reported Su-27 ACS Landing Mode broken

[Ironhand]

This has been going on for a long time, perhaps since the introduction of the PFM. Maybe before. At any rate, the ACS Landing Mode is mostly broken, often leading to severe oscillations in the pitch channel that make landing impossible. I've attached a TRK file demonstrating an attempt to land at Mozdok.

There is no wind in the mission. I approach the final waypoint and some of the Return Mode using what the sim refers to as "Route Following Mode" (LAlt+6) alone because I want to stay at roughly the insertion altitude for the glidepath. If I had used "A" or had added "BaroAlt Hold" the AP would have added some extreme altitude changes to confuse the issue after the switch to Return Mode. Once Return Mode commanded the altitude drop to the glidepath insertion altitude, I add "Barometric Altitude Hold" (LAlt +4). Lowered my landing gear. Lowered my flaps. Hit the glidepath insertion and the oscillations began and got worse. Eventually, the system directed me back to the mid-insertion point and things got even more exciting from there.

I've tried landing on 4 of the 6 PRGM equipped runways--the default runways for Mozdok and Krymsk and the similarly equipped runways at Kransnodar Center and Maykop. Not all of the runways yield similar results. For some reason, I seem to have much better luck landing at Maykop.

Also want to mention that 100% of the time, on the occasions that the system works without extreme oscillations, the system holds you above the glidepath.

Flanker AP Landing Test PRGM-Mozdok-Fail2.trk

[Flappie]

From my point of view, it's the speed control that goes AWOL: I took control over your track, I enabled Alt+6 mode, then I raised the speed whenever the aircraft would go down. I lowered the flaps, then the gear, then I was able to land without touching the stick during the whole flight. The only problem when land is the speed, which is way too high : my tyres explode!

Before I report this, do we agree on the diagnosis?

Su-27 overspeed Alt6.trk

[Ironhand]

13 hours ago, Flappie said:

From my point of view, it's the speed control that goes AWOL: I took control over your track, I enabled Alt+6 mode, then I raised the speed whenever the aircraft would go down. I lowered the flaps, then the gear, then I was able to land without touching the stick during the whole flight. The only problem when land is the speed, which is way too high : my tyres explode!

Before I report this, do we agree on the diagnosis?

Su-27 overspeed Alt6.trk 616.29 kB · 0 downloads

LAlt+6 is route following without altitude control. It simply keeps you pointed at the waypoint or, in landing mode, the runway centerline. You climb or descend based on your throttle setting. If that is how it now is meant to work, then there's no issue. I suppose you should ask.

OTOH, Landing Mode had always, in the past, been done with Barometric Altitude hold engaged, what you get with the "A" key. The combination brought you smoothly down the glideslope. In my testing, that works most of the time, if you keep your landing gear and flaps stowed. Using either the "A" key or "LAlt+6" with "LAlt+4" (route following and barometric altitude hold both engaged) works equally well. It seemed to be the extension of the flaps especially that really screwed things up.

Edited May 21, 2024 by Ironhand

[Flappie]

21 minutes ago, Ironhand said:

If that is how it now is meant to work, then there's no issue.

No, I'm sure you are right. Since DCS is a thing, I don't fly FC modules anymore. Please pardon my lack of FC knowledge.

I'll try your procedure (and I'll RTFM too, that helps ).

Edited May 20, 2024 by Flappie

[Ironhand]

@Flappie

Here's the TRK file I posted above but, this time, I take over prior to where I had lowered my landing gear and leave everything alone. Notice how smooth the transition is. This is done with Nav, Auto, and Baro Alt hold selected on the ACS. A simple pitch down to start down the glide path and, then, none of the oscillations that occurred when the landing gear and flaps were extended in the original TRK. I lower the landing gear part way down the glide slope without so much as a wobble. Later, I extend the flaps. All is well until right around 1 km out, when the system suddenly commands an unnecessary pitch up at which I disengage the AP and land. That pitch up with the flaps extended should not have happened. Nor should the oscillations occur when the gear and flaps are extended from the start as they were in the original track.

