Jump to content

Need ED input: is yaw AP channel bugged or not?


GregP

Recommended Posts

8 months after early access began, multiple discussions in both the English and Russian forums still haven’t clarified the intended behavior and use of the autopilot yaw channel.  The current behavior appears to be bugged in that the yaw AP seems to occasionally remain in 'heading hold' mode when it should be switching to 'stabilization mode'.

This is important because in these cases it appears that the yaw AP is resisting players’ pedal inputs when in fact it should be assisting them.

Unfortunately our conjecture about this will continue without any resolution until ED comments on the following:

  • How is the yaw channel behavior modeled in the DCS Hind?
  • What yaw channel behavior should users be experiencing for both:
    • ‘Rudder Trimmer’ checked, presumably for users of spring-centering pedals?
    • ‘Rudder Trimmer’ unchecked, presumably for users of pedals without a centering spring?

Most of us seem to agree that in the real Hind:

 

  • The yaw AP channel is intended to be used throughout the entire flight, from takeoff to landing.
  • When the pilot intends to take control of the yaw angle, he puts his feet on the pedals, which press the microswitches, which places the yaw AP into ‘stabilization mode’ and deactivates ‘heading hold mode’.
  • When the pilot intends the yaw AP to maintain the set yaw angle, he pulls his feet off the pedals, releasing the microswitches, which places the yaw AP into either ‘heading hold mode’ (where the yaw AP only uses its 18% separate control authority) or ‘displacement mode’ (where the yaw AP physically moves the pedals so that it can make use of more than its 18% separate control authority).

But since the pressing and releasing of the microswitches does not appear to be currently modeled in the DCS Hind, the yaw AP has no foolproof way of ‘knowing’ what the pilot wants it to do, and thus in some situations users believe they are experiencing the yaw AP fighting their inputs – in other words, that the yaw AP is remaining in ‘heading hold mode’ rather than switching to the ‘stabilization mode’.

So can someone from ED clarify these points for us and confirm whether or not this is a bug?
 

  • Like 6
  • Thanks 2
Link to comment
Share on other sites

It is easy to tell the difference in a coordinated turn while periodically moving pedal from center to either direction. 
 

When in the 9% region to center and AP is in heading hold it will give full input opposite the turn, once it reaches the limit it will switch to displacement mode and you see a yelllow cryllic letter.

When you pass the 9% threshold it will enter stabilization, you will see it quickly jump to counteract your turn but as a few seconds pass and the AP system realizes the yaw rate is constant it will settle gradually to 0 AP input. 

The only way to turn fast enough that the stabilization mode makes inputs as big as heading hold is to turn in slow speed or a hover or a barrel roll. 
 

At normal flight speeds of 200-300 kmh rolling and yawing will only introduce minor inputs as long as it is stabilization mode and pedals are deflected more then 9% from center. If you watch the R control + enter menu the difference is easy to notice after a while between the two modes 
 

Sorry if I misunderstand your question. But to me this is all intended behavior as it is just like Mi-8 with addition of stabilization outside of the heading hold boundary. Trim has no effect on yaw AP in any mode whatsoever. Trim interacts with pitch and roll AP but only those two. 
 

So I don’t think any of this is a bug and works entirely as intended, as this approach to simulating pedal micro switches was determined during Mi-8 release and remains consistent with Mi-24. In Mi-8 trim also has no effect on yaw AP. I think there is no bug, my wish was only for a better simulation 


Edited by AeriaGloria

Black Shark Den Squadron Member: We are open to new recruits, click here to check us out or apply to join! https://blacksharkden.com

E3FFFC01-584A-411C-8AFB-B02A23157EB6.jpeg

Link to comment
Share on other sites

I believe that there is a bug. Or rather if there isn't I don't understand what the purpose of the 9% from threshold is in the DCS Hind. The Hind recognizes pedal inputs less than 9% from center, if you do small corrections with the YAW AP channel on within that threshold the heading hold function will fight your inputs. This can lead to all sorts of wonky flight issues. 

Specs: Win10, i5-13600KF, 32GB DDR4 RAM 3200XMP, 1 TB M2 NVMe SSD, KFA2 RTX3090, VR G2 Headset, Warthog Throttle+Saitek Pedals+MSFFB2  Joystick. 

Link to comment
Share on other sites

I'm afraid that this is going to go nowhere, it's not a real bug report and as such will not get looked at. 

