8 simple rules of good gyro design [Archive]

darrellwittke

04-23-2004, 05:37 AM

I just read a post of G. Gremmingers concerning MOI and tandem seating that contradicts something Dan Haseloh told me once (Surprise.) I also have some other things which may or may not be true that I have picked up lurking around Norm's forum for years. So, in the interest of argument, entertainment and education, here are the rules and the progenitors (I think) thereof. (Forgive me if I misconstrue....and feel free to correct.)

1) Lightness counts in all things that fly, from birds to Boeings. Ounces count. (C. Beaty)

2) Ideally, the thrustline should pass through the center of gravity and the center of aerodynamic pressure, although difficult to do for all speeds. (C. Beaty, D. Riley, others)

3) Gyrocopters optimally should have tailfeathers that produce 4 times the force (through sq. footage and lever arm) than any destabilizing force in front of the rotor thrust line (the dividing line). (C. Wall)

4) The rotor thrust line should pass through or behind the center of gravity, not in front of it. (D. Riley)

5) The 2/rev vibration is simply a product of the path the rotor must fly, and should be accommodated by a flexible pylon, rubber mounting, or a slider rotorhead if possible. (C. Beaty)

6) Gyrocopters are one of the most inefficient ways to fly, but properly designed gyrocopters offer one of the safest ways to fly. (J. Mayfield)

7) Returns in performance are greater with aerodynamic drag reductions which reduce to the 3rd power (cube law) than in adding horsepower and thrust which operates to the 2nd power (square law). (C. Wall)

8) Gyrocopters with suspension should have adequate damping to preclude ground resonance with rotor. (such as A&S 18A). (C. Beaty)

9) Full cabin enclosures are very difficult to do in gyrocopters, if attepted they should be analyzed thoroughly for any lifting moments they may produce ahead of the center of gravity and mitigated by more thatn adequate tailfeathers behind the center of gravity. (G. Gremminger)

10) Powerful prerotators are limited by the ability of the brakes to stop forward movement as the rotor lifts the gyro. Depitching of the rotor blades are necesary for higher prerotation speeds but generally not worth the complexity and weight on simple gyrocopters. (C. Beaty)

11) The center of gravity should be kept as close as possible to the rotor thrust line. (D. Haseloh)

12) The center of aerodynamic pressure is what is most important for the thrust line pass through with speeds in excess of 40 mile per hour. (C. Schneider)

I know that's over 8 rules but I couldn't give up the catchy title! I know there is so much more, aircraft design being a complex task, and simple rules like these do not contain enough depth and probaly will make for a long thread, but it is a start. I check in infrequently but will try to reply.

darrellwittke

04-23-2004, 06:31 AM

Some background on rules 11 and 12, of which I have questions about. Rule 11 was told to me by Dan Haseloh circ 1990,1991 when I went to Kindersley from my home in Montana. I received my first gyroplane ride from Dan and I remember him stating that the weight of a gyroplane should be kept close to the mast. In his words, as I remember, "Gyrocopters should be short and stubby."

I see no reason for gyrocopters to be kept to side-by-side versus tandem (stubby vs. leghtened) as long as there are adequated tail feathers to compensate.

I know the above statement might generate more anti-RAF rhetoric, but an a personal not I found Dan H. and interesting and affable fellow with several interests akin to my own (ie. weightlifting and hockey.) Dan was by far the most friendly there and, as Chuck B. has alluded to, if Dan was still alive things might have changed at RAF. (Dan Haseloh was killed while flying formation (for pictures) with another RAF in 1993?)

Rule 12 comes from a short post by Carl Schneider stating center of aerodynamic pressure is what is important after 40 mph. I don't remember the reasons for the statement and hope someone expounds on it or points me in the direction of the relevant posts if I missed it.

Again, Sorry if I have confused the authors, I know many of the statements had more than one progenitor.

My thanks to all who vet these statements as to correctness. Special thanks to those with education through college, experience, or both, who share and educate freely.

Darrell Wittke

Udi

04-23-2004, 08:34 AM

You are confusing two separate issues, Darrell. Rule #11, as stated, simply means that the sum of all the pitching moments acting about the CG should be as close to zero as possible. When the sum of all the moments is zero, the rotor thrust vector (RTV) is passing right through the CG.

In your second post you say:
the weight of a gyroplane should be kept close to the mast. In his words, as I remember, "Gyrocopters should be short and stubby.

This relates to the gyroplane moment of inertia about the pitch axis. Some people say that a lower MOI is desirable because a high MOI is a dynamically destabilizing factor (Beaty et al). Other people maintain that a high MOI is desirable because this helps the gyro resist quick changes in pitch (Gremminger et al). I am sitting on the fence waiting for the one who can prove it either way.

Udi :cool:

raghu

04-23-2004, 08:34 AM

All else being equal increasing MOI makes the control situation worse- more lag and overshoot. Not sure why Greg G. would recommend higher MOI.

Common notions of prefering longer coupled versus more shorter coupled crafts are based on the length of tail arms and thus damping (damping increase with the square of the increase in tail length)- damping reduces overshoot.

cheers
Raghu..

raghu

04-23-2004, 09:49 AM

Udi said, " I am sitting on the fence waiting for the one who can prove it either way".

Udi,

Let me see if I can tip you over to the"right" side of the fence. Now, all things being equal, two issues should concern you vis-a-vis incresing MOI: 1) What does it do in terms of control, 2) what does it do in terms of the stability or in other words, how is the response of the aircraft to a disturbance affected.

Control: Assuming the same control power, a gyro with higher MOI will take longer to respond than the one with a lower MOI ( it is simply a matter of inertia). So, more lag and overshoot. If the response is too "hot" the procedure is to reduce the control power versus damping ratio, not play with MOI

Stability: As you are aware, when a gyro encounters a ( vertical) disturbance the resulting pitching moment in a stable gyro is in a direction to minimise the magnitude of the disturbance. You want this pitching moment to take effect as soon as possible so it can minimize the effect of the disturbance. So, what would you select a high MOI or Low MOI? Obviously the lower MOI gyro.

Here is the catch. If you are flying a unstable gyro, the pitching response to a disturbance amplifies the disturbance, so you are better of with a higher MOI gyro than a lower MOI one as the higher inertia will give you more time to react to the disturbance. However, even in the case of the unsable gyro the control situation is worsened with an increase in MOI, so you will have to deal with incresed lag and overshoot, but atleast you have the time.

Hope this helps!

Nb. when I refer to stability I a refering to AOA stability, though the result is the same for other static stability derivatives as well.
cheers
Raghu..

Udi

04-23-2004, 10:53 AM

It is interesting, Raghu, that you have used "nb" rather than "ps" for an afterthought - we are doing the same in my native language.

Although I understand your arguments for a low MOI, let me play the devil's advocate (until the devil himself steps in) and argue for a higher MOI.

Devil's advocate:
Lets start with control. I would argue that a higher MOI has a negligible effect on the control of a gyroplane equipped with an effective stabilizer. A cyclic input results in an almost instantaneous change in the flight path. The stab aligns the airframe with the relative wind, also almost instantaneously. I think the effect of MOI on pitch control is negligible.

Stability. Higher MOI increases the period of short-term oscillation, thus making the aircraft more controllable by, and forgiving for, a human pilot. When the gyro is hitting a vertical gust, although you want it to react in the right direction, you don't want it to react "too fast". Besides, it's more pleasant to fly in an aircraft that is damping away vertical disturbances. You want to plow through the gusts, sort of speak, not ride them!
End Devil's advocate.

So, who's right?

Udi :cool:

gyrogreg

04-23-2004, 02:22 PM

MOI:

I agree, as stated above, that higher MOI has benefits and possible negative issues as well. The lag of reaction of a higher MOI gyro MUST be accommodated by an adequately high power HS. That way the airframe IMMEDIATELY aligns with the changing free-air direction – adequately overpowers the MOI. Higher MOI also decreases the NATURAL oscillation frequency of the airframe - so for an unstable gyro, or one with inadequate HS, the quicker natural pitch oscillations (quicker than 5 second periods), would make that gyro susceptible to PIO by lagging pilot corrective cyclic inputs. If a high MOI gyro has little or no HS damping, it MAY make the higher MOI machine less desirable. However, as noted below, this does not seem to be born out by even the tandem unstable gyro accident record!

In general, just like in overall stability of a gyro, the end product is the combined result of many factors. If higher MOI is to be used to advantage, it should be "harmonized" with the other aerodynamic parameters, such as the "damping" rate of the HS. This can only be determined by "results" flight testing. But, IMHO, if done right, higher MOI results in a much more stable and pleasant flying control.

But, even for the more unstable tandem gyros - i.e. original tandem Air Commands, SnoBirds, and Parsons - there are fewer PIO incidents reported than for the short-coupled, single-seat versions. I maintain this is because the airframe reactions (wrong-direction but slower) do not excite either pilot over-control or out-of-phase corrections as easily.

By "unstable", I mean those gyros who's "Sum of Static Moments" results in the CG aft of the RTV. But, for those gyros who's Sum of Static Moments result in the CG exactly on the RTV, those are not stable gyros either. This would be the Static G-Load neutral stability point. Neutral stability will not inherently correct for G-Load disturbances - the pilot must do some correcting. So, for even this neutral case, the issue of higher or lower MOI is an issue. I do contend, that in this neutral case, the higher MOI will present less influence for the pilot to either over-control or support PIO.

A stable gyro, IMHO, should have the CG, in flight, positioned by the Sum of Static Moments forward of the RTV. G-Load stability is only one of the static pitch stability issues, but it is a most important static issue. If a gyro is G-Load unstable (CG aft of the RTV), it is highly susceptible to a rapid progressive buntover with only an adequate nose-down disturbance required to initiate the forward bunt. IMHO, the true safety of a gyro will be accomplished only if the Sum of Static Moments causes the CG to be statically held significantly forward of the RTV. (Note, I did not say that thrustlines or drag lines needed to be centered. CLT or CLD is not absolutely required, it is how the resulting Sum of Static Moments, including that of the HS and other airframe and fuselage components, causes the CG to be aligned, in flight, with the RTV.

gyrogreg

04-23-2004, 02:54 PM

Center of Drag:

There is a persistent mis-understanding about the "center of pressure" or drag line or “Center of Drag”. The airframe drag line presents a static moment about the CG, much as does the propeller thrustline present a static moment about the CG. This “drag” moment is not about the prop thrustline, the rotor head, or even the pilot – it is about the CG of the aircraft!

It is often stated, as above, that the drag line must be aligned with the propeller thrustline. What counts is how the aerodynamic center of pressure (drag line) aligns with the CG - not with the prop! It is perhaps a sign of technical naitivity to say the drag line must be aligned with the prop thrustline!!!???. To say this suggests that one does not really understand the concept of static moments. All static moment arms are relative to the CG of the object.

Both the prop thrustline moment, and the airframe drag moment (about the CG), are static moments that must add into the Sum of Static Moments, that ultimately determine the position of the CG relative to the RTV - the real issue in gyro static stability. But, there are other moments, that must add into this "Sum of Static Moments" as well - i.e. HS moment and fuselage/airframe/windscreen moments also (the aerodynamic forces created by the enclosure, windscreen, etc., are the "lift" elements of parasitic lift and drag. Drag of these components presents a static moment about the CG, but the "lift" factors of these components also present a static moment about the CG that must be taken into account or “balanced” by the HS.

These other aerodynamic moments, including the drag line moment, are square functions of relative airspeed - so they DO become very essential issues at higher airspeeds. It is good to be aware of the propeller thrustline offset, but this must also take into account all the other static aerodynamic moments as well. At different combinations of airspeed and prop thrust, their "Sum of Static Moments" must still statically position the CG forward of the RTV to maintain positive static G-Load stability throughout that entire flight envelope of speed and power.

In summary, consideration of the drag line or "center of aerodynamic pressure” is a good thing in the design of the gyro. And, these particular moments may require a HS designed to "balance" their effects - as a result of both prop thrust AND airspeed - so that the total "Sum of Static Moments" properly aligns the CG forward of the RTV. AND, in addition, the HS will have to also "balance" any enclosure/windscreen destabilizing (nose-down) static moments as well!

So, it is not truly adequate to try to "cookbook" the design of a gyro. These can be "rules of thumb", as long as you consider all the "rule of thumb" issues. To think in just terms of propeller thrustlines, or even prop thrustlines and drag lines, is to not take everything into account. But, taking all of these issues into account is very difficult to do “on paper”. The true determination as to whether all this is taken into account properly is to do the flight test. In this case, the G-Load Static Stability flight test!

G-Load Static Stability flight test:

Trim the gyro in level steady flight, then to bank into a spiraling turn of about 30 degrees or more. To maintain the original (straight and level) trimmed airspeed and power, is aft stick pressure and position required? If it is, this verifies that the CG is forward of the RTV – pulls the nose lower with the higher Gs of the banking turn – under that condition of power and airspeed. This is all that counts, but it should be true at all combinations of power and airspeed and aircraft loading! You can talk thrustlines and drag lines all you want, but in the end, this simple test tells you if you did it all right. And, if you experiment with different prop thrustlines and drag lines (and windscreens and enclosures), you will find that you don’t necessarily need CLT or CLD. It is the “Sum of Static Moments” that counts, and there is more to this than just CLT and CLD!

- Greg Gremminger

gyrogreg

04-23-2004, 03:27 PM

RTV aligned with CG

I’d like to comment on the thought that the best gyro stability is when the CG is aligned (in flight) with the RTV. Although there are some rotor/cyclic issues that may add some degree of static stability in even this condition (as I’m sure Udi is willing to suggest) I maintain that it is much better to have a gyro’s CG forward of the RTV in flight. This is accomplished by a HS that adequately aligns the attitude of the airframe a bit nose-up relative to the free air flow. This means that the “Sum of Static Moments” is such that the CG is statically held a bit forward of the RTV under all combination conditions of power, airspeed and loading. (This does not necessarily mean the gyro flies nose-high – on such gyros, with the CG statically held forward of the RTV, the keel angle can be adjusted to take this into account, so that the airframe actually flies in a level attitude!)

