Autopilot going crazy

Hi everyone !

I recently fly back on the Mi-8 and I was facing an issue I didn't have before.

I do the starting procedure as always, no problem. Before taking off I activate the autopilot for the pitch & roll I can fly without any issue.

But usually I used the autopilot heading too. In my memory it helped me to keep the chopper with the rudder, was not active when you give some rudder and is automatically set to the direction when you don't put rudder anymore (I can still see that with the autopilot light turning off and on when I apply some rudder and then release it).

Now, I don't know why, when I release the rudder after a turn, most of the time the "rudder/heading autopilot" is going crazy and set the rudder full left or full right... Making the flight absolutly horrible.

Is there anything I'm doing wrong ? Because some months ago, I was flying it so easily I don't understund why it's happening now...

Thanks for your answers.

Just saw a video about autopilot. Seems my rudder is trimmed a lot sometimes. Is it needed to reset this trim frequently in order for the autopilot to not fight it ?

Why the deleted message ? You were suggesting to turn off the rudder trim in the options. What is the closest to reality, option on or off ?

well I trim out before I ever activate the auto pilot, then any small rudder movements never seem to bother it, but if I'm going wildly of Couse say due to the AP trying to fight me or the wind lets say I will stop it, re align the re activate, never had a problem that way

well I trim out before I ever activate the auto pilot, then any small rudder movements never seem to bother it, but if I'm going wildly of Couse say due to the AP trying to fight me or the wind lets say I will stop it, re align the re activate, never had a problem that way

I'll try this asap thx ! If possible, I'll upload a video of my problem if it persists.

hope it works for you

This is the same problem I'm having with the Shark. Rudder seems to pull HARD when maneuvering and trimming often. Seems to build up energy until it becomes uncontrollable and seems to just "snap". Must be something in the new engine. As I'm trying to lock up an air target, the helicopter suddenly seems to just loose control and start spinning hard and wildly, no matter how much I push back against it with the rudder.

EDIT: Well, I found out that we are supposed to hold down the trimmer while we are moving the cyclic, and then release it. Seems to work well in slow, easy flight. I'll test tomorrow how it does in fast maneuvering and trimming, and whether I still get the mysterious build up of torque energy. Might be better to do fast maneuvers in FD mode, where there is just damping.

I tested for the Mi-8. It seems to work a little more when keeping the trim button pushed when changing course. But the most effective is to reset the trim after a turn and then trimming the aircraft.

I'll do some more tests to find the most effective way. For the KA-50 I was already doing this. Especially when you pilot with the trim setting your current autopilot course.

I think a lot of trimming is totally up to the individual player on how they trim or what method, I continually trim so I suppose I don't see so many errors, but yes even in a hover the old shark will move a bit

Why the deleted message ? You were suggesting to turn off the rudder trim in the options. What is the closest to reality, option on or off ?

As a matter of fact I WAS suggesting just that. I deleted because I started to think I misinterpreted your issue.

My pedals do not self center, so I prefer no rudder trim.

Speaking as a user of self-centering pedals, I was curious and had to test the AP yaw channel with Rudder Trimmer off (I've only ever had it on). At least in Central Position Trimmer Mode, it seems to do absolutely nothing. So if you want the benefit of the heading hold, it seems Rudder Trimmer must be enabled. I personally would never forgo the yaw AP channel because it's actually pretty amazing; once it reaches its 20% authority limit, it switches to "displacement mode" (as the Dash 10 refers to it [2-46]), allowing the system to move the pedals to adjust the trim through its entire range, effectively giving the yaw channel 100% authority. If you take it a bit slow, it's possible to transition through all flight regimes from ground to cruising at altitude to landing while not touching the pedals once.

[...]You were suggesting to turn off the rudder trim in the options. What is the closest to reality, option on or off ?

Certainly on. Unless I'm mistaken and it's still possible to take advantage of the yaw AP channel with Rudder Trimmer off...

[...]Might be better to do fast maneuvers in FD mode, where there is just damping.

Yeah, trying to fight the Ka-50's AP is just painful. I consider FD mode so essential I have it bound to my stick's weapon release button and I press it whenever I need to do, well, nearly any maneuver that isn't just adjusting my trim a bit. Keep in mind that holding the trim button has basically the same effect on the AP as activating FD mode, which is why it makes the process so smooth. Hold trim, adjust attitude/heading, step on the ball, release trim, perfect coordinated flight :D (even works with the Mi-8 most of the time as well for me)

I tested for the Mi-8. It seems to work a little more when keeping the trim button pushed when changing course. But the most effective is to reset the trim after a turn and then trimming the aircraft. [...]

