The "Torque Fix" and the MU2...

[jcomm]

I've been successfuly applying the "Torque Fix" to many of my prop aircraft, which is based on a very simple LUA script directive . This includes the LES dc-3, which is Gizmo-based, and it worked there too.

 

When I tried to apply it to the MU2, v1.5, the results were rather unpredictable... It doesn't work and even produces very strange effects.

 

Could Tom Kyler or someone with experience in Gizmo advise me on how to make it work?

 

I am aware that this "tweak" explores an undocumented legacy feature of oe of the datarefs in X-Plane 10, inherited from previous versions, and allows for decimal values where

a 1 or a -1 were actually the only values expected, but it works beautifully so far, and I would really like to try it in the MU2.

 

Ah!  I even took care to set it to -0.3 for both props since the MU is CCW!

 

Thx in advance for any suggestions ;-)

Edited August 21, 2013 by jcomm

[tkyler]

Could Tom Kyler or someone with experience in Gizmo advise me on how to make it work?

 

 

Hi jcomm.    I do not subscribe to Dan's  'fix',...there is a reason BenS cautioned him and I would not make any recommendations or expend any effort to alter the torque in the MU2 as I've flown the darn thing and it simply has a ridiculously strong roll tendency; however, I fully understand the desire for users tweaking to their comfort level though, after all, it is your personal enjoyment you are seeking, but folks are on their own on this one.   The torque in x-plane is fine, its real, and accurate and mathematically I can demonstrate it.  What is not real and accurate though are a whole host of other factors that exacerbate the perception of the effect though so I can easily understand why some folks would want to "turn it down" to compensate....its just happens to be one of those areas, cognitively, that I find accurate and acceptable and therefore am not the one to expend time to investigate the 'fix' or the implementation.

 

Do post any results you find though...I'm sure other folks are definitely interested.  Certainly swapping one of the prop rotation directions is the fastest way to lose the MU2 roll...but perhaps you're only seeking to minimize it a bit.   If you can't get it to work though.....then there is another tweak you can do and ironically, its what we do in the real MU2.  Use aileron trim

 

TomK

Edited August 21, 2013 by tkyler

[LA]

I'll have to also get into this later. Must leave to work now. I know that Tom has some right seat time. I've read several years of Tom's MU2 development from the org. I've also read every MU2 pilot report I could find on the internet, and watched many videos. I did this for the MU2, and many other twin turbo's. This almost became an obsession, and I was banned from Avsim for a few weeks, due to my continued observations on the subject. I'm NOT going along with the strong roll tendency. If nothing else, it was good for several months of night reading. In the meantime, having to adjust for roll, means drag. Dropping a wing on rotation is not good. If the lift from the wings (and possible engine offset) can't override a want to roll, airplanes would have been designed with counter rotating props for twins, and contra-rotating props for singles a long time ago. It was learned early on, that applying different incidences for wings, to compensate for torque, wasn't the best of ideas, as it creates an opposite roll, when power is pulled back. Having to use trim as a standard measure, would be doing the same. I got into this, because of my real life observations as a pilot of higher performance singles........which could have lot's of torque on the takeoff roll.

 

LA

[jcomm]

Tom has mentioned something I fully agree with - mathematically the torque calculations in X-Plane 10 are probably near perfect, but the fact is that other factors that could help turning it all a lot more into yaw aren't that perfect. So we are left with two options - either accept that the torque is correctly modeled, which I am able to accept it is and forget about the other factors not being yet there or well simulated, or try to, while those other factors don't get better modeled, overcome the roll tendency using tweaks... The torque fix is one of those tweaks.

 

Turboprops and twins, even with non-counter rotating props are better known for the lesser effects of the turning props and engines. There are a few easy t understand explanations for why a Baron58, a Beech 1900D, etc, require no aileron input or aileron trim on takeoff and climb - only rudder / rudder trim.  Given that the MU2 has such a small wing area / high load factor, I admit the torque effects will be present at least when power variations occur.

 

Of course I look fwd for the day Austin announces that those other factors that come into play are being correctly modeled, just as the torque !