There are two really essential static stability criteria to be considered – Static G-Load (maneuvering) stability, and Static Airspeed Stability. Just as in a FW aircraft, that requires that the CG be forward of the lift of the wing in order to have both airspeed and g-load static stability, the gyro CG must be positioned in flight by the “Sum of Static Moments” to be forward of the RTV of the gyro.

For Static Airspeed Stability (nose responds in proper direction to stick force/movement), the CG must be forward of the lift vector and “balanced” by the down-loaded HS. In this way, the airframe reaction to an airspeed disturbance will be in the proper direction to correct for the airspeed disturbance. This is basic airplane aerodynamics 101 that all private airplane pilots learn! In the gyro version, this requires that the CG be forward of the RTV – not just aligned exactly on it or, heaven forbid, aft of the RTV!

For Static G-Load (maneuvering) stability, it is also essential that the CG be forward (in flight) of the RTV. In this way, the airframe reaction to a G-load disturbance will be in the proper direction to correct for the g-load disturbance.

To fly a gyro without the inherent static stabilizing effects of the forward CG will require a good degree of pilot corrective action to a disturbance. That disturbance, even means the disturbance of a pilot “commanded” input! To fly a gyro with the CG exactly aligned with the RTV would be something akin to flying a competition aerobatic airplane – something, no general aviation pilot would attempt to do without adequate training and practice!

Positioning the CG in flight forward of the RTV has little or nothing to do with the loading of the gyro. If the CG is loaded forward, this mostly means that the gyro will fly a bit nose-down. This does not necessarily indicate that the CG is aft of the RTV! It is where the CG is statically positioned in flight, relative to the RTV, that counts! This is mostly a function of the “Sum of Static Moments”, not of how heavy the nose of the gyro is! Of course, a nose-heavy condition may limit the range of the cyclic stick controls, that is why we check the control range with a “hang test”. But, if the control range is not an issue, a gyro that is flying nose-low, may actually be more statically stable because the HS is then forced into more of a down-loaded AOA. (But, a nose-low flight attitude could also aggravate other nose-down static moments such as that of a large windscreen.

Again, the proof of the pudding is in the flight test. The static stability flight tests should be conducted over the full loading range – as well as over the full power/airspeed envelope.

Static Airspeed Stability test:

From the straight and level trimmed condition, without changing power, does it take forward cyclic stick pressure and position to increase airspeed? Does it require aft stick pressure and position to slow the aircraft from trimmed steady-state condition? The “positive slope” of this airspeed vs. stick pressure/position indicates positive static airspeed stability. If the CG is exactly aligned on the RTV, there will likely be no slope to this line! - - little or no stick pressure required to maneuver in pitch!

raghu

04-23-2004, 03:52 PM

Udi,
Ok we are making some progress! I think we agree on the control part of things, increased MOI is bad for control, although the degree is debatable but the trend is definitely adverse.

Now, your assertion (or the devils) that you dont want the gyro to be pitching about in response to disturbances and so prefer a higher MOI is flawed. Stoping the pitching motion is not hard at all if that is the goal- just design a neutrally pitch(AOA) stable gyro! It is going to be immune to any and every disturbance wrt. pitch. It is not the pitching that is of concern but the actual effect of the distrubance or gust in terms of CHANGING THE PATH OF THE CG. Let me clarify by going back to my earlier example of the higher and lower MOI gyros; it is true that the higher MOI gyro will respond less dramatically (in terms of pitch) to the disturbance but on the other hand the result of the lethargic response is that the path of the gyro (CG) is changed more dramatically compared to the lower MOI gyro, as the gust will cause a change in AOA which will result in a slow down or spead up and climb/decent.

I think qualitatively what you describe is a less gust sensitive gyro and the only way to get there the parameter to tweak is the ROTOR/WING LOADING of the machine, not MOI! Remember we are talking of stable gyros.

Hope this helps! There are some other comments I want to make re. the desired frequency response to a disturbance and I will address those in my response to the devil ( oops I ment Greg G. :-) )post

The assertions I make are true for all aircraft and are well proven and documented in the handling qualities literature, there really is no debate about this among the theorists and practioners, its not something Beatty and gang cooked up!

cheers
Raghu..

PS. interesting note on nb versus ps in the context of hebrew(or was it another language). I use Nb. ( widely used in text books and academia) to mean
"note" rather than an afterthought. PS is used to indicate the latter.

raghu

04-23-2004, 04:47 PM

Greg in addition to my comments to Udi here are a few more specific comments. You said: MOI:

I agree, as stated above, that higher MOI has benefits and possible negative issues as well. The lag of reaction of a higher MOI gyro MUST be accommodated by an adequately high power HS. That way the airframe IMMEDIATELY aligns with the changing free-air direction – adequately overpowers the MOI. Higher MOI also decreases the NATURAL oscillation frequency of the airframe

Now Greg, you are falling prey to the proverbial bake your cake and eat it scenario. You assert rightly that increasing MOI decreases frequency of the pitch response ( short period motion) and further go on to say the negaive aspects of MOI namely lead/ lag can be overcome by a suitably sized HS, which is correct as well. So your prescription is increase MOI and damping and get the best of both worlds. But remember higher damping increases the frequency of the short period pitch response. So, while the incresed MOI decreses the frequency the higher damping increses it; back to square 1 in terms of frequency response and you need to go through all that effort of extra damping. Increased MOI has no benefits wrt stability and control and definetly has no place as a design guideline in isolation or as part of some feng shuiesque harmonic concoction.

The other point I would like to make is that you seem intent on increasing the time period of all the modes. In fact a high frequency and high damping of the short period mode is often prescribed in handling texts. A lower frequency will give a chance for the pilot to hose around with pitch digressions, leading to a possible PIO situation. In the case of the longer period mode you cannot get much damping except by making the aircraft more aerodynamicaly "dirty" so a longer period is desired so that a pilot can compenate it with small enough (instinctive) reactions on the stick so as not to notice the excitance of the longer period mode.

cheers
Raghu..

gyrogreg

04-23-2004, 09:10 PM

Raghu,

I'm not sure of some of your assertions. While increased MOI increases the period of the natural oscillations, the damping provided by the HS does not change the natural frequency of the oscillation. So, for an increased MOI, the required powerful HS does not affect the period of oscillation, only the damping of that oscillation. The MOI vs. damping rate are not dynamically canceling mechanisms.

I think you implied to Udi that the goal is to eliminate pitching in response to disturbances. And to do that you say we could just design the gyro to be statically neutral in (AOA) pitch. (This would be the condition if the CG were caused to be exactly on the RTV.)

I am maintaining that the CG must be forward of the RTV, so there IS pitch response to disturbances. This pitch response of the airframe then couples into the rotor (via the spindle) to cause the rotor disk to also pitch with the airframe – in the direction to counteract the disturbance. [This is for both an airspeed disturbance (first order) and a g-load disturbance (second order)]. The goal of this (as in FW airplanes also) is that the external disturbance be corrected or cancelled.

This is actually a high gain closed loop, negative feedback control system. The result of this is that there is minimal airframe pitching – doesn’t take much rotor disk pitching to counteract the external disturbance. The external disturbance has essentially a one-to-one effect on the airspeed and g-load impact on the aircraft. But, the high effect of the cyclically pitching rotor disk is much more powerful, and very easily negates the original disturbance. The result is a minimal airspeed or g-load divergence. There is some airframe pitching, but barely noticeable except for severe external disturbances.

It is the high frequency (short period) natural responses (rate of pitching and/or frequency of natural oscillation) that contribute to PIO. This is for the reason you mention also – the pilot cannot properly keep up enough to “dampen” the response. But, a high frequency, natural oscillation will not be a PIO factor if the oscillation is quickly damped – without exciting over-reaction of the pilot. In fact, the ideal situation is that the higher frequency natural pitch oscillation tendencies of the airframe would be that they are “critically” damped so as to not really present an oscillation that might incur an incorrect pilot response. With those high frequency natural oscillations critically damped, or non-existent, the gyro inherently does not excite any reaction from the pilot. If the pilot were excited into a reaction at those rates, they could likely be out-of-phase which would present a positive feedback control loop that could dynamically destabilize the system.

The slower, longer period natural oscillations or reaction rates are a bit better handled by the pilot and not as easily sustained into PIO. PIO is not really an issue for the slower oscillations where the pilot can readily damp any oscillations that are not readily and inherently damped by the airframe passively. Therefore, the higher MOI airframes present less possibility of badly out-of-phase pilot reactions (positive feedback and PIO).

I’m not sure what assertions you are saying are well documented in the aircraft literature. I think you are asserting that a high MOI gyro cannot respond quickly to a gust. That might be true for a FW aircraft because the wing will respond only at a basic one-to-one corrective feedback rate to the disturbance. But, the rotor cyclic control power is much more powerful than a wing – to the pitch of the airframe – essentially a much higher gain negative feedback loop. When the negative feedback of a control loop is very high, the control accuracy is extremely high – does a better job of countering external disturbances. If the assertion is that neutral static stability is the objective – so that the airframe will not pitch in response to a disturbance, then the aircraft will feel the full effects of the disturbances – because it is not causing the rotor to take any corrective action. This would be subjecting the aircraft and pilot to the full impact of a disturbance. For instance, we need any g-load disturbance to be immediately counteracted and returned to normal 1g loading. This is so because we do not want the rotor slowing down because of an extended down-gust and resulting extended lower than 1g loading – it slows down quickly and could exceed the flapping limits (precession stall).

The truly stable gyro will be caused by either an airspeed (wind gust) disturbance, a g-load disturbance, or a pilot strong commanded input, to pitch the airframe (and therefore the rotor disk) in the direction to immediately correct for the disturbance and restore 1 g loading. And, because of the extremely high negative feedback power of the rotor disk AOA, the actual pitching of the airframe is barely noticeable for even the most extreme external disturbances.

You said in one of your posts that when you refer to stability you are referring to AOA stability. AOA stability is not the goal of a stable gyro. The goal of a stable gyro is g-load and airspeed stability (and dynamic stability). To achieve this, AOA of the rotor must change to restore airspeed and g-load back to normal steady-state. Neutral AOA would result in the full impact of g-load and airspeed disturbances.

Now, theoretically, you may have all kinds of disagreements with this. And, I might not be so strongly convinced of these principles if I did not experience the results of all this every time I fly my Magni M-16 on windy days. The M-16 is a long tandem design – a very high MOI. Its natural long-period oscillation is in the order of 13-15 second period. This is a rate that is readily “damped” by the pilot if it were not so readily damped by the large HS. This gyroplane has no shorter period oscillation tendencies – or if it has any, they are critically damped. The HS provides very high inherent and passive damping of ANY oscillations. The result is a gyro that essentially corrects for all external disturbances on its own – passively. It does this so well that even a very low-time student pilot can readily fly in very turbulent winds. A 6 hr student pilot (only powered parachute flying experience!) this past weekend essentially was able to master gyro landings in winds that were gusting from 25 mph to 40 mph – reported by the weather bureau and recorded on the airport wind indicator system. I’m not bragging, I’m just pointing out that the proof is in the pudding – a very high MOI gyro which is essentially immune to the effects of the wind – at least it doesn’t require any or much pilot input to correct for disturbances. And, I would challenge anyone to try to detect any sluggishness in control authority or correction to wind turbulence!

So, I invite anyone who does not believe this to come fly with me to see how this works. If you bring a copy of this discussion, I will not charge you for the first hour of training. My goal is to demonstrate how a gyro can achieve its full potential to be the safest form of sport aviation available. My goal is not to sell gyros – I know most of you cannot afford a Magni. But, if we can better understand what can make gyros so inherently stable and safe and controllable, we will improve the safety of the whole sport.

Thanks, Greg Gremminger (“The Devil”????)

KenSandyEggo

04-23-2004, 10:21 PM

"I know most of you cannot afford a Magni."

Why would that stop any gyro-head who wanted one from buying one? Never stopped me.

C. Beaty

04-24-2004, 09:07 AM

Stability of gyroplanes is a subject of so much debate because there is no authoritative reference.

We attempt to transfer fixed wing and helicopter principles and gleam what we can from the golden age of Autogiros; more often than not confusing and contradictory.

But the laws of physics are chiseled in stone and apply without exception.

MOI AND DAMPING

The period of a torsion pendulum (a flywheel suspended by a spring wire) is:

T = 2*pi*(I/t)^0.5

T = time of one cycle
I = MOI
t = spring constant (should be Greek letter tau), the ratio of torque to twist

If the ratio of restoring force to moment of inertia remains constant, the period doesn’t change. The restoring force comes both from rotor damping and horizontal stab volume. Increasing only stab volume decreases the period. It also appears that rotor tip weights will decrease the period.

Long period motion is an entirely different subject and some of the things that influence the short period have exactly the opposite effect on long period.

CG Vs. ROTOR THRUST LINE

A primary influence here is the ratio of lift slopes of rotor and stabilizer airfoil. If the lift slope of the horizontal stabilizer is steeper than that of the rotor (it normally is) and stabilizer volume is sufficient, CG on or slightly aft of the rotor thrust line can be perfectly stable.

Ratio of lift slopes forms the basis for the 15% stab volume rule.

Fixed wing aircraft have to deal with a wing that has a much steeper slope than a rotor and the rule for them is 50% tail volume. All fixed wing aircraft fall within a narrow range of 50%; (horizontal stabilizer area*moment arm)/(wing area*chord).

Another difference of course is that rotor volume and wing volume don’t directly relate.

Udi

04-24-2004, 01:46 PM

A primary influence here is the ratio of lift slopes of rotor and stabilizer airfoil. If the lift slope of the horizontal stabilizer is steeper than that of the rotor (it normally is) and stabilizer volume is sufficient, CG on or slightly aft of the rotor thrust line can be perfectly stable.

Well said, Chuck. If this were not the case, stabbed gyros would not have been stable at low airspeed/high airframe AOA. But they are. Now, if only Greg could modify his preaching with this little detail... Slowly but surely, we are converging.

Udi :cool:

Al_Hammer

04-24-2004, 01:50 PM

It seems to me that the torsion pendulum is not necessarily a valid model for the phugoid mode.
I do not have a formula at hand to give you for the phugoid frequency of a gyro, (although I have tried to learn some of the math for a simulator program I am working on.)