As the manual states (pg 221),

"In the Mi-8 depression and holding of the trimmer button do not engage alignment mode of the autopilot, as it is done in the Ka-50. I.e. trimmer as object is not interacting with autopilot at all."

The trimmer shouldn't affect the AP whatsoever; the smoothness of your trimming only has to do with your trimming method and settings. You really shouldn't have to reset the trim at all unless you suddenly need to do some hard maneuvering and need the entire range of control.

If you have a self-centering stick which returns to center quickly and accurately when you let go, I'd recommend giving Central Position Trimmer Mode a try if you haven't. As far as smoothness goes, it's the next best thing to a springless or actual FFB stick.

And now the big one: It took me a little while, but I figured out why the yaw AP channel is sometimes inconsistent in setting a new heading, and it's rather strange; if you look at the АП-34Б control panel, at the centering knobs with the small windows just above them, you'll notice that when any particular channel isn't engaged, the indicator disks in the windows rotate on their own, corresponding to the attitude of the helicopter (e.g., when the helicopter's heading is steady, the indicator dial under "НАПРАВЛЕНИЕ" won't be rotating). If you turn and change your heading quickly, you'll notice that the indicator takes some time to catch up to the new detected heading. Halting the disk's rotation too early by activating the autopilot channel (or releasing the rudder pedals if the AP was already active) will cause the system to lock to a point between its last setting and the new heading (for example, making a rapid turn from a heading of 5° to a heading of 90° and then immediately activating the yaw AP channel could cause the system to "lock" onto a heading of 30°).

In practice, what this means is that every time you make a major adjustment to your attitude/heading, you must watch the indicator window to make sure the dial has stopped rotating, otherwise you may inadvertently set the channel to a point that's too "early" in the arc. If the yaw AP setting is too far off your actual heading, it will indeed cause the pedals to go massively out of alignment and destabilize the helicopter.

Now I have a mouse with a freewheeling scroll wheel, so I wanted to mess with the AP a bit since I can shift+scroll like nobody's business :P

As it turns out, the indicator might not actually be realistically limited in how much it can lag behind. Turning the centering knob with a good, solid flick of the wheel resulted in the system taking over a minute and a half to realign. The dial can also be sped up to expedite realignment by turning it in the direction of your turn.

Unrelated, but the autopilot indicator that shows up with the controls indicator seems to feature Cyrillic characters regardless of language or cockpit settings, including the 'Л' and 'П' indicators which show up in the heading AP channel when the system enters "displacement" mode to move the pedals. Further trivia: this seems to be the only indicator of when the pedals are being moved by the AP (other than seeing them move of course).

OKAY, this is the last of the text wall; it's a question for any devs or experts who may be reading: In the Dash 10 (2-46), it states "When the pilot’s feet are on the pedals, the sub-pedal microswitches activate, the force trim disengages, and the yaw channel operates in stabilization mode."

So, my question is, is the "stabilization mode" implemented in the sim in any way? As I understand it, the pedal microswitches are simulated by the AP disengaging when the pedals are moved (always wondered why that couldn't be implemented with an additional option of having the toe brakes activate the switches), but I'm not noticing any additional "stabilization" when the pedals are off-center with the AP on. Is the Dash 10 or my reading comprehension just off base here?

OKAY, this is the last of the text wall; it's a question for any devs or experts who may be reading: In the Dash 10 (2-46), it states "When the pilot’s feet are on the pedals, the sub-pedal microswitches activate, the force trim disengages, and the yaw channel operates in stabilization mode."

So, my question is, is the "stabilization mode" implemented in the sim in any way? As I understand it, the pedal microswitches are simulated by the AP disengaging when the pedals are moved (always wondered why that couldn't be implemented with an additional option of having the toe brakes activate the switches), but I'm not noticing any additional "stabilization" when the pedals are off-center with the AP on. Is the Dash 10 or my reading comprehension just off base here?

The -10 is incorrect. While I love the TSMO manuals, they do occasionally misinterpret the Russian manuals. Stabilization duty is when the pilot's feet are OFF the pedals and the yaw channel is maintaining aircraft heading.

The -10 is incorrect. While I love the TSMO manuals, they do occasionally misinterpret the Russian manuals. Stabilization duty is when the pilot's feet are OFF the pedals and the yaw channel is maintaining aircraft heading.

Thanks a bunch - really appreciate all you contribute here, especially the presentations which have been invaluable to my understanding of these systems. And such good pictures too!