On 6/26/2022 at 6:18 PM, Car147 said:

Well since there was an update for the CRJ the other day, i hoped it would solve some of the nav issues it displays during flight. Sadly it still cannot follow a route without constantly zig zagging along the route. Why can't they sort this? Every other plane i own works great!

 

On their forums they said that LNAV will be reworked in the future.

https://forum.aerosoft.com/index.php?%2Ftopic%2F173016-what-is-going-on-with-the-crj-update%2F

Quote

As I promised to be as open as possible, I am afraid to reveal here that there will not be any LNAV related fixes on this update, which I know some of you are eagerly waiting for. We are aware of these issues and plan to work them out eventually. But they require deep dive into the code as well changing some of the fundamentals, and we did not want to delay this update any further.

 

8 minutes ago, -Belga- said:

On their forums they said that LNAV will be reworked in the future.

https://forum.aerosoft.com/index.php?%2Ftopic%2F173016-what-is-going-on-with-the-crj-update%2F

 

For those that choose to deny there is an issue.. Please read on.

Edited June 30, 20223 yr by Car147

Can someone post a route where they see this zig-zagging? I haven't had a chance to try the CRJ yet since the most recent update.

I am a beta tester for the CRJ and a r/w CRJ mechanic and avionics specialist. 

The CRJ can still “zig zag” under certain conditions, but the LNAV behavior is greatly improved over the past year. When it does occur, it is usually happens at high altitudes, (in 1/2 bank mode) with a significant crosswind and a significant course change at a waypoint. We did find and correct one serious bug that specifically happened on courses close to due north with a strong crosswind from the left. That bug would cause constant zig-zagging that would never stop. That was fixed.

The focus in the current update cycle has been on correcting several issues relating to entering and modifying complex SIDS and STARS, and improving the DIR INTC function, but LNAV will be revisited at some point. In general, it works well most of the time. It would not be correct to say that LNAV “always” zig zags, because it definitely does not.

I would point out that the LNAV functionality of the autopilot of the real CRJ is rather “primitive” compared to a Boeing or Airbus. The latter two aircraft can (and do) use a dynamically varying bank angle during turns while the CRJ cannot. The CRJ autopilot can only do two things: it can fly level, or it can turn at one of two fixed bank angles - either 25 or 12.5 degrees. Above 31,500 feet it is limited to only turning at a bank angle of 12.5 degrees. For this reason, the real CRJ will often use an extremely long turn anticipation distance compared to more advanced airliners. At high altitude, the real CRJ may have to begin a course change up to 11 miles before reaching a waypoint if the course change is significant. 

The limitation of a fixed bank angle why no r/w CRJ can ever be certified to fly an approach procedure with complex RNP legs. The real aircraft can definitely overshoot localizer captures if the intercept angle is greater than about 30 degrees and there is a crosswind. Most CRJ pilots will intercept an ILS initially in HDG SELECT if there is significant crosswind. The real aircraft can also have problems with mandatory flyover waypoints, requiring use of HDG SELECT to get back on course efficiently before re-engaging LNAV. Mandatory flyover waypoints (when part of a SID or STAR) seem to be more common in Europe than the US.

I have shared my observations with Hans regarding some of the scenarios where zig-zagging can still happen - especially at high altitude and high airspeed where there is a significant course change. It is on the “to do” list, but an important guideline for software development is “don’t fix one issue, and break something else in the process”. In general, the LNAV function is working well 90 percent of the time, and any future changes will have to be thoroughly tested to avoid unintended consequences. 

 

 

On 6/27/2022 at 1:14 PM, Car147 said:

Really? I think you will find, the usual " We could not replicate it " as the usual cop out.they announced. but not to worry or continue this thread. plenty other decent aircraft to fly, just sadly thought they may have fixed an issue that affects more than just you and the other word not allowed.

Carl you've had several people explain to you a very different experience than you do.  You're the exception so I'll go back to my original guess:  you don't know how to operate it.

3 hours ago, JSmith2112 said:

Can someone post a route where they see this zig-zagging? I haven't had a chance to try the CRJ yet since the most recent update.

KCHA HCH VXV KTRI at FL210. Takeoff Rwy 2. While it's not an automatic, the zig-zag usually happens after the first waypoint HCH.

I just got the newest update 1.0.17 and have not yet tried this route but I always fly it first going back to the P3D versions of the CRJs.

19 minutes ago, Noel said:

Carl you've had several people explain to you a very different experience than you do.  You're the exception so I'll go back to my original guess:  you don't know how to operate it.

You really have a nerve, But hey ho, if you say so lol

And yet again confirmed by the Dev, last post of the thread . Here

Edited June 30, 20223 yr by Car147

36 minutes ago, Car147 said:

You really have a nerve, But hey ho, if you say so lol

And yet again confirmed by the Dev, last post of the thread . Here

I might really have nerve, but you're just flat out obstinate.  Go read JRBarrett's educated post you might learn something.  Until then maybe you should volunteer to beta test the fix:  "The issue is known but not easy to reproduce."  For you apparently it's real easy to reproduce--for me, nope, for most others on the thread, nope, for JRBarrett, up to 10% of the time he can tease out the issue.  Whoopdeedoo.  Go find out why YOU can't make the plane do what it's supposed to do and most everyone else can.   It's a great plane I love it.