Airplanes are not gyros, but mathematically, the phugoid dynamics seem to be similar.
potential and kinetic energy are exchanged in a cyclic manner. In this mode, the aircraft, at nearly constant angle of attack, climbs and slows, then dives, losing altitude while picking up speed.
Interestingly, the phugoid frequency of an airplane can be shown to be inversely proportional to the trimmed airspeed and independent of MOI.
Even though the aircraft pitches about its cg in a sinusoidal manner, the period has nothing directly to do with the mass or MOI.
I suspect that its the same with the gyro.
As speed increases, the gyro pitches up and slows down while gaining altitude.
As it slows, it then pitches down and gains speed. Same dynamics.

Rotor inertia, on the other hand, may indeed affect the period, as evidenced by the experience of the designers of the Masquito helicopter:
from their website:
"A helicopter is dynamically unstable in hover. This creates a real problem if the period of longitudinal or lateral oscillation is low, say less than 10 seconds. With the teetering rotor "Masquito" this problem has been resolved by using a relatively heavy rotor, with 2 kg. (4.4 lbs.) of lead at each tip to increase the inertia, and a Hiller servo rotor. This results in an increase in the oscillation period to about 24 seconds which results in a very stable aircraft in hover. This 24 second oscillation period has a direct and favourable influence on the time the phugoid motion takes, (in forward flight), to complete 1 cycle."
-------------------
Some links:
Phugoid equations
http://www.math.sunysb.edu/~scott/Book331/What_do_solutions.html#secWhatDoSolutionsLookLike
phuugoid flight path
http://www.math.sunysb.edu/~scott/Book331/Seeing_flight_path.html

phugoid math model

http://www.mit.edu/afs/athena/org/a/aa-math/www/modules/node4.html

http://adg.stanford.edu/aa241/stability/dynamicstability.html

raghu

04-24-2004, 02:28 PM

raghu

04-24-2004, 02:33 PM

It seems to me that the torsion pendulum is not necessarily a valid model for the phugoid mode.
I do not have a formula at hand to give you for the phugoid frequency of a gyro, (although I have tried to learn some of the math for a simulator program I am working on.)

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

Al, for the most part we have been talking about the effect of MOI on the short period mode and not the phugoid. MOI as very little effect on the long mode.

Al_Hammer

04-24-2004, 02:47 PM

You're right, Raghu; "my bad", as they say.
I should have re-read the posts before going off thinking about torsion pendulums.
I'll save those phugoid factoids for another day, when they ,no doubt, will come in handy.

C. Beaty

04-24-2004, 03:09 PM

I'm afraid afraid Raghu beat me to it with an answer, Al. He may not have stayed in a Holiday Inn Express last night but I suspect he slept with an aeronautical engineer; -his wife.

The long period mode is mostly set by rotor characteristics and is more complicated than a helicopter because rotor rpm is another degree of freedom.

gyrogreg

04-24-2004, 09:51 PM

A lot of good but confusing technical discussion going on here. First let me say that this subject is far from precise, but interesting to conjecture about. Also, I am often confused about the lack of clarity of specific conditions or parameters being commented on - high speed stability, or low speed stability, mixing of dynamic and static effects, phugoid or short-period responses, stable or unstable gyro, etc. Not complaining, just saying this is confusing without very specific parameters identified for each comment! But, at least we are somewhat beyond just saying a gyro “feels” stable or not! My primary position is that, with the universe of complicating configuration combinations, the proof of the pudding is in objective flight test results. And, with the high degree of unknown factors and combinations of factors, the best design tool, I believe, is flight test iteration upon iteration until the desired flight test stability criteria are met - or at least identified. These deeper technical theories here may provide clues as to how to improve a design, but the flight test results are the only true verification.

Our efforts to establish flight test results criteria for the LSA rules have progressed without benefit of this full level of theoretical discussion. Perhaps considering these discussions, we might further evolve the current thinking on these results criteria. But, I am convinced that the current flight test stability criteria are at least better than the old stability criteria in the UK Section T gyroplane standard.

I think I'll start a new thread on stability flight testing criteria and methods. With the level of expertise assembled on this thread, perhaps we could critically advance these flight test results criteria. On that new thread, I will summarize currently proposed LSA gyroplane flight test stability criteria.

I'd like to comment on some of the comments above:

Chuck's comment on CG vs. RTV: I do agree that the relative lift slopes of the HS and rotor can extend the CG range slightly aft of the RTV for some degree of static AIRSPEED stability only. But, we might also need to consider that the lift slope of the rotor varies significantly depending on airspeed (shallow rotor disk AOA vs. high rotor disk AOA), whereas the lift slope of the HS is relatively constant over the speed range from MPRS to high airspeed. This simply means that this Airspeed Static stability mechanism can change greatly in its effect over the full speed range. But, I do still suggest that the CG forward of the RTV IS necessary for G-LOAD static stability - the stability parameter that is important to progressive forward bunts under decreasing G-load conditions. When I suggest the CG must be forward of the RTV to be statically stable, I am implying this for the condition of higher airspeeds where stability issues are most important. At slow speed, all gyros seem to not have stability issues, and this might be explained by many mechanisms - possibly including the relationship of HS and rotor lift slopes. I have not, however, rationalized how G-Load Positive Static Stability can be achieved without the CG being forward of the RTV – I see this as simply a moment issue that may be somewhat complicated by the G-Load reaction of the offset gimble and trim spring! But, the G-Load stabilizing effects of the trim spring/offset gimble mechanism are also difficult to factor in! Again, the results of all these interacting complicating mechanisms might be somewhat predicted by these theories, but testing must verify the final results. As a good design guideline, I still do recommend that the in-flight Sum of Static Moments should produce a CG physically forward of the RTV to assure positive G-Load static stability. I believe this is the true reason that gyros such as the Dominator and Magni are so stable – especially at higher airspeeds – the CG is well forward of the RTV! The Sparrow Hawk, with relatively CLT, may be an example of positive static stability without the CG being forward of the RTV – I don’t know, but this will be very interesting to study – where the in-flight CG of the Sparrow Hawk is and just what it’s degree of positive static Airspeed and G-load stability is! The Sparrow Hawk airspeed and G-Load static stability may be the verification that I am wrong about this issue!

Raghu, from your's and Chuck's arguments, I grant you that the HS may influence the short-period natural oscillation frequency. And, I do agree that the MOI issue may only be really important for “UNSTABLE” gyros anyway. In this instance I am referring to gyros without adequate HS power to adequately damp short period oscillations. If a gyro is dynamically “STABLE” – no very short period oscillations – the issue of MOI is less significant and may mostly just influence the solid or comfortable control “feel” a pilot subjectively prefers. But, for an unstable gyro – one without adequate HS power – I maintain it is an advantage to have a high MOI because that will likely result in longer period natural oscillations that are more readily compensated by the pilot. The very worst case would be an unstable, or especially a “tailless”, gyro that has a small MOI. In this case, the lower MOI without damping may require the pilot to provide “damping” control inputs at frequencies that are difficult for the pilot to properly phase the corrective inputs. I believe this is born out by the fewer number of PIO accidents in even unstable tandem gyros – older Air Commands, Parson Trainers, SnowBird tandems. I agree that MOI would not be more than a subjective “feel” issue if all gyros had any very short period oscillations critically damped – dynamically stable! But for an unstable gyro, without adequate damping properties from the HS, a very long tail boom to increase the gyros MOI to a slower response rate would make PIO in that gyro much less likely because the pilot could more properly compensate and “dampen” the slower pitch oscillations.

Houston model disparity – I’m not sure how this is pertinent to this MOI discussion, but: I’m not sure what you are saying about the linearized and non-linearized model. I may be wrong, but as I interpret the Houston graphs, they are a comparison between the computer model and the physical model tested. Dr. Houston does try to explain, without investigation, why there are disparities between the phugoid periods between model and real gyro. That leaves me in doubt that he really has an accurate computer model from which to draw accurate conclusions. But there are other issues in his study and reports that leave me less confident. His computer model reports do not consider or vary any HS parameters – volume, position, AOA or prop immersion factor. The single variable in his model that he did explore was the prop thrustline – from which he concludes 2 inches above the VCG is the safe limit. (This is an actual criteria that has subsequently been adopted in the UK Section T gyroplane standard!!??) I think his conclusions are faulty on this simply because he did not explore any other HS parameters or really any other gyro configurations such as immersed HS! I do also question his determination of the VCG relative to the prop thrustline on the test gyro (VPM-16). I agree with many that the VPM-16 and M-16 prop thrustlines are somewhat higher than the .79 inch reported in the Houston study. The computer model study of the VPM-16 reports that short-period oscillations are not critically damped, but the graphs present only long-period, less than critically damped responses. I have not been able to excite any short-period oscillations in the Magni M-16. The phugoid oscillations of the M-16 are not critically damped, but seem to match his computer damping rate that he reports for short period oscillations! Even though I should be pleased with the study conclusions (I concur in the stable performance of the Magni M-16), such disparities dampen my confidence in the conclusions he draws from his computer model, such as the prop thrustline should be no more than 2 inches above the VCG. And, I am afraid, this incomplete analysis has regrettably been the foundation of the UK Section T stability criteria – even to the point of prescriptively specifying a maximum high prop thrustline offset while not even considering the very low prop thrustline issue! I don’t disparage Dr. Houston’s work, I just wish he would have continued to study these disparities to verify the models’s accuracy and then to study other variables that influence gyro stability other than prop thrustline – especially in light of the considerable new understandings of gyroplane stability as represented by people such as those contributing to this thread!

I’ll let everyone know when I start a discussion thread on stability flight testing criteria and methods.

Thanks, Greg Gremminger

Udi

04-24-2004, 10:03 PM

...But remember higher damping increases the frequency of the short period pitch response....

Although I can't contradict this statement, I find it counter intuitive. Why would higher damping increase the short period frequency? I thought damping forces counteract static stability forces. If short period frequency is dierctly proportional to static stability, it should be counter proportional to damping... :confused:

Udi :cool:

Udi

04-24-2004, 10:39 PM

... What makes things tricky is that unlike FW gyros ( rotorcraft) in general have what is termed velocity stability, this is different from the speed stability, though on the surface it may look the same. The velocity stability in rotorcraft is due to rotor blowback , and this may in the first instance seem like being helpful but it actually messes about with the AOA stability and if strong enough may interact with the AOA stability and make the phugoid unstable.

There is a wide spread misconception among many between the desired speed stability ( which is really a function of AOA stability), and velocity stability . You can have a perfectly stable gyro if you make it AOA stable and minmise any velocity stability. Note tough the velocity stability is useful if you have an unstable gyro...

Raghu (or Chuck),

Would you please explain the difference between airspeed stability and velocity stability? Isn’t blowback (flapping angle) THE primary mechanism for airspeed stability in rotorcraft? FW aircraft rely on AOA stability for airspeed stability, but this mechanism is, unlike blowback, an indirect one.

Thanks!

Udi :cool:

Hognose

04-25-2004, 06:33 AM

Great thread. I do have to take issue with (or really, clarify) one statement.

8) Gyrocopters with suspension should have adequate damping to preclude ground resonance with rotor. (such as A&S 18A). (C. Beaty)

Ground Resonance happens when the rotor blades get out of phase with one another. Immediate action is to take the aircraft aloft muy pronto. However, it's important to note, that for the rotor blades to get out of phase, you need a fully-articulated rotor head, specifically with lead and lag hinges. (Which does describe the Air & Space 18A).

As a quick rule of thumb, these rotor heads are found in multi-bladed rotors, but almost all two-bladed rotors are teetering Bell-type rotors. The two gyroplanes I can think of with three-bladed, fully articulated rotors are the A &S 18A, the McCullough J-2, and the all-but-forgotten Canadian machine that someone is now trying to revive as the Pegasus. (http://www.pegasusrotorcraft.com/) All were 1960s-vintage attempts at a certified, jump-takeoff gyroplane and had three-bladed rotors.

Ground resonance is not a factor with a teetering system. So most older design Bell Helicopters, Hiller helicopters, and practially all experimental gyroplanes are immune. (The teetering head has its own negatives; everything in engineering is a trade-off).

For more on Ground Resonance, read the helicopter parts of the FAA's Rotorcraft Flying Handbook, available for free in PDF format online (Try this, although I'm not certain: http://av-info.faa.gov/data/traininghandbook/faa-h-8083-21.pdf). DON'T just read the gyro parts, as the FAA's writer assumes you know all the material in the helicopter chapters, which come first!

cheers

-=K=-

darrellwittke

04-25-2004, 10:19 AM

Thank you Mr. O'Brien, I have wondered about ground resonance with teetering systems for quite a while. Two episodes (as posted on Norm's forum) bring these questions to mind. One was David Holmes ill-fated attempt to arrive in style at a golf course, in which he related as he went to flare vibrations started and increased in intensity until they contacted with the ground in the rear. This of course caused the machine to roll over and then caught fire (which became a classic picture on his website).

The second anecdotal example of possible ground resonance is a machine built (I believe) by Ernie Boyette as a true ultralight. This machine had simple spring leafs out to the main wheels and I seem to remember (forgive me... I do not save posts to hard drive like Al Hammer only to the wetware between my ears) post to a possible buyer warning of problems to ground resonance.

What I am asking is are you quite certain ground resonance is not possible with a teetering rotor system? It seems to me any rotor can get out of phase with another, although admittedly harder with a two blade teetering system, and without just a little damping it may reinforce itself. I have no technical training, this is simply (false?) intuition on my part. Any actual technically educated insight is appreciated.

Thanks again, Mr. O'Brien

Al_Hammer

04-25-2004, 10:34 AM

Darrell, here's another snippet of a post I saved, this time on ground resonance:

Its mainly germane to helicopters which have
> shock-struts as the main rotor rotational rate crosses the
> resonant frequency(ies) of the struts. It tends to be exaggerated
> and more dangerous in machines with four struts and three blades.

>The reason that a JetRanger can not have ground resonance is that it does
>not have lag hinges or other source of flexibility in-plane. If the rotor's
>natural lag frequency is higher than the rotor speed, as is the case with
>the Bell teetering rotors, it is impossible for it to have a ground
>resonance instability. Rotors with this characteristic are called
>stiff-inplane. You can have a stiff-inplane rotor with more than 2 blades
>(BO105, BK117) and they too are immune from ground resonance. (If look at a
>BO105 or BK117 hub you will not see any lag dampers.)

