Elevator trim too aggressive

[GregP]

21 minutes ago, Zabuzard said:

Almost. The keyboard file, not joystick. Then search for "trim". You should see the commands being setup with a value, let's say 0.1. Reduce that number and see if it helps in your case.

In the example above I was referring to the SavedGames folder, but now I think you're talking about the main DCS folder, as the only file in my SavedGames 'keyboard' folder is 'keyboard.diff.lua'.

So, in my main folder (\DCS World OpenBeta\Mods\aircraft\F-4E\Input\F-4E-Pilot\keyboard\default.lua'), there are two lines for pitch trim:

 { name = _('Trim - Nose Up (Hat Aft)'), up = iCommandPlaneTrimPitch, pressed = iCommandPlaneTrimPitch, value_up = 0, value_pressed = 1, category = { categories.flight_controls, categories.stick } },
 { name = _('Trim - Nose Down (Hat Forward)'), up = iCommandPlaneTrimPitch, pressed = iCommandPlaneTrimPitch, value_up = 0, value_pressed = -1, category = { categories.flight_controls, categories.stick } },

But if I change the 'value_pressed' to anything other than 1, it seems to disable pitch trim altogether...

Edited May 26, 2024 by GregP

[Zabuzard]

In the example above I was referring to the SavedGames folder, but now I think you're talking about the main DCS folder, as the only file in my SavedGames 'keyboard' folder is 'keyboard.diff.lua'.
So, in my main folder (\DCS World OpenBeta\Mods\aircraft\F-4E\Input\F-4E-Pilot\keyboard\default.lua'), there are two lines for pitch trim:

 { name = _('Trim - Nose Up (Hat Aft)'), up = iCommandPlaneTrimPitch, pressed = iCommandPlaneTrimPitch, value_up = 0, value_pressed = 1, category = { categories.flight_controls, categories.stick } }, { name = _('Trim - Nose Down (Hat Forward)'), up = iCommandPlaneTrimPitch, pressed = iCommandPlaneTrimPitch, value_up = 0, value_pressed = -1, category = { categories.flight_controls, categories.stick } },

But if I change the 'value_pressed' to anything other than 1, it seems to disable pitch trim altogether...

I see. Then the amount is likely on our backend, ill have a check :)

[Hawkeye_UK]

Damn i just left some feedback in another thread about trim, missed this main post.  I completely agree reminds me of when the mossie came out where a quick press and release would add way to much. In the other thread i would be interested to hear the SME feedback as i am suspicious that the real aircraft would move the control surface that much (looking at the F2 view during a momentary push).

There is quite a few keybinds like this where a brief push goes way to far, reticle depression is another one, literally momentary push adds 10-20 units.  Also instrument lights that control the gun/heat/radar and other features in the front cockpit, its really either on or off.  Someone at Heablur needs to sit down and map and test all the keybinds really as quite a few are off.

[Zabuzard]

Are you playing in MT? It is likely that these issues are caused by that. Could you try in ST and see if its still present?

Also, try a keyboard instead of joystick hardware. Could possibly be the hardware sending the command too long as well.

[Sile]

While I find that trimming during level / cruise flight works quite well, I am not able to achive a stable trim when landing / base turn configuring the aircraft at low speed with flaps, …

steps seem huge at low speed

Edited May 27, 2024 by Sile

[Alpha]

Agreed. While the real jet required lots/frequent trim inputs it was always short klicks - a lot of times multiple klicks. In DCS (on my system) I find that a single, short klick already has quite a lot of effect. A bit less effect per klick would be closer to what I remember - but it´s not really "worlds apart" . Using MT.

[Sile]

vor 4 Minuten schrieb Alpha:

Agreed. While the real jet required lots/frequent trim inputs it was always short klicks - a lot of times multiple klicks. In DCS (on my system) I find that a single, short klick already has quite a lot of effect. A bit less effect per klick would be closer to what I remember - but it´s not really "worlds apart" . Using MT.

exactly what our squadron "sme" - a former F4 pilot - describes.
frequent trim inputs with multiple short clicks. But short clicks have too much effect as it is now.

