Jump to content

Zig zag lines


Recommended Posts

2 hours ago, JRBarrett said:

The problem I occasionally (but not always) see, is that the autopilot sometimes rolls out of a turn too soon on a significant course change (before the FMS CDI centers) then works its way over to the new course by a series of bank/level bank/level maneuvers that could be described as “zig zagging”. The real CRJ autopilot by contrast can only use a constant radius turn initially, and will then level out on an intercept heading to the new course and hold that until the CDI centers. (More like the way a localizer intercept works). I am not sure how easy that would be to implement, as the aircraft has to use core MSFS autopilot functions behind the scenes, which is a current limitation one of the WT developers now working on the core sim discussed in a recent post on the MSFS forum.

 

YES.  This is happening all the time for me now (I honestly think more so than prior to update).  

 

Thanks for the IRL explanation on the turn situation I mentioned.

Link to comment
Share on other sites

  • Replies 64
  • Created
  • Last Reply
50 minutes ago, DFWSupertrooper said:

 

YES.  This is happening all the time for me now (I honestly think more so than prior to update).  

 

Thanks for the IRL explanation on the turn situation I mentioned.

If you can give me a full flight plan where the glideslope does not work for you, I will be glad to give it a try.

Link to comment
Share on other sites

23 hours ago, JRBarrett said:

I maintain avionics systems on real CRJs, and the autopilot can indeed begin a course change as much as 11nm before a waypoint. The limits are 7 nm for full bank mode, and 11 nm for half bank mode.
 

The FMS calculation of where a turn should begin is based on airspeed, the bank angle to be used and the number of degrees of course change required. At high airspeed, with a significant course change and limited to a fixed 12.5 degree bank, turn anticipation of as much as 11 miles would not be unusual. 
 

The actual CRJ autopilot is not very sophisticated in its lateral nav functions compared to something like a 737NG. Above 31600 feet, the CRJ can use only one bank angle which is 12.5 degrees, and it is “all or nothing” - the aircraft is either banked 12.5 degrees or it is flying level. It rolls into and out of banks at exactly 3 degrees per second, and that is also a fixed value. Below 31,600 feet it will normally use 25 degrees, but it can use 12.5 degrees if a minor heading change is required.

 

In the sim, the bank values are 30 and 15 degrees.

 

By contrast, a 737 autopilot can use multiple bank angles between 30 and 2 degrees, and it will often change the bank angle dynamically as a turn progresses based on required ground track, meaning that it can use less turn anticipation even on significant course changes.
 

The problem I occasionally (but not always) see, is that the autopilot sometimes rolls out of a turn too soon on a significant course change (before the FMS CDI centers) then works its way over to the new course by a series of bank/level bank/level maneuvers that could be described as “zig zagging”. The real CRJ autopilot by contrast can only use a constant radius turn initially, and will then level out on an intercept heading to the new course and hold that until the CDI centers. (More like the way a localizer intercept works). I am not sure how easy that would be to implement, as the aircraft has to use core MSFS autopilot functions behind the scenes, which is a current limitation one of the WT developers now working on the core sim discussed in a recent post on the MSFS forum.

 

The image below is the flight path of an actual SkyWest CRJ-700 on a flight from Fresno to Phoenix making a left course change at Palmdale VOR. It began the turn 10 miles before the waypoint, and did not fully rejoin the new course until 18 miles past the waypoint. The dashed blue line is the flight plan course (as you would see on the MFD), and the green line is the actual flight path.

 

Unfortunately, I cannot give you an answer on the glideslope issue. This was probably one of THE most intensively tested things I have ever been involved with as a beta tester. I myself sometimes did as many as 10 or 15 ILS approaches during a single session after the fix Hans came up with, and never saw it fail to begin descent or “lose” the descent path, which was a definite problem in the previous version.

 

 

 

 

Please login to display this image.