Mike Jackson

04-25-2004, 10:36 AM

Although I can't contradict this statement, I find it counter intuitive. Why would higher damping increase the short period frequency? I thought damping forces counteract static stability forces. If short period frequency is dierctly proportional to static stability, it should be counter proportional to damping... :confused:

Udi :cool:

After dusting a few manuals, I found that the damping (damping ratio) is directly proportional to pitch damping (Cmq). If we increase this term we decrease the time to damp - period decreases and means the damped frequency increases. Greg appears correct.

Cheers,

Mike

darrellwittke

04-25-2004, 12:33 PM

I would like to draw your attention to rule #6 which states a properly designed gyroplane is one of the safest forms of flight. I believe that is our goal, simply stated.

Whether tandem design or side-by-side, I haven't seen any evidence yet that precludes one design or the other, both if properly designed seem to be able to fulfill the above mentioned goal. As long as there are more than adequate tail feathers, it seems to be (practically speaking) a question of pilot preference.

Ground resonance seems to be impossible on our simple gyrocopter's.

Horizontal stabilizers stiffen frame response and shorten period of oscillation.

Are there any discrepancies with the other general guidelines?

( Just a note to Greg. G., your reliance on pilot flight tests for confirmation of stability and/or good gyro design reminded me of Dofin F./Duane H. adamant arguments regarding RAF's design. I prefer to see the numbers (ie. why the design and calculations supporting that) although I cannot understand them. (slowly...over time) Can you possibly contact J. Tervamaki and ask why he prefers the RTV ahead of CG?)

With greatest respect. Darrell Wittke

Al_Hammer

04-25-2004, 12:57 PM

In the general case of a damped harmonic oscillator, it appears that the damping coefficient has only a small influence on frequency. Damping is assumed proportional to velocity.

Here is a good page with some diagrams:
http://hyperphysics.phy-astr.gsu.edu/hbase/oscda.html
Lower down on the page, it gives the formula for frequency as a function of the damping coefficient.

The formula indicates that the freq of oscillation will slightly decrease as damping goes up.

Here is a java applet that lets you try different damping coefficients(always less than 1).
http://www.chem.mtu.edu/~tbco/cm416/qsecord_2k1.html

C. Beaty

04-25-2004, 08:57 PM

VELOCITY STABILITY

Velocity stability, Udi, is defined as the derivative of pitching moment with respect to flight path speed. Velocity stability is a primary factor increasing the frequency of the long period mode.

With an increase of speed, the rotor blows back, generates a nose up pitching moment that causes a rotorcraft to climb and slow down. As it slows down, the rotor tilts forward, generates a nose down pitching moment and the cycle repeats.

Angle of attack instability and the main cause of long period oscillation. The fuselage MOI plays a negligible role.

Rotor properties play a dominant role; the ratio of inertial force to aerodynamic force. Kurt Hohenemser, one of the German helicopter engineers brought to this country following WW II, patented the concept of pitch-cone coupling; an increase of rotor thrust increases coning angle which in turn reduces collective pitch and velocity stability, improving angle of attack stability. The scheme was tried on the compound helicopters developed by McDonnell and reportedly worked well. Pitch-cone coupling should not be confused with pitch-flap coupling.

Unfortunately, stabilization by mechanical/aerodynamic methods was overshadowed by the arrival of the integrated circuit. A box full of ICs and a couple of gyroscopes do everything that needs doing for million dollar helicopters.

DAMPING Vs. FREQUENCY

There’s a definition problem here. Dissipative (friction) damping does not affect frequency; i.e., the hydraulic dampers on an automobile or a resistor shunted across a resonant circuit.

Damping that produces a restoring moment; horizontal stab, rotor lag following an airframe excursion, etc., does however increase the frequency of the short period mode as a function of its magnitude. The suspension wire of the torsion pendulum is stiffened.

raghu

04-25-2004, 09:30 PM

After dusting a few manuals, I found that the damping (damping ratio) is directly proportional to pitch damping (Cmq). If we increase this term we decrease the time to damp - period decreases and means the damped frequency increases. Greg appears correct.

Cheers,

Mike

That was my contention, I believe greg initially though otherwise.

cheers
Raghu..

darrellwittke

04-25-2004, 09:46 PM

Sorry, Chuck can you state it more basically? IE. derivative of pitching moment with respect to flight path. I assume pitching moment to mean flapping angle (or axis of blade tip rotation). With respect to flight path I assume flight path of gyro.

How do you minimize velocity stability? (I'm still hung up on words...velocity to me implies kinetic energy with direction)

What I got out of Al Hammer's attachment is that an oversized damping force works almost as well as critical damping. Would this be analogous to horizontal stabilizers? (ie. it may in practice do very limited harm, and of course do great benefit in comparison to none or undersized) Also, can you acheive critical damping (which I assume to be the quickest suppression of oscillations and return to steady state) with a simple non-moving correctly sized horizontal stabilizer?

raghu

04-25-2004, 09:58 PM

Raghu (or Chuck),

FW aircraft rely on AOA stability for airspeed stability, but this mechanism is, unlike blowback, an indirect one.

Thanks!

Udi :cool:

Udi, just to add to Chuck's lucid (as always)explanation. I would like to comment on your assertion that in the FW the aoa stabilty mechanism is somehow a more indirect mechanism. It is indirect if you think the goal is keeping a fixed speed; actually the goal is and should be maintaining AOA. This is true as well in a gyro. One trouble ( there are others as well) with the gyros velocity stability mechanism is it breaks down in not uniform flight ( a turn). When you bank you increase the load and consequently the gyro decents which in turn increases the AOA of the rotor. Now if we had only the velocity stability ( assume for a moment neutral AOA stability), the blowback would actually increase(due to increased aoa) thus reducing speed in a turn rather than speeding up. On the other hand AOA stability will maintain speed during uniform flight and do the right thing in circular flight as well by pitching down and speeding up.

A good example of a strongly AOA stable gyro is ron herrons LW2 (elevator controlled model). AOA stability is the holy grail to rotor craft stability

nb. in a turn, the velocity stability may eventualy do the right thing when the decreased speed due to the increased AOA slows the gyro sufficently that the blowback decreases, and the gyro pitches down. This is infact indirect and not reliable- if there is any AOA instability it may not actually happen.

gyrogreg

04-26-2004, 06:07 AM

Mike, Rahgu is correct. My original contention was that more damping power (HS) did not affect (or quicken) the natural frequency of oscillation. It appears my position is being out voted by some very technically deep thinkers. This still does not argue against the advantages of a larger, more effective HS. I am very interested in seeing a valid computer model for these gyro issues – to better understand how MOI, HS volume, etc. affect the static and dynamic performance of a gyro – I encourage Al Hammer to continue his work on this model.

Darrellwittke, I sincerely hope my arguments for flight test confirmation of stability factors is not consistent with Duane and Dofin’s arguments for why the RAF design would be stable. The difference is that their arguments are based on subjective pilot “feel”. Subjective evaluations of stability, especially by pilots who are skilled in flying that particular gyro (they are the stabilizing computer in the loop!), are not quantitative. They may verify that pilot’s ability to make that gyro fly stable, but they do not evaluate the gyro’s inherent stability or its aversion to instability accidents. What we are proposing in the new LSA gyroplane standards, same as all aircraft standards, are flight tests that objectively measure the quantitative stability qualities of the aircraft, not the qualities of the aircraft/pilot combination.

I will start a thread when I get time to ask for discussions on what these objective quantitative criteria should be and how to measure them. The ASTM gyroplane subcommittee has developed criteria for the LSA gyroplane standard. I would like to get wider peer review whenever we can - safety is the goal - not theory! I hope we can engage the level of expertise in that discussion as is being engaged in this discussion.

Also, Darrell, you asked about why Jukka T. prefers the CG ahead of the RTV. I’m not sure why you bring up Jukka on this matter, but I have had numbers of conversations with Jukka, and I don’t recall any converstations as basic as this. Please don’t think that my persistence in this CG/RTV matter is from Jukka. My perspective on this is more basic and perhaps does not fully consider the more involved rotor issues that are being brought up in this discussion. My perspective was evolved originally from what Chuck has always preached about the CG/RTV, and from Sum of Static Moments analysis around the CG. In my KISS concept of these issues, it is much more understandable and assuring if the CG can be maintained in flight forward of the RTV – this provides all the static stability requirements without relying on the more complex and possibly less predictable concepts or rotor factors such as “blowback” and lift curves for such stability. It seems to me that some of these more complex rotor concepts are addressing dynamic reaction issues, which are by nature more complex that the static stability issues. I think it is very likely, that once a good degree of static stability is achieved in a gyro – this requires significant tail feathers – the dynamic issues are mostly satisfied as well.

My persistence on this basic CG/RTV concept may come from more of a desire to help the general gyro community better understand and respect basic gyroplane stability safety issues. Even static moments are beyond many people’s perception, but I tend to avoid much more complex technical issues which may leave J.P. Gyro Pilot more confused than convinced. Although I find the more intricate rotor/gyro issues interesting and challenging, I tend to prefer to communicate stability issues in language that the less technically astute might understands and accept. That does not say these discussions are not valuable or wrong, it only says I might try to simplify the total picture into something more people can really grasp. My mistake may be when I espouse these simpler concepts in a forum where more technically astute people then raise the discussion to a higher technical level!

That is why I prefer to say that these issues are very complex, and that the final “proof of the pudding” is in flight testing. I am hoping that the simple flight tests, and comprehendible supporting theories, will influence more of the gyro crowd to respect the need for stable gyros.

As far as Jukka and Vittorio Magni’s design concepts, I think they prefer to not overly rely on “paper design” analysis, but, apply basic concepts and then an iterative flight test technique to hone a design to required results criteria. This is partly why I will always say that the many interacting factors in a gyro’s stability may not be able to be fully “paper designed” from a set of cookbook rules – too complex to fully analyze. The final design should be verified by flight testing; flight tests that I believe can be very simple and quick to test. As far as CG/RTV relationship in a Jukka or Magni design, my impression and experience in the M-16 is that the HS is just so powerful that it will not allow the CG to be displaced from its slightly forward position by either propeller thrustline or airspeed static moments. This is what assures static stability throughout the speed, power and loading flight envelope. This is most likely a prominent factor in Ron Herron’s LW and may even be a factor in the Bandit as well – an over-powering HS. For some, I know, that is too simple of an answer.

Thanks, Greg Gremminger

raghu

04-26-2004, 07:44 AM

It appears my position is being out voted by some very technically deep thinkers. This still does not argue against the advantages of a larger, more effective HS.

Greg,
Absolutely, I agree! Just to emphasize, larger and effective stabilizers are good. While there may be a slight increase in frequency, the damping makes control crisp and damps the short period mode. In general increasing the frequency of the Short period mode is not bad (in fact desired) as long as you damp it sufficiently. After all the other "good thing" namely CG ahead of RTV ( aoa stability) also increases the frequency of the SP mode- further behind the RTV you move the CG more you increases the frequency of the SP mode, but there is no problem with this, in fact like you say CG behind RTV is desired.

My argument is specifically against your recomendation of increasing MOI. It has no place in a list of good stability design practice (under the assumtion we are designing stable gyros). The CG ahead of RTV and large HS are definetly good practice. I have some other comments/arguments on your presentation of velocity stability but I will save it for another post.

The key to good dynamic stability in a gyro is AOA and rotor G static stability.

cheers
Raghu..

Al_Hammer

04-26-2004, 03:17 PM

DAMPING Vs. FREQUENCY

There’s a definition problem here. Dissipative (friction) damping does not affect frequency; i.e., the hydraulic dampers on an automobile or a resistor shunted across a resonant circuit.

Damping that produces a restoring moment; horizontal stab, rotor lag following an airframe excursion, etc., does however increase the frequency of the short period mode as a function of its magnitude. The suspension wire of the torsion pendulum is stiffened.

Chuck, thanks for the clarification. I refreshed my memory on this and realize my mistake in lumping the two forms of damping together.
There is indeed, a definition problem.
We keep talking about stab "damping". The damping comes in two forms; one is the "restoring force" , which is , as you explained,equivalent to the spring in the torsion pendulum.
The stab generates lift force in a direction opposing the pitch disturbance.
The more the pitch diverges from a direction that is aligned with the relative airflow, the more restoring force the stab produces.
The restoring moment is linearly proportional to the stab moment arm and to the stab area.
Since it acts to stiffen the spring, it does increases the natural frequency of the system, no argument.

The other form of damping is proportional to the moment arm squared.
I disagree with calling it dissipative, or resistive damping. The stab produces a force proportional to the rate of pitch change of the aircraft. This is similar to the mechanism of rotor flapping. As the stab is moved up or down due to pitching, it experiences a vertical velocity component which changes the AoA and therfore changes the lift.
This is not frictional damping or viscous damping. The airfoil is not stalled and simply stirring air.
This form of damping has little effect on freqency, but does reduce the time needed to damp out the pitch disturbance.

These two forms of damping are always present.
In a short coupled pusher gyro, the moment arm is short and so the second form is not a major factor.
In a tractor gyro, it becomes a major factor.

Martin Hollmann left this term out of his gyro math model, but he told me (when I called to inform him about a problem on his web page) that it would be interesting to model a gyro like the Little Wing and that he would definitely need to include the rate sensitive damping in that model.

(from a post on a fixed wing forum)
The other side of this issue is the tail size. Dynamic stability is linearly proportional to tail area, and proportional to the SQUARE of the tail moment arm. In other words, a 10% increase in horizontal tail area increases dynamic pitch stability by 10%, but a 10% increase in tail moment arm increases dynamic stability by 21% (i.e.: 1.10^2 = 1.21). This is because the increase in tail moment arm not only increases the tail's leverage, it also increases the tail's vertical velocities during the oscillations, which increase the tail's lift forces that are trying to damp those oscillations.

Mike Jackson

04-26-2004, 04:09 PM

C. Beaty

04-26-2004, 05:17 PM

Rotor angle of attack stability can be improved in several ways, Darrell.