12 minutes ago, Noel said:

I might really have nerve, but you're just flat out obstinate.  Go read JRBarrett's educated post you might learn something.  Until then maybe you should volunteer to beta test the fix:  "The issue is known but not easy to reproduce."  For you apparently it's real easy to reproduce--for me, nope, for most others on the thread, nope, for JRBarrett, up to 10% of the time he can tease out the issue.  Whoopdeedoo.  Go find out why YOU can't make the plane do what it's supposed to do and most everyone else can.   It's a great plane I love it.

Utter rubbish and you know it. end of listening to your word not allowed rhetoric

2 hours ago, Twenty6 said:

KCHA HCH VXV KTRI at FL210. Takeoff Rwy 2. While it's not an automatic, the zig-zag usually happens after the first waypoint HCH.

Going to try this now. Is the problem more apparent on one of the specific variants?

2 minutes ago, JSmith2112 said:

Going to try this now. Is the problem more apparent on one of the specific variants?

I use the 700 since forever. Takeoff Rwy 2 and climb at 240kts. Should put you at FL210 about HCH but keep 240kts as it's a sharp turn, then accelerate to 290. Just did the flight and zig zag still there, but seeing as Aerosoft didn't do anything to fix this I didn't expect anything different. Version 1.0.17

43 minutes ago, Twenty6 said:

I use the 700 since forever. Takeoff Rwy 2 and climb at 240kts. Should put you at FL210 about HCH but keep 240kts as it's a sharp turn, then accelerate to 290. Just did the flight and zig zag still there, but seeing as Aerosoft didn't do anything to fix this I didn't expect anything different. Version 1.0.17

Ok, confirmed. I saw quite a bit of slalom action at HCH but it did settle eventually. It overshoots the turn and makes some steep bank changes to try and get back on course.

2 hours ago, Noel said:

I might really have nerve, but you're just flat out obstinate.  Go read JRBarrett's educated post you might learn something.  Until then maybe you should volunteer to beta test the fix:  "The issue is known but not easy to reproduce."  For you apparently it's real easy to reproduce--for me, nope, for most others on the thread, nope, for JRBarrett, up to 10% of the time he can tease out the issue.  Whoopdeedoo.  Go find out why YOU can't make the plane do what it's supposed to do and most everyone else can.   It's a great plane I love it.

So you understand the issue is confirmed in this very thread by a beta tester, but still think Carl fails to operate the aircraft? I'm really starting to worry about your condition.

57 minutes ago, Fiorentoni said:

So you understand the issue is confirmed in this very thread by a beta tester, but still think Carl fails to operate the aircraft? I'm really starting to worry about your condition.

Just let him live in his own little fantasy world…

2 hours ago, Fiorentoni said:

So you understand the issue is confirmed in this very thread by a beta tester, but still think Carl fails to operate the aircraft? I'm really starting to worry about your condition.

Do we know for sure it has nothing to do with how we each operate the CRJ?  I have to think we’re in the same MSFS version, same CRJ version, probably same Windows version.  Different operators.  This does not imply one is doing it "correctly".  In my case just happened upon what it takes to get it to cooperate.  It may well be that I'm blissfully word not allowed--of the way it is supposed to work.  I just know it works for me so when I read a comment like this from Carl or whomever I struggle to imagine what's different:  Carl said,

“I don't think it's a case of it works fine for some but not others., unless i am making a mistake somewhere but i don't think so….”

And yet, here are other replies besides my own in the thread:

“No zig zag for me. LNAV is quite stable….”

“I used to have tons of problems in P3D with the LNAV but I rarely have problems in MSFS”

“i haven't had any problems with the CRJ tracking in LNAV. every so often it might overshoot a turn, requiring a bank to correct back on course, but these are very small and don't happen all the time.”

“not had any issues with it. Granted it maybe a day or two behind on updates.

"As for the zig zagging, this is something I haven't experienced….”

“…same, LNAV stable, VNAV tricky but most likely pilot error 😉”

Etc.

When I look at Carl’s overview I see a difference right away:  my initial climb rate is much lower.  I stay well below 250kts until over 10,000ft, around 220knts tops.  I’m using a flight monitoring software that dings me for exceeding 250 below 10K feet so I avoid that. Often turns are more acute immediately after takeoff so my untrained, total non-RW pilot simmer thought was you don’t want to be moving too fast if you have a sharp turn—just like in any other moving vehicle.  Perhaps this is where is starts falling apart, even if it shouldn’t.  The plane certainly turns just fine at high speeds it’s LNAV tracking we’re talking about.  Perhaps JRBarrett’s comment comes into play:

“I would point out that the LNAV functionality of the autopilot of the real CRJ is rather “primitive” compared to a Boeing or Airbus. The latter two aircraft can (and do) use a dynamically varying bank angle during turns while the CRJ cannot. … and much more.”

I understand some have trouble with it but a whole bunch of others don't and my guess is it's how it's operated, or perhaps even conflicting software like FSUIPC or something making our experiences different.  I think the AS CRJ is a fabulous plane and am sorry to see someone unable to enjoy it.