Hand Flying

[Hardy Heinlin]

I have the Thrustmaster Warthog. The centering force near the center is very high; this always provides a clean neutral signal which is fine. But it doesn't feel like a real aircraft stick where the centering force rises gradually with the distance from the center. The Warthog feels like a snapping switch when approaching the center. Does this effect occur on the TM 16000 as well?

[G-CIVA]

Hardy,

Thanks for this update...& to the real world pilots & former pilots that assist you in keeping PSX up there as THE best flight simulation product.

[Cbf]

The centering force of the TM 1600 is not really high. I use 15 for elevator center zone. At A/P disconnect, pitch has a tendency to decrease (about 1/2°).
The advantage of this joystick is the ambidextrous option.

[Hardy Heinlin]

At A/P disconnect, pitch has a tendency to decrease (about 1/2°).

There's abolutely no logical link between A/P disconnection and pitch change. Yes, a change may occur; the atmosphere is alive, the engines are alive. But depending on the scenario and the finest butterfly movements, the pitch may also rise -- or just remain constant for some seconds. The air mass the aircraft is flying through is not a frozen, steady world.

Also, the A/P adjusts the stab trim from time to time when a certain tolerance limit is exceeded. When you disconnect the A/P in a moment where the A/P is applying an elevator input and the stab trim isn't perfect, then you will, of course, need to apply the same elevator input that the A/P has been applying until this moment.

[Panos Bilios]

Hardy,

thanks for the modification, my landings have greatly improved, crosswind too,
thanks also to all the real world pilots for their input.
The best flight simulator gets only better.

Panos

[Cbf]

Hardy, I agree with you but most often it is pitch down.

[Tsuru]

The nose down pitch moment when closing the throttles in the flare was one of the things which I found a bit off before, I'm very pleased that this was resolved with revision 182.

One thing which has always "irked" me about the 747 (the aircraft as represented in our level D sims, not just PSX...) in particular is that in terms of speed stability it seems to be just so happy sitting at VREF +15 or -10 on finals, without really telling me what she wants through pitch reaction or elevator feel as much as something smaller, say a 737 or a large bizjet. This removes a bit of a lack of feel of speed stability which I always found extra challenging, having to pay extra attention to the speed tape to make sure I hit the numbers. PSX captures this feeling so well for me. I don't suppose over the years you didn't think to dabble a bit with with trying to run QTGs with it to see how it matches up with the response of the real thing?

As a general comment, I have been a fan of Hardy's work for many years and have recently gotten back in to home simming after buying PSX, and I just couldn't be happier. I continue to be amazed at PSX's realism, its crispness, adaptability and the continued support and evolution of this fantastic product not just by Hardy but also the community which surrounds it
I recently started simming again in earnest, with PSX as a flightmodel and instrument panel, MSFS2020 as an image generator and created a little multipanel to change the radios, baro setting, autopilot parameters and other things out of a raspberry Pi, a dot matrix display and a rotary encoder since of course no touchscreen can beat a real knob. Pondering what to do next...

[Hardy Heinlin]

Gentlemen, thank you all for the kind comments.

I don't suppose over the years you didn't think to dabble a bit with with trying to run QTGs with it to see how it matches up with the response of the real thing?

QTG = Flight Simulator Qualification Test Guide?

I want the PSX price to stay below 350 Euro :-)

[Alex268]

Hey guys,

Yes Mark, checkride was completed at the end of April. Im about half way through line training at the moment. Im currently on vacation but as soon as I get back I will download the new update and do a few flights. Ill give my feedback as best I can based on my experience so far. She really is a beauty to fly and im very much enjoying her. Please dont hesitate to ask any questions or for any particular information and I will accommodate as best I can.

Alex

[Tsuru]

Another thing that I found that might be worth considering for a future release: I did find PSX fairly easy to over-rotate and tailstrike on takeoff. Rather than looking inward, I suspected that PSX seemed to lack the proper tail-in-ground-effect-effect: When the tailplane gets deep into its own ground effect on takeoff rotation, most airliners with a conventional tail typically have a very noticeable cushioning effect. A reduction in pitch rate, requiring more elevator input to overcome to continue the proper pitch rate to V2.

This is backed up with objective flight test data: whilst I can't show you the numbers since they are proprietary data I can provide you with a qualitative comparison. I conducted a practical experiment where I perform a takeoff under identical conditions: max takeoff weight at sea level, with 20% CG and 7 units of trim. Since the ERAU output does not capture the flight control inputs but to still generate usable output I input a constant elevator input and hold, rather than modulate as you would normally, and compared the pitch rate characteristic through rotation, unstick and the first segment of the climb.