The rpm of a gyroplane rotor depends upon load. As load in increased, the rotor
speeds up and cyclic flapping is diminished. A low inertia rotor will respond
more quickly, reducing blowback and velocity stability which improves angle of
attack stability.

But that’s a double-edged sword. A low inertia rotor also reduces damping by the
rotor. A high inertia rotor lags an airframe displacement and supplies a damping
moment.

Another way is to overbalance torsionally flexible rotor blades. With CG forward
of the aerodynamic center, an upward gust causes the blade to twist nose down
and the rotor disc to tilt into the relative wind instead of away from it. I
imagine this is the way Wing Comdr. Wallis achieves what some claim to be
remarkable stability.

Another way is the pitch-cone coupled rotor invented by Kurt Hohenemser. As the
rotor cones up with a load increase, collective pitch is pulled out and blowback
is reduced.

Pitch-cone coupling requires a floating hub with 3 or more blades or skewed
coning angle hinges on a seesaw rotor.

The attachment below appears in the J. of the AHS, vol. 5, no. 4, Oct. 1960. I
don’t have that volume but someone gave me a draft copy of the original paper;
“Comparison of Current Operational Rotor Systems and a Rotor Having Floating Hub
and Offset Coning Hinges” by the chief engineer of McDonnell Aircraft’s
helicopter division.

raghu

04-26-2004, 05:25 PM

Chuck, thanks for the clarification. I refreshed my memory on this and realize my mistake in lumping the two forms of damping together.
There is indeed, a definition prob

The other form of damping is proportional to the moment arm squared.
I disagree with calling it dissipative, or resistive damping. The stab produces a force proportional to the rate of pitch change of the aircraft.........
.....
This form of damping has little effect on freqency, but does reduce the time needed to damp out the pitch disturbance.

Al,
The form of damping (the other form is not really damping it is the restoring force due to static stability) you refer to is commonly denoted by the damping derivative Mq and is dominated by the paddle action of the HS. It turns out even this (damping) increases the frequency ( at least in the case of fixed wing). I can try and dig out a reff. but most standard texts on fixed wing stability should give you a formula for the SP mode and you can see that it increases the frequency. I believe there is no reason why this result should not hold for gyros as well though I will need to think through it before I can stick my neck out and say so.

cheers
Raghu..

C. Beaty

04-26-2004, 05:30 PM

The other form of damping is proportional to the moment arm squared.
I disagree with calling it dissipative, or resistive damping. The stab produces a force proportional to the rate of pitch change of the aircraft. This is similar to the mechanism of rotor flapping. As the stab is moved up or down due to pitching, it experiences a vertical velocity component which changes the AoA and therfore changes the lift.

We could chase our tails from now until Doomsday Al, and still not agree of the precise meaning of damping.

For an airfoil to generate a force, it must produce a change of momentum in the airsteam. This requires energy that is lost for all eternity as far as the vehicle is concerned.

Al_Hammer

04-26-2004, 07:11 PM

[I]
We could chase our tails from now until Doomsday Al, and still not agree of the precise meaning of damping.

For an airfoil to generate a force, it must produce a change of momentum in the airsteam. This requires energy that is lost for all eternity as far as the vehicle is concerned.

True enough, Chuck, but if both forms of damping ultimately produce a change in momentum, why call one form dissipative and not the other?

BTW, Damping that is proportional to the pitch rate is apparently called pitch damping.
http://www.flightlab.net/pdf/6_LongitudinalManeuveringStability.pdf

Stability refers to a force that arises depending on the position of the system; damping refers to a force that arises depending on
the velocity.
from http://www.av8n.com/how/htm/equilib.html

Raghu:

In an underdamped oscillator, the damping has only a small effect on freq.
http://hyperphysics.phy-astr.gsu.edu/hbase/oscda.html
If that model applies to the gyro, in short period mode, then the pitch damping should not affect freq. At least that is my reasoning.

C. Beaty

04-26-2004, 09:10 PM

It may be, Al, that the word damping is an unfortunate choice.

Perhaps the rate damping of a horizontal stabilizer should be called derivative compensation.

Al_Hammer

04-26-2004, 10:29 PM

hmm, derivative compensator. I like it.
With a name like that, even the marketing department in Kindersley could go for it;
except it would have to be magic derivative compensator. :D

gyrogreg

04-27-2004, 04:59 AM

Al and Chuck,

Your points on the HS vertical velocity as a "damping" factor is enlightening to me. Some time back I had started to develop an Excel spreadsheet that attempted to determine the time response of a disturbance to a gyro. This “spreadsheet integration” incremented very small time periods with calculations at each time increment from the results of the previous time increment. F=MA and MOI formulas, with drag and thrust moment calculations as each row incremented time. The result was a big matrix of cells with formulas that calculated their value from the previous time incremented values. The lift factor of the HS (and other configuration parameters such as MOI and prop offset, drag and thrust factors) is adjustable in this model. A disturbance could be introduced at any point to see the time reaction response after the disturbance. This response is plotted graphically.

I know this model was not correct or finished. One inadequacy is that I had only modeled the position (AOA) lift of the HS, not the vertical velocity induced (“paddle action”) lift. Considering these discussions, I went back to look more closely at the effects of changing MOI and/or HS power. I know this model is not correct – the HS factor is programmed only as a basic “spring” function – more displacement = more static restoring force.) This model does not program the HS vertical velocity “damping” factor. On this simple model, when the HS factor is critically small (model starts getting dynamically divergent), an improvement of the HS factor does quicken the natural frequency of oscillation – up to the point where the oscillations become very minimal (no reaction from a disturbance). This seems to me to support the conclusion that improved HS effectiveness actually shortens the period of oscillation – until the oscillations basically don't happen. Also, increasing the airframe AOA – at least at the critical point, does seem to decrease the natural oscillation rate, but my incomplete model does go unstable quickly at higher MOIs, and disturbance oscillations do quicken but basically disappear with lower MOIs. (It is interesting that this model does seem to show, for poor configuration parameters, dynamic oscillations diverging until the point where static instability is encountered and then it basically "bunts over"! - PIO and buntover!)

I had noticed MOI and HS effects on the oscllation rates that I considered not correct. I had discounted these because I realized that I was getting no "damoping" effect from my hS model - I had not yet programmed the vertical velocity damping effects of the HS – that was going to be a big challenge to program that velocity factor – and I was not sure that factor would be significant enough to actually provide damping. I had recognized though, that the simple “spring” model of the HS would not provide any damping from the HS – so I did not trust this model much!

These discussions have made me reconsider this model, and the HS damping. The vertical VELOCITY component of HS lift occurs essentially 90 degrees ahead of the POSITION HS component – essentially slowing the HS vertical motion in that portion of the oscillation – damping action. The straight “spring” model for the HS, without damping, just tended to keep on oscillating in response to a disturbance – if the other configuration parameters provided static AOA. (This model did show static instability if the configuration parameters were inadequate (prop offset, HS power, etc.). I am encouraged, that including the vertical velocity factor of HS lift might provide the “damping” that my model did not seem to have.

Anyway, I might find time some time to try adding the HS vertical velocity to my model, to see if it might start looking a bit more realistic. I can’t pretend to keep up with all the higher math and dynamic physics analysis (“derivative compensator”) some of you are exploring, but I might be able to understand the simpler thrust, force, inertia, acceleration velocity formulas I am using in my Excel cells to model responses. That way I can let the small time increments of the spreadsheet integrator do the higher math work for me.

If anyone would like to see my incomplete Excel model, let me know, I can email the file.

Thanks, Greg

raghu

04-27-2004, 06:23 AM

Al,

Here are the factors that affect the short period mode

f proportional to (MqZw/m -MwU)/I
and damping is proportionl to (Zw(I/m)-Mq-MwdotU)/fI

where f =frequency of the sp mode
Mq= is the paddle effect that we have been talking about, i.e.dM/dq; rate of change of resistive moment m with pitch rate q. Two factors will contribute to this derivative HS (paddle) and rotor damping.

I = moment of inertia on the pitch axis

Zw = the lift slope

Mwdot= ignore for this discussion as it has to do with the downwash interfering with the HS

Note by convention both Mq and Zw are always negative so both frequency and damping increase with an increase in magnitude of Mq ( i.e. the paddle effect of HS and rotor damping)

raghu

04-27-2004, 06:52 AM

A dynamic model should not be that hard to build. What is hard is getting the stability derivitive to plug into the model so as to to predict the exact stability behaviour of a specific gyro design . That said, most of our discussions have been on trends and effects of the various parameters on stability so the actual values should not matter.

A year or so ago I built a excel model to study the interactions of the stability derivatives on frequency and damping of a fixed wing. I used a freely available plug in to solve the stability quadratic.

If we ignore rotor inertia, I believe the standard fw model of pitch dynamics can be used to study gyros, though some of the stability derivative definitions will need to be updated, but the relationship between all the variables remain the same. Here is why if the rotor is inertialess then it will change speed instantly to any change in AOA, producing the resultant torque immediately (i.e. acts like a wing). Note also velocity stability is neutral in the case of a inertialess rotor and so can be ignored. The rotor offcouse has the freedom to move in a different plane to the rest of the gyro and damp fusalage pitching but this can be accounted for by a separate derivitave for rotor damping ( or clubing the effect into Mq ( the HS paddle effect)). At least this way everone can be clear on all the interactions. I recall shapiro saying this as well in his book. Chuck can you confirm?

Then one could add to the model rotor
speed variations to get a more complete model. Though to some extend this is reinventing the wheel- this as been done by stewart houston and he in fact has made a comparison between ignoring and not ignoring rotor speed and he has shown that the rotor speed degree of freedom impacts the phugoid damping, specifically the rotor torque derivative.

Al_Hammer

04-27-2004, 11:25 AM

Raghu,

I can see that the formula you posted indicates that the HS paddle damping has an effect on frequency, although it is hard to decipher some of the terms..

I see from your post to Greg that your approach has been to use classical control theory in which linear differential equations representing the behavior of the system are transferred into algebraic equations via the Laplace transform.
You then determine the transfer function and work in the s-plane, plotting complex conjugate pairs representing poles. The phugoid is controlled by the slow poles and the short period is controlled by the fast poles.
The transfer function of a DC-8, for example, has been found to be:
G(s) = 4:57(s + 0:154)(s + 0:735)/
(s2+ 2:16s + 9:928)(s2+ 0:0128s + 0:00061)

the denominator contains the 2nd order, slow poles, giving rise to the phugoid.
In general, fast poles are farther away from the stability boundary than the other system poles. Transients associated with fast poles extinquish faster.

The spreadsheet model that Greg is developing does not use transfer functions or diferential equations. Rather, he simply represents all the forces and moments acting on the gyro as vectors and sums the moments. Using a discrete time approach, a series of iterations results in discretized output in the time domain.
Displacement, velocity and acceleration are found by applying the vector sum of moments and forces to the equations of motion of a solid body.
My only concern with Greg's model, which he sent me a working draft of, is that I think his vector analysis is incomplete.
Hollmann's model,to me is more accurate in that it uses 3 coordinate systems; one for the gyro body, one for the rotor, and one absolute system to which the other two are referred through transformations.
The vector math needed to do the transform is a bit sticky, but without it, you get into trouble when you rotate the gyro and try to figure out what the forces are in the x and y directions.
The rotor in Greg's model would be represented with parametric values for inertia, RTV angle, etc. This is simpler than trying to use blade element theory to sum all the forces and moments along the span and integrate.
The advantage of using blade element theory is that flapping , rotor thrust and RPM, etc, could be derived using the characteristics of the particular airfoil and blade dimensions.

PS: I have a shockingly limited grasp of transfer functions and control theory, so all of the above could be a complete crock. :)

raghu

04-27-2004, 01:33 PM

Al,
Well I was describing a model using a system of partial differential equations that represent the longitudinal motion ( three of them one each for the forces in the forward and upward direction and the other is the pitching moment). You can use l'place transforms, alternatively, just obtain the charecteristic equation matrix and solve for the roots ( it is a algebric quatratic-power 4 equation and there are plugins to solve these in excel) . The roots will indicate the nature (frequency and damping of the oscillations). This was what I was suggesting.

I am aware how greg has modelled the problem. The trouble is it is more of a simulation than a model, you need additional methods (sampling) to work out the frequency and damping and this can be quite un wieldy.

Yes you have to be careful with the co-ordinate systems as newtonian laws do not apply in accelerating frames. A popular way is to use the so called wind axis. Not sure though about Hollmans multiple axis...seem unnecessary. Not sure about his model either, remember reading about it a while back in custom planes and it did not make much sense to me.

Al_Hammer

04-27-2004, 03:14 PM

Thanks, Raghu, I'll have to take my time absorbing all that. ;)

on the choice of a coordinate system:

Probably the most efficient way to model the dynamics of a complex moving system is to:

1. Define axis systems that are related to certain parts of the object and its environment.
2. Define transformation angles and corresponding transformation matrices between the different axis systems.
3. Describe each position, attitude, velocity, angular velocity, force and moment in its "own” axis system, in order to get vectors with a maximum number of zeros.
4. Transform all the vectors via the transformation matrices into the same axis system.
5. Apply Newton's second law in this axis system.

The quote above comes from a kite simulator site.
http://buchholz.hs-bremen.de/revsim/mkreal/paper.htm
This page also shows the use of transformation matrices to change from one coordinate system to another.

Greg: You make a good point about HS damping being necessary in the model.
The kite sim is no exception, and lists damping as one of the variables.

darrellwittke

04-28-2004, 10:53 PM

To Greg Gremminger. You're right Greg, about the proposed flight tests (and other things) to be both quantitative and not "seat of the pants" as Dofin F./Duane H. espoused (to my perceptions.) I would also like to go on record how much I admire your post's on this and Norm's conference, which is by far the main source of my (limited) gyro knowledge.

I have questions, however, about the accuracy of what I've picked up over the years and thought it might be good to say what I think I know and have it checked by more knowledgeable people. (Such as yourself.)

Therefore, I concur that tandem machines can be designed with more than adequate horizontal stabilizers to be as stable as practically necessary. I disagree that a tandem design with good tailfeathers should be promoted as the best design practice. More than adequate, certainly, and with flight qualities that people I respect (from this forum...namely you and C. Wall) desire. My perception is the most optimal design is "short and stubby" as Dan H. suggested (applying to the cabin area, not tailfeathers). This simply minimizes the forces the horizontal stabilizer has to control. I also believe the thrustline should pass through the center of gravity and center of pressure for the same reasons (for an optimally stable gyro design.) Is this correct? If not, why not? (I really don't know, I'm just parroting things I think I've picked up on the conferences and finally decided to ask.)

