Trimming seems to be a fiddly business in a sim because you can't feel the pressure ease off the yoke to let you know when you have it right. Just work slowly and carefully until the plane is more or less where you want it.

FSX is fairy trimmable except in the pitch axis which is always hunting more than it should.. In xplane all axis are hunting all the time.

 

You are correct about control pressure, none the less when one uses hardware such as the go flight trim wheel ( and there is a plug in or it) it can and is very close to rw. The problem is the hunting that occurs after a trimmed condition, and that minor corrections in such a trimmed condition should not produce major results.

 

 

FSX is fairy trimmable except in the pitch axis which is always hunting more than it should.. In xplane all axis are hunting all the time.

 

You are correct about control pressure, none the less when one uses hardware such as the go flight trim wheel ( and there is a plug in or it) it can and is very close to rw. The problem is the hunting that occurs after a trimmed condition, and that minor corrections in such a trimmed condition should not produce major results.

 

I think that best describes the feeling Geof, hunting. It never really seems to find a happy spot, trim wise. I don't seem to have issues with elevator trim, it's aileron / rudder (I imagine) and the rolling side to side effect that's needs to be "nulled" out? You should be able to in any given flight, get to a point where you're able to counter act any wind and torque effects on the aircraft. With Tom's suggestions / Planemaker adjustments its better. Only you real world fliers would know for sure.

 

Maybe I could make a Trim wheel with a really high resolution digital potentiometer for fine adjustment on the aileron. Will look into it.

 

If someone is well versed or educated in this field of electronics I would welcome a nudge in the right direction. May save me some time and money. I have built a few things with rotary encoders, but never a potentiometer for flight simming. There must be something that guys like to use. Will scour some of the other forums in the meantime. Can't be too hard to make a nice trim wheel, can it? I already have another BU0836X controller to hook up to, just need a good potentiometer I guess.

 

Actually just went back to the website where I got the controller card and he is selling pots for trim wheels. Just need to buy one and make a wheel.

 

http://www.leobodnar...2fa78036786528b

 

For anyone looking to make their own stuff for flight simming, or anything for that matter, his control cards work great. All "plug and play", no soldering to the controller.

The problem is the hunting that occurs after a trimmed condition, and that minor corrections in such a trimmed condition should not produce major results.

 

I agree with you Geofa, its always a bit overkill with default settings in the flight models. Now x-plane DOES have a way to tweak that, but the final feel is probably a bit different for everybody so each would have to give it a try themselves. The technique is called "phase out"...and what x-plane does is reduce the control surface deflection relative to the yoke input. You set the speeds for this effect to occur and a "low end" and "high end" and x-plane will interpolate. For example, at 100 kts...you set the phase out to be 1.0....so full yoke means full control surface deflection up to 100 kts. Then you can set at 150 kts, the phase out ratio to be 0.1....and at full yoke deflection at 150 kts and greater...the control surfaces will only deflect 0.1 of their full travel with full deflection of the hardware. In between 100 and 150 kts, control travel is interpolated between 1.0 and 0.1 in this example. This gives a better "feel" through the normal operating range at higher speeds. The trade off is "realism" at the extremes though. Jerk the yoke to full roll deflection and hold at 150kts in a real aircraft and you are rolling over hard....but with phase-outs set low in x-plane, you'd just get a gentle roll....but due to roll momentum in a real plane also...you can deflect quit a bit very quickly as in Geofa's video and not get an overkill response

 

Now in a real plane, being "hard wired" with cables, this just can't happen....BUT we all agree that there are sensations and feedback you cannot get in a joystick. Full deflection at 150 kts would take incredible muscle.....but what this will do is make the stick less twitchy at high speed. You still have to fly the aircraft after trimming, but hopefully you can limit the "major results" to small inputs this way...giving up accuracy at full yoke deflections.

 

The phase outs are found in the menu area: "Standard > control geometry > Phase-Out tab.

 

If you set the first row, "AILERON 1"...to a high end speed value of say, 160 kts...and the control travel to 0.1....you can minimize the over-control at higher speeds. At least that's the first approach I'd take and see how that "feels" in the normal operating regime. I think messing with phase-outs and the roll intertia together would help tune this to "feel better".