Thx for the reply, but, for me its not very clear about the zig zag line when the plane should be flying straight. Is this not possible to solve? or the devs are working on it? i ask because it is a bot anoying when you buy a plane and cant do your flight because of some problem like this... i understand that developing something is very very hard, but this problem of zig zag happens in any flight that i'm flying from south to north and, in long straight legs, it happens a lot of zig zags :( please help us at about 2 weeks since i bought the CRJ, i couldnt do a flight without some issue :( 

Link to comment
Share on other sites

2 minutes ago, FHSim said:

Thx for the reply, but, for me its not very clear about the zig zag line when the plane should be flying straight. Is this not possible to solve? or the devs are working on it? i ask because it is a bot anoying when you buy a plane and cant do your flight because of some problem like this... i understand that developing something is very very hard, but this problem of zig zag happens in any flight that i'm flying from south to north and, in long straight legs, it happens a lot of zig zags :( please help us at about 2 weeks since i bought the CRJ, i couldnt do a flight without some issue :( 

Do you have the new update for the CRJ that just came out? There was a confirmed bug in the previous version of the aircraft where the aircraft would bank left and right continuously when flying a heading very close to north with a crosswind. That particular bug was definitely and conclusively fixed in the new update. If you still seeing constant unending zig-zagging on north headings, you either do not have  the new update, or it did not install correctly.

 

Other kinds of “zig zagging” can occur when the aircraft makes a significant course change at a waypoint, levels out too soon, and will make a series of banks to work its way back to the new course. There is a need for improvement here, but the developer is somewhat limited by the amount of custom control can be exerted over the default MSFS autopilot system that has to be used “behind the scenes”.

Link to comment
Share on other sites

4 hours ago, JRBarrett said:

Do you have the new update for the CRJ that just came out? There was a confirmed bug in the previous version of the aircraft where the aircraft would bank left and right continuously when flying a heading very close to north with a crosswind. That particular bug was definitely and conclusively fixed in the new update. If you still seeing constant unending zig-zagging on north headings, you either do not have  the new update, or it did not install correctly.

 

Other kinds of “zig zagging” can occur when the aircraft makes a significant course change at a waypoint, levels out too soon, and will make a series of banks to work its way back to the new course. There is a need for improvement here, but the developer is somewhat limited by the amount of custom control can be exerted over the default MSFS autopilot system that has to be used “behind the scenes”.

i installed the new update to test, i reinstalled the aircraft. The bug in these situations, usually with more than 50kts of crosswinds flying straight north, the plane sometimes goes off the route making a zig zag. The last pics i sent in this topic was just after the update, (i think there´s no way of instaling wrong with the Marketplace but just to make sure i reinstalled the plane from zero) but the problem remain. Flying from LFMN to EDLP it still very zig zaggy, but i did a flight from LFMN to LFPG and i experienced this issue not constant during the flight (of course, i'm not talking about the turns, but straight legs!!!). And i did experienced issues with the FD. I already tested this being with the plane heavy and only with fuel for the route... various cruise levels... the problem remains! But i noticed that it only happens when i'm stablished in cruise, if i'm climbing it does not happen! I hope it can help you guys to solve this annoying issue. 

Link to comment
Share on other sites

On 5/29/2021 at 2:23 PM, JRBarrett said:

If you can give me a full flight plan where the glideslope does not work for you, I will be glad to give it a try.

 

@JRBarrett These were the first three flights using 1.03.  Have not yet flown using 1.04 (just downloaded and hopefully will give it a shot in the morning but not expecting ILS issues to be remedied if they exist since they weren't mentioned):

 

EDDB 25L LURO1N LUROS M725 HDO GOLOP VOZ LANUX RUPET GRZ PETOV PETO2Z 04 LDZA

 

ENVA 27 UTUN1W UTUNA ARTOR ALASA AGATI RIMET T157 KERAX KERA3R 07R EDDF

 

EDDL 05R DODE6Z DODEN Y853 BOMBI ESATI BARSU ENITA PEPIK NIT M748 ARFOX ARFO2H 31R LHBP

 

Thanks for giving it a try and seeing.  Also, to clarify, I simply pressed APPR before intercepting the FAP and prior to 1.03 could generally (not always) get a GS intercept and descent even if it wasn't perfect it was close to green dot the whole way if I was careful with throttle control).  Since 1.03, that descent never happens (LOC lock does), the A/C simply continues on as if GS never intercepted.  If I should change my process (pushing other buttons, etc) please advise.