I also have a deep curiousity about the veracity of rule #4 which states it is best for the rotor thrust line to pass through or behind the center of gravity not in front of it. I seem to remember Doug R. stating this, and perhaps I have it completely backwards, but possibly you or others could verify this for me? (My perception is optimally it should pass through or as close as practically possible to center of gravity.) That is why I asked about Mr. Tervamaki's design idea regarding this, as my perception is he was the progenitor of the design, which was later modified by Mr. Magni. (I also remember your post (which became rule #9) in which you stated you had contact with him.) If it is at all possible, and not to much trouble, I would still love to know why he designed it that way.

As you can see I still have a lot of questions, but be assured that I am following the trail that you and others are illuminating for me (though admittedly...slowly.)

With greatest regard, Darrell Wittke

P.S. Gentlemen (Raghu, Al H., Udi etc.) I personally believe reinventing Dr Houston's work would be enlightening to many of us here. But who pays for such work?

gyrogreg

04-29-2004, 06:03 AM

Darrell,

I’ll try to keep this short (not likely!): I think the jury is still out on whether “short and stubby” or “long and narrow” would be easiest to make stable. My intuition is having a hard time accepting that the higher MOI with necessarily slower pitch reactions would be less stable – especially since it appears that the real dangerous area for gyros is where they get into rapid pitch action – PIO and incipient buntover. But, I do emphasize that either configuration can be made stable with good design and verification testing. I also would not conclude that short and stubby requires less that the HS has to do. Other possible configuration issues can complicate HS requirements, such as a SBS model might have a windscreen or enclosure that presents significant aerodynamic download forward of the CG that the HS moment must over-power.

RTV/CG: I would not characterize CG forward of RTV as a Tervamaki or Magni principle. Their designs may have originally arrived at this condition from other development means, and I would not even say that they would place as much emphasis on this as I and others have. CG forward of the Lift Vector and a down-loaded HS is a basic principle in airplane stability, and applies to gyroplanes for most of the same reasons – and for other reasons. I will see if Jukka has any thoughts on how this might have influenced his original JT design and how he feels about the current emphasis on this parameter now. Mr. Magni has always dismissed such simplex reliance on single concepts such as CG/RTV relationship – preferring to emphasize that the overall stability and handling properties are a complex “harmony” of many issues. I would interpret this to include many of the very complex (static AND dynamic) issues that have been broached in this thread. But, I doubt that either Jukka or Vittorio delved that deeply into such “paper designs”, recognizing the technical challenge, preferring in both cases to evolve final characteristics through iterative modifications and flight testing.

As I have said, the emphasis I have placed on the CG/RTV concept is that it presents stability concepts in technical terminology that I think the general masses can understand more readily. I am a gyro CFI, and it is important for me to be able to help my students understand and respect the need for stable gyros, or at least to have enough knowledge to be able to make good decisions about what and how to fly their gyro. The CG/RTV relationship readily explains static airspeed and g-load stability – important concepts for gyros and analogous to the basic stability concepts for airplane students. Many of the concepts broached in this thread address dynamic responses, much more complex subjects involving derivatives, differential equations and Laplace transforms, etc.

Rotor effects such as what Udi readily espouses can probably contribute to some stability – perhaps providing even some positive static stability even if the CG is on or slightly aft of the RTV. However, even this effect could be represented as an effective re-alignment of the RTV as a simpler way to understand static stability. For the less technically astute students, the CG/RTV concept is much more easily visualized for pilot instruction purposes. I encourage those who can, to continue to drill the technology to deeper levels – but for most of the gyro community, I maintain, those levels of intricacy would add to confusion and decrease acceptance of gyroplane stability imperatives for the general community. The first requirements for gyroplane stability are the static stability issues. I feel these are readily flight tested to determine, and most readily explained in terms of CG and RTV. If we need to start saying “effective RTV”, then we might have to start trying to explain what an “effective RTV” is – more confusion and less acceptance?! I prefer to leave that intricacy for those who dig deeper!

As far as DYNAMIC stability that most of these complex concepts address – even if thoroughly “paper designed”, would really require testing to verify as well. But, that dynamic testing requires more professional expertise and can possibly be dangerous if done on dangerous designs. But, I suspect, that when a gyro employs an adequately effective HS to achieve the static stability requirements, that HS is probably more than adequate to provide safe dynamic characteristics - more testing is necessary to support this. My suspicion is that the deeper intricacies of a “paper design” to address dynamic properties would be more than over-shadowed by other vagaries, and most designers would resort to iterative flight testing for final “tuning” anyway.

For your list of design goals, it may be misleadingly simple to just say the RTV should be designed to be on or aft of the CG. This is not something you can just draw on a side view drawing. The position of the RTV relative to the CG in flight is the result of the Sum of all Static Moments acting on that gyro in flight. So, this is not a simple “cookbook” guideline. An understanding of static moments, including those from propeller thrustline offset and aerodynamic drag offset – and others – are necessary to understand what it means to have the CG on or forward of the RTV – in flight. But, this concept of CG/RTV relationship is relatively easy to visualize or draw and understand, and is easier to determine by proper flight testing. To suggest the RG/RTV relationship is a guideline, requires more than just saying to do this!

I maintain that it may also be misleadingly simple to say that thrustlines and draglines should be aligned or aligned with the CG - CLT, etc. This does not account for other static aerodynamic moments, and would imply that simple neutral stability is adequate. If the "effective RTV" concepts apply, perfect CLT may be somewhat positively stability. But, a Sum of Static Moments that results in the CG further forward of the RTV can add the improved gyro stability that makes such "gyroplanes" as easy to fly and learn to fly as true safety of the sport probably requires.

Thanks, Greg Gremminger

RHerron

04-29-2004, 12:29 PM

Greg,

Have a look at all the Cierva designs, even the earliest. They clearly show the rotor thrustline to be located aft the C of G. Old drawings actually give the linear distance for example.

Later, around 1928, he introduced propeller thrustlines which intesected C of G as a principle factor for achieving longitudinal stability, in addition to horizontal tails and RTV .

gyrogreg

04-29-2004, 05:22 PM

Ron, the RTV is not always able to be simply depicted on a gyro drawing - it depends on how low or high the nose flies in flight - to determine where the CG actually is in flight. Also, the RTV angle changes at different airspeeds.

On Cierva type designs, or any design with a huge HS volume, it can be reasonably assumed that the airframe will align with the relative wind – airframe level - so we can pretty well tell where the CG will be in flight. The RTV at various airspeeds and loads might be fairly predicable also. But, without the rotor RTV information, you can't just look at a side view of a gyro to say how the RTV aligns with the CG.

The Cierva designs probably do have the CG (in level attitude) well forward of the RTV at all times. The HS down-load ability of the huge tail volume would be holding the airframe in near level attitude for all flight regimes - therefore the CG would probably remain well forward of the RTV in all flight regimes but especially at high airspeeds where the RTV would bve farther aft yet! This is probably the same for the LW as it is also for the Magni design.

Under this condition of the CG well forward of the RTV in level attitude, the HS will be providing a lot of static moment to hold the nose up. This is one reason those designs are so stable - a lot of airspeed and g-load static stability because the CG is so far forward of the RTV and a significant amount of HS down-load. By using CLT on such designs, the HS just has to be doing a little less static "balancing" work - it only has to hold the heavy nose up - it doesn't also have to also "balance" the offset prop thrustline. The prop thrustline is just one of the static moments the HS has to statically balance, but with CLT, it doesn’t have to do that! Removing the necessity of it having to also hold up the nose because of a high prop thrustline, just makes the HS job a little easier - a bit less nose-down AOA of the HS. Because the HS is so far aft of the prop on the tractor design, the HS would have a lot less effect from the propwash, and therefore CLT makes it easier to hold the nose level because the HS does not have to do that job in reaction to the propwash part of the airflow on it - lower propwash immersion factor. For pusher gyros, the HS has more effect from propwash, and it makes configuring the HS to "balance" a high prop thrustline a bit easier. I would expect a high prop thrustline on a tractor design could be a bit problemsome – statically - just because the HS has a lot less propwash to work with to balance that offset prop thrustline.

But, I think the point you are making is that, with the CG well forward of the RTV, you have gyros as stable as those Cierva tractor designs. The farther forward the CG is of the RTV, the more statically stable the design is - the point I have been making. A good reason the LW is probably the most stable gyro available today. I would sure like to see the static and dynamic flight test numbers on the LW - probably the "gold standard" for stability!

Thanks, Greg

RHerron

04-29-2004, 06:39 PM

Greg said:
<<Because the HS is so far aft of the prop on the tractor design, the HS would have a lot less effect from the propwash, and therefore CLT makes it easier to hold the nose level because the HS does not have to do that job in reaction to the propwash part of the airflow on it - lower propwash immersion factor. For pusher gyros, the HS has more effect from propwash, and it makes configuring the HS to "balance" a high prop thrustline a bit easier. I would expect a high prop thrustline on a tractor design could be a bit problemsome – statically - just because the HS has a lot less propwash to work with to balance that offset prop thrustline.>>

Greg,

Interestingly, Cierva's center-thrust design was to correct for a low,(not high), prop thrustline. Prior to this, his machines had a tendancy to pitch up with application of power and down with reduction. From his and Pitcairn's experiments it was determined that overall longitudinal stability was improved with the thrustline centered on CG.
Also, from my experiments with elevator control I would have to say that horizontal tail power is extremely powerful. While not in the immediate prop blast, it is still immersed in substantial flow from the prop as the slipstream spirals around the fuselage.
I have been able to lift up the tails of these machines (on the ground) using "down elevator" and throttle . No different from a J-3 Cub in that regard. That represents a tail weight of over a hundred pounds. That tells me that the horizontal surfaces are getting plenty of air from the prop. What little is lost in propeller velocity is made up for by shear leverage.

No doubt about it, lots of horizontal tail surface is just good business.

Hope to see you soon.

Ron

Al_Hammer

04-29-2004, 09:38 PM

While static and dynamic stability are the two basic kinds of stability, there are additional classifications.
One of these is maneuvering stability. Greg might call it g load stability, but he says(in gyro glossary):

In a gyroplane, g-load stability means
that the Rotor Thrust Vector (RTV) is physically located aft of the CG of the aircraft.

The point of pulling back on the stick is to move the RTV ahead of the cg so the nose will pitch up.
Having the RTV forward of the cg is unstable, so some means must be provided to prevent a divergent pitch up.

Here is what a test pilot has to say about maneuvering stability in a helicopter.
In a maneuver, the lift on the tail switches from download to
upload. What we hope is that the tail always has enough "lift" to counter the
rotor's wish to just keep digging in and flipping the aircraft.
Usually aft cg is the worst case, and we test for positive maneuvering stability, looking
for the tail to keep the nose in place in a high G pullout. The way we can
tell is to note if we need to push forward on the stick as we hold the G in a
turn. If we have to use cyclic to keep the nose from accelerating its pitch
rate, then the tail is not big enough yet. We then make it bigger, or reduce
the aft cg, or add a Gurney flap to make it more effective.
-N. Lappos

In a pusher gyro we can't get the HS far enough back to produce significant pitch damping. ie, rate damping.
However, we can simply make it larger than needed for static stability as mentioned above.
Also, it helps to have enough pitch pivot offset in the rotorhead to provide a positive stick feel. As more g load is pulled, the rotor will exert forward pressure on the stick.
Also, it is possible to use blades that have a pitching moment such that pitch is taken out as angle of attack increases. This, again, pitches the rotor forward and adds stability.
Another factor is that as the gyro pitches up, the mass below the rotor will tend to swing back down with gravity. In this sense there is pendulum static stability. HS damping helps limit the overshoot on the way down.

So, damping = dynamic stability. We could stop using the term dynamic stability altogether and just call it damping.
to sum up: Static stability acts to return the aircraft to trimmed condition, damping makes sure it doesn't get there too fast and overshoot.
Maneuvering stability insures stick forces don't reverse in tight turns and climbs and prevents divergent pitch ups. :)

quadrirotor

04-29-2004, 09:57 PM

Al, what could be the contribution of hydraulic dampers in the command links?...Let's say those air springed dampers used for levering the hood (or back door) of a car, to be used for triming the rotor with an electric actuator (for example)...

Al_Hammer

04-30-2004, 12:14 AM

Good question, André. Maybe flybars would be another way to do the same thing.

quadrirotor

04-30-2004, 04:19 AM

Sorry Al, what are flybars?

gyrogreg

04-30-2004, 05:49 AM

Al, you commented on Maneuvering or G-Load static stability. I agree with some of your points. But, you said “The point of pulling back on the stick is to move the RTV ahead of the CG so the nose will pitch up.” I take issue with this. The RTV doesn’t have to be moved forward of the CG to cause the nose to pitch up. When the Sum of Static Moments are in equilibrium, any forward movement of the RTV will be a temporary imbalance and cause the nose to pitch up – the nose doesn’t have to be PULLED up! If the CG is statically far enough ahead of the RTV, there is no need for the RTV to be ahead of the CG to pitch the nose up.

You go on to say: “Having the RTV forward of the CG is unstable, so some means must be provided to prevent a divergent pitch up.” Even if the RTV is commanded to be forward of the CG, there is no concern that it might diverge in a pitch up. The nose will pitch up – as quickly as the temporary imbalance of the Sum of Static Moments causes it to pitch – but, quickly, the CG will move ahead of the RTV and restore static G-load stability – no divergence. This is all dependant on whether the static G-load condition is stable – the “effective” RTV is aft of the CG.

