Dutch roll

[blackhand]

Could someone explain, in words of two syllables or less, the effect of dutch roll in a helicopter and how it is offset.


Cheers

Blackhand

[the delaminator]

never heard of it in helos.

In a fixed wing, dutch roll refers to the tendency of light twins at low speeds ( ie take off) to roll over in the direction of the failed engine if the failed prop is not feathered promptly.

[Mad (Flt) Scientist]

never heard of it in helos.

In a fixed wing, dutch roll refers to the tendency of light twins at low speeds ( ie take off) to roll over in the direction of the failed engine if the failed prop is not feathered promptly.

I also can't help wrt helos and "Ducth Roll" but that above most certainly is not fixed-wing a/c Dutch Roll.

Dutch Roll is a roll-yaw coupled oscillation inherent in any conventionally configured fixed wing aircraft, typically at frequencies in the same range as piloting tasks, and therefore a potential handling issue. Often addressed on larger aircraft with dedicated damping systems which provide inputs to the rudder.

[blackhand]

Dutch roll is an oscillation involving mostly roll and sideslip.Dutch roll can be prompted by flying through a gust or by a twitch of the rudder pedal. Following the input the oscillation will occur over a two to four second period no matter how large or small the aircraft.


Blackhand

[John Eacott]

blackhand,

I'll try a simple explanation: imagine a 'corkscrew' effect, where the aircraft is rolling and yawing at the same time. The BK117 is known for Dutch roll, due to the short moment arm of the tailrotor, and the easiest solution is to put on "opposite rudder" (yaw) to create out of balance flight. When the dutch roll damps out and becomes yaw only (almost instantaneous), feed in corrective pedal to get back into balanced flight.

BK117 is not liked by pax in the back rows, for this reason :yuk:

[Brian Abraham]

Helicopter Dutch Roll is addressed here

Helicopter flight dynamics: the ... - Google Books (http://books.google.com.au/books?id=ewVI57iyLRwC&pg=PT379&lpg=PT379&dq=definition+helicopter+dutch+roll&source=bl&ots=IWf8TOgQJE&sig=ZlHbkJN199YpdAmOtWXtKVEO5Tg&hl=en&ei=jwVDSonBOYOPkAX_ycSgDw&sa=X&oi=book_result&ct=result&resnum=2)

[Shawn Coyle]

Dutch roll is yaw coupling into roll, causing a motion that is a combination of roll and yaw, and is found in forward flight.
The name supposedly comes because is seems like the motion of a speed skater, and at the time it was coined, most of the speed skaters were Dutch (at least that's the reason that appears most often)
A sideslip typically starts the motion, due to a gust of wind upsetting the torque balance between the main rotor and fuselage. The helicopter yaws, and this produces a change in the sideslip. The sideslip produces a side wind on the fuselage and main rotor, which generates some rolling. Most of the time, the motion damps down immediately, and all you would see or feel is a slight yaw and perhaps a small change in roll attitude.
In swept wing airplanes, the motion can continue, and for this reason yaw dampers are fitted to stop any yawing motion.
Some helicopters are more affected by it than others, and changes in CG position (both longitudinal and vertical CG) can change the effect of a small change in sideslip.
It's not very prevalent in helicopters - I just had the pleasure of sitting in the back of a Bell 407 for several hours yesterday and the mild turbulence caused noticeable yawing motion, but almost not perceivable roll.
And any helicopter with an attitude hold in roll will also mask the problem.

[blackhand]

The name supposedly comes because is seems like the motion of a speed skater,


Mmmmm it is the motion of two skaters arm in arm moving forward.

John Eacott

I was sitting on top of a BK117 looking rearward at the offset of the vertical stabs. When questioned the pilot(high time) said it(the offset) assisted to prevent dutch roll. This sent me on a search about dutch roll and spiral dive in helicopters.
The explanations I have found about the phenomenom in helicopters is very technical, involve angle of attack of MR, forward airspeed and relationship with TR torque.
The fact that a MR has dihedral was enlightening to say the least.

Thanks for the Readers Digest version.

Blackhand

[John Eacott]

blackhand,

The tail boom on the 117 is the same part as the Bo105 tail boom, which makes it a tad short for a machine with such a substantial increase in weight and power! The increased size of the vertical stabs are to assist in overcoming this, plus the sizeable offset of the stabs helps further.

IIRC, the main rotor head titanium casting is also the same part as the 105: it must have been well engineered for the Bo105 in the first place :ok:

[212man]

The fact that a MR has dihedral was enlightening to say the least

Well, it's more like 'dihedral effect' but the outcome is the same, just like the roll-yaw instability is not following the same mechanism as a fixed wing but generates the same 'dutch roll' effect.

[DennisK]

We Schweizer (Sikorsky) 300 instructors at the lower end of the pile will know of the problem especially applicable to the type, which in a developed state is quite alarming for the new pilot.