 

Thx Tom for the reply, and meanwhile I think I'll keep using the aileron trim on the MU :-)

[tkyler]

yea..it's been debated endlessly but its a matter of perspective IMO.  Take a car rolling backwards slowly next to another car.  For an instant, a person questions whether he/she is rolling back or the other vehicle is moving forward...and for that instant, you don't know and have to look a little futher to make a final determination.  Some folks would say the car is moving back, others would say the adjacent car is moving forward.    The whole "perception" of too much torque is attributable to multiple factors and each person has a different perception of what the cause is (I know I'm resolute on my conclusions) and my whole point during the forum conversations was that the torque reaction is not the mathematical problem...but rather the lack of other factors fidelity that put torque in the spotlight.   BUT.....of all factors, adjusting a torque effect in sim would be the easiest solution to compensate for other algorithmic limitations no doubt.  So all this time, I just don't like the words "torque is wrong" or "torque is broke"  or "fix the torque".   Torque is one of the easiest and most fundamental quantities to calculate.   Its right, but is the final motion of the aircraft "realistic" is the "effect" realistic?  Definitely not always as Larry can attest with his experience......should torque be "tunable to compensate"?  yea...I think so.  Desktop simulation is a compromise in many areas and I like that fact that folks can fine tune things.  It just so happens I prefer not to tune the torque, at least not on the Moo   I'd tune it down on some other aircraft though, but I'd never say the torque was implemented wrong in x-plane.

 

TomK

Edited August 21, 2013 by tkyler

[jcomm]

but I'd never say the torque was implemented wrong in x-plane.

 

I agree Tom,

 

And, I can't forget how other aspects are so much more detailed in terms of simulation in X-Plane. It's nice to see things like the propwash being calculated ( only yesterday by Austin's suggestion did I enable it in the Data output in order to better tune a prop aircraft using one of the parameters that asks for cruise speed + 1/2 propwash), the control surfaces moving with the unstable air around them, even the effects of the asymmetric hit of different aircraft surfaces by the spiraling slipstream etc..., the myriad of hypothesis we have to edit the airfoils using Airfoil Maker, etc...

 

It certainly is a different approach altogether, and the fact that I'd like to see some aspects done better doesn't mean I can't recognize the complexity and quality of the sim, as you know :-)

 

Yesterday I got yet another aircraft into my X-Plane 10 hangar - the X-Aviation C400 was my latest acquisition, and I really like this model!

[wim1976]

The whole problem is that we behind our computerdesks we can't feel the rolling of an aircraft, so it suprised us many times. If you feel the roll, then your natural reaction will be compensating with the controls. The problem is not X-Plane, it is the simulated situation. We have to live with it, maybe you can compensate this with force feedback yokes or build/by a motion-simulator-base  .

[LA]

I'd tune it down on some other aircraft though, but I'd never say the torque was implemented wrong in x-plane.

 

The torque itself, could be quite correct. Have you ever seen the video of the turbine powered Lancair, that had it's first engine run without the wings installed? As the pilot added in more power, the airframe twisted to the left, collapsed the landing gear, and shredded it's very expensive mass of engine and airframe all over the ramp. That's the torque effect, that X-Plane reproduces. It's there!  But add the wings, and especially some airspeed, and lift will now counter that torque that wants to twist the fuselage. As physics says, lift squares with airspeed. And "this" is what X-Plane isn't doing the best job of.

 

The torque fix is a compromise. You're actually reducing the simulated torque, rather than X-Plane being able to simulate the lift that counters it. Problem is, X-Plane is creating some false aircraft physics. The question comes up all the time ( just as it has today at the org.) and the one who inquires, is told to use aileron trim, or now the "torque fix". It's then implied, in some cases, that the torque fix, is for those who don't want to be bothered to fly an airplane properly.

 

So, yes, I do have a problem, in the way that X-Plane produces the final result. We shouldn't have to be constantly trimming for roll. There are companies that specialize in re-rigging airplanes, to get rid of any roll or pitch, that adds to drag. When you use aileron trim, or fixed tabs to compensate for roll, you are adding drag, at a cost of airspeed & fuel. You should be able to handle any initial tendency to roll, with just aileron use. The sensation shouldn't be a heavy one. There will always be a need to use aileron trim for heavy wings, due to passenger, load, or fuel imbalance. Either wing could be heavy. And that's if.......you do have aileron trim. Many planes don't.

 

A few months ago, I mentioned this on our experimental aircraft builders forum. A RV6 & 10 owner, who flys a crop dusting ag plane with a Pratt & Whitney turbine, initially replied, that yes, torque is there and you have to compensate with aileron. A week later, he comes back, after a recent flight and says, you have to use whatever control is nessesary to control the airplanes attitude, but when it got right down to it, the torque effect was just inconsequental. He just hadn't really gave it that much thought before. IMO, that's much closer to the answer that Tom got, from MU2, pilot #2. 