 

Tom K

Laminar / IXEG

 

Here's another modified Baron with some phase out set:

 

https://dl.dropbox.c...80/Baron_58.acf

 

For anyone looking to make their own stuff for flight simming, or anything for that matter, his control cards work great. All "plug and play", no soldering to the controller.

 

Leo's card doesn't work well with Mac. It only supported one analog axis for some reason...even with multiple ones hooked up. I took it up with Leo, who answered me one time only about it being x-plane's problem and I never heard from him again. I think for windows users, you probably won't see this issue though as most hardware is pretty solid for windows...but for mac users, be cautious. If the situation has changed, I'd be elated.

Thanks for the long answer! Ill play around a little myself now and see what I come up with.

Thanks again Tom for the detailed response and help. I will play around as well. Geof, if you come up with something that you really like, please share.

 

That's a shame about Leos board, and I apologize if I got anyone’s hopes up. I was not aware of the issue on Macs, having never owned one. They are a great addition to anyone's simming experience, adds so much more functionality with custom build hardware. I haven't had any issues with the board with Windows and Xplane, just so others know.

 

Glen

 

Is it, will it ever to be possible to accomplish what the video shows e.g. the pilot in moderate turbulence is anticipating the aircraft movement and using aircraft momentum-making minute but rapid control inputs after a trim (which right now in xplane for the most part would create deviations/oscillations ) with the end result being the nose does not yaw, pitch much at all, and the instruments (altitude and heading indicator) look like they are glued on e.g. smooth flight even in non smooth conditions-the hallmark of a commercial pilot. It actually is not that hard to do rw-in the sim it is 100 times harder.

 

Thanks for posting the video. I'm not a real life pilot but I think I can say one or two things about what I see. Of course, I see the same as you do, i.e the pilot anticipating the aicraft movement. Thank you for pointing this, IMHO this is fundamental for the ongoing debate about stability. But it happens I don't draw the exact same conclusions (or wants, needs for our sims).

 

I've seen the same yoke handling in a video taken in a MU2, posted by Tom Kyler years ago, with the added singularity that the real MU2 really wants to bank a lot at take off. That's apparently a characteristic of this aircraft.

 

So, the pilot anticipates. But how ? What sense tells him that the aircraft is going to bank right or left, nose up or nose down ? As you point out, the visual outside the window seems perfectly stable. So I guess, the clue is not visual. By the time the pilot would notice the move with his eyes, would he react too late to keep the perfectly straight attitude. The fact is, forces (from turbulence in that case) start acting on the aircraft just before the actual move becomes perceptible. And as human beings, we are able to detect tiny forces on our body instantaneously. And react consequently. From my interpretation, he instinctively overcorrects a bit to catch up and comes back some way toward the neutral position, which is surely the right thing to do, and accounts for the fast "back and forth" yoke movements.

 

When I think of it, this equally applies for cars or any other drivable vehicule: we aim with our eyes, and we drive with our body !!!

 

Now, what happens in a desktop sim, where obviously the very limited visuals clue is our only effective sense (limited visual, read: no peripheral vision) ? You guess: it is absolutely not possible to anticipate as effectively. Simply because at the very moment the (virtual) turbulence hits the aircraft, we are not aware of it.

 

This brings up the question as to what desktop sims are supposed to simulate:

1) the aircraft behaviour in an airmass, or

2) the aircraft behaviour coupled with the human instinct, both element which are needed to account for a smooth ride as in reality ?

 

With the second one I mean: does the simulator has to compensate for the lack of possible instant reaction from us, by overstabilising the flight model ?

 

Both are possible, I for one strongly prefer the first option, i.e only simulate the physics, even if I'm aware that as good as I could be, I'll never ever be able to anticipate and fly as smoothly as it's possible in reality or in a full motion sim.

X-plane is clearly an engineer's work, and as such, seems to take the first option. (this does not imply that it does it right or wrong, this is another question).