In my experience, the worst case situation occurs in 'air-taxying' manoeuvres with a stiff 2 o clock breeze. The rapidly alternating yaw angles frequently reach 20 degrees either side of the normal axis which, accompanied by the lateral rolling and wrong corrective handling, risks loss of control. Newer pilots become quite frustrated in their attempts to control the condition and many times, I've seen the late application and out of sequence corrective yaw pedal worsen the situation. I don't think I've actually discussed the problem with another instructor and would like to hear their experiences.

In post lesson briefs, I use Shawn Coyle's dynamics to explain this odd quirk of handling but as noted the deliberate 'out of balance' sideslip using yaw pedal is always effective.

Safe flying to all,

Dennis K

[GANNET FAN]

Slight thread creep, but the phrase Dutch Roll is common in sailing parlance. Usually taking place on a downwind course, spinnaker set with the (strong) wind pretty much dead aft and possibly helmsman aided, a yacht can commence rolling one way, dunking the end of the boom in the water and then reverse the roll so that (sometimes) the boom can be pointing upwards 45 degrees. And if not controlled can end in a very spectacular massive gybe with the mast near horizontal. Very uncomfortable and very hairy!
I have no idea where the Dutch implication came from
GF

[blackhand]

Shawn Coyle

Does the rotorcraft right itself or does the correction have to be made by the pilot??


Blackhand

[firebird_uk]

Having trained in the Schweizer I can confirm that, at hover taxy speeds, the dutch roll this machine can experience can be quite severe. Back in my student days I almost called the tower to say I was unable to control the aircraft!

Part of the problem is that students are told to always taxy with their skids in the direction of travel (obviously sensible for engine failures, inadvertent ground contact etc). A long hovertaxy along a runway with the wind from the wrong direction can be an unpleasant experience as the dutch roll builds and the student chases the controls to no avail.

Put the machine out of balance thus changing the angle the wind hits the aircraft and the dutch roll disappears almost instantly.

As I got more proficient I'd sometimes set it up so I could see if I could control it. I never could, but then again I couldn't hover within 5 minutes. :E

[Nigel Osborn]

I've only ever had Dutch roll when flying a Bolkow 105D on the North Sea. It came in on fairly short finals to an oil rig when the ias was low & was quite frightening on a black night until I got used to it with more experience. However when North Scottish bought 2 new 105D, probably less than 25 hours old, I never got Dutch roll. I ran this past the engineers & they reckoned the roll was caused by a slight wear in the linkages but still well within limits.
This roll is not the same as the Sikorski shudder in the older machines, so I'm told.:confused:

[Sir Niall Dementia]

Firebird and DennisK;

I well remember the S300c dutch roll, I scared myself to death with it! The EC135 can be put into dutch roll if you mis-handle it enough.

[ramen noodles]

Dutch Roll is an old sailing term that described how Dutch warships moved. Because of the shallow draft needed for their home waters, the ships had small shallow keels, and thus little roll and yaw stability. They would roll in one way and yaw in the opposite way, then reverse the motion.

Similarly, an aircraft with too little yaw stability and too little roll stability can enter a coupled motion where a yaw movement causes a roll in the opposite direction. If the yaw oscillates, than the roll does as well.

For helicopters, the reason is usually a high tail rotor and weak roll stability. If the aircraft hits a small gust, the high tail rotor causes the restoring yaw moment to be coupled with a strong opposite roll moment (the high tail rotor makes the yaw come back home, but rolls the aircraft in the opposite way.)

As several posters have said, the best way to handle a dutch roll is to apply small pedel motions to stop the yaw. This keeps the roll in check as well. If a yaw SAS is available, it will magically make the roll behave.

[Simonta]

As a low time PPL student, on perhaps my 3rd or 4th solo, I got into exactly the situation Dennis describes in an R22 so I guess it might be more common in helis than may be thought, if indeed it was dutch roll I experienced.

It scared the cr*p out of me. I'd just turned and moved forward from the heli playground at Thurxton to face the runway and call for crossing when I got into this state. My response was instinctive and wrong - I hauled in power to get away from the ground and stuffed the nose down to add speed. Ended up crossing the middle of the runway at about 50 feet and 40 kts rolling and waying madly, zooming over Fast and pretty much doing something like a descending quick stop to get back into some sort of shape, with of course no call to ATC. How I didn't stuff it up, I don't know. Happily, no-one noticed, or at least, no-one said anything :bored:

ILAFFT! Quartering winds always get my attention.