(mt client here)

Edited May 27, 2024 by Sile

[Alpha]

Yeah, that´s me, Sile  

[lengro]

I tried mapping the trim to a rotary encoder on my throttle where I can define the hold time for each click.

Default is 100ms and I found that a bit to sensitive, so I reduced it to 40ms hold time pr. click and that is way better. Now I can click several times for small adjustments.

On the hat switch on the stick, the hold time begins when pushing the hat and ends when releasing the hat. This means it's impossible to make consistent hold times for each click.

I think it would be better if it's possible to change the code, so a hold on the hat only sends one pulse of a predetermined length. A continued hold should then start repeating pulses to accommodate larger trim adjustments.

[Mainstay]

On 5/27/2024 at 12:19 PM, Alpha said:

Agreed. While the real jet required lots/frequent trim inputs it was always short klicks - a lot of times multiple klicks. In DCS (on my system) I find that a single, short klick already has quite a lot of effect. A bit less effect per klick would be closer to what I remember - but it´s not really "worlds apart" . Using MT.

Same problem here with a WinWing Orion Base 2 in combination with F-16 Stick. Trim button clicks ever so short cause massive input ingame.

Doesnt really feel correct and glad a F-4 driver confirms its a bit overdone.

[CF104]

On 5/27/2024 at 9:10 AM, lengro said:

I tried mapping the trim to a rotary encoder on my throttle where I can define the hold time for each click.

Default is 100ms and I found that a bit to sensitive, so I reduced it to 40ms hold time pr. click and that is way better. Now I can click several times for small adjustments.

On the hat switch on the stick, the hold time begins when pushing the hat and ends when releasing the hat. This means it's impossible to make consistent hold times for each click.

I think it would be better if it's possible to change the code, so a hold on the hat only sends one pulse of a predetermined length. A continued hold should then start repeating pulses to accommodate larger trim adjustments.

 

MT user with the same issue. I've also had to assign the pitch trim to a rotary encoder on my throttle base which allows for much finer control of the trim inputs. It's a bit of a pain on the approach since the encoder isn't on my throttle levers themselves. Lets hope that HB can make a fix for this as it does detract from the flight experience of the F-4E.

Cheers,

John

 

[GregP]

On 5/27/2024 at 5:24 AM, Zabuzard said:

Are you playing in MT? It is likely that these issues are caused by that. Could you try in ST and see if its still present?

Also, try a keyboard instead of joystick hardware. Could possibly be the hardware sending the command too long as well.

Just wanted to note that in my case, I usually use MT but tried ST and got the same result: using a hat for pitch trim is too rough, but rotary encoder works very well.

[Zabuzard]

At this point it would be very helpful to see a quick video of what the "shortest input" of trim causes on your end (inflight, not on ground).

Because it is otherwise very difficult to differentiate between "this is normal, it is supposed to be difficult to trim" and "oh, wow. okay, that is way beyond what I am seeing on my end", if that makes sense :)

[jojo]

Same problem for me.

MT Client.

Virpil WarBRD base + Thrustmaster Hornet stick.

The slower it gets, the more difficult it is to trim, at landing AoA I’m fighting the jet. I can’t let the stick go even half a second. I’m always one click too much or one click not enough with the Trim HAT.

I can trim the Tomcat in landing configuration fine.

Edited May 28, 2024 by jojo

[CF104]

I'm not sure of the actual technical manual figures for pitch trim times but using the stick hat I can go from full nose down trim to full nose up trim in 10 seconds. That seems a bit too fast and could be a part of our problems. 

Cheers,

John

[Beirut]

I find the trim to be very sensitive on my Warthog HOTAS. That said, I lack any kind of expertise to say if that is the way it should be.

 

Be nice if it was cut in half, but maybe that's not realistic, I don't know.

[skywalker22]

Personally, I think trim work fine, or I should say, I have no issues with trimming the plane the way I want to. Maybe, maybe each button press the trim should really have a bit less of an effect, for lets say half as @Beirut said. So for an effect that we have now per one button click, there should be two clicks.

[average_pilot]

It's hard to quantify objectively. During approach it seems easier to trim with keyboard presses rather than with the trim hat, but I'm not quite sure. Overall, pitch trim feels a bit too sensitive.