 

LA

[jcomm]

At least one thing is for sure:

 

At least in X-Plane we can effectively create the real world compensations ( less those aerodynamical that I believe have to be fine tuned in future releases and LA mentioned, such as a dynamic pressure building up and than compensating, on most aircraft, for the torque), contrarily to the other sim I have used too and where you either completely tune down the torque or, if you have it, and all prop aircraft in FSX have the roll due torque!, no matter what combination of tuning parameters you use, the results are going to be very weird.

 

In FSX you can cant the thrust axis up/down and left/rigt just like in X-Plane, but don't expect any near the same plausible consequences. You can also play with tuning parameters for p-factor and torque on roll, and even make them depend on AoA, which looks great as a start - problem is the way those parameters affect the flight characteristics, which is very unrealistic to say the least...

 

I have learned to live with this detail, as well as others in X-Plane Ground handling under crosswind is also not very well done in X-Plane ( it's overdone IMO ) but! ground handling as a whole in X-Plane is a lot more realistic in terms for instance of sideways and rolling friction, than in FSX, even using some tweaks for that other sim!

 

Taking off on a good X-Plane prop model, with well done sounds and flight dynamics designed to do the best at matching the real thing is, for me, a much better experience than in that other sim, at least for some aircraft I have had the chance to right seat, and even left seat :-)

 

I am a glider pilot who once considered "adding engines" to his license... A few weeks later a fellow pilot died and the other got seriously injured on a crash with the same Tecnam I was using for my instruction... :-/  From there on, engines only in X-Plane or DCS, and in true prop aircraft, or when flying on an airliner, as opposed to those so called "ultralight aircraft" with engines that when observed/inspected with attention and good sense should bring some thoughts of getting away from it - sorry for the ULM pilots around here.... you are all kamikazes somehow...

 

EDIT:  I would really like to add just a bit more...

 

The other day I bought another X-Aviation excellent aircraft - the C400. I am really enjoying everything about it. It's also Austin's aircraft ( or at least was before the accident in the hangar.. ) and many times when I exchanged oppinions with him regarding the roll due to torque - let's not call it "torque bug" again, at least I won't because it's NOT A BUG in the torque modelling! - he told me he certainly made use of the aileron trim when flying it.

 

Well, the Corvalis I bought comes with very complete documentation, and reading the aircraft POH it really suggests that the electric elevator/aileron trim is there exactly for it!

Also readng a descrition frm a C400 pilot on the net, an interesting review of the real aircraft, I found exactly the mention to the roll developing at high power settings, not only while climbing!!!! Well....

 

Then, there is yet another source if respectable information for me, which on some of the included aircraft ( the Bonanza A36 at least) does show evident roll due to torque or other prop effects!

ELITE Premium, my 3rd simulator, and one I use often when I am in the mood to pretend I am IFR-rated :-), includes a "hangar" of nice GA representatives. Among them are Cessnas, even equipped with rudder trim, Pipers, Mooneys, Beechraft and even the Trinidad T10 and T20 models. Well, while some clearly show only yaw at high power / AoA settings, the A36 shows a lot less yaw, but a very noticeable roll due to torque! 

 

Now, I know ELITE excels at replicating the flight, engine and other important characteristics of the modeled aircraft included in the package. You can actually fly by the POH... so, having never flown in an A36, but having ELITE's "oppinion", I have to accept that most probably the real thing does show a rolling moment due to torque!

Edited August 22, 2013 by jcomm

[LA]

Then, there is yet another source if respectable information for me, which on some of the included aircraft ( the Bonanza A36 at least) does show evident roll due to torque or other prop effects!
ELITE Premium, my 3rd simulator, and one I use often when I am in the mood to pretend I am IFR-rated :-), includes a "hangar" of nice GA representatives. Among them are Cessnas, even equipped with rudder trim, Pipers, Mooneys, Beechraft and even the Trinidad T10 and T20 models. Well, while some clearly show only yaw at high power / AoA settings, the A36 shows a lot less yaw, but a very noticeable roll due to torque!

Now, I know ELITE excels at replicating the flight, engine and other important characteristics of the modeled aircraft included in the package. You can actually fly by the POH... so, having never flown in an A36, but having ELITE's "oppinion", I have to accept that most probably the real thing does show a rolling moment due to torque!

 

I just have to bring it up again.  Will never let it go, I guess......