Another point: You said: “it is possible to use blades that have a pitching moment such that pitch is taken out as angle of attack increases.” I think you are suggesting blades that are able to twist from root to tip as load is increased. This would suggest that the center of lift on the blade is also a bit aft on the blade to cause it to pitch to lower AOA (possible blade flutter or weave??). I don’t think we want to encourage more flexible (twistable) rotor blades. “Twistable” blades can lead to something I call “Blade Runaway”. Blade runaway is the phenomena that causes the blades to speed up further as the blade tips start getting more compressibility influence closer to the speed of sound. As tips speed up, the center of lift moves aft, twisting a flexible blade to lesser blade pitch, further speeding up the RRPM and getting the blade tips into more compressibility effects. This further moves the CL aft on the blade, more twist, and faster RRPM. Rotor blade designers concern themselves with this issue – one reason some longer rotor blades are spun at slower RPMs – to avoid compressibility affects at the tips and possible blade twisting to lower AOA as a result. This is an issue if the blades are “twistable”.

You suggest calling “dynamic stability” as “damping” instead. I may be wrong in this, but we have come to define “dynamic stability” as two issues – both important for safer gyros. “Damping” is one of these factors for sure – all natural oscillations should be “damped” to at least some degree. In the LSA standard we define this as simply “damped” – do not require a damping rate. A commonly accepted damping rate would be damped to 10% amplitude within three cycles, but for the gyro standard, we are simply requiring that they be damped to some degree!

But, we also include another parameter under the general term of “dynamic stability”: No natural oscillation tendencies at periods less than 5 seconds. The tendency to oscillate too quickly – even if damped – should be avoided in gyros. We know that PIO is a major safety issue in gyros. We know that most pilots, can “self-damp” a slower oscillation. But for faster oscillations, pilot reactions may be “out-of-phase” and therefore actively cause the oscillations to diverge – PIO. Even if quick oscillations are inherently damped by aerodynamic damping, those nose-pitching oscillations might excite the pilot into reactive “wrong phase” control inputs – overpowering the inherent damping and causing dynamic pitch divergent oscillations. For PIO reasons, we have included both a damping function and a minimal natural oscillation period under our term of “dynamic stability”. This may not be technically accurate terminology, but for gyroplane safety, we should understand that a gyro should also be “dynamically stable” – no tendencies to excite the pilot into reactive over-control or “wrong-phase” commanded inputs. Note: a natural oscillation tendency may be “critically damped” and would therefore not be a tendency to oscillate. In other words, any natural oscillation tendencies at periods shorter than 5 seconds should be critically damped. When both conditions of this “dynamic stability” condition are satisfied, there will be little or no tendency for either PIO or over-reaction into an incipient buntover.

Again, my purpose for "pushing" stability issues is not purely for the scientific fun of it. I believe that if we can communicate the basic issues of stability to the whole gyro community, gyro people will be better equipped to make good gyro flying decisions. I tend to not stick to purely text-book terminology if clearer language will help more people understand and respect the important safety/stability concepts.

Thanks, Greg Gremminger

Thanks, Greg Gremminger

Udi

04-30-2004, 07:07 AM

Definitions…

In a gyroplane, g-load stability means that the Rotor Thrust Vector (RTV) is physically located aft of the CG of the aircraft.

Having the RTV located aft of the CG is one of a few mechanisms that make a gyro G-load stable. Having the RTV aft of the CG is helping (again, one way) to make the AIRFRAME G-load stable. This is helping to make the gyro, as a whole, G-load stable through trim-spring feedback and friction in the control mechanism.

The other important contributor to G-load stability is, of course, the offset gimbal head. This is a more direct mechanism. The horizontal stabilizer is also helping G-load stability by keeping the airframe pointed into the wind even when the RTV is forward of the CG. This is described nicely in Al's post above.

So, damping = dynamic stability. We could stop using the term dynamic stability altogether and just call it damping.

Damping is one of the handles that help us tweak dynamic stability. Dynamic stability is the end result of all the components working together. Other mechanisms that affect dynamic stability are airframe MOI, horizontal stabilizer, rotor dynamics, offset gimbal head, trim spring, etc.

Different systems pull in different directions. Dynamic stability is the sum of all these systems working together. Dynamic stability is measurable.

Udi :cool:

gyrogreg

04-30-2004, 07:27 AM

Udi, I can agree that the definition of “G-Load or Maneuvering stability in the Glossary of Gyro terms might need a better or more complete definition. I do agree that the offset gimble/trim spring arrangement can also be a static G-Load stabilizing mechanism.

But, it has always been my understanding that the offset gimble/trim spring is mostly a mechanism to neutralize the natural static instability of the rotor. It does so by adjusting rotor disk AOA in the corrective direction as a result of a change in G-Load on the rotor.

The old tailless Air Command, as I am sure a number of other tailless gyros do, required some forward stick pressure to maintain trimmed airspeed in a banking turn. This would be an indication that the (“effective”) RTV is forward of the CG enough to even over-power the offset gimble action. I know there are additional complicating mechanisms to all of these concepts, but, again, I tend to try to keep it simple enough to be understood, accepted and respected by the less technically patient masses. It is great for those of us who might be able to delve into these issues more deeply, but it is the understanding of the general gyro public that will help to reduce fatalities related to stability issues.

I would accept any suggestions for updates to any of the terms in the “Glossary” – just remember this is intended for general public absorption, not PHD Mathematicians.

Thanks, Greg

Udi

04-30-2004, 07:30 AM

...I think the jury is still out on whether “short and stubby” or “long and narrow” would be easiest to make stable. My intuition is having a hard time accepting that the higher MOI with necessarily slower pitch reactions would be less stable – especially since it appears that the real dangerous area for gyros is where they get into rapid pitch action – PIO and incipient buntover...

A higher MOI is acting to slow down the short-term period. i.e. make a lower short term frequency. From a practical point of view, this means the gyro will react more slowly to pitch upsets, and (the other edge of the sword), react more slowly to pilot inputs.

Having a longer short-term period may be helpful in gyroplanes that are inherently unstable, and require the pilot to be the stabilizing mechanism. A slower frequency allows the trained pilot to react to upsets before the amplitude gets out of hand. The pilot is the stabilizing force and making the short term period longer helps the pilot react in time.

So what's the problem with a higher MOI? If you take the pilot out of the loop it makes the gyro even more dynamically unstable. This gyro will require more damping than a lower MOI gyro. In a gyroplane, more damping = more stabilizer. More stabilizer = higher short term frequency. Back to square zero.

So, for the purpose of designing a new gyroplane, a higher MOI does not make a lot of sense. But for the older lead sleds, which are statically and dynamically unstable, a higher MOI may help the pilot "stay in the loop" and keep the oscillations from diverging.

Udi :cool:

Udi

04-30-2004, 07:39 AM

raghu

04-30-2004, 07:51 AM

So, for the purpose of designing a new gyroplane, a higher MOI does not make a lot of sense. But for the older lead sleds, which are statically and dynamically unstable, a higher MOI may help the pilot "stay in the loop" and keep the oscillations from diverging.

Udi :cool:

Udi, so we tipped you to the right side of the fence afterall :) . I am afraid greg has been more resistent....perhaps I am flogging a dead horse but to further emphasize.

As Greg says PIO is a major concern in gyros. All else being equal increasing MOI will make PIO worse as the lead and lag to control inputs gets worse- think dougs water temperature example of the shower.

raghu

04-30-2004, 08:13 AM

Definitions…

Having the RTV located aft of the CG is one of a few mechanisms that make a gyro G-load stable. Having the RTV aft of the CG is helping (again, one way) to make the AIRFRAME G-load stable. This is helping to make the gyro, as a whole, G-load stable through trim-spring feedback and friction in the control mechanism.

Udi :cool:

Udi, I agree with the import of your statement on G loading viz. that the definition of G loading must be generic and not prescripitive ( saying what it means rather than how to get there) but I want to emphasize some details. G load stablity is one of the static stabilities in a gyro. It tells you the response in terms of pitching moments when the speed of the rotor is increased. It is stable if an increase in rotor RPM results in nose down piching moment ( as is intuitive). The obvious way to do this is to place the RTV behind the CG, but there are other ways too. For example the rotor can be made nose heavy so that increase in rotor speed will decrease the magnitude of the RTV and consequently make even a RTV ahead of CG gyro G loading stable. However, HS or other aerodynamic and non rotor surfaces do not have any direct impact on G loading stability.

The two static stabilities that have the biggest impact on a gyros dynamic response is the AOA stability and the rotor G loading stability. You would be hard pressed to find a gyro in a conventional configuration that has both of these and manges to be (longitudinally) dynamically unsuitable.

C. Beaty

04-30-2004, 08:19 AM

Another point to consider vis-a-vis the MOI/damping relationship is that pilot control about the roll axis seldom presents a problem.

One reason is that roll axis MOI is always smaller than pitch axis MOI and although the only damping is the result of rotor lag, it is evidently enough to be near critical.

Of course, not having to confront the devil of translation is another.

Udi

04-30-2004, 10:45 AM

...As Greg says PIO is a major concern in gyros. All else being equal increasing MOI will make PIO worse as the lead and lag to control inputs gets worse- think dougs water temperature example of the shower.

Raghu - you are ignoring the pilot as part of the loop. Take two statically and dynamically unstable gyroplanes. One has a short period of 2 seconds, and the other 5 seconds. Although both gyros will PIO as a result of out of phase pilot control inputs, I say, the pilot of the 5 seconds gyro has a better chance of LEARNING to control his gyro (with properly-timed inputs); even though his machine has worse lead/lag responses. The 2-second pilot has no time to think; they have to be pre-programmed with correctly timed inputs, or they die.

Here's an example everyone can relate to: take a broomstick and balance it on your finger. You can do it with no problem. Now, take a pencil and balance it on your finger - more difficult (impossible?). Both systems are unstable, but the broomstick has a higher MOI/lower period.

Udi :cool:

raghu

04-30-2004, 10:55 AM

Absolutely udi! I think we are saying the same thing...to quote from my first post to you on this subject

Here is the catch. If you are flying a unstable gyro, the pitching response to a disturbance amplifies the disturbance, so you are better of with a higher MOI gyro than a lower MOI one as the higher inertia will give you more time to react to the disturbance. However, even in the case of the unsable gyro the control situation is worsened with an increase in MOI, so you will have to deal with incresed lag and overshoot, but atleast you have the time.

Also, your broom stick example is not relavent as all things are not equal between the pencil and the broom stick ( both the moi and the degree of instability are different between the pencil and broomstick) . Consider 2 broom sticks of equal length. One has a 1 pound weight at the tip and the other a 2 pound weight at the middle of the broom stick. Now both have exactly the same degree of instability ( heavy side in the air) but the one with the 1 pound weight has a higher MOI. Which one would you find easier to balance on your finger tip?

Udi

04-30-2004, 11:43 AM

...G load stablity is one of the static stabilities in a gyro. It tells you the response in terms of pitching moments when the speed of the rotor is increased. It is stable if an increase in rotor RPM results in nose down piching moment ( as is intuitive). The obvious way to do this is to place the RTV behind the CG, but there are other ways too. For example the rotor can be made nose heavy so that increase in rotor speed will decrease the magnitude of the RTV and consequently make even a RTV ahead of CG gyro G loading stable. However, HS or other aerodynamic and non rotor surfaces do not have any direct impact on G loading stability...

Raghu - I am a macro kind of guy... Although the stab has no "direct" effect on G-load stability, it does have a very significant indirect effect on G-load stability. I may be using wrong terminology to describe this stability, so let me explain.

A gyroplane with the RTV forward of the CG is entering a turn. The nose will want to rise, slowing down the gyro, requiring forward stick. Negative G-load stability. BUT, this gyro has a large stabilizer. The stabilizer has a steeper lift curve than the rotor. When entering the turn, the rotor and the stab AOA increases, but the stab has a steeper lift curve so the NET static moment would be nose down. Also, the initial AOA of the stab was smaller than the initial AOA of the rotor, so the delta AOA of the stab, in percent, is much larger than the delta AOA or the rotor.

Direct or indirect - I think this gyro would be G-load stable.

Let's examine a practical example. I like to use the Sparrow Hawk as an example because it is such a clean design. The SH is a true CLT gyroplane and it has a huge stab, mounted with no angle of incidence. If we consider all the moments about the CG in the SH, the location of the RTV relative to the CG depends mostly on airspeed. At high airspeed, the rotor disc angle is shallow, and the RTV moves back. The huge stab does not let the nose go down, and so the RTV is located solidly aft of the CG.

At low airspeed the opposite happens. The rotor disc has a high AOA, and the RTV moves forward. The gyro wants to fly nose high, but the stab is now pulling the nose DOWN, moving the CG behind the RTV. In this flight regiment, the airframe is definitely G-load unstable. BUT the gyro is G-load stable! Proven! It was tested.

Call it G-load stability or maneuvering stability - the stab is a major contributor.

Udi :cool:

Udi

04-30-2004, 11:50 AM

...Consider 2 broom sticks of equal length. One has a 1 pound weight at the tip and the other a 2 pound weight at the middle of the broom stick. Now both have exactly the same degree of instability ( heavy side in the air) but the one with the 1 pound weight has a higher MOI. Which one would you find easier to balance on your finger tip?

Uh... I don't know. Don't they have the same MOI (mass times distance)?

Udi :cool:

darrellwittke

04-30-2004, 12:12 PM

The broomstick with the weights on the end are analogous to a tight rope walker with the long stick right? I found my physics 101 book the other night... I'll be back.

raghu

04-30-2004, 12:29 PM

Uh... I don't know. Don't they have the same MOI (mass times distance)?

Udi :cool:

No, MOI varries with the square of the distance, while the destabilizing moment is the same- force * distance!

Al_Hammer

04-30-2004, 12:34 PM

Udi, MOI in this case is proportional to mass times distance squared.

Darrell, the tight rope walker with the long stick is different. In that case the heavy stick puts significant mass below the cg of the person, lowering the overall cg and making it easier to balance. The tall broom is easier to balance than a short one because the angle of tilt changes more slowly, allowing more time to make corrections.The corrections need to be larger, however.
An inverted pendulum controller using fuzzy logic. http://www.aptronix.com/fuzzynet/java/pend/pendjava.htm

Raghu, now I'm confused. Your definition of g load stability seems to be quite different than the one for maneuvering stability.
One is in terms of rotor speed, the other in terms of reversal of stick forces.
Which one are we using?

Greg,
Judging from the following description of blade weave, it would appear that it should not be a problem if there are no pitch links.

