Discrepancies in the Su-27 longitudinal fly-by-wire system (SDU-10)

[DummyCatz]

I'll be quoting a publicly available SDU-10 block diagram to elaborate two discrepancies in the implementation of Su-27 FBW (SDU) control logic. The document is titled as 苏－２７飞机电传操纵系统 (Su-27 aircraft fly by wire system), published in 1993 at 第五届飞行控制及操纵学术交流会 (The 5th Flight Control Academic Exchange Conference) held in China. Link below:

https://www.zhangqiaokeyan.com/academic-conference-cn_meeting-35806_thesis/020221359205.html

The diagram in question is Figure one and Figure two, which is obviously written in Russian and I'll explain in detail later.

The first issue is about the aircraft behavior when pressing the 'S' key, which puts the FBW into DIRECT mode. In DCS, there's a noticeable pitch transient when switching the mode back and forth. But it shouldn't, as switching to DIRECT mode only removes a lag filter in the middle portion of the diagram, and the pitch rate and Ny feedback in the right portion of the diagram, which shouldn't be causing any pitch transient if flying straight and steady.

In DCS it just looks like the open-loop trim schedule (lower left corner of the diagram) is completely removed when in DIRECT mode, which it shouldn't as it's added to the stick input, without any switches.

Track for the first issue below:

su27 direct link control test.trk

 

The second issue is about the stick limiter system (OPR) control logic.

As shown in the diagram, the integration (1/p) block processes the inputs from the control stick position, AOA and Ny feedback.
When AOA or Ny is within safe limits, the feedback to the integrator results in a continuous adjustment of the OPR actuator rod's position, allowing the stick limiter to move along with the stick, with a gap between them. If the stick moves fast enough, it will also hit the rod, as the OPR actuator rod position is integrated at a limited rate.

As AOA or Ny approaches its limit, the feedback signals to the integrator decrease the rate of integration.
If AOA or Ny reaches the limit, the feedback effectively stops the integration rate, causing the OPR actuator rod to halt in its current position. This creates a physical stop for the control stick, and also the vibration is transmitted to the stick.
If AOA or Ny exceeds the limit, the rod moves forward and pushes the stick forward.

Now, the issue in DCS is that there's no gap between the stick limiter and the control stick whenever the stick is pulled back a little bit. So the stick will always hit the stop and the vibration will always be transmitted to the stick. IRL it should be moving along the stick with a gap. The gap only closes when the limits are exceeded.

Track for the second issue below:

su27 stick limiter test.trk

[Yo-Yo]

As far as I can see in DCS, OPR works exactly as it should and according your expectations.
1 - Limiter rod is in its forward limit
2 - The stick goes aft quite slowly, watch the gap.
3 - One or two parameters are out of their limits, shaker on, watch the gap (no gap), pilot overrides the OPR
4 - Stick moves too fast, and pilot overrides it.

"The gap", if you mean the control window, is a distance between the center of the stick circle and the pushrod symbol.

 

[Yo-Yo]

Quote

The first issue is about the aircraft behavior when pressing the 'S' key, which puts the FBW into DIRECT mode. In DCS, there's a noticeable pitch transient when switching the mode back and forth. But it shouldn't, as switching to DIRECT mode only removes a lag filter in the middle portion of the diagram, and the pitch rate and Ny feedback in the right portion of the diagram, which shouldn't be causing any pitch transient if flying straight and steady.

In DCS it just looks like the open-loop trim schedule (lower left corner of the diagram) is completely removed when in DIRECT mode, which it shouldn't as it's added to the stick input, without any switches.

Yes, and it was done intentionally. We still have no confirmed data on whether trim schedule block is operative during DIRECT MODE, but we can make an educated guess, that it zeroed, using several facts:
DIRECT is an emergency mode for catastrophic SDU failure while scheduler signal can be false too. Using this signal deadly connected to the system seems to be very optimistic.;
the scheduler can be switched off or zeroed by means, for example, of switching of its supply.
this scheduler is a mean to mimic natural positive velocity-stable trim characteristics when SDU is operative and maintain neutral velocity stability within the whole range of Mach number.
all sources mention "direct link using separate electric supply" but no source mentions this scheduler active.

But if some different and confirmed information is available - things can be changed.

 

[DummyCatz]

8 hours ago, Yo-Yo said:

Yes, and it was done intentionally. We still have no confirmed data on whether trim schedule block is operative during DIRECT MODE, but we can make an educated guess, that it zeroed, using several facts:
DIRECT is an emergency mode for catastrophic SDU failure while scheduler signal can be false too. Using this signal deadly connected to the system seems to be very optimistic.;
the scheduler can be switched off or zeroed by means, for example, of switching of its supply.
this scheduler is a mean to mimic natural positive velocity-stable trim characteristics when SDU is operative and maintain neutral velocity stability within the whole range of Mach number.
all sources mention "direct link using separate electric supply" but no source mentions this scheduler active.

But if some different and confirmed information is available - things can be changed.

 

Thanks. This does sound reasonable.

[MA_VMF]

в режиме жесткой связи передаточный коэффициент задается летчиком в ручную. При включении этого режима отключается автомат устойчивости и демпфер тангажа. Сигнал минует вычислитель 

[DummyCatz]

On 1/4/2025 at 1:15 AM, Yo-Yo said:

As far as I can see in DCS, OPR works exactly as it should and according your expectations.
1 - Limiter rod is in its forward limit
2 - The stick goes aft quite slowly, watch the gap.
3 - One or two parameters are out of their limits, shaker on, watch the gap (no gap), pilot overrides the OPR
4 - Stick moves too fast, and pilot overrides it.

"The gap", if you mean the control window, is a distance between the center of the stick circle and the pushrod symbol.

 

Yes I just meant the distance between the center of the stick circle and the pushrod symbol.

Now I noticed the gap. It is indeed hard to notice.

[DummyCatz]

14 hours ago, MA_VMF said:

в режиме жесткой связи передаточный коэффициент задается летчиком в ручную. При включении этого режима отключается автомат устойчивости и демпфер тангажа. Сигнал минует вычислитель 

это правильно в соответствии со схемой

[DummyCatz]

Now that we have a clean view of the graph, we can use the realistic numbers from the Kwz and Kny schedules. I think there's a FM options file out there that allows us to change those values.

[Yo-Yo]

Why do you think they are not used?

 

[DummyCatz]

33 minutes ago, Yo-Yo said:

Why do you think they are not used?

Maybe I was wrong but I'm not seeing any values being used by the FMOptionsSu27.lua?

cs_q_ny_max = 17000
cs_q_ny_min = 1600         
cs_Kny_max = 0.15
cs_Kny_min = 0.031

whereas in the diagram cs_Kny_min = 0.5, cs_Kny_max = 3.0, cs_q_ny_min = 0, cs_q_ny_max = 3000.

Like wisely,

cs_q_wz_max = 26000
cs_q_wz_min = 2000        
cs_Kwz_max = 0.72
cs_Kwz_min = 0.4

In the diagram there are three split sections of scheduling where q = 0, 4500, 10000 and Kwz = 0.85, 0.32, 0.1.

I'm also not sure where does cs_Kwz_g = 3.8*0.7 come from.

Edited January 6 by DummyCatz