AUTO STEP CLIMB oscillation

[19AB67 16]

Hi folks, 

I just completed QFA16 KLAX-YBBN with the following flight plan. 

LAXX9 SLI V21 SXC  ROSIN  FICKY B577 BELAN 16N42  12N48  

WAYSE  05N58  02N62  04W70 05W71  IDEMU  12W80  16S74  ELSAX  

21S66 ISTEM B578 MULET Q93 SAVER SAVER1P  

SIM BRIEF suggested FL STEPS KLAX/0280/BELAN/0300/05N58/0320/IDEMU/0340/ELSAX/0360

and I programmed them into the FMS as 280S etc. 340S including ELSAX. 

After ELSAX the tiny capital font suggested FL380 and I left it like this until TOD (not FL360). 

I left the cockpit with auto step climb ON for a while, came back, the bird was nicely cruising at FL380,

completed the flight and checked the FSX Flight Analysis to watch my manoeuvering around some thunderstorms mid way.

Now to the topic's title: The altitude plot below the FSX chart showed 

that the final step seems to have happened by 4 (sic!) climbs to FL380, but also 3 descends back to FL340. 

Judging from the plot the oscillation was not dramatic, but as if the auto step function set back to 340, 380 and so on. 

Has anyone experienced this behaviour as well? 

Thanx in advance. 

[downscc 1,327]

Do not use a hobby-grade flight planner to set step climbs. I have programmed steps where appropriate such as NATS crossing and never had a problem. Stick around next time and observe the flight, this gives you an opportunity to investigate if something odd happens; however trying to do it postpartum (does that work?) is not going to reveal anything.

[19AB67 16]

Hi Dan, 

On 11/04/2017 at 0:48 AM, downscc said:

Do not use a hobby-grade flight planner to set step climbs. I have programmed steps where appropriate such as NATS crossing and never had a problem. 

Yup, on NATS also program sticky flight levels for the whole crossing via (e.g.) 310S. 

Which flight planner do you use? 

On 11/04/2017 at 0:48 AM, downscc said:

Stick around next time and observe the flight, this gives you an opportunity to investigate if something odd happens; however trying to do it postpartum (does that work?) is not going to reveal anything.

Right, will check the function witnessing the auto-step. 

[downscc 1,327]

3 hours ago, 19AB67 said:

Which flight planner do you use? 

I use PFPX, which has its quirks but uses current navdata from Navigraph, same as PMDG, and has good ETOPS and Re-Dispatch features.  It includes performance data on most popular types for reasonable time and fuel burn plans... it's not good enough to ignore the PMDG FMS but is certainly worth learning its use.  Some say it's too complex for their needs... it's certainly not as complex as a commercial quality dispatch tool but I guess it's not as easy as free on-line flight planners.

[19AB67 16]

Hi folks, 

I am just on the way as 5X58 PHNL to PGUM, left the bird with auto step climb ON, had some steps programmed 320S and 340S along my way on A450. 

When I came back I was cruising at FL340 and non-waypoint FMC chosen S/C to FL360 proposal was some miles ahead. 

As usual some miles before this indication the FMC flipped to NOW, the altitude  window showed 36000 and the aircraft started to climb to FL360. 

We had reached FL360 when the A/P altitude window flipped back to 34000 and we descended again to FL340. 

After landing I check how often the FL oscillation took place during my absence (watched the movie "Going in Style"...)

Maybe somebody can test the behaviour as well? 

I think it is ok that she climbs after the programmed step climbs once or twice again along the OPTimum FL,

but she should descend again, should she? 

[downscc 1,327]

If you are going to use manual step climbs I recommend you set the step to zero in the FMS and only use manual step climbs.  I've always seen FMC derived steps missing after a manually inserted step, kinda surprised that you had the FL360 step in there to begin with.

FCOM 11.31.23: Entry of a step climb overrides the FMC predictions.  Yeah, don't know why it continued to give you predicted steps.  Keep that step size set to zero and you ought to be okay.  Just my opinion here, but I do not see a reason to use manual steps in the Pacific.

[19AB67 16]

18 minutes ago, downscc said:

If you are going to use manual step climbs I recommend you set the step to zero in the FMS and only use manual step climbs. 

Yes, that's true. I had set it to 2000. 

18 minutes ago, downscc said:

kinda surprised that you had the FL360 step in there to begin with.

You mean that might be a bug in the simulation? 

The PMDG auto step function should go up, but never down, should it? 

20 minutes ago, downscc said:

Keep that step size set to zero and you ought to be okay. 

Will do next flight with programmed steps. 

(BTW: 9 ups and downs ... then I switched the auto step function off.)

20 minutes ago, downscc said:

Just my opinion here, but I do not see a reason to use manual steps in the Pacific.

You mean crossing the Pacific is not organised as the NAT area, where you keep a given FL for the whole crossing? 

[downscc 1,327]

Yeah, the appearance of the FL360 FMC derived step climb after using manual steps is probably an error. I encourage you to submit a trouble ticket on this and get it into the bug tracking system.

Step functions can go down, such as leaving metric airspace, but not a FMC derived step climb.

Crossing the Pacific is not like NATs in that step climbs are restricted.  You are still going to ask Oakland or whomever for climb clearance before the climb, but except for busy periods on the West Coast / Hawaiian routes there is unlikely to be any contention.  You don't need to fix your speed in the Pacific either, just keep track of the position reports to make sure you are pretty close to the ETAs crossing fixes.  Timing is important.

[19AB67 16]

Trouble ticket opened. 

I keep you posted. 

Hello PMDG, 
thank you for this great simulation 747 Queen of the Skies II. 
Sometimes I use the auto cruise -- auto step climbs function. 
Doing so in combination with programmed step climbs via e.g. 320S, 340S on the LEGS page, the function works fine. However, I experience a flight level oscillation after the last programmed step: 
Along further FMC proposed steps (with e.g. STEP SIZE 2000) the aircraft climbs futher, e.g. FL360 with 36000 in the A/P altitude window. 
1. In the forum a member claimed that any programmed step (FL320S etc) should invalidate FMC proposed step climbs. (Only the 777 offers E/ to allow further steps after defined ones.) Is this the true 747 behaviour? Or must STEP SIZE be set to "0"? 
2. What puzzles even more is that after a successful climb to e.g. FL360, the PMDG auto step function flips back to e.g. FL340 (i.e. 34000 in the A/P altitude window) and the aircraft descends again. Assuming that this is not the intention of the auto-step function, this might be a minor bug, is it not? 
3. This up and down happens again and again, so that the VIEW FLIGHT ANALYSIS graphics of FSX show a flight level oscillation, in my example between FL340 and FL360. 
(See also https://www.avsim.com/forums/topic/508397-auto-step-climb-oscillation/)
Thanks, 
Andreas Berg