Autopilot oscillations

[HaziDesigns]

Hi, when in VS or VNAV, the plane climbs past and descends past the target crz alt by 300-400 feet at a time, and doesnt stay at crz alt, and power is continuedly pushing forward and back. as you can see in the pic, my CRZ is FL240, ye tits at 2000 VS and climbing. Both the FMC and MCP are set at FL240

[Litjan]

I would need to see the whole panel - well, at least the EADI (with the green writing on top) to begin at guessing what you did wrong ;-)

Cheers, Jan

 

[HaziDesigns]

1 hour ago, Litjan said:

I would need to see the whole panel - well, at least the EADI (with the green writing on top) to begin at guessing what you did wrong ;-)

Cheers, Jan

 

here you go

Edited December 23, 2020 by HaziDesigns

[HaziDesigns]

2 hours ago, Litjan said:

I would need to see the whole panel - well, at least the EADI (with the green writing on top) to begin at guessing what you did wrong ;-)

Cheers, Jan

 

what it is, is its not following the flight directors properly, im not sure how to fix it

[Litjan]

Ok - I don´t know why your power is "pushing back and forward" because your autothrottle is completely off.

I see the flightdirector bars being completely centered - this is odd, because it looks like the plane is following its commands (the flight-director bars are actually always getting centered by the autopilot, because it flies exactly like the bars indicate).

Did you "push" the ALT HOLD button? You have to be aware that the ALT HOLD command will maintain exactly the altitude at which it was pressed - not the altitude that is selected in the window. If you push it in a rapid descent (or climb) it will "go back" to that exact altitude, but it will take a second because otherwise the acceleration loads are too uncomfortable for the passengers.

Cheers, Jan

 

 

[HaziDesigns]

2 minutes ago, Litjan said:

Ok - I don´t know why your power is "pushing back and forward" because your autothrottle is completely off.

I see the flightdirector bars being completely centered - this is odd, because it looks like the plane is following its commands (the flight-director bars are actually always getting centered by the autopilot, because it flies exactly like the bars indicate).

Did you "push" the ALT HOLD button? You have to be aware that the ALT HOLD command will maintain exactly the altitude at which it was pressed - not the altitude that is selected in the window. If you push it in a rapid descent (or climb) it will "go back" to that exact altitude, but it will take a second because otherwise the acceleration loads are too uncomfortable for the passengers.

Cheers, Jan

 

 

when the pic was taken, i had turned off the AT due to the weird acts of it, and i pushed hold as i was in a GA and vnav wouldnt hold it there. Im in the middle of another flight now and all is fine atm, but at the start it was doing the up and down, and i noticed that the FD's would move up, then a few seconds later, the plane would follow, but by the time the plane had got to the correct envelope, the FD's had gone down to level, causing the plane to do a big dip, then due to that, the FD's shot up, followed by thr plane a few seconds later etc etc 

[Litjan]

Hmm, this sounds odd. Note that you should not use VNAV if you experience any problems with it - it does not work right in some circumstances (and never in a missed approach).

Make sure that your framerate is sufficiently high (> 25fps) to avoid autopilot oscillations.

Other than that you can also watch my tutorial videos on how to operate the autopilot system correctly - since you are the only one reporting these problems I am inclined to think that it must be something out of the ordinary you are doing ;-)

Cheers, Jan

 

[HaziDesigns]

6 minutes ago, Litjan said:

Hmm, this sounds odd. Note that you should not use VNAV if you experience any problems with it - it does not work right in some circumstances (and never in a missed approach).

Make sure that your framerate is sufficiently high (> 25fps) to avoid autopilot oscillations.

Other than that you can also watch my tutorial videos on how to operate the autopilot system correctly - since you are the only one reporting these problems I am inclined to think that it must be something out of the ordinary you are doing ;-)

Cheers, Jan

 

thank you  im using the AP in the same was as the -800, also the FD issue was the same with VS too

[Litjan]

5 minutes ago, HaziDesigns said:

im using the AP in the same was as the -800

I was never rated on the 737NG but I know there are some substantial differences between the NG and the Classic as far as the autopilot/flight-director system is concerned. Especially with regard to VNAV and but also some things like "arming" LNAV before takeoff and others...