 

Was talking to another pilot friend the other day. He's owned a Beech Debonair for many years. Okay, it's just 225 HP versus 300 for the A-36. Both planes have a near identical wing span. Naturally, I had to ask about torque, and the need to use aileron and perhaps trim to counter it. His response, was that it's right rudder thats required on the takeoff run, and just use any aileron pressure to keep the wings level.....if required. As to a sensation of roll, and perhaps some trim to counter it?  The answer is no. Geoff Applegate at Avsim, also owned a Debonair. He said the same thing. He seldom touched the aileron trim. Granted, there is a 75 HP difference here. But when another friend, said it was all rudder, when flying the 2700 HP single engine Skyraider in Vietmam...............and not aileron, I can only speculate. The Skyraider pilot specifically told me............aileron is the wrong control. It's rudder! BTW, this same pilot has also been a flight instructor for many years.

 

P.S. --- I just happened to end up on this thread again, in my continual search of "torque and airplanes". The never ending quest..

[Goran_M]

Holy thread revival, Batman!

[Hueyman]

I don't understand why you say this is the wrong action to add ailerons ...

 

Well, it's really function of an airframe and if pilots feel the need to add a tiny bit of ailerons after the rudder action, it is because it's requested.

 

Different airframes, cowlings shapes, wing surface etc will of course be determinant factors, but globally, the slower you fly, the more noticeable this effect is ...

 

For instance, try doing a go-around with a powerful GA plane ( the Cap 10B is 180HP, injection engine for a lightweight but coarse fixed pitch two bladed propeller, for a small and light two seated plane, the weight/power ratio is very good  )

 

 

For example this is me doing some traffic patterns and so go-arounds. You'll notice the great ammount of right rudder and the tiny ailerons input. As I said, it's also a function of speed, so the faster you are, the less you need to input. I was climbing 1500ft/min and a slow max slope speed of 75IAS

 

So, I wonder why you say it's non existant ... it is the exact same torque effect that apply on a helicopter, by the engine torque and around the rotor mast axis ... I must admit it is most of the times overly present in the sim but this is a real behavior

[LA]

I don't understand why you say this is the wrong action to add ailerons ...

 

It's not a wrong action to use ailerons. Airplanes are not  comparable to helicopters, as far as torque is concerned. Helicopters don't have wings, and the helicopter blade is much larger than an airplane's. What's wrong about X-Plane, is the assumption that there is a strong enough force, to roll the airplane left after takeoff. It's as if this force needs to be corrected with constant aileron pressure or trim.........just as elevator and perhaps rudder might be (if there is rudder trim). By the time the airplane has gained enough speed for a normal takeoff, the effects of torque, will be greatly diminished by other forces that over power it. An airplane wing has the helical prop wash hitting the bottom of the left wing, as well as pushing down the top of the right wing. Any time that there is dihedral or anhedral, you have rudder forces that will counter roll. The engine "offset" (cant) is doing the same. And it's forces vary with power. More torque due to power, and more offsetting forces at the same time

 

At slow airspeed, high power applications such as a go-around, it could feel very normal to see and feel the left landing gear push towards the runway. This would naturally be offset with right aileron. Depending on rotation speed, from this quick go-around might even cause a quick wing dip...........to once again be countered with right aileron.

 

My response, here, is due to the edit, of the last comment......before the thread went dead. As usual, it's just a case of ----- well that must be it, then. It's wrong, it's the wrong assumption. It's why I'm always willing to ask, for other pilot opinions, in addition to my own pilot opinions. It's why I keep reviewing all of my WWII fighter instruction videos, as well as my flight manuals for these high powered piston airplanes. For example, the P-47 has aileron trim at neutral. It very much mentions rudder for the takeoff sequence to counter torque, but not one word about aileron. It's the same for a P-51 takeoff video, that recently watched. It's a WWII training film, and discusses rudder, but not a word about a heavy roll or need to combat a roll motion on takeoff. Just rudder, as well as a trimmed right rudder for takeoff. The F4U Corsair calls for 6 degrees of right aileron trim before takeoff. That's not a lot of trim, but could certainly ease the push on that left wheel.

 

All of the pilot reports, in all of these planes, never mention the need to watch for, and counter roll motions on takeoff. A lot of them certainly discuss the need to watch go-around power, to prevent a loss of control, and roll, because the flight surfaces don't yet have enough authority to counteract the torque.  It's an accepted fact, just to use aileron to counter any roll motion, in either direction.