Link to comment
Share on other sites

2 hours ago, DFWSupertrooper said:

 

@JRBarrett These were the first three flights using 1.03.  Have not yet flown using 1.04 (just downloaded and hopefully will give it a shot in the morning but not expecting ILS issues to be remedied if they exist since they weren't mentioned):

 

EDDB 25L LURO1N LUROS M725 HDO GOLOP VOZ LANUX RUPET GRZ PETOV PETO2Z 04 LDZA

 

ENVA 27 UTUN1W UTUNA ARTOR ALASA AGATI RIMET T157 KERAX KERA3R 07R EDDF

 

EDDL 05R DODE6Z DODEN Y853 BOMBI ESATI BARSU ENITA PEPIK NIT M748 ARFOX ARFO2H 31R LHBP

 

Thanks for giving it a try and seeing.  Also, to clarify, I simply pressed APPR before intercepting the FAP and prior to 1.03 could generally (not always) get a GS intercept and descent even if it wasn't perfect it was close to green dot the whole way if I was careful with throttle control).  Since 1.03, that descent never happens (LOC lock does), the A/C simply continues on as if GS never intercepted.  If I should change my process (pushing other buttons, etc) please advise.

I will fly the EDDB LDZA flight in the morning and report back.

Link to comment
Share on other sites

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.

Please login to display this image.

 
Link to comment
Share on other sites

Mine was like that before the recent updates, now it's behaving very well, with only a slight hint of a correction. All other turns are spot on. I think a lot of issues are just due to everyone's differing PC's  (ie no two are the same) Same with ctd issues, a few months ago now, i was getting constant ctd's, along with some others. Yet a lot of people were getting non in the sim.

Link to comment
Share on other sites

4 minutes ago, Car147 said:

Mine was like that before the recent updates, now it's behaving very well, with only a slight hint of a correction. All other turns are spot on. I think a lot of issues are just due to everyone's differing PC's  (ie no two are the same) Same with ctd issues, a few months ago now, i was getting constant ctd's, along with some others. Yet a lot of people were getting non in the sim.

Out of curiosity which version of the sim do you have Microsoft Store or Steam, also which version of the CRJ do you have, FS Marketplace or Aerosoft? I have the MS Store, and FS Marketplace versions... Honestly, my mfiles are not being recreated either, I don't think this has to do with differing pcs... I think this may have something to do with a potential coding error for the FS Marketplace version.... As Asobo continues to do small coding overhauls on every update. I do agree some problems are potentially different pcs... However in this case where mfiles are not being recreated and the livery for UAL is missing for the 550, I am thinking something might be sus with the FS Marketplace installer per se... 

Link to comment
Share on other sites

4 minutes ago, quasiguru said:

Out of curiosity which version of the sim do you have Microsoft Store or Steam, also which version of the CRJ do you have, FS Marketplace or Aerosoft? I have the MS Store, and FS Marketplace versions... Honestly, my mfiles are not being recreated either, I don't think this has to do with differing pcs... I think this may have something to do with a potential coding error for the FS Marketplace version.... As Asobo continues to do small coding overhauls on every update. I do agree some problems are potentially different pcs... 

MS Store for the sim, and the CRJ was purchased through Aerosoft. I am on a AMD based machine Ryzen 3800x, 32gb DDR4 ram, Radeon XT5700XP GPU 8gb. MSI Pro 570x M/Board. Hope it helps, as i know there's nothing more frustrating.

Link to comment
Share on other sites

2 hours ago, quasiguru said:

Out of curiosity which version of the sim do you have Microsoft Store or Steam, also which version of the CRJ do you have, FS Marketplace or Aerosoft? I have the MS Store, and FS Marketplace versions... Honestly, my mfiles are not being recreated either, I don't think this has to do with differing pcs... I think this may have something to do with a potential coding error for the FS Marketplace version.... As Asobo continues to do small coding overhauls on every update. I do agree some problems are potentially different pcs... However in this case where mfiles are not being recreated and the livery for UAL is missing for the 550, I am thinking something might be sus with the FS Marketplace installer per se... 