Specs: Win10, i5-13600KF, 32GB DDR4 RAM 3200XMP, 1 TB M2 NVMe SSD, KFA2 RTX3090, VR G2 Headset, Warthog Throttle+Saitek Pedals+MSFFB2  Joystick. 

Link to comment
Share on other sites

On 2/9/2022 at 9:40 AM, Lurker said:

I believe that there is a bug. Or rather if there isn't I don't understand what the purpose of the 9% from threshold is in the DCS Hind. The Hind recognizes pedal inputs less than 9% from center, if you do small corrections with the YAW AP channel on within that threshold the heading hold function will fight your inputs. This can lead to all sorts of wonky flight issues. 

The 9% deflection threshold is a DCS workaround. Whenever your physical pedals are beyond that threshold, the YAW AP recognizes the pedal microswitches as pressed and will go into stabilization mode. This creates a weird situation when you deflect your pedals inside the 9% threshold and your inputs fight the APs inputs, which can not happen IRL unless the pilot moves the pedals without touching the microswitches.

Also note that this applies regardless of whether the pedal trim option is active or not. This basically renders YAW AP useless if you do not use the pedal trim option, since in this case, your physical pedals will almost always be more then 9%off center and heading mode will not engage.

i5-8600k @4.9Ghz, 2080ti , 32GB@2666Mhz, 512GB SSD

Link to comment
Share on other sites

5 hours ago, sLYFa said:

Also note that this applies regardless of whether the pedal trim option is active or not. This basically renders YAW AP useless if you do not use the pedal trim option, since in this case, your physical pedals will almost always be more then 9%off center and heading mode will not engage.

Right, and so this again highlights the inadequacy of either pedal position or movement being the determining factor that triggers the yaw AP to move out of heading hold mode and into stabilization mode.  Instead, what is needed is some signal of pilot intent, which in the real thing is handled by the pedal microswitches: when feet are on, the pilot desires to control the yaw angle; when feet are off, the pilot desires the yaw AP to hold the yaw angle.

Thus it seems like some kind of simulation of the microswitches themselves is needed in order to get the yaw AP behavior right for both spring-centered and springless pedals.


Edited by GregP
  • Thanks 1
Link to comment
Share on other sites

But why 9%, why not 1% or something? I don't get it. It's absolutely killing me that AP hold will fight me and my pedal inputs. It shouldn't be doing that at all. It's why I fly with AP YAW off, and why I removed the springs from my pedals. 

  • Like 2

Specs: Win10, i5-13600KF, 32GB DDR4 RAM 3200XMP, 1 TB M2 NVMe SSD, KFA2 RTX3090, VR G2 Headset, Warthog Throttle+Saitek Pedals+MSFFB2  Joystick. 

Link to comment
Share on other sites

On 2/16/2022 at 5:21 AM, Lurker said:

But why 9%, why not 1% or something? I don't get it. It's absolutely killing me that AP hold will fight me and my pedal inputs. It shouldn't be doing that at all. It's why I fly with AP YAW off, and why I removed the springs from my pedals. 

We should probably caveat that the 9% number came from AeriaGloria's tests and have never been actually quoted by anyone from ED, right?  So that may not be the actual number.

Either way, your point stands, that -- at least following their presumed logic for implementing it in the first place -- the 'deadzone' chosen should've been smaller.

Link to comment
Share on other sites

I personally believe the 9% comes from assuming the 18% percent AP authority goes 9% in each direction of pedals are centered. I don’t think many people but those with the dirtiest POT pedals would mind it being smaller, except for springless users that have hard time finding center. 

Sure my number might be off a bit, but it’s close recently tested again, I can’t test fractions of a percent. I can test 8 at which I get heading hold. And it’s always off when Yaw AP engaged above 9%.  Could it be between 8 to 9%, or even 9-10, sure

 

Black Shark Den Squadron Member: We are open to new recruits, click here to check us out or apply to join! https://blacksharkden.com

E3FFFC01-584A-411C-8AFB-B02A23157EB6.jpeg

Link to comment
Share on other sites

  • 2 weeks later...

I have been a way at work for a long time and just hopped back in the hind and the rudder yaw AP still seems a bit of mess. I can't notice any difference since release.

Kinda stage as the Black Hawk Mod has a similar system with micro switches in the pedals and they work fine. When the Hind launched there was loads of feedback on the issue but it seems to have gone no where unfortunately. 

 

 

  • Like 3
Link to comment
Share on other sites

  • 1 month later...
2 hours ago, NineLine said:

Hi, can someone give me a track of what exact issues are seeing and I will see if the team can comment?