Edit:

Some more information. I'm not using any joystick software, and I'm still in Windows 10.

Joysticks are polled and keyboards sent events, isn't it? Maybe that makes a difference in this module, in some user configurations, for some reason? Maybe everything is fine but I'm too slow while doing short presses with the trim hat but not with the keyboard for mechanical reasons?

Some more edits: not a show stopper in any way, I'm more than happy with the Phantom, but I'm curious whether there is a problem or not with this.

Edited May 29, 2024 by average_pilot

[Andrew8604]

I'm not having any trouble with trim.  Just short, normal presses of the hat forward and back for pitch trim.   It's not doing anything crazy.  It feels about right, to me.  Pretty much consistent with the trim response of all the other modules in DCS.  Definitely not experiencing the wild trim changes you guys are describing.  I mean, it sounds like something's definitely wrong with your trim functioning by what you describe.  But WHAT is going wrong?  That's the question.  I don't think it can be Heatblur's coding, because it works for many of us...but there are always possibilities until the real cause is found.  But then why does trim apparently respond normally for you guys in other modules?  Does it work normally in the F-14, for you guys?

Your trim hat *IS* mapped to pitch and roll TRIM and NOT to pitch and roll, for the F-4E, right?  And if you have it mapped for the WSO, is it the same?  Actually, I tried to map it wrong like that, and it doesn't seem to replicate the problem you have.

Maybe try a full joystick calibration in Windows Game Controller?  It doesn't seem like that should make any difference with the buttons and POV hat, but who knows.

Is there a chance the firmware in the stick/base is set to send a stream of button clicks at some high rate which each press of the hat?  I can't imagine why, though.  Or that some intervening joystick's software app is doing it?  Or that the firmware on the joystick is set to send high rates?  I wouldn't think so, but why are we players seeing such seemingly wildly different results using the same DCS version and the same module?  I'm running it in single player.  I think I have multithreading enabled but e-cores disabled, if that makes any sort of difference to this issue.  Not sure if that's what "MT" and "ST" are referring to in this thread.

I'd hate to see it changed when the Heatblur test team, SMEs and approximately half the users think it's fine.  Somehow have to find the root cause.  Maybe take one user and do a lot of investigating/troubleshooting to find the cause?

[Grundar]

It definitely feels more sensitive than some other modules and the F4 definitely requires those frequent trim changes. Its significantly more noticeable at slow speeds.  If I stick to a light touch on the trim (i.e - small, deliberate clicks) it functions as I would expect it to. If you hold it down too long it certainly has a big throw of the aircraft - whether that is accurate to the F4 airframe - I don't know.

If trimming issues are on the ground, it's the bellows system quirk, trim nose down 3 for takeoff and avoid the immediate aggressive trim chasing saga on lifting.  So are the bellows modelled to be affected by barometric changes and from there is influencing the trim? I'm honestly not sure of how elevation changes would affect that particular system in the F4 or if it is already compensated for.

It would also be interesting to see/find out if the trim on hard wing F4's was more sensitive or whether it is a quirk of this particular model.

[tusler]

Yep, I agree about too sensitive, I am running a warthog, all stock no software, I have input curves, tried what i could, it will not fly level for me and if I get it as close as I can and let go of the stick to look in the cockpit, it is in a increasing bank when i look up. Might have been how the real acft flew and pilots had to feel the seat of the pants thing which we can not do in a sim, but it is no fun for me in a game. I am flying with 2 fuel tanks on and 4 Aim9's on.

Edited May 30, 2024 by tusler

[Super Grover]

Hey everyone, thank you for all your messages.

First of all, I wanted to ask everyone reporting if you could be a bit more precise - I mean, please include the conditions in which you performed your observation, like gross weight, fuel and stores (or simply CG location), altitude, and speed. Those things have a super high impact on trimming.

Second, please remember that the trimming speed in the F-4 was dictated mainly by the requirement to cancel stick forces during quick force changes, usually when accelerating. Also, the requirement for trim speed consistency wasn't very high. It took approximately 8 to 13 seconds to move the actuator between the extremes. I just wanted to note that 13 is 163% of 8 . I mean, don't shoot the messenger; send letters to McDonnell.