 

2 hours ago, quasiguru said:

Out of curiosity which version of the sim do you have Microsoft Store or Steam, also which version of the CRJ do you have, FS Marketplace or Aerosoft? I have the MS Store, and FS Marketplace versions... Honestly, my mfiles are not being recreated either, I don't think this has to do with differing pcs... I think this may have something to do with a potential coding error for the FS Marketplace version.... As Asobo continues to do small coding overhauls on every update. I do agree some problems are potentially different pcs... However in this case where mfiles are not being recreated and the livery for UAL is missing for the 550, I am thinking something might be sus with the FS Marketplace installer per se... 

 

 

I just found out that the M files do indeed exist in the Marketplace version. Beginning with SU4, They are now pre-compiled by Microsoft before installation, (rather by the sim itself on first run), encrypted, and put somewhere in the Aerosoft folder in OneStore. So apparently, deleting then recreating the M files as a troubleshooting step will no longer be possible in the Marketplace version.

Link to comment
Share on other sites

2 minutes ago, JRBarrett said:

 

 

I just found out that the M files do indeed exist in the Marketplace version. Beginning with SU4, They are now pre-compiled by Microsoft before installation, (rather by the sim itself on first run), encrypted, and put somewhere in the Aerosoft folder in OneStore. So apparently, deleting then recreating the M files as a troubleshooting step will no longer be possible in the Marketplace version.

I did check the One Store Aerosoft-CRJ directory and sub directories... They are absolutely nowhere to be found with hidden files on... Wondering if they stuffed them elsewhere??? Heh

Link to comment
Share on other sites

On 5/30/2021 at 8:38 PM, JRBarrett said:

I will fly the EDDB LDZA flight in the morning and report back.

@DFWSupertrooper I finally had the chance to try the EDDB to LDZA flight. I really need to do more flights in Europe - the scenery is beautiful.

 

I did use the clear sky weather preset to insure I would not have a tailwind or excessive crosswind while testing the approach.

 

As I expected based on all my testing over the past few weeks, there was no problem with the GS capture on the LDZA ILS04 approach. I descended via the STAR as published. On the base leg from ZA601 to EKSON, I descended to 4,000 feet, and 180 knots with flaps 8. The radio auto-tuned to the localizer frequency. My own personal preference is to switch to green needles manually with the Nav Source dial, rather than using Nav-To-Nav, and I did so this time after the aircraft made the left turn at EKSON still in FMS NAV mode.  
 

When I went to green needles, the localizer was already centered, so I pressed APP, which gave an immediate green LOC1 and a white GS since I was still below the glideslope. I decelerated to 170 knots and flaps 20. When the GS was one dot above center I extended the gear, went to flaps 35 and 160 knots. The GS captured and the aircraft descended with no problem. At the OM I went to flaps 45 and set power to maintain 140 knots. The GS tracked perfectly to decision height, at which point I disconnected the AP and landed.

 

Unfortunately, I really can’t give you any clues or suggestions as to why the aircraft will not descend on the GS in your installation. The technique described above is how I fly almost every ILS, and it always works. I could try your other flight plans, but I don’t think the results would be any different.

Link to comment
Share on other sites

I understand why this would be frustrating to deal with. I've encountered this a few times prior to the update but I haven't had any issues since the update (in very limited use though). But in the mean time until the issue is improved, the best thing to do is to go into heading mode and turn the plane onto a heading that will re-intercept the desired course. Once the wings are level, re-engage NAV mode and it should re-intercept the course nicely.

 

@JRBarrett did a beautiful job of explaining why the plane is having issues maintaining a track after a large turn due to turn anticipation. Going into heading mode and putting the plane on an intercept course resets the turn anticipation and should help make things run a little smoother.

 

I know it's not ideal but it's better than nothing. I've had to do this in the real plane before so it's not unheard of. 

Link to comment
Share on other sites

12 hours ago, JRBarrett said:

@DFWSupertrooper I finally had the chance to try the EDDB to LDZA flight. I really need to do more flights in Europe - the scenery is beautiful.

 

I did use the clear sky weather preset to insure I would not have a tailwind or excessive crosswind while testing the approach.