Thanks.

Yes, one moment. I believe it works as intended, but is still awkward. Will upload a short track shortly 

Black Shark Den Squadron Member: We are open to new recruits, click here to check us out or apply to join! https://blacksharkden.com

E3FFFC01-584A-411C-8AFB-B02A23157EB6.jpeg

Link to comment
Share on other sites

9 hours ago, NineLine said:

Hi, can someone give me a track of what exact issues are seeing and I will see if the team can comment?

Thanks.

    Here are two tracks. In the first one, I focused on how the yaw automatically trims the pedals from center, and to visually show how it can mess with turns. I am aware that in reality when reaching its limit the yaw autopilot does use the pedal dampener to continue holding heading past its authority, but the trimming it does i believe throws many off as well as the real Mi-24 not having any pedal trim. 
 

    The second track is similar, but I tried to focus on the border of pedal deflection that causes the yaw autopilot to go into heading hold mode versus dampening mode. It takes about 9% deflection from center pedal position to change the yaw autopilot from heading hold to dampening, it is same in Mi-8, except in Mi-8 there is no dampening mode. This is the main complaint people seem to have, many desire this 9% deflection zone to be smaller (though this number does make sense as total authority of autopilot is 18%), or the method of heading hold activation to be different such as dedicated pedal micro switch binds. As in the real helicopter heading hold is only activated when feet are off the pedals (releasing pedal micro switches), and feet on pedals (pushing micro switches) yaw autopilot switches to dampening mode, which never interferes with turns or trims you. 

   I believe what most people are desiring to be looked at, is if this method of simulating the pedal micro switches(determining if the yaw autopilot should hold heading or do dampening by amount of deflection) can be revised. As it makes more sense back when Mi-8 released in 2014, where the yaw autopilot ONLY does heading hold. Whereas in the release of Mi-24, the yaw autopilot has an additional dampening mode likely there to combat Dutch roll oscillations, and real life literature says to have the yaw autopilot on at all times from take off to landing, with the only exception being for training.
 

     The idea being that yaw autopilot is more essential in Mi-24 to the entire flight where it helps dampen and stabilize yaw and Dutch roll oscillations throughout all maneuvers. Whereas in Mi-8 there is nothing lost by just turning it on as needed, as it’s only use for Mi-8 is as a heading hold in straight and level flight. 
 

   I apologize if this is redundant to explain again, I wish to just provide context as I know you can’t read everything, in the hope that perhaps this can be looked at. 
I have attached an additional screen shot of the export Mi-35P manual saying the yaw autopilot is essential from taxing and all of flight until landing, Section 2.1 “Preparation for Taxiing,” I can PM you the details of the full document to make sure which specific document it is. And here I will link a thread where the same document has been used in a bug report, to show that I am not using sensitive/classified material that would go against the rule 

So to re iterate, I am pretty sure this is working as intended, as it’s same method as Mi-8, but if it could be looked at or improved in any way or additional options, like delicates pedal switch bonds or special options, that would please many people and allow it to be flown like it is in real life

 

 

yawautopilotheadinghold+trimissue.trk

 

yawautopilotheadinghold+trimissue2.trk

    Here is the screenshot from section 2.1 of the export Mi-35P manual, “Preparation for taxiing.” To show that unlike Mi-8, having the yaw autopilot on is vital to flight, and that having heading hold activate the way it does interferes with that real life function 

42D7C41E-F82F-486F-9D10-1ACFBD76CA17.jpeg


Edited by AeriaGloria

Black Shark Den Squadron Member: We are open to new recruits, click here to check us out or apply to join! https://blacksharkden.com

E3FFFC01-584A-411C-8AFB-B02A23157EB6.jpeg

Link to comment
Share on other sites

  • ED Team

Hi, 

we have spoken to the team, they are aware of a trim issue and are working on a tweak. 

thanks

  • Like 2

smallCATPILOT.PNG.04bbece1b27ff1b2c193b174ec410fc0.PNG

Forum rules - DCS Crashing? Try this first - Cleanup and Repair - Discord BIGNEWY#8703 - Youtube - Patch Status

Windows 11, NVIDIA MSI RTX 3090, Intel® i9-10900K 3.70GHz, 5.30GHz Turbo, Corsair Hydro Series H150i Pro, 64GB DDR @3200, ASUS ROG Strix Z490-F Gaming, HP Reverb G2

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...