Should the blade AC be ahead of the blade CG, then when the blade sees a disturbance it will try to diverge and pitch up or down. It is kept from doing so because of the pitch linkage. However before this diverging pitch change is stopped, the pitch linkage is loaded up and stretched. Now once the blade pitching stops, the stretched linkage pulls the blade back and it pitches in the opposite direction until the pitch linkage is again loaded up. If the blade oscillation amplitude does not dampen out, then a blade weave will develop which shows up as a severe stick shake. The stiffer the pitch linkage the higher the blade weave frequency and the higher the blade RPM can turn before a blade weave will develop. -Jay Carter

While I am aware of speculation about blade runaway(it was discussed on Norm's forum), it would be useful if you could provide a reference.
I don't see why the rpm isn't self limiting given the fact that drag goes up with the square of airspeed, and as mach 1 is approached,it goes up even faster.
Also, if the blades were to pitch down in flight, wouldn't lift decrease and the driving region grow smaller?

I think Chuck has mentioned how the Bensen wooden blades were torsionally flexible and this made them inherently more stable as I recall.

RE: the RTV moving forward during maneuvering: Given that head can move about 9 degrees or so off of neutral, it is easy to see that for any moderate back stick movement the RTV is going to swing well forward of the cg no matter what the static balance is.
If the HS is just exactly balancing the RTV in trimmed flight, then obviously it may tend to be inadequate to stabilize the gyro when the RTV is well forward. It certainly holds true for helicopters, and they have the advantage of a long moment arm on the HS, and they have the same lift slope ratio of rotor/HS as gyros, more or less. (The lift slope of a rotor is less because the AoA is less affected by translational speed- most of the wind the blade sees is from rotation.)

gyrogreg

04-30-2004, 03:45 PM

Al, Udi and Raghu, you guys are leaving me in your dust – I don’t think I can keep up – but I want to!

Someone stole the yardstick balanced in the hand from my static stability forum presentation last year at Mentone. In that demonstration, I used both a long and a short stick, and clamped a weight at various positions on the stick to demonstrate the easier balancing with the higher MOI. The shorter stick's, more negative static stability, were definitely beyond the ability of the “pilot” to stabilize – no matter what MOI!.

I tend to try to relate what you are suggesting to real world flying in a number of gyros I have experience with. So, my impressions of some of these proposed concepts are either beyond my comprehension or adjusted by real-world flying experiences. I’m not sure I can buy that the RRPM changing under loading is a major contributor to static G-Load, maneuvering stability. The Magni has a very heavy rotor – slow to change RRPM. The rotor disk AOA and airframe nose reacts to commanded cyclic input much quicker than the RRPM. For commanded disturbances, the RRPM is in reaction to the g-load change – after the nose has already reacted in pitch to the commanded input. The changing RRPM is another element, but does not appear to me to be the fastest or major element in maneuvering.

The long tandem Magni M-16 has a very high MOI – especially with me occupying the front seat! I have never been aware of any tendency to be slow to respond – in any axis – and it has not demonstrated any tendency to over-shoot in control. It does demonstrate some over-shoot at it’s natural oscillation frequency – 12-15 second period. The damping rate is between 1 and 3 cycles – depending on the severity of the initiating pitch disturbance - but this is not noticeable when normally flying and has apparently not been an issue for even novice, first time pilots. I have had many first time pilots on the controls in even rough air – with no tendency to over-control or be sluggish or laggish in control. This may not be a fair assessment, because, in even the roughest air, there is no commanded response excited in even the most novice pilots. But, at low heights above the ground in very rough air, the pilot is required to apply stick inputs to maintain the height over the ground, and there are still no difficulties for even the most novice pilot to control the height at even high airspeeds (100 mph!). So, if I seem a bit biased on some of these concepts, it is because I think this real-world example verifies some of the theories – such as high MOI is a good thing, even for stable designs.

Blade weave or flutter – ala pitch links. Al, I think the statement you quote verifies what I am saying – a torsionally flexible rotor blade that has its AC ahead of the CG is susceptible to blade weave or blade flutter. Our rotor blades do not have pitch links that can stretch or compress, but if the blade can twist, this is the equivalent of stretch pitch links. It is dangerous to design the rotor blade tail heavy!

I have no references for “Blade Runaway”. I’m not sure there are gyroplane textbooks or studies on blade runaway. Believe it or not, maybe some gyroplane principles are not yet mature! But, for rotor designers, this has been a real concern and a major reason they don’t want their longer rotors turning too fast. But, there may be a real-world example of such. Remember the Marchetti that drove into the ground at Mentone a few years ago. This is not verified, and I don’t mean to open up old sores – but this was very likely a condition of “blade runaway”. The RRPM on these new blades was noted to be higher than the designer was comfortable with – for just that “blade runaway” concern. The pilot and passenger noted the stick forces kept pulling harder and harder forward as the gyro picked up speed and impacted the ground. Both pilot and passenger noted the stick forces quickly built to much higher than both persons, pulling hard, could overcome. This incident cannot really be explained by any other mechanism – at least I have heard no other mechanisms proposed. This subject isn’t a real issue in the stability subjects you are discussing – I was just making the point that we should not design some form of static stability that requires the blades be very flexible – for both blade weave and flutter considerations, and for “blade runaway” considerations.

If you use torsionally flexible blades for stability improvement, I’d suggest they be shorter for less actual twisting, and slower tip speeds – as in the wooden Bensen blades you suggest. But, I would still find this a risky design on the heavier and higher performance gyros of today!

I seem to be the one playing devil’s advocate here, but maybe it is inspiring more insight. Please talk among yourselves, but I’m having trouble keeping up with the wide range of issues and may not have time to keep responding so much for a while.

Thanks, Greg

Udi

04-30-2004, 04:07 PM

...The long tandem Magni M-16 has a very high MOI – especially with me occupying the front seat! I have never been aware of any tendency to be slow to respond – in any axis – and it has not demonstrated any tendency to over-shoot in control. It does demonstrate some over-shoot at it’s natural oscillation frequency – 12-15 second period. The damping rate is between 1 and 3 cycles – depending on the severity of the initiating pitch disturbance - but this is not noticeable when normally flying and has apparently not been an issue for even novice, first time pilots. I have had many first time pilots on the controls in even rough air – with no tendency to over-control or be sluggish or laggish in control. This may not be a fair assessment, because, in even the roughest air, there is no commanded response excited in even the most novice pilots. But, at low heights above the ground in very rough air, the pilot is required to apply stick inputs to maintain the height over the ground, and there are still no difficulties for even the most novice pilot to control the height at even high airspeeds (100 mph!). So, if I seem a bit biased on some of these concepts, it is because I think this real-world example verifies some of the theories – such as high MOI is a good thing, even for stable designs...

Greg - nobody said that the Magni MOI is too large. Your tests prove that, even with this high MOI, your Magni is well damped! But - if your Magni was closed coupled (i.e lower MOI), you could have had a smaller stab, with equally good results.

Udi :cool:

PW_Plack

04-30-2004, 05:11 PM

Guys, I love reading debates such as this. I'm not qualified to interject much, but I am left to ask for comments on this assumption...

If you believe that a particular design can bunt spontaneously within one half-cycle after a sudden loss of rotor thrust, before PIO is even a factor, it would seem that increasing MOI and lengthening the period of oscillation might allow the pilot a chance to intervene, or at least delay the point at which the rotor impacts the airframe.

If, on the other hand, you believe that PPO generally happens after a pilot unwittingly worsens a gyro's tendency toward divergent pitch oscillations, reducing MOI and shortening the period give the pilot a better shot at avoiding overcontrolling.

Am I correct in sensing that this is one reason for the different starting points in this debate?

Al_Hammer

04-30-2004, 06:33 PM

Paul, yes, I think you've identified one part of the argument.
You remind us of how futile the debate really is, since making a gyro stable in the first place will obviate the need to play games with the distrubution of mass. It's like asking what would happen if I wore a lead suit while riding my unicycle? Would I fall over any slower?
not likely, but a better idea might be to get a bicycle or maybe buy a Segway! :)

GyroRon

04-30-2004, 07:11 PM

I have only followed little bits and pieces of this thread but this is where I am at with it....

Dean_Dolph

04-30-2004, 07:50 PM

Paul, I have absolutely NO credentials that I can offer when posting to a thread such as this but…. I suspect that this debate will go on until everyone tires of the subject without ever reaching a consensus. I base this on C. Beaty's quote in this thread:

Stability of gyroplanes is a subject of so much debate because there is no authoritative reference

We attempt to transfer fixed wing and helicopter principles and gleam what we can from the golden age of Autogiros; more often than not confusing and contradictory.

This quote is taken somewhat out of the context of his message but I think what we have seen demonstrated here is that our well respected experts do not agree on all aspects of the subject. And Chuck's quote indicates why. So they have to base their presentations on another quote from Chuck's message:

The laws of physics are chiseled in stone and apply without exception

While this last quote is true, it seems that they are struggling to define the application and the result with no agreement in sight. The only thing that seems true to me at this time (someone who fits Greg G. description of the gyro layman) is that Greg is correct in saying the proof is in the puddin'! In other words an effort can be made to cook book a design but flight testing is still going to be needed to determine if it actually works.

I've had a well respected gyro person tell me that he would like to write a book on gyro theory at some point in time. And another well respected gyro person say that he doesn't believe anyone can write a book until the theories have been tested and proven, which makes sense. However, sometimes the best way to get something tested is to make a bold statement and then wait for someone to take exception and test it and have the results either prove the theory and/or show what really happens.

It would be extremely valuable to people like me if the energy we are seeing exhibited by our expert enthusiasts was directed to laying out hypothesis and a test program(s) and then finding a academic institution to setup, manage and run the program(s).

I was feeling pretty comfortable in accepting that CLT and a properly designed horizontal stabilizer was enough to ensure that a gyro would be a stable, safe, machine. But now it seems that there are a lot of interacting issues that are not easily defined or fully understood. I'm not going to lose any sleep over it but I almost wish I had not started following this thread!

PW_Plack

04-30-2004, 08:31 PM

Dean,

I think too much is made over the CLT "debate." I don't think there's anyone without a vested financial interest in older designs who doesn't think CLT (or near-CLT with a good horizontal stab) is a good thing. My frustration lies in the fact that a couple of the older designs which don't meet today's higher standards of stability have other safety features I do want.

One of the few upsides of having my financial life placed in limbo by my estranged wife's attorney, is that better stuff keeps coming on the market while I wait for the courts to tell me what I have left! I look forward to seeing what Larry Neal's new machines look like up close, and there are some amazing things about to happen in the two-place world.

CLS447

05-01-2004, 03:56 AM

Always look on the bright side of life ! I am anxious to see this stuff to. We got the Butterfly 2 place, the new SxS sportcopter & from what Ron A. says... something great coming from Dominator. Everytime I talk to Doug at Air Command, he says they have new stuff coming also. I just hate to think that my new machine will be outdated before I even get some good hrs on it.

GyroRon

05-01-2004, 04:51 AM

I am not letting the cat completely out of the bag, Mike or Ernie Boyette who both come to this forum can do that - but the machine coming from Dominator in the near future - it is in testing stages now - will do EVERYTHING you would ever want a gyro to do..... Think Fast like any of Dick Degraws machines, Roomy SxS like a sparrowhawk, CLT of course, and it will not need any runway to take off. Plus target price is less than most luxury cars..... I just hope he doesn't paint them yellow!

GeneWeber

05-01-2004, 10:32 AM

Hi Ron,

I’d like to see us keep the forum “G” rated (off-topic probably won’t be, but the rest would be nice). Could you please do me a favor and delete the picture in Post #78. I’m not real keen on your current signature line either, but it’s a free country.

Thank You & Regards,

Gene

darrellwittke

05-01-2004, 02:11 PM

And threads (although I wasn't expecting a deep technical discussion of MOI benefits, simply a yea or nay vetting of the simple rules) such as this are intellectually challenging to my "not technically trained mind." I would like to extend my thanks yet again to those with higher levels of education and understanding who take the time and effort to share their knowledge.

I would like to caution anyone against belittling what's going on here. This stuff is neither intuitive nor easy. There are numerous examples of aeronautical engineers who should have known better (I. Bensen, D. Fetters. others) but didn't! Striving for deeper technical understanding through any means may prevent a poor and unsafe gyro design killing or disabling a person.

"A properly designed gyro offers one of the safest forms of aviation" being our goal, I echo Greg and Udi and Al's assessment that there are many designs (both tandem and side-by-side) capable of achieving that goal. (More than adequate tailfeathers being the paramount design factor required.) I would hate to see only one optimal design for stability permitted. (Which we've more or less had since the thirties, a tractor with fixed rotorhead.)

My perceptions have been that Greg G. was promoting the Magni Gyro with high MOI and more than adequate horizontal stabilizers as the absolute best solution to proper gyro design and stability in pusher gyros. I now see that isn't true, Greg G. is stating the Magni is more than practically stable to pass objective flight tests of stability and is only one of many possible designs to attain this. (To my perceptions now.) My own simple belief the best design to promote gyrocopter stability is to minimize offset of thrust and drag vectors through center of gravity is simply that, my own belief. I standby that belief and still have not seen anything technical to make me change my mind. (And I'm not technical but I enjoy trying to be!) It's possibly technically simplistic of me to think not only the easiest but the best way to achieve stability is to not introduce any destabilizing factors in the design to begin with. (ie. thrust offset from cg or center of drag) Any Yea's or Nay's greatly appreciated.

This is a technical discussion however, with argument over what is technically the best, not practical discussion of what is sufficient to meet the above goal.

On that note, can anyone enlighten me regarding rule #12, in which C. Schneider once said "After 40 mph, what's important is for the center of thrust to pass through the center of aerodynamic pressure, not the center of gravity."

I suspect he is correct (ie. that the aerodynamic pressures grow in magnitude at 40 mph and above that they easily overshadow center of gravity misalignments) but again, adequate tailfeathers correct any problem.

In the interest of argument, entertainment and knowledge, (not necessarily in that order!) can anyone tell me if it is correct that aerodynamic center of pressure misalignments with thrust can generate the same kinds of torque (ie. 500 ft-lbs in RAF?) as thrust to center of gravity misalignments?

Thank you to everyone for the enlightening and civil discourse. Darrell Wittke