Flight model Over Trimed if trim pressed

[GumidekCZ]

Flight model Overtrimed state if trim button pressed.

Resulting overtrim result depends on direction and size of actual cyclic correction when trim button pressed.

I can repeatable bad results when takeoff with cyclic set to stable hover possition, and than press Trim button.

Also when accelerated into straight flight with cyclic set to stable bank and pitch, if trim pressed, the pitch will quickly decrease (nose down) and if near to ground and not corrected, it will result into crash. If than corrected back by cyclic and pressed trim once again, it will result in 5-10° more pitch up then was in time of trim press. So I have to do two more repeats with trim pressed and correcting overtrim, after only minor cyclic correction is needed, Im able to stabilize my flight with just minor cyclic corrections as usualy.

This wrong quick trim press behaviour BUG is present now also in Black Shark 2 (probably copied from it into BS3).

Ka-50_2_quick_press_overtrim.trk Ka-50_3_quick_press_overtrim.trk

[Schlomo1933]

With only 3 flight director modes ( pitch, roll and bank ) this is a realistic* behavior when u just „tap“ the trim button in the shark. It’s only recommended when u make small adjustments or u have the flight director mode on. 
Without flightdirector u need to press and hold the trimmer for large corrections. 
With flight director on , the shark will not overcompensate. But don’t forget to switch it off , if u need the automatic hoverhold. 

The reason for it is the 20% authority of the autopilot. If u move your cyclic, the autopilot tries to maintain his last known attitude. It means it will fight u till 20% and u need to push the cyclic over the 20%. If you tap then the trimmer, the autopilot will reset to zero and your cyclic position is too far in your chosen direction. So the autopilot overcompensate. 
It’s not like in the Mi8 or Mi24. This have different systems. 
 

not easy to explain… but i hope it will help u.

*It was discussed often in the BS2 forum. If it’s realistic or not , I‘m not shure.

Edited December 29, 2022 by Schlomo1933

[GumidekCZ]

Thanks very much @Schlomo1933for your effort of explaining me things how they are right now. But still, let's keep it as a Bug, till someone from ED come up with proof that this is how it should work, even if it make no sense to behave like it is now. So for now I'm just patiently waiting.

[AeriaGloria]

18 hours ago, Schlomo1933 said:

With only 3 flight director modes ( pitch, roll and bank ) this is a realistic* behavior when u just „tap“ the trim button in the shark. It’s only recommended when u make small adjustments or u have the flight director mode on. 
Without flightdirector u need to press and hold the trimmer for large corrections. 
With flight director on , the shark will not overcompensate. But don’t forget to switch it off , if u need the automatic hoverhold. 

The reason for it is the 20% authority of the autopilot. If u move your cyclic, the autopilot tries to maintain his last known attitude. It means it will fight u till 20% and u need to push the cyclic over the 20%. If you tap then the trimmer, the autopilot will reset to zero and your cyclic position is too far in your chosen direction. So the autopilot overcompensate. 
It’s not like in the Mi8 or Mi24. This have different systems. 
 

not easy to explain… but i hope it will help u.

*It was discussed often in the BS2 forum. If it’s realistic or not , I‘m not shure.

 

Mi-24 actually works same way as it’s trim interaction is similar, since trim causes AP and attitude hold to reset completely. So you get similar jerk, that I similarly solved by either holding the Trim button down or trimming multiple times. 
 

I would say it might be even worse becuase when you press trim you have no AP. But In Ka-50 you still have dampening. 
 

@GumidekCZ, it’s definitly really. It’s been something virtual pilots have had to learn and struggle with since Black Shark first came out. You will see many old black shark tutorials and lessons about this. For the Mi-24, there is documentation I have seen that backs this up.

Apache has a similar function where you hold force trim forward for 3 seconds or so. But becuase it’s a long press I think it’s more gradual and less intrusive then when AP instantly cuts out 

[Schlomo1933]

vor 19 Minuten schrieb AeriaGloria:

 

@GumidekCZ, it’s definitly really. It’s been something virtual pilots have had to learn and struggle with since Black Shark first came out. You will see many old black shark tutorials and lessons about this. For the Mi-24, there is documentation I have seen that backs this up.

 

Thats just true for BS2. In the BS1 there was no overcompensating. And i´m just not shure if this is realistic. Because all discussions went around if theres a micro switch who will detect when the cyclic is moved away from the trimmed position. there is no such a switch thats fact.

But if the flight computer has the possibilty to detect when the cyclic is moved away from the trimmed position and will switch the autopilot to standby (that will act the same way as we hold the trimm release or the new pedals - heading is working ) was never cleared. I believe in the real shark it is used this way. because how it is implemented at the moment is feels just unnatural and dangerous. 

But i got i used to it since many years, and dont want to start a new discussion. We or Ed (i believe) dont have anything what will tell which way is implemented in the real shark. 

