PilotEdge Forums

I'm hoping this post will be the start of an experiment. It's one I need help doing from folks who fly light aircraft in the real world as well as online.

I've been working with Keith Smith on a project that uses X-Plane 11. I had been away from the sim for a while, so this was a good opportunity to look with fresh eyes, as well as get used to this new Thrustmaster brushless joystick. Not being an X-Plane user I wasn't happy with the initial control response. I tried Keith's settings, which were better, but still didn't feel right for a light aircraft. I finally found something that felt right — and then had an epiphany looking at the numbers in front of me.

That where I need help. Is this just me? Or am I really onto something here?

See the attached images of the control sensitivity from XP11. My understanding is the control response (left column) is how fast the stick “ramps up” in sensitivity. Zero would be a linear, half stick motion equals 50-percent control deflection, while 100 would be a serious exponential increase in control deflection for a given stick position, which gives more fine control at the center. The stability augmentation essentially delays the response of the sim to allow for a feeling of feedback. Zero would be a technically accurate flight model, but might feel wrong because there’s no breakout force or air-loading. A discussion with someone who worked with Level D sims brought up this same concept. Those sims have stability augmentation in pitch and roll beyond the technically accurate model to get the feel correct.

The default settings are 50 percent for everything. That sucked. After a bunch of experimentation, I came up with the 30-15-50 (Pitch-roll-yaw) you see in one of the pics, with about 12 percent stability augmentation for pitch and roll. None needed in yaw. This felt like a GA light aircraft to me. The Thrustmaster is super-precise (I highly recommend these) so I ended up dialing the settings down to 20-10-40 and seven percent augmentation.

That’s when I realized I had reverse engineered the 1-2-4 Roll-Pitch-Yaw control pressure relationship of a classic well-harmonized airplane. This ratio comes out of studies in the late 40s and work NASA did in the 60s. It says an airplane generally feels best to pilots if it's twice as heavy in pitch as roll, and four times as heavy in yaw. When I used to fly airplanes for Aviation Consumer magazine for review, this was one of the things we tested (with a digital spring scale in flight). Incidentally, the problem many pilots have with LSA is they are too light in pitch. Most have a roll-pitch ratio of about 1-1.2 or so.

Again, it's not the absolute numbers that matter so much as the ratio. I'm referencing X-PLane here, but I'm curious about people's experiences and preferences on all sims.

So:
a) What control sensitivity setting do you like?
b) What about stability augmentation? Do you use it at all? How much on which axises?
b) Note your settings, and then give a 1-2-4 (roll-pitch-yaw) sensitivity setting a try. What do you think? I'm particularly curious about how it feels compared to real-world airplanes.

Thanks in advance for any insights.

Jeff is my cohort from Real World VFR. I know that, at a glance, this may not be the most exciting topic, but when the rubber hits the road, the 'feel' of a sim is largely defined by the control settings that Jeff is talking about in his post. If people can roll up their sleeves and take a look at how they've configured their sim, the results would definitely be interesting when viewed in aggregate. Who knows you might end up with a sim that flies 'better', too?

I loved this post and am quite eager to get the feel in my sim as close to real world as I can. I'm away from the sim for a few days but will pitch in when I'm back.

This post was talked about in my Twitch stream this morning.

Hey Jeff, good to see you here!

I think you're onto something, but there probably are a number of factors that make each sim setup unique. My home setup uses a really nice PFC yoke which feels great at fairly linear settings on each axis. There's also naturally more centering force on the pitch axis than roll so 1:2 control response settings might not be appropriate. Before that I used a cheap Logitech 3D Extreme joystick, which I think is fantastic for the price, but required more exponential settings. That stick has about the same centering force in any direction, so maybe a 1:2 setting relationship would be right. That all depends on the flight model though, too, since it's not really the relationships between control travel we care about, but the pitch/roll/yaw generated by control deflections. Changes to the elevator specs or phase-out in Plane Maker would alter the relationship, for example. I use a plug-in called X-Assign that lets me save control profiles for each aircraft, and I've settled on different numbers for different aircraft just by feel and trial and error. Probably, what I'm “feeling for” is something in the ballpark of the 1:2:4 you describe. I know that on the TouchTrainer at the flight school, which uses a modified Saitek yoke, I've settled on higher numbers for pitch control response than bank for many aircraft, but I don't have it in front of me to check exact numbers.

I'm curious, what measurement was roll-pitch ratio based on when you were review flying? Is it something like force required to change pitch by X degrees in Y seconds versus the same for roll?

Hi Ryan. Yes, I think the goal is a 1:2:4 ratio. If you have asymmetric stick forces to begin with, I'd think the software adjustment would be less. To be clear for everyone: this relationship is for a "well-harmonized" aircraft. If you want a realistic feel for a fighter, or many high-performance kit planes, or many pre-golden-age aircraft, this wouldn't be the correct solution.

I think the underlying concept here is that control response + minor augmentation might be a good proxy for control forces themselves. If you know the control forces you're shooting for, getting the right *ratio* of control response could be a great start. Since most general use aircraft shoot for 1:2:4, that's a good general starting point.

As for measuring the actual stick forces, I just used a simple digital "spring scale." It had a memory for max sustained weight (force), so I could connect the scale to the yoke or stick and pull the plane through 90 degrees of roll (45 left to 45 right, for example) and 25 degrees of pitch (+15 to -10 and back). Averaged over several tries you could get pretty good numbers. Yaw forces were a guess. We tested phugoid, trim band and many other items. A flying mentor of mine was a DoD test pilot for many years. He helped me develop the review sheet. He also had the distinction of flying the last P-51 in service to the U.S. government ... as a chase plane on an experimental helicopter program. After that, he flew the Mustang to the bone yard, shut it down, and walked ... slowly and sadly ... away.

Hey...

Gonna bring back this thread alive. I have been looking for something like this...and low and behold, I find it here on Pilotedge forums. How cool

So which is the better settings - the one listed on 2nd image of the Op post or the 3rd image which as the 1-2-4. I too have a Thrustmaster T.16000.

Fascinating post