Correct AP HDG Mode behavior?

[SharpeXB]

This autopilot mode will not work, track attached. Stable 2.8.4.39731

HDGAP.trk

Edited June 29, 2023 by SharpeXB

[BIGNEWY]

Hi, 

it seems to be working for me when I take control of your track. 

I did notice your stick seems to have some micro jitter, try a slightly larger dead zone and see if it helps. 

thanks

[SharpeXB]

5 hours ago, BIGNEWY said:

Hi, 

it seems to be working for me when I take control of your track. 

I did notice your stick seems to have some micro jitter, try a slightly larger dead zone and see if it helps. 

thanks

I thought about that. The BALT also mode gets affected by any micro input from your stick, I don’t imagine that’s very realistic though. But the HDG selection is colonized implying that it’s engaged. Again that doesn’t seem realistic that the AP disengages without warning. 

[BIGNEWY]

If you have evidence it should be different please DM me, as far as I have been told it is correct as it is. 

thanks

[SharpeXB]

Any SMEs know whether this is the correct behavior for the AP HDG Mode?

Will it not engage if there’s any small micro input on the stick? Does it colonize and show as engaged when it’s not? That seems like very odd behavior for a real aircraft. 
 

 

Edited June 29, 2023 by SharpeXB

[BIGNEWY]

thread merged no need for different topic. 

Please try the larger dead zone, we can not account for inaccuracy in hardware

thanks

[SharpeXB]

21 hours ago, BIGNEWY said:

thread merged no need for different topic. 

Please try the larger dead zone, we can not account for inaccuracy in hardware

thanks

A dead zone interferes with precise flying like AAR so that’s not a good solution. Such a workaround wouldn’t be necessary if the AP was modeled correctly. I can’t imagine this behavior is accurate. 

Edited July 1, 2023 by SharpeXB

[SharpeXB]

So adding a 1% dead zone in the roll axis fixed the problem. But that means the autopilot is being tripped up by a <1% input in one axis of the stick, the equivalent of touching it with a finger. I have trouble believing this behavior is correct. 

[Brun]

I'm also very sceptical that the real aircraft requires such absolute precision. The stick being 'centered' does not explicitly mean both axes have to read exactly zero.

Claiming nothing can be done about it despite several people independently suggesting the same perfectly obvious solution is insulting.