[Graviman]

I'd like some (good or bad) feedback on whether i have understood this properly. Apologies in advance of the slightly technical nature of the post. I figure some of the TPs out there will have their own thoughts on this. :ok:


I'm not convinced helicopters demonstrate dutch roll:

Fixed wing dutch roll is an undamped oscillation resulting from the lateral dihedral trying to keep the aircraft "wings level" after a disturbance. The roll-yaw coupling comes from the vertical stabiliser trying to keep the heading correct while the aircraft is sideslipping port-starboard-port. This means yaw direction should lag roll attitude by about 90 degrees, with a larger rudder area reducing tendancy to dutch roll (until you get to spiral divergence).

The problem in a helicopter is that there is a delay before the sideslip induced lateral flapback causes the fuselage to change roll attitude. This delay is caused by the relatively low effective hinge offset (compared to a fixed wing) requiring the rotor to change attitude before generating the necessary moment above the centre of mass. So the lateral dihedral has a response lag which swamps any dutch roll time constant.


The practical upshot of all this is that the instability is more likely to be PIO from the cyclic roll response lag than instability in the aircraft. So in a quatering wind, any delay in pedal position results in a cyclic input. Since the cyclic input was not really required then when the pedals are moved the cyclic needs to be re-corrected. All this means that cyclic effectively sees a disturbance, so the helicopter starts to oscillate.

Any thoughts?

[212man]

So in a quatering wind, any delay in pedal position results in a cyclic input. Since the cyclic input was not really required then when the pedals are moved the cyclic needs to be re-corrected. All this means that cyclic effectively sees a disturbance, so the helicopter starts to oscillate.

Any thoughts?

My thoughts are that all the talk of quartering tailwinds, and student pilots losing control while hover taxying, has nothing to do with the subject.

A classic demonstration is a high power low IAS climb - say around Vy - with a pulsed collective input, or rudder doublet, and then observe the behaviour with the controls fixed. A divergent Dutch-Roll-like motion will ensue if no SAS enagaged. As I said in a previous post, the mechanism is different to FW - there's no equivelant to swept wings for instance, which is where the problem really appears generally (hence yaw dampers) - but the motion appears the same.

[Graviman]

212man, good feedback...

What (roughly) is the time period of the oscillation? Is it similar in time constant to hover instability? This is mostly in roll?

It sounds to me like a more extreme version of the hover dynamic instability from the delay in roll response to a lateral cyclic input. As the blades are operating at higher pitch than hover then the rotor (optimised for level flight) produces more lateral flapback for any given sideslip. So the dynamics of the whole rotor-fuselage interaction change to the point where stability augmentation is required.

Out of interest do you find that the longitudinal trim position for Vy changes with climb when compared with level flight? If what i mentioned is correct then i would expect cyclic trim position to be more rearward in climb.

[Hughes500]

DennisK
Yup Hughes300's will demonstrate it very well, never really sure what causes it, Iam convinced it is the wind and main rotor vorticee hitting the horizontal stab ( not horizontal so causes the problem for the tail) and then going into the tail rotor that causes the problem. Teach my students to stick the nose forward and pull power and it goes away.:ok:

[Shawn Coyle]

What's happening with the Schweizer and others in the low-airspeed environment wouldn't qualify in the classical sense as Dutch Roll - it's a weird oscillation where the response in roll gets out of phase with the pilot inputs.
Just wondering if this might be the cause of several otherwise unexplained accidents in the hover / low airspeed environment...

[tistisnot]

Sorry, but you all have it wrong as there is only one Dutch roll - Ham en Kaas, and some foreigners were always spluttering in their best attempts at Cloggie demanding Geen Kaas! But give 'em their due - 20 years ago they had DDAFCS, Pratt & Whitneys, FMS and floppy downloads of flight logs ...... munch on that.

[blackhand]

Thanks everyone for your thoughts on dutch roll in rotorcraft.

I read recently that the "new" AW139 has a pronounced yaw/roll couple with autopilot turned off. Must be inherent.


Cheers
Blackhand

[carholme]

I am surprized that no one has referenced the installation of the ventral fin on the Super Puma (all variants) as compared to the standard Puma without the fin.

This was to counteract the D/R characteristics which existed on the standard Puma and could certainly be felt in cruise (fishtailing) when occupying the most rearward seats.

carholme

[Shawn Coyle]

The ventral fin on the Super Puma was added in part to counteract the problem found at aft cg. If you rolled in to a turn and the ball was not centered (i.e. it was to the outside of the turn) in the original Puma, adding pedal would roll the helicopter quite quickly. Lateral cyclic would not stop the roll...
Aerospatiale called this 'le virage Anglais' - (the English turn) as it first happened to a Royal Air Force Puma which was fitted with the very large particle separators well in front of the CG.

[carholme]

Mr. Coyle;

With all due respect, your promotion may have been a result but it was not the primary reason.

carholme

[[email protected]]

The issue with the Puma has always been described as a 'yaw/roll divergence' rather than a dutch roll since it is very unstable.

I think the Puma fishtailing is just a case of yaw instability and a desire to weathercock because of the extra area forward of the C of G; this is why the large