 

I might be wrong, but I suspect that Flight Simulator lends more toward the second one. That would partly explain some misunderstanding and why both sides feel the debate is biased. We should really clarify what we want in a sim before talking about such a thing as "realism". Which kind of realism?

 

 

The choice is more significant than it seems. I always have the bitter impression that we don't define what we really want when we talk about "stability". For example, you'd have two exact opposite ways of trying to mimic what we see in the video:

1) let the commands react faster, so that the really talented simmer gets a chance to apply it's corrections more quickly despite the inherent lateness of the visual clue.

2) let the commands react more sluggishly, and add more inertia and/or artificial stability to the aircraft.

 

and both options has some really big drawbacks !

 

or

 

3) Just accept the fact that a desktop sim is very limited. Accept the challenge. Eventually, start building a 180° visual system in order to enhance every possible visual hint and try to pilot ahead of the plane.

 

Hope this makes sense, and someone actually reads that in an already long thread and neverending debate.

 

Pascal

Actually stability is spelled out by the FAA and can be defined by their criteria.

 

I agree with most of your assessment-however it seems like we should be able to meet somewhere in the middle-if we into desktop simulation getting better and better. An aircraft like a Baron once trimmed is very stable-and once in this stable condition can be hand flown with autopilot precision with extremely small corrections as shown in the video.

 

Once again-Carenado seems to have gotten very close.

Actually stability is spelled out by the FAA and can be defined by their criteria.

 

I agree with most of your assessment-however it seems like we should be able to meet somewhere in the middle-if we into desktop simulation getting better and better. An aircraft like a Baron once trimmed is very stable-and once in this stable condition can be hand flown with autopilot precision with extremely small corrections as shown in the video.

 

Once again-Carenado seems to have gotten very close.

 

My RV wasn't what you'd call "as" stable as compared to some larger GA airplanes, but at least it didn't take a lot of constant & annoying effort to move in the same direction & altitude.............once it was also in it's trimmed state. Cruise flight shouldn't be as hard as some sim aircraft make it out to be. When it is, my sim flights are always short lived...............or just on A/P while I do something else.

Please remember that some of us left MSFS due to the constant tweaking of parameters to get the sim to operate correctly. I really don't want to start that with XP. If something is broken or not dialed-in correctly in XP, can't we let the developers fix it?

If something is broken or not dialed-in correctly in XP, can't we let the developers fix it?

 

Whether or not it's "broke" is a matter of perception in the topics we are dealing with here though. We are not talking fuel consumption, or top speed or gross weight, things that can be measured. We are talking "feel" and Pascal has that exactly right....each persons perception is different. I think...as Geoff says though, there is some middle ground where we can probably get reasonably close. Having spent a lot of time testing the XP Baron lately, I'd say it's close in the "normal category" range in which its intended to operate. I've had a few "real" high time buddies test it. The business manager of Laminar owned a Baron for a time too...he said it behaved as he expected....so I consider this section of "feel" handled with the latest adjustments that will make its into the next beta.

 

Tom K

Then it should be able to pass FAA stability standards? Have your baron guy test the milviz b55 and the xplane Carenado's bonanza as a compare. I agree this adjustment is a good improvement-what I guess I am asking is can improvement be made in control that FSX also could not accomplish that will allow precise aircraft control in a trimmed condition?

 

P.s. thanks for this dialog-I find this exciting and renews my faith in xplane.

Interesting, who told you that the FD of MSFX is not realistic?

Not sure who you are addressing but I don't think anyone said the fd of msgx is unrealistic. I do find in the pitch axis it is too sensitive and pretty much impossible to trim. I think we are all searching for if something even better is possible-after all 6 years have passed.

I think someone with knowledge should tell us what the standards are for an aircraft to pass FAA Flight Test regarding the flight control effectiveness, stability, range of motion, etc. This would make it much easier to develop and fine tune addons, if we know what they need to do, and most important, what is not acceptable behavior for an aircraft that is supposed to be flown without autopilot. There must be specific tests to determine these. People who have RL Flying experience knows something is not right because it doesn't feel right, but that's not enough for a developer to go and change the FM, he needs more information than that.