Finally, please let me quote some test pilot ratings of the pitch trim. They come from various feel trim configurations, but please remember that those configuration changes were a slow evolution rather than a revolution, so most of the comments on the controls can be applied to all versions of the F-4.

Quote

In the highlift configuration, the combination of low friction and breakout forces, weak stick centering, and large trim speed bands accounted for a derogation in trimmability in the airplane with downsprings removed. Pilots noted a tendency to trim out forces at the slow speed end of the trim speed band...

Quote

Longitudinal trimability was qualitatively evaluated during climbs, cruising flight, and landing approaches. Because of the extremely weak static stability... it was nearly impossible to trim the airplane 'hands-off for a desired flight condition (Rating C3). Any slight disturbance would cause the airplane to vary from trim, and if not stopped by the pilot, proceed into a divergent long period oscillation. An improvement in longitudinal trimability during climbs, cruising flight and landing approaches is desirable for improved service use.

Quote

Longitudinal trim rate was sufficient to maintain trim requirements during MAX A/B accelerations to 750 KIAS. However, non-linear trim and high control system sensitivity make precise trimming difficult.

I can promise that I'll take a look and maybe increase the time for the actuator motor to get to full speed, but it may be a bit difficult to achieve the results you'd like to have without sacrificing realism. However, I don't see a reason why we can't look for some not-so-realistic optional settings for those of you who prefer usability over historical accuracy. Also, please don't feel bad if you like such settings - it's a sim/game, so the most important thing is to have fun, and with the hardware we have, it's impossible to get the same accuracy and feedback as in the real aircraft.

Edited May 30, 2024 by Super Grover

[average_pilot]

Thank you for the detailed answer. I for one enjoy historical accuracy. In my case what I'm really trying to find out is if there is a difference in how the simulation reacts to inputs depending on the peripheral (joystick buttons vs. keyboard presses). I'm not having much time to test anything lately unfortunately.

[schmiefel]

12 minutes ago, average_pilot said:

In my case what I'm really trying to find out is if there is a difference in how the simulation reacts to inputs depending on the peripheral (joystick buttons vs. keyboard presses)

I think that there is at least some connection to how much of a delay each button click e.g. on the trim hat buttons of a certain joystick type has. As the configuration tool from Virpil allows to set various delay timings I tried it first with the default settings (approx. 100ms) and then moved to much smaller delays (approx. 20ms) and that helped a lot with small trim control inputs. Modern gaming keyboards may be even  faster with their delayed action.

See the configuration tools screenshots in my comments in another thread about the same 'issue':

Default Virpil settings:

Shorter delay timings for button action:

Plus: additionally I set a larger dead zone tha I usually use and added some curve for the X and Y axis of my stick in the DCS controls setup.

[Relayer]

Unfortunately, without FFB, how we interact with the trim system is always going to be different to how it was in the actual cockpit, since we simply don't have stick forces. My understanding is that, IRL, the general procedure is to move the stick to where you want it, and them trim to relieve the stick forces. Since we can't do that without FFB, trimming is inherently different. Firstly, trim is additive - if I hold the stick slightly aft, and apply aft trim - instead of the trim relieving the force I'm applying with my arm, it is added in addition to the force I'm applying. Which means I need to release pressure on the stick concurrently with the trim, without being able to feel anything, and then see if the nose is still moving, and apply further stick deflection +/- trim until the nose is where I want it.

I've generally found the pitch trim to be very twitchy, and with it bound to a hat switch on my X-56, I can either have 'too much' or 'not enough' trim for any given circumstance, meaning I still need to apply constant stick pressure. Which, yes, I get it, the Phantom was a very hands-on aircraft and hands-free flying is not going to be a thing. But a bit more granularity in trimming would make it less of a chore.

My general point is that striving for perfect realism is all well and good - but since we are interacting with the simulation with very imperfect tools, some concessions to absolute realism in the aim of achieving relative realism are probably worthwhile.

Edited May 30, 2024 by Relayer