 

To ask these questions, I'll tell the pilot, that I'm interested in some information for "home use" flight simulation purposes. In reality, most everyone I've asked, just hasn't used a desktop flight simulator. I have a tendency to ask high time pilots, or at least ones who have flown planes such as the Skyraider or B-17. I'm an old guy, a past airport "bum", and have been around the flying scene for a long time. It's not as if I'm a kid, asking strangers for a few unknown answers. I specifically ask about the effects of "torque" roll, and the need to counter it.

 

The answers are always the same. It's right rudder! They feel the same way as I do! It's the same reason that many small airplanes don't even have aileron trim. They have that right engine offset. Even radio controlled planes do. Changing the incidence of the wings, or adding fixed trim surfaces to counter-act roll, have gone out of style. It just means "drag". And it also means oppsite aileron anytime your desending at a decent airpseed with power pulled back. The normal response for this is "left rudder" on descent, not aileron. My P-51 training video discusses the same.

 

Let's go to the MU2 for a minute. There is some very interesting pilot reports for that airplane. For owners who maintain a well rigged airplane, it's a dream to fly. For those cargo carriers who may be flying a high hour, somewhat beat, and probably out of rig airplanes, it seems to always be a handful. There is some good discussions about flaps being slightly out of adjustment, which causes roll and trim headaches. And at the same time, the pilot/owner with a well rigged plane, mentions fuel flow adjustment to keep the engines at the same rpm, and little or no need to even use those electric "aileron tabs" mounted on the flaps for trim.  There is certainly a love/hate relationship for MU2 pilots. Those who own and maintain them with pride, seem to love them, as is in evidence with the FAA reports, that came about, due to an investigation because of frequent accidents. The MU2 was aquitted.

 

In the meantime, I as an airplane builder, would get the "roll" removed, if there is a heavy wing. Austin Meyer, and one of his test pilots for his turbine Lancair experimental kitplane, did the same. Austin talks about removing a heavy wing during test flights. In other words, "roll" due to torque, just shouldn't be there. It's over ridden by greater powers....so to speak. If it wasn't, planes would have been designed much differently for all of these years. Wing dipping on takeoff, isn't the best flight characteristic, to have to deal with. It would seem that something is "wrong"......and needs to be fixed.

 

Note: it's left rudder, if the props turn the other way. And I'd still like some anwers about the anheadral on the MU2, since it's used as a "anti stabilizing" effect, as with many high wing airplanes. What are the exact effects of rudder and anheadral on this airplane, as to the direction of roll? Going to the net.................it's a bunch of pilots disagreeing about these "force" effects. Or a engineer and a pilot disagreeing on the same subjects. Take your pick, as you usually won't get the perfect answer to the question you've wondered about.

 

Until next time.....

Edited October 12, 2013 by LA

[Hueyman]

Hey LA your post is very instructive and a pleasure to read !

 

So the final word is that this tendency is greatly reduced by rudder and also some wing position/engine mountslight orientation.

 

I agree, I just said that during high power/low airspeed operation, such go around, you may feel on some airframe that tendency more than others. And it's less here during take-off because the power requirement is not so " sudden " and as the plane gain airspeed, the pilot naturally find the correct stick position for, when light on wheels , it would lift off approximately " balanced "

 

And I know for the differences about helicopters vs airplanes, it was just a very crude example of the principle.

 

Thanks for your long time investigation, that has helped I think to demystify that " issue "

 

Cheers

[OT2]

If I might join in this conversation, I find that the most uncontrollable aspect of the MU2 is landing.  When coming in on glideslope and disengaging autopilot, the plane takes a sudden dive to the left. I have found that adjusting aileron trim to near neutral nearly does it, but leaves a surprise at almost every landing.  I have also found that moving aileron trim to the left slightly before take-off reduces the roll, but it is critical to adjust during flight to make sure that it is back to near neutral prior to landing.  It improves, but being ready with yoke and rudder is essential.

 

John

[tkyler]

When coming in on glideslope and disengaging autopilot, the plane takes a sudden dive to the left

 

Indeed you must manage "roll trim" all the time, even with the AP on, including coming in for landing....the autopilot will attempt to level the airplane to counter any roll trim inputs.  When approaching for landing, the torque is typically very low since the power levers are retarded, hence the roll tendency will be low, hence the roll trim input should be very low.  If you mind the details, then turning off the AP will not be very eventful.  A good rule of thumb...is when the autopilot is on, adjust the roll trim to keep the yoke relatively level.  If the yoke is turned with the autopilot on, then too much roll trim will be the culprit.

 

TomK