[AeriaGloria]

8 minutes ago, Schlomo1933 said:

Thats just true for BS2. In the BS1 there was no overcompensating. And i´m just not shure if this is realistic. Because all discussions went around if theres a micro switch who will detect when the cyclic is moved away from the trimmed position. there is no such a switch thats fact.

But if the flight computer has the possibilty to detect when the cyclic is moved away from the trimmed position and will switch the autopilot to standby (that will act the same way as we hold the trimm release or the new pedals - heading is working ) was never cleared. I believe in the real shark it is used this way. because how it is implemented at the moment is feels just unnatural and dangerous. 

But i got i used to it since many years, and dont want to start a new discussion. We or Ed (i believe) dont have anything what will tell which way is implemented in the real shark. 

In Mi-24 it is this way. There is compensation sensors. They detect cyclic deflection from last trimmed point. They then bias the auto pilot to that direction, that way the AP does not fight you after trimming. It’s also the chief way to check the AP on the ground by moving the cyclic and looking at AP deflections 

Too bad Shark has no such way to see what AP is doing 

[GumidekCZ]

On 12/30/2022 at 1:42 PM, Schlomo1933 said:

I believe in the real shark it is used this way. because how it is implemented at the moment is feels just unnatural and dangerous.

Simple question:

Would any helicopter producer military or civilian implement system, which is very very frequently used and near ground (where attack helicopter mostly operate) would bring pilots near death if accidently only quick pressed?

Edit: I can imagine behaviour like if quick pressed, that not all amount of cyclic offset would be accepted to the new trimmed state. But OVERTRIM?

Btw., every other rotorhead in DCS will NOT overcorrect if quick pressed. May faults are there, not here

Edited December 31, 2022 by GumidekCZ

[tae.]

As far as I understand it, this has been raised as a bug a few times in the past, and has been confirmed as such, simply not fixed.
 

 

In the older thread, we can even see BIGNEWY confirm it as a bug with no ETA for fix.

So, from my understanding, unless someone can please correct me, this has been a bug for a while, is confirmed as such, and we are waiting on a fix for it. I would maybe have expected that to come with BS3, but for some reason that did not happen.

[Schlomo1933]

Sorry but no. This thread was about something different. And as Big Newy stated in one of the latest posts , it’s already fixed. 

Edited January 1, 2023 by Schlomo1933

[tae.]

1 hour ago, Schlomo1933 said:

Sorry but no. This thread was about something different. And as Big Newy stated in one of the latest posts , it’s already fixed. 

 

I am experiencing exactly what the OP is experiencing in my Black Shark, and it was previously a bug in some of the other helos too, but was fixed there. It is 100% still present, and very easy to test and reproduce. I'll even happily create tracks of it, if needed, but I think some have already been supplied.

Edited January 1, 2023 by tae.

[J-man]

Yes, this is really dangerous. Whenever I am in level flight and press my trim button, the helo doubles the pitch down. I´m using the Re-Center trim mode since I have no FFB Stick.
This does not occur when I hold the trim button before moving my stick away from the center position.
It almost appears like trimming takes the stick input pre-trim and adds the stick input when the trim button is released.

So when I pitch down 20° manually and then press trim, it trims for 40° (20° Pre-Trim Stick Input + 20° Trim Stick Input).

And if my stick is centered and I hold trim and move to 20° down, I get 20° trim (0° Pre-Trim Stick Input + 20° Trim Stick Input).

Since I´m using the Re-Center trim mode, what I would expect to happen is that when I press the trim button, the current input is stored as the base input value each frame until I release the trim button, at which point the base input value remains what is was before I released the trim button and my actual stick input value is set to 0 until I recenter the stick.

Edited January 3, 2023 by J-man

[AJaromir]

There are more causes of this problem. But it is not a bug.

First of all: you have to take into account the autopilot authority. The propper way how to trim is:

1) Press and hold the trim button

2) Set new cyclic position, attitude, heading

3) Release the trim button.

Doing it different way will cause troubles because:

 

The cyclic output = Pilot input + Autopilot + Damper.

lets say the actual input is 0%(pilot)+0%(autopilot). Damper is not important.

Now you move the stick like 30% left. The result is: 30% Left (pilot) + 20% Right (autopilot have 20% authority) = total output 10% left

So in fact only 10% left input is needed but you did not turn off the autopilot, so you had to move stick 30% left.

Now you press and release the trim button. This causes turning off the autopilot for very short time. That means the total output is 30% left for very short time and results in overtrimming. That's why is important to turn off autopilot first. To turn off autopilot press and hold the trim button or turn on the flight director mode.

This is how it works in BS1 and also in BS2.

 

Second issue is caused by Non-force feedback stick. You have to understand how trim works in real helicopter. Please see this video where everything is very well explained. The video is universal for all helicopters in DCS:World

 

A nebo se, Gumídku, prostě zeptej na českém fóru. KA-50 mám od jeho vzniku. 

Edited May 18, 2023 by AJaromir