 

As I expected based on all my testing over the past few weeks, there was no problem with the GS capture on the LDZA ILS04 approach. I descended via the STAR as published. On the base leg from ZA601 to EKSON, I descended to 4,000 feet, and 180 knots with flaps 8. The radio auto-tuned to the localizer frequency. My own personal preference is to switch to green needles manually with the Nav Source dial, rather than using Nav-To-Nav, and I did so this time after the aircraft made the left turn at EKSON still in FMS NAV mode.  
 

When I went to green needles, the localizer was already centered, so I pressed APP, which gave an immediate green LOC1 and a white GS since I was still below the glideslope. I decelerated to 170 knots and flaps 20. When the GS was one dot above center I extended the gear, went to flaps 35 and 160 knots. The GS captured and the aircraft descended with no problem. At the OM I went to flaps 45 and set power to maintain 140 knots. The GS tracked perfectly to decision height, at which point I disconnected the AP and landed.

 

Unfortunately, I really can’t give you any clues or suggestions as to why the aircraft will not descend on the GS in your installation. The technique described above is how I fly almost every ILS, and it always works. I could try your other flight plans, but I don’t think the results would be any different.

 

@JRBarrett I appreciate your feedback on this and I'll keep trying, paying close attention to your described procedure.  I rarely had issues before 1.03/4 with ILS and if I captured, I always got the A/C to descend along GS, so i's definitively weird that I can't seem to now, but...alas.  It may be something to do with my Bravo throttle.  I noticed I can still select the VS using the dial while in GS mode to control the descent FPM, and I could not before (before 1.03 once in GS mode, the dial wouldn't cause any change to the descent, now it does).  Thanks bud!

Link to comment
Share on other sites

13 hours ago, JRBarrett said:

@DFWSupertrooper I finally had the chance to try the EDDB to LDZA flight. I really need to do more flights in Europe - the scenery is beautiful.

 

I did use the clear sky weather preset to insure I would not have a tailwind or excessive crosswind while testing the approach.

 

As I expected based on all my testing over the past few weeks, there was no problem with the GS capture on the LDZA ILS04 approach. I descended via the STAR as published. On the base leg from ZA601 to EKSON, I descended to 4,000 feet, and 180 knots with flaps 8. The radio auto-tuned to the localizer frequency. My own personal preference is to switch to green needles manually with the Nav Source dial, rather than using Nav-To-Nav, and I did so this time after the aircraft made the left turn at EKSON still in FMS NAV mode.  
 

When I went to green needles, the localizer was already centered, so I pressed APP, which gave an immediate green LOC1 and a white GS since I was still below the glideslope. I decelerated to 170 knots and flaps 20. When the GS was one dot above center I extended the gear, went to flaps 35 and 160 knots. The GS captured and the aircraft descended with no problem. At the OM I went to flaps 45 and set power to maintain 140 knots. The GS tracked perfectly to decision height, at which point I disconnected the AP and landed.

 

Unfortunately, I really can’t give you any clues or suggestions as to why the aircraft will not descend on the GS in your installation. The technique described above is how I fly almost every ILS, and it always works. I could try your other flight plans, but I don’t think the results would be any different.

And what about the zig zags? i still have it :( 

Link to comment
Share on other sites

4 minutes ago, FHSim said:

And what about the zig zags? i still have it :( 

 

I still get some but for whatever reason, my one flight yesterday they were not as pronounced (the A/C tracked better).  About to fly some more and will report back.

Link to comment
Share on other sites

3 hours ago, FHSim said:

And what about the zig zags? i still have it :( 

 

I just flew KIAD-KFLL and didn't have major issues with zig zags, aside from initial climbout on a vector approach where I DIR to the first waypoint en route and the A/C had to make a major turn to intercept - it was a little sloppy, but the remainder of the flight (like all others thus far) the A/C tracks much better laterally.  Overall, the flying feels much more realistic (as much as I know anyway - what I would expect as a passenger). 

 

The only issues for me are the known bug with programming an RNAV approach (which has been confirmed and being looked into [reference in Systems]) and the afore mentioned GS change.  I can capture the GS fine, but I have to use VS knob to keep it on the actual path (this could be the Bravo not the A/C itself - there are some weird bugs using the selector dials some times but once they're identified generally able to be worked around such as the GS for me).

