How to counter tail rotor roll moment

[kimmik]

I came across this thought today:

The tail thrust produces a moment in the roll axis, rolling to the right. however due to precession the actual action is pitch nose up.

i noticed today on maidening my FBL SK540 700 that there is a slow pitch up. The swash is level to within 0.2degrees, so that isn't the cause of the slow pitch up. There was almost no wind.

So my question is, instead of making a level swash, should I set a small amount of pitch nose down in order to counter the moment produced by the tail rotor. Some people have said that setting some ROLL on the swash will counter the tail thrust, but that sounds erroneous to my logic.

[npomeroy]

I'm not big on precession physics. But I don't think you can automatically assume all the moments of rotation are going to be expressed through a 90 deg twist. The main rotor creates a reaction in the yaw axis and the tail rotor counteracts it in the yaw axis.

[kimmik]

Yes thats what the tail rotor is designed to do, counter the yaw moment from the main rotor.

But because the tail is on a different horizontal plane to the main rotor, there is also a roll moment. Which acts on the pitch control surface due to precession.

[npomeroy]

Yes thats what the tail rotor is designed to do, counter the yaw moment from the main rotor.

But because the tail is on a different horizontal plane to the main rotor, there is also a roll moment. Which acts on the pitch control surface due to precession.

OK but what I'm getting at is that precession (i.e. the reaction being offset 90 deg) comes into play when there as a significant mass spinning, and not when there are simply two opposing forces in balance. Otherwise when you made a see-saw tip there would be a significant sideways force at the other end! The torque/antitorque of the main vs tail workes within single plane: If there was significant precession the tail rotor would need to be offset 90deg but it isn't. Similarly the slight roll moment of the tail rotor being on a different plane has no angular momentum to it (it's not spinning in the roll plane) - it's only a static force moment in the roll plane and I don't think precession effects are involved.

[Gedexas]

Just to clear something up, I have a question that's not really a question. To create a roll moment, you need two points of reference, one pushing let's say left from below and the other one pushing right from above. The theory claiming that tail rotor causes roll provides the lower left pushing vector, but fails to produce the upper right pushing vector source, so my question is: what pushes the top of your helicopter to the right?

[npomeroy]

Just to clear something up, I have a question that's not really a question. To create a roll moment, you need two points of reference, one pushing let's say left from below and the other one pushing right from above. The theory claiming that tail rotor causes roll provides the lower left pushing vector, but fails to produce the upper right pushing vector source, so my question is: what pushes the top of your helicopter to the right?

Has to be inertia. Doesn't really need a right push from the top. In fact the so-called right roll will only happen if the horzontal plane of the tail rotor is lower than the centre of mass of the helicopter.

[kimmik]

Gedexas you don't need the two points to be symmetrical about the axis of roll. E.g. A spanner could tighten the bolt by you pushing on one side of the roll axis only.

[kimmik]

Npomeroy, I'm not talking about precession of the tail rotor. The tail blades are so light and small compared to the main blades, that the entire Heli is effectively one single gyroscope.

Because the main gyroscopic effect of the Heli is centred at the head hub, the tail thrust creates two moments on the Heli. One is yaw, which is NOT subjected to precession, as it's in the same axis of rotation as the big gyro, the head.

The second moment is the roll, which is
Roll Moment = tail Thrust x tail offset distance from the plane of main blade rotation.

This is the moment that cause slow pitching action on the Heli.

[Gedexas]

Gedexas you don't need the two points to be symmetrical about the axis of roll. E.g. A spanner could tighten the bolt by you pushing on one side of the roll axis only.

No, you are pushing two sides of the wrench, your hand is pushing the handle and the bolt is pushing the top of the wrench in the opposite direction.

In response to the inertia argument. Inertia not only prevents things from gaining speed, it also prevents things from slowing down, meaning that once the force has been applied for a short duration, it does not need to be applied to maintain motion. Since that helicopter is motionless, any moment acting against center of mass does not count in this argument.

[Gedexas]

Ponder this: When you fly the said helicopter upside down, it does not lean the opposite way, thus throwing the whole tail induced roll argument straight out the window.

[kimmik]

Gedexas you didn't realize that the Heli doesn't hover horizontal. There are three forces that form a triangle of vectors to keep the Heli still, one pair of opposing moments, and one pair of opposing but unstable moments.

A - Three vectors: tail thrust, gravity, and main rotor thrust, which leans to the right to counter tail thrust.

B - Two static moments: main rotor torque and tail yaw moment in balance

C - One static and one dynamic moment: tail roll moment towards the right, and main rotor gyroscopic force to counter it. The gyroscopic force only exists when the gyro is changing in its axis of rotation, thus the slow pitch up that dynamically balances the static roll moment from the tail.

Either way, it's a subtle motion of the Heli that's largely made more subtle by the stabilization mechanism. That's why you don't really notice it.

[kimmik]

I just did a vector diagram of a Heli at equilibrium in hover, I think you're right gedexus that the tail does not induce a roll moment.

In a hover, the clockwise moment from horizontal thrust is countered by the anti clockwise moment from the vertical component of thrust.

http://img.tapatalk.com/3cf2b7c3-5d7a-c945.jpg

[extrapilot]

OK

So a few misconceptions here guys.

First is, you cannot have precession if you don’t change the angular momentum of the gyro. Applying a torque does not imply work is done. If you don’t rotate the gyro out of plane, you don’t change its momentum. It the momentum doesn’t change, it doesn’t precess. If it did, gyros would not be stable reference platforms…

Second is, the rotor is NOT a gyro. That myth just will not die. The rotor blades fly to position, they do not precess to position. This is easy to disprove, simply because the phase angle is not 90deg. And, the phase angle changes with blade Lock number, and with air density.