In PSX I can use this constant amount of elevator input through the sequence and hold this all the way through to the first segment climb: the pitch rate with continuous elevator input does reduce slightly for 3-4 seconds after unstick speed, after the pitch angle reaches 11-12 degrees, and then continues to its original value again. Compared to the real thing there are a few noticeable differences:

The effect, while present should start a few degrees earlier, well before unstick speed. In PSX I have to apply at least 50% of full elevator input to start the rotation at this weight and CG and keep this throughout, whereas in the real thing you would need only around 25% to initiate the rotation, and then increase to 50% to overcome the effect during the rotation once it kicks in. This is not entirely dissimilar what you would experience if you do not set the correct amount of stabilizer trim, so I put in a bit extra (+0.5, +1.0) during subsequent tests to put it right at the edge of the green band, but to no avail.

So it seems that the issue might actually be more related to the ground model (too much elevator needed to start rotation, drowning out any noticeable cushioning effect before unstick) rather than the aero model. I'm curious to hear from the lucky few who use a full control column to fly PSX and have had pilots fly their sim: the difference should be very noticeable to an experienced 747 pilot.

I hope this information is useful in further honing the product, let me know if you need more information 😊

[Hardy Heinlin]

Thanks for your feedback.

As you have noticed, the effect is actually modelled; now the question is: Where should the effect begin?

When you say 50 % elevator deflection: Does that refer to your hardware or to the EICAS status page?

You can set an elevator hardware bias on Instructor > Preferences > Basics.


Here's the modification I made seven years ago:
--------------------------------------------------------

Version 10.15 - 12 January 2018

15.08. Takeoff aerodynamics: Pitch rotation rate on the ground now starts decreasing slightly when pitch rises above 5 degrees, requiring more elevator deflection when rotating through 8 degrees. The effect is caused by the reduced distance between the stabilizer and the ground, and is fully washed out at 100 ft AGL.

-------------

Here are some further comments from 2018:
https://aerowinx.com/board/index.php/topic,4191.msg48757.html#msg48757

[Tsuru]

Hi Hardy,

The elevator deflection is based off of the EICAS page. The effect should start to kick in at 7 degrees of pitch and be a lot more pronounced, actively requiring you to "pull through" to maintain positive pitch rate.

I'm experimenting a bit with doing comparison to real flight test data and am still building a more detailed flight data recorder capable of capturing more parameters. At the moment the ERAU format built into PSX is the only reliable means I have, here is the result of one of my tests which shows us the exact effect of the ground effect on the elevator. Elevator input of ~50% of full is constant through the rotation, data shown is only from PSX:


Note that I only tested 1 condition so far (394 tons, 20% and F20), there is another case done with F10 at 224T/30% where the input required to rotate is even a bit less, and almost returning to zero after reaching unstick since the CG is so far aft.

[Tsuru]

Made a short test with revision 181 to see if this is a possible regression caused by the fix for the thrust reduction in the flare, but the result is the same

[Hardy Heinlin]

By the way, that second increase starting at circa 15° is caused by the rising airspeed which makes the elevator more effective.

[skelsey]

I'm curious to hear from the lucky few who use a full control column to fly PSX and have had pilots fly their sim: the difference should be very noticeable to an experienced 747 pilot.

I've asked the instructors at 744 for their thoughts; generally I would say that most type-rated pilots tend to overcontrol the sim somewhat, especially in the flare, as they feel the sim is less responsive than they would expect (usually expressed as feeling like there is a delay or lag in response, although the PSX elevator moves instantaneously if viewed on the STAT page). This would seem to be borne out if the elevator is not as effective (particularly near the ground) as it perhaps should be.

[Hardy Heinlin]

I think if I reduce the pitch axis inertia a little bit in general, that would solve Tsuru's issue better than changing the above mentioned starting point from 8° to 7°. I guess you guys wouldn't notice any difference if I changed that by just 1 degree.

[skelsey]

Thanks Hardy, be interested to try it out! It's an interesting problem because having used PSX both on a desktop with a joystick and in the full sim with real controls the "feel" is obviously very different and (in my experience) much "tighter" and more responsive on the desktop (which I suppose is to be expected since the controls are that much lighter and have much less travel), so I can appreciate it is a balancing act since those of us in full size sims are certainly a minority.

Feedback I have received so far re: rotation below:

It does seem to need a bit more elevator than I remember to unstick, after that the ground effect is about right. Was a long time ago though!

[Hardy Heinlin]

Well, that sounds like a matter of stick centering force rather than inertia. If the centering force needs to be adjusted for heavy yokes, you can set the elevator bias, as you know, on Instructor > Preferences > Basics. The linear bias is for average joysticks. Peter has set the bias to a more positive value so that small elevator inputs have a greater effect. That makes sense.

[Tsuru]

I think if I reduce the pitch axis inertia a little bit in general, that would solve Tsuru's issue better than changing the above mentioned starting point from 8° to 7°. I guess you guys wouldn't notice any difference if I changed that by just 1 degree.

That sounds like a good idea Is there a way for us mere mortals to see the polar moments of inertia and the relevant inertia product from the simulation? For this particular case the value for Iyy in either slug*ft^2 or kg*m^2 is of interest.

[Hardy Heinlin]

Is there a way for us mere mortals to see the polar moments of inertia and the relevant inertia product from the simulation?

No, but please notice my last message re "centering force".