 

Overall 1.04 is a marked improvement over prior to 1.03 in terms of climb and cruise. 👍

Link to comment
Share on other sites

37 minutes ago, DFWSupertrooper said:

 

I just flew KIAD-KFLL and didn't have major issues with zig zags, aside from initial climbout on a vector approach where I DIR to the first waypoint en route and the A/C had to make a major turn to intercept - it was a little sloppy, but the remainder of the flight (like all others thus far) the A/C tracks much better laterally.  Overall, the flying feels much more realistic (as much as I know anyway - what I would expect as a passenger). 

 

The only issues for me are the known bug with programming an RNAV approach (which has been confirmed and being looked into [reference in Systems]) and the afore mentioned GS change.  I can capture the GS fine, but I have to use VS knob to keep it on the actual path (this could be the Bravo not the A/C itself - there are some weird bugs using the selector dials some times but once they're identified generally able to be worked around such as the GS for me).

 

Overall 1.04 is a marked improvement over prior to 1.03 in terms of climb and cruise. 👍

But KIAD to KFLL is flying to south, the issue usually happens flying north...

Link to comment
Share on other sites

36 minutes ago, FHSim said:

But KIAD to KFLL is flying to south, the issue usually happens flying north...

 

Fair enough...when I get back from my meeting, this afternoon, I'll fly a S-N route and see what happens, and let you know what I observe.

Link to comment
Share on other sites

33 minutes ago, FHSim said:

But KIAD to KFLL is flying to south, the issue usually happens flying north...

The specific issue of continuous roll oscillations on headings very close to north with a crosswind was definitely fixed. I was directly involved in testing that. We were able to duplicate the issue consistently. Hans found the exact bug in the code that was causing it, and after the fix, the aircraft would hold a north heading even with a 100 knot crosswind.

 

Again, this relates only to the specific case of a course very close to due north with a crosswind. “ZigZags” that occur at other times or scenarios likely have a different cause.

Link to comment
Share on other sites

1 hour ago, JRBarrett said:
 

The specific issue of continuous roll oscillations on headings very close to north with a crosswind was definitely fixed. I was directly involved in testing that. We were able to duplicate the issue consistently. Hans found the exact bug in the code that was causing it, and after the fix, the aircraft would hold a north heading even with a 100 knot crosswind.

 

Again, this relates only to the specific case of a course very close to due north with a crosswind. “ZigZags” that occur at other times or scenarios likely have a different cause.

Well... or the problem is not solved 100% as you guys think... or there is something wrong with the product downloaded throught Marketplace! Having this in mind, would be great having a review from Aerosoft on the files that Microsoft provides inside the sim, because something is very very wrong :( 

Link to comment
Share on other sites

3 hours ago, JRBarrett said:
 

The specific issue of continuous roll oscillations on headings very close to north with a crosswind was definitely fixed. I was directly involved in testing that. We were able to duplicate the issue consistently. Hans found the exact bug in the code that was causing it, and after the fix, the aircraft would hold a north heading even with a 100 knot crosswind.

 

Again, this relates only to the specific case of a course very close to due north with a crosswind. “ZigZags” that occur at other times or scenarios likely have a different cause.

 

I can concur.  I've see no oscillation at cruise.  The flight I'm on right now, KSFO-CYLW at FL340, current leg RBL-LKV (heading 015): ZERO issues with oscillation.  Beautiful departure and climb out actually as well. 👍

 

To the comment of @FHSim, I am using the store-downloaded version, not  Marketplace, so I dunno...

Link to comment
Share on other sites

36 minutes ago, DFWSupertrooper said:

 

I can concur.  I've see no oscillation at cruise.  The flight I'm on right now, KSFO-CYLW at FL340, current leg RBL-LKV (heading 015): ZERO issues with oscillation.  Beautiful departure and climb out actually as well. 👍

 

To the comment of @FHSim, I am using the store-downloaded version, not  Marketplace, so I dunno...

Maybe the marketplace has some bug in the files that is getting downloaded... 

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