Cheers, Jan

 

[HaziDesigns]

Just now, Litjan said:

I was never rated on the 737NG but I know there are some substantial differences between the NG and the Classic as far as the autopilot/flight-director system is concerned. Especially with regard to VNAV and but also some things like "arming" LNAV before takeoff and others...

Cheers, Jan

 

ill defo give your video a watch, thank you!  

[Litjan]

 

 

 

Edited December 23, 2020 by Litjan

[Torbinator]

Hi Jan, I have been having this [relatively new] oscillation issue specifically when using the VNAV and often in LNAV based on the current IXEG v1.33, and maybe since v1.30 onward? Switching to HDG, LVL CHG, and VSI modes tends to smooth everything out. I know this wasn't an issue in prior versions. I am providing two video clip examples here.

I actually planned to report it sooner, but quite often I have been doing my own interventions since I do a lot of online flying with PE/Vatsim ATC, and they often ask me to use other altitudes in the STAR. Since the VNAV T/D has stopped being able to re-calculate since V1.32 (?), I haven't been able to rely on the T/D.

Anyway, today I did happen to let the VNAV do it's thing, and here are two clips that shows what I've been experiencing. The winds were not gusting, nor was there significant turbulence. Perhaps the OP is speaking about this? In this first clip you'll see the pitch command bar, the VSI, and the green vertical predictor on the ND all moving in response to the airframe pitching up & down: 

https://www.twitch.tv/torbinator97/clip/UnsightlyAdorableCardChefFrank

Here it is in LNAV. It's a bit subtle in this example sine it was a relative short heading change, but you'll see the roll oscillates about 3°, the roll command also shows left/right movement, and the airplane/horizon movement. If the turn was longer, it would have continued throughout most of the turn.

https://www.twitch.tv/torbinator97/clip/PlausibleTrappedDonutKappa

Hope these help!

Edited December 24, 2020 by Torbinator

[Torbinator]

BTW, I hope it was OK to piggyback on this thread, I now realize it may have been describing a different issue from @HaziDesignsdescription. Either way I hope my report helps, let me know if I should copy paste it into a separate topic. Thank you!

Edited December 24, 2020 by Torbinator

[Litjan]

Hi Torbinator,

yes, I can see those oscillations...

The pitch oscillations are new to me - I have not seen those before...and really have no idea where they are coming from, especially if you say that they disappear in other autopilot modes. The VNAV pitch mode targets the the vertical deviation bar - and that one is not moving (at least I can´t see it in your video), so I don´t know where that input is coming from. It would be interesting to see where the stabilizer position (trim) is. It could be that it is at the limit of autopilot trim, so the autopilot keeps pushing against it?

The LNAV oscillations in turn are looking similiar. We used to have a bit of those since turns are really just a segment of straight lines, we tried to dampen that out, though. I don´t recall seeing it this obvious on my end, I have to check again (sometimes one gets blind to these things!).

Thanks for your report, Jan

 

[HaziDesigns]

Hi, sorry to continue the thread but i followed all tuts but it still happens (doesnt follow the FD asap)

[Litjan]

Ok, does it follow the FD eventually? You wrote "asap" - there is sometimes a bit of delay between the plane reacting to the commands of the FD...

[HaziDesigns]

the pink lines will go up or down, then a few seconds later the plane will dramatically follow, climbing or descending at sometimes 6000fpm to catch up

Edited December 28, 2020 by HaziDesigns

[Litjan]

Hmm, I really have no idea what could be causing it, sorry. It seems to work for pretty much everyone else :-(

[Diogo.jmr]

Hello! I´m having the same problem with pitch oscilations.. did you solve the problem?

[Litjan]

No, the reason for this happening were never discovered or reproduced.