to work just like the real aircraft. It is the final interception of the new course where I sometimes see a problem. What sometimes happens is that the aircraft continues the the turn until arriving at the calculated new heading before the CDI needle is centered, meaning that the aircraft has not yet fully intercepted the new course. It then executes a series of corrective banks to “work its way onto” the new course.

 

The real aircraft will turn at a fixed bank angle until the heading is within approximately 10 to 12 degrees of the calculated new course heading. If the CDI needle is still deflected, (meaning the new course is not yet intercepted), it will then roll level, (and remain level) until the CDI centers, at which time it will execute a final bank to roll out on the new course. 
 

This can be emulated in the sim by going to HDG mode during the final part of a course change turn, and manually rolling out on the new course when the CDI centers, then re-engaging NAV mode, and it seems to work well.
 

Even the real aircraft will occasionally have to make a couple of corrective banks after intercepting a new course in LNAV to acquire a wind correction angle, especially with a very strong crosswind.

 

There may be limitations in how the sim CRJ LNAV routine is implemented (using default simulator autopilot commands behind the scenes) that would make it difficult to implement a turning course intercept the way the real aircraft does it, without potentially causing issues in other parts of the LNAV system - especially those responsible for maintaining course track once acquired - which in general works very well.

 

I will send my suggestions to Hans on our private Discord to see what he thinks.

 

 

  • Like 4
  • Upvote 2
Link to comment
Share on other sites

On 6/13/2021 at 9:35 AM, JRBarrett said:

I will send my suggestions to Hans on our private Discord to see what he thinks.

 

 

Thank-you very much for having patience and taking the time to follow up on this. Would love to see this fixed.

Link to comment
Share on other sites

On 6/13/2021 at 8:35 AM, JRBarrett said:

I flew your example flight plan, and I agree that the aircraft is taking an excessively long time to get established on the leg from SCAPA to DANDE. I did the flight with the clear sky weather theme, to insure winds aloft was not a factor.

 

I do not know the details of how the autopilot is coded, but am somewhat familiar with the mathematical algorithms used for doing great circle navigation. Two factors that might make this particular leg problematical are the length of the leg from SCAPA to DANDE (242 miles) and the relatively southern latitude of the two waypoints (approx 15 and 17 degrees).

 

I am using the equivalent of the Aerosoft version for testing, so I do not think this is an Aerosoft vs. Marketplace issue.

 

I am in the process of putting together some information from the real Proline 4 autopilot systems manual which I will send to Hans this week, with some suggestions on possible ways to improve the capture of an outbound course leg when executing a turn at an enroute waypoint. 
 

Many users have remarked that the aircraft turns “too soon” when turning at high altitude, but this is actually a characteristic of the real CRJ, which can only use a single fixed bank angle of 12.5 degrees when turning under autopilot control above 31,600 feet. The real aircraft can begin a course change turn as much as 11 nautical miles before arriving at a waypoint, and the turn anticipation distance the sim version is using appears to be very close to the value the real aircraft  would use - the distance is calculated by a specific algorithm that uses current airspeed, expected bank angle, and the number of degrees of the required heading change as inputs.

 

Because the CRJ can only turn using a single fixed bank angle, the long turn anticipation distance (in the real CRJ) is used to insure that the aircraft will not reach the new course line prematurely, causing an overshoot, which could easily happen if the aircraft did not start the turn until closer to the waypoint. 
 

This initial part of the turn sequence in the sim appears to work just like the real aircraft. It is the final interception of the new course where I sometimes see a problem. What sometimes happens is that the aircraft continues the the turn until arriving at the calculated new heading before the CDI needle is centered, meaning that the aircraft has not yet fully intercepted the new course. It then executes a series of corrective banks to “work its way onto” the new course.

 

The real aircraft will turn at a fixed bank angle until the heading is within approximately 10 to 12 degrees of the calculated new course heading. If the CDI needle is still deflected, (meaning the new course is not yet intercepted), it will then roll level, (and remain level) until the CDI centers, at which time it will execute a final bank to roll out on the new course. 
 

This can be emulated in the sim by going to HDG mode during the final part of a course change turn, and manually rolling out on the new course when the CDI centers, then re-engaging NAV mode, and it seems to work well.
 

Even the real aircraft will occasionally have to make a couple of corrective banks after intercepting a new course in LNAV to acquire a wind correction angle, especially with a very strong crosswind.

 

There may be limitations in how the sim CRJ LNAV routine is implemented (using default simulator autopilot commands behind the scenes) that would make it difficult to implement a turning course intercept the way the real aircraft does it, without potentially causing issues in other parts of the LNAV system - especially those responsible for maintaining course track once acquired - which in general works very well.

 

I will send my suggestions to Hans on our private Discord to see what he thinks.

 

 

Seriously, thank you!!!! :) It is nice to see it is not just the Marketplace version having this issue!!! 

Link to comment
Share on other sites

  • 2 weeks later...

I just want to confirm I'm still seeing this issue with the latest non-marketplace version. My flight plan was MYEH FOWEE FLIPR7 KMIA and I just experienced this zig-zagging back and forth for a few minutes after making the turn at FOWEE. I didn't think to check the wind values right at that time but shortly after they were displaying in the MFD as 082/07.

Link to comment
Share on other sites

  • 1 month later...

Seconded, the zigzagging at high altitudes (i.e., the description below) still happens as of the latest version on the marketplace (as of Jul 24).

On 6/13/2021 at 9:35 AM, JRBarrett said:

This initial part of the turn sequence in the sim appears to work just like the real aircraft. It is the final interception of the new course where I sometimes see a problem. What sometimes happens is that the aircraft continues the the turn until arriving at the calculated new heading before the CDI needle is centered, meaning that the aircraft has not yet fully intercepted the new course. It then executes a series of corrective banks to “work its way onto” the new course.

 

The real aircraft will turn at a fixed bank angle until the heading is within approximately 10 to 12 degrees of the calculated new course heading. If the CDI needle is still deflected, (meaning the new course is not yet intercepted), it will then roll level, (and remain level) until the CDI centers, at which time it will execute a final bank to roll out on the new course. 

image.thumb.png.a4bbd68c44bf244295d5c0bed097478c.png

Link to comment
Share on other sites

  • Root Admin

I am not sure of the scale of that image but to be honest, I have seen worse in very modern 737's and I doubt any pilot would worry about that.  These are seriously complex calculations with a lot of factors (air density, wind speed and direction) that are not stable. 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...