Also notice that in both tracks, the AP kept me above the glideslope rather than exactly on it. That's another, separate issue.

Flanker AP Landing Test PRGM-Mozdok-Late Gear&Flaps.trk

Edited May 21, 2024 by Ironhand

[draconus]

From the DCS manual p. 44:

Quote

 

ACS provides:

...

Automatic control of climbing and descending

...

Automatic return to the programmed airfield and landing approach down to the altitude of 50-60 m in accordance with input from the navigation system

 

So it might be that this system is not supposed to take you safely all the way to touchdown.

Also is it part of the procedure to turn on barometric altitude hold? Doesn't make sense if we want it to descend down the GS.

[Ironhand]

40 minutes ago, draconus said:

So it might be that this system is not supposed to take you safely all the way to touchdown

It's not. You're supposed to take manual control no later than 50-60 meters above the ground. But that's not the issue. The issue is the extreme oscillation you often encounter in the sim.

In the real world manual do I not see an instruction to turn off Barometric Altitude hold prior to hitting the transition from Return to Landing mode when using the ACS in full automatic mode. So I think it's supposed to work as it used to in the sim, and still does, if you wait to extend your landing gear and, especially, your flaps. But, like I said earlier, maybe Flappie should ask the question.

Edited May 21, 2024 by Ironhand

[Volator]

On 5/21/2024 at 3:31 PM, draconus said:

Also is it part of the procedure to turn on barometric altitude hold? Doesn't make sense if we want it to descend down the GS.

Hope you don't mind me jumping in here, just want to say that IRL flying an autopilot ILS approach is initiated with altitude hold as well, and once the glideslope is captured, the autopilot automatically switches from altitude hold to glideslope and rides it down nicely, so assuming PRMG basically works similarily, it would make sense to fly to the glideslope capture point with altitude hold. 

Edited May 24, 2024 by Volator

[Ironhand]

21 hours ago, Volator said:

Hope you don't mind me jumping in here, just want to say that IRL flying an autopilot ILS approach is initiated with altitude hold as well, and once the glidepath is captured, the autopilot automatically switches from altitude hold to glidepath and rides it down nicely, so assuming PRMG basically works similarily, it would make sense to fly to the glidepath capture point with altitude hold. 

Thanks for that.   ED has been modeling that as the method since the early 2000s (?) even before they decided to add the PRGM window dressing on the HUD and HSI for the Russian FC3 aircraft. The fact is that, with those aircraft, you can make a precision landing on any ILS or PRGM equipped runway.

Edited May 22, 2024 by Ironhand
edited for clarity

[Flappie]

I've re-tested with Alt+4 and Alt+6 altogether, and now I see what you're seeing.

I think the problem does not only affect landing. If you set a series of waypoints with different altitudes, you'll see the problem too. The aircraft is unable to stay at given altitude. If the new WP says "3000 m", it will go down to 2940 m then will stabilize at 2980 m. Once or twice it will go up, until it reaches 3020 m, then it will go down again and stabilize at 2980 m again.

The landing mode makes it only worse since PRGM causes the wanted altitude to change constantly.

Issue reported.

EDIT: the Mi-29 is also affected.

 

Su-27 ACS.trk

Edited June 3, 2024 by Flappie

[Ironhand]

Thank you.
 

I’ll download your trk as soon as I am able to.

Edited June 2, 2024 by Ironhand

[Ironhand]

Watched your trk and I see why you concluded what you did. It's interesting that, while in Enroute mode, barometric stabilization keeps you below the required altitude. OTOH, while in landing mode, it keeps you above the required altitude. You won't notice the latter, of course, unless you don't extend your landing gear and flaps. It's having the aircraft in landing configuration that causes the oscillations.