Consider what happens to a spinning rotor on a heli which is on a moving platform (typically, a marine heli on a ship). If the ship is rocking left/right, the ‘gyro’ rotor should precess forwards/backwards. It doesn’t, it tracks the swash, and the swash, unless commanded by the pilot, tracks the rocking of the chassis. So, the rotor rocks with the machine/ship.

Why? Because the rotor is flown to a position defined by the swash. As the rotor deviates in plane from the swash for whatever reason (rocking, gusts, etc), the AOI is changed automatically, and the blades fly themselves to where they are supposed to be.

When you add an FBL in, the system’s task is to counter uncommanded roll. This changes the way the rotor will behave on a machine strapped to a rocking deck (because the swash is being manipulated by the autopilot)- but ‘if’ the TR roll torque actually caused chassis roll (which which be required for any roll to be imparted to the rotor via the rotor shaft/head), the FBL would detect that and apply counter cyclic to correct it.

On the whole torques question, just remember that a law of statics is that all torques sum at the COG. There are a whole lot of ways to get lost in this. There are several very real torques in action, the impact of which depends upon lots of factors. But, the net question here is- what is the result at the rotor? Is the chassis trying to torque the rotor out of plane? You have do to a lot of math on that one guys.

Here is the problem though. The solution is specific to a single condition. Consider a situation where you have some lateral cyclic to counter drift in hover, and you have a low TR, so this delta in arm length yields a roll torque on the chassis. Fine. You now accelerate, and you are slipping the machine slightly nose right in FF. Now, body lift is offloading the rotor, so there is no lateral force from the rotor required… The TR still must react the rotor torque though, and so the arms don’t change (head vs tail- you would need to find where the body lift arm is- but you get the idea...), but the magnitudes do (net change in roll torque). Then, as you accelerate further, the vertical stab begins to offload the TR. If the center of pressure of the vertical stab is not at the TR hub, the net arm changes again, etc.

And in inverted flight, you reverse the cyclic correction for drift, but not the TR thrust direction (torque doesn’t reverse with thrust reversal). So, again, a different situation, where now the main rotor is torquing the chassis the opposite direction…

Gets confusing fast…

Cheers
R

[kimmik]

Thanks extra pilot, I knew you'd have some insightful thoughts on the matter.

could I ask, if the rotor disc does not act as a gyroscope, why is the phasing not 0degree? Most helis have a cyclic phase of something like 85-90deg, as is predicted by precession. I.e. to pitch forward, the left blade goes negative while right blade goes positive.

I also do want to ask for some practical advice. My Tdr during its maiden has a precessing wobble in the fuselage. I'm using 88deg at the moment, ie the blade deflection is 88deg before the intended action. What phasing is typically used and how does rpm affect this?

[extrapilot]

Hi Kim

There is a gyroscopic component of force to all rotating things, but in a heli rotor operating in air, it is small relative to the aero forces.

Phase lag is a function of point of maximum rate of change vs point of maximum amplitude; many things in nature are this way- sine waves, kids in swings, etc. These are 90deg out of phase. Doesn’t mean they are gyroscopes.

Here is the thing with heli phase lag. The blades flap in resonance, like the kid in the swing. Flap is a function of the blade length, and the acceleration (G). The math is ugly, but the flap period ends up as one cycle/rev for a pivot at the hub, because the acceleration imposed by the rotation yields that period. As the flap rate is inversely proportional to acceleration, and acceleration is a function of RPM, the flap period stays locked to the RPM. Absent factors mentioned below, the system should show 90deg phase lag in a pure teetering head.

But, with a blade in air, things change a little. Air damps the flap, so the peak amplitude is going to be < 90deg after peak rate of change. This is expressed by the blade’s Lock number (the ratio of the aero vs inertial forces of the blade). Different blades may have different Lock numbers, and as a result, different phase.

Also, there will likely be a complex component of ‘flap hinge’, where for whatever initial flap your dampers allow the head to teeter at the hub. Once travel there is exhausted, any additional flap is now offset further out (grips/blade/both). This increases the natural flap frequency of the rotor (reduces the phase lag). Dr. Leishman has a good chapter on this (Principles of Helicopter Aerodynamics, section 4.7, Dynamics of Blade Flapping with a Hinge Offset). There, he reports that for hingeless rotors, similar to ours when we use stiff dampers, phase lag is typically 75-80deg.

I wish I could be more helpful with your config, but so much depends on these variables. Maybe ping some TDR pilots who are flying your blades/dampers?

Regards
R

[kimmik]

That's very interesting, I'll see if I can procure a copy to do some light reading. I have mentioned the phasing issue in the Aussie forum but they don't think the precession wobble is a phase issue, so I guess they're all running 90deg standard. I tend to disagree, since retarding the phase by 2deg has improved the wobble. I will try retarding to 86deg next though weather has been moist lately. With what you said about a rigid head having 75-80deg phasing, I will have confidence in exploring this setting a little further.

[Gedexas]

Kimmik, there's a little problem with that vector diagram, you drew the heli at an angle and then made all vectors based on that, that looks really confusing and hard to see what causes what. If you drew the heli perfectly level and then drew the vector from the tail rotor you could see that there's nothing causing the heli to tilt, only to slide sideways.

What causes the tilt? Pilots's thumbs, otherwise the heli would slide away to the left. Do not confuse cause and effect, tilt is not caused by the tail, tilt is caused by the pilot to combat the tail rotor.

[Just Another Flyer]

Can't you just try your original solution and see if it fix is it. Also what about helis with a raised tail, would that change anything?