PDA

View Full Version : Let's talki about phasing


dpinvidic
05-04-2012, 10:06 PM
Since this was asked on another thread, I thought I would start a new thread on this topic.

I understand that phasing is the relationship of the swash tilt to the direction that the heli goes. I "Think" the high point of the swash should be exactly at 90/270 deg for right/Left aileron.

Now it would seem to me that since most of us have 3 cyclic servos, the phasing is really determined by the Transmiter, and how it mixes the servos. If I look at my swash while giving full right aileron, I can see where the high point would move around
the main shaft depending if I made the ail-servo move just a little bit more/less.

Does this seem reasnable ...to those

terrabit
05-04-2012, 10:20 PM
Actually, phasIng refers to the position of the leading edge of the blade relative to the swash. The blades can trail or lead the swash which affects the way the blades bite as pitch is introduced. Here's a NASA (http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19760025036_1976025036.pdf) study on the affects of phasing on performance.

It's the relationship between the swashballs and the axis of the feathering shaft.

extrapilot
05-04-2012, 11:30 PM
This thread belongs in the Aero/Engineering forum.

Phasing is not lead/lag- it is the relationship between the tip path plane and the swash. You can set lead/lag at zero by removing the hinge, and you still have phase offset.

It is not 90/270, that is a popular myth- where people assume the lag is a result of precession (the rotor as a gyro). It is almost always less than 90, regardless of how your head is rigged, and changes with blade Lock number, and airspeed. There is a reason many heads have phase angle adjustment, and all FBLs do…

Phase is not the result of your TX- it is the result of blade flap resonance. Normally, the head is rigged mechanically to compensate for it. There is no reason the swash could not be mixed differently, such that there was no mechanical compensation, but rather, electrical, but that sort of complexificates some basic preflight (i.e. cyclic forward would yield left or right tilt on the swash, and that would be reversed depending on the direction of main rotor rotation)… So, probably for the sake of consistency, the rigging handles most of the compensation.

terrabit
05-04-2012, 11:55 PM
Okay, so the plane of the disc is not exactly parallel to the plane of the swash? And this is a result of blade flap resonance. Umm, what's blade flap resonance? I'm guessing that the two planes can be tuned by modifying the position of the feathering shaft relative to the swashballs?

laminarjack
05-05-2012, 12:02 AM
Less talki, more phasi...

extrapilot
05-05-2012, 12:15 AM
Okay, so the plane of the disc is not exactly parallel to the plane of the swash? And this is a result of blade flap resonance. Umm, what's blade flap resonance? I'm guessing that the two planes can be tuned by modifying the position of the feathering shaft relative to the swashballs?

Well, just to be fair, there are different ways people describe phasing. The official term is Gamma, which is the relationship between the swashplate tilt angle and the rotor tip path angle. Some people might define it as the relationship between the cyclic stick input and the peak rotor displacement- which is not correct, but is a very practical concern…

When you apply cyclic, the blades resonate in flap. That is how you get so much displacement with so little energy input. And like a pendulum, the rate at which they flap is tied to gravity. As RRPM increases, so does gravity- and so the flap frequency relationship remains tied to RPM. But, the phase angle/error is a function of a variety of factors, including properties of the air, so phase angle changes slightly in flight (can be 5deg or so for conventional machines), and with different blades.

terrabit
05-05-2012, 03:22 AM
Gama, interesting. Thanks for the explanation. That makes more sense to me than what I've read so far. I am not an engineer, so some of the language is outside of my vocabulary. What I've understood is that phasing error results in nasty things like nonaxial flips and rolls which can be adjusted via the swash follower or in software, like vbar for example. Would you care to take a stab at delta? I understand that data adjustments can be used to make a heli more responsive or more stable.

extrapilot
05-05-2012, 04:06 AM
Yea, phasing is seen as a coupling between pitch and roll in this context. So if you pull back on the cyclic (i.e. a back flip), and you have 5deg phase error, your machine will roll a bit in the process.

For FBL-based systems, the FBL will see that, and add opposite roll automatically, but that also has phase error (now in pitch), the correction for which is in roll, etc. That coupling decays fast- since the error terms are pretty small. But, the net result typically is a wobble, and often that leads to tail instability. FBLs seem remarkably tolerant of this sort of error- I have seen tailcases come a bit loose in flight and cant 30-40deg, and the machine is still (marginally) flyable…

Delta is just cross coupling between blade pitch and some other blade motion (lag, flap, etc). So, for example, with Delta3, the control system is setup such that a blade moving up or down affects its feathering pitch. The idea there is that if you have turbulence, and the blade grabs a bunch more lift than you desired, it subtracts some pitch automagically to offload some of that lift. There are a few mechanical approaches to this- high level, the flap hinge is not aligned with the pitch horn. Makes for a more stable rotor, which is good for passengers, bad for acro.

This all gets really complicated really fast, because everything is intertwined. For example, Delta3 will impact phase angle. And, blades have some inherent stability from the flap itself (i.e. as the blade flaps up, its AOA drops, and vice versa). Generally, Delta3 is seen on full scale TRs, just because it the only method they have to relieve dissymmetry of lift. We don’t generally see it on our models on either rotor; since we don’t have a flap hinge or cyclic on the tail, it is very difficult to implement there.

xspare
05-05-2012, 09:25 AM
... As RRPM increases, so does gravity- and so the flap frequency relationship remains tied to RPM...

Huh? Could you explain this gravity control thing a bit?

terrabit
05-05-2012, 10:26 AM
If I may, I believe that refers to COG and centrepetal force as rotor rpm changes.

extrapilot
05-05-2012, 05:32 PM
Hey guys

No, ‘gravity’ in this context means, the acceleration felt by the blades due to RPM.

Here is one way to think about it-> a kid in a swing. The kid is able to achieve very large displacement, given some time, by inputting a small force at the right position in the cycle. The period of the swing is a function of its length and the local force of gravity, not the mass of the kid. That period is its resonance.

The peak speed of the kid is at the bottom of the arc, and the peak amplitude is at the top of the arc- 90deg out of phase.

If you take this same kid, and put him and the swing on the moon, the period of the swing drops, as gravity is reduced there. Conversely, on Jupiter, it is much higher… The ‘restoring’ force of gravity changes its period.

Now, take that model, and tilt it 90deg back. The swing is the rotor blade, and the swinging motion is the blade flapping up and down. As the rotor RPM increases, so does the gravity force the blade experiences. Its resonance is, like the swing, exactly tied to this force, since its length doesn’t change. So, if you double the RPM, you also exactly double the period of the blade flap.

With that in mind- that the blade flap resonance is just like a swing, and it is synchronized exactly to the RPM of the rotor, it is easier to see how adding a little force at the right place in the cycle can lead to high flap amplitude, if that force is added at the right time (phase) in the cycle, each cycle.

Fredo0709
05-06-2012, 12:29 AM
If I may add, extrapilot is correct when the phase lag is almost always less than 90 deg because most rotorhead designs have flapping hinges that are offset from the center of rotation. This hinge offset changes the phase lag by reducing it from 90 deg because a hinge offset results in increasing the resonant frequency of flapping. If we take a look at extrapilot's analogy of the kid on the swing, the period is proportional to length over gravity or L/g. If we look at the blades flapping about the hinge offset, the effective "length" of the blades is reduced without the diameter changing. This reduction in length leads to a shorter period. As a result frequency, which is inversely proportional to period, increases.

npomeroy
05-06-2012, 01:46 AM
I neeed to take more time to digest the explanations given thus far.
However, given that the setup geometry is normally recommended to provide what I would call a 90 degree offset (i.e. zero pitch change in response to elevator command with blades parallel to boom), is there any generalisation that can be made for when the phase angle should be more or less than 90 deg?

extrapilot
05-06-2012, 04:06 AM
The flap hinge offset is only one of several core factors that define the phase lag. Another is Delta3, another is the blade’s Lock number, and another is the fluid. For a given rotor on a given machine, the phase angle changes in flight, simply because of variations in the airspeed/altitude.

It is often assumed that teetering rotors always have a 90deg phase lag. They don’t.

Nelson- there is no generalization, because it is specific to your machine. It is possible to measure this on a test stand, but it probably makes more sense to just back down the phase angle from 90deg while practicing very clean front or back flips (i.e. no crosswind, no cyclic roll or rudder input), until you see what seems the most stable tail/rotor. Usually, the value will be between 75 and 85, typically more towards the higher end of that range, but again, that is entirely a function of lots of complex things.

terrabit
05-06-2012, 09:42 AM
This has been very interesting. Thank you for taking the time to explain this stuff. By the way, why do you know this stuff? Seems like more than typical hobby level knowledge.

Fredo0709
05-06-2012, 09:27 PM
Yes, the hinge offset is not the only parameter that affects phasing. However, it is an important one.

Here are some definitions for those who aren't completely familiar with the terms. A quick look up in a helicopter theory book would cover these terms. I use Helicopter Theory by Wayne Johnson. Dover Publishing 1980

The lock number is basically the ratio of aerodynamic forces compared to the inertial forces acting on the blade. When looking at equations of motion for flapping, it is in the damping term. While we have variations in airspeed and altitude which can affect the air density (greek letter ro) term in the lock number definition, I argue that for RC helicopter purposes, we won't see ro changing too much. What we do see is that lock number changes with changing lift slope.

A lift slope is the variation of the lifting coefficient (C_l) used to measure amount of lift generated for an airfoil while varying the angle of attack. For our RC blades, our symmetric airfoils are very close to NACA 0012 shaped which has a lift slope that ranges from negative numbers to positive numbers of C_l. This of course means that it is possible to have negative lock numbers.

Delta3 is an angle that can be measured by drawing a line from the flapping hinge to the end of the pitch horn and measuring that line with another line that is orthogonal to the flapping motion. Positive delta3 indicates a stable relationship between flapping displacement and blade pitch motion. As the blade flaps up, it tends to pitch downwards thus acting as an aerodynamic spring to restore the blade back towards zero flapping angle (greek letter beta). Please note that the factory heads that come with our helicopters are rigid and do not contain a flap hinge. There is a however, some flapping motion allowed because of the damper in the hub arm. As a result, there will be some delta3 effect.

So what does this all mean? It means that everything that has been discussed here affects the phasing or rigging of a swashplate. Some such as hinge offset and delta3 affect the static phase offsets while others like the effect of lock number change in flight. As was said before, it's too complicated to account for all the factors that affect the phasing... especially the ones that vary with flight regimes or changing flight dynamics. Thus, it's best to set the phasing to a value below 90 deg that is a good compromise.

Please feel free to correct me if I didn't get something right. I am doing this from memory.

If I had to guess though, a good phasing compromise should be between 80 and 85 degrees for our RC helicopters.

A lot of this knowledge can be self taught with a good textbook. It helps if your work is related to this. =)

Austin

terrabit
05-07-2012, 12:21 AM
I met an aeronautical engineer today while flying my 700E who is in the process of designing a multiprop heavy lift craft for industrial and possibly defense application. We discussed some of these very terms at length. His description of blade-flap resonance was sufficient to make me understand it finally (that wingtip vorteces and other light topics). He's interested in the hobby so I directed him to this site. I hope he finds it so he can participate. Decent guy.

ehernan3
07-23-2012, 11:23 AM
The flap hinge offset is only one of several core factors that define the phase lag. Another is Delta3, another is the blade’s Lock number, and another is the fluid. For a given rotor on a given machine, the phase angle changes in flight, simply because of variations in the airspeed/altitude.

It is often assumed that teetering rotors always have a 90deg phase lag. They don’t.

Nelson- there is no generalization, because it is specific to your machine. It is possible to measure this on a test stand, but it probably makes more sense to just back down the phase angle from 90deg while practicing very clean front or back flips (i.e. no crosswind, no cyclic roll or rudder input), until you see what seems the most stable tail/rotor. Usually, the value will be between 75 and 85, typically more towards the higher end of that range, but again, that is entirely a function of lots of complex things.

Just ran into this thread, and I find it fascinating. I'd like to know how you adjust this phase angle. If it matters, I fly a TREX450 with an RJX head and BeastX using a Futaba 7C.

npomeroy
07-23-2012, 02:51 PM
Just ran into this thread, and I find it fascinating. I'd like to know how you adjust this phase angle. If it matters, I fly a TREX450 with an RJX head and BeastX using a Futaba 7C.

The phase angle is altered by loosening the swash-driver (that sits on the shaft above the swash, with 2 arms) and rotating the swash relative to the head. If you are already flying this you must have already set it up, presumably to the recommended 90 deg. Do a search on the "scale" forum for lots of threads on phase angle.

extrapilot
07-23-2012, 04:23 PM
Last I checked, RJX has the swash driver arms attached to the head, so there is no provision to bias the swash. You can adjust phasing electronically via the BeastX GUI; requires the dongle interface, but you can borrow one at a local field, etc, as it is not bound to a particular unit.

ehernan3
07-23-2012, 10:53 PM
Last I checked, RJX has the swash driver arms attached to the head, so there is no provision to bias the swash. You can adjust phasing electronically via the BeastX GUI; requires the dongle interface, but you can borrow one at a local field, etc, as it is not bound to a particular unit.

Thanks, you are correct: the RJX head does not have a mechanical adjustment for this. I do have what I need to do this on the PC for the BeastX, though, so I will explore it.

Azhar
07-26-2012, 04:14 AM
Good Day

I am working on a project which involves developing the mixing algorithm for a single rotor (5 bladed) helicopter. I dont have much experience with rotoray wing aircraft and have spent the last few days getting aquainted with the various control characteristics pertinent to such aircraft.

In particular I need to obtain the relationships between control inputs (i.e. collective & cyclic) and the actuator displacements which are attached to the lower non-rotating swash plate such that the rotor blades are pitched by the correct amount. This is necessary to fully understand the process. Additionally, the second objective is to develope a technique for reducing cross-coupling between inputs. Essentially, the goal is to determine the mathematical relation/s which would allow me to develope the algorithm and simulate the system using Matlabs Simulink addon.

Could anyone kindly assist by referring me to appropriate references or point me in a the correct direction to tackle this problem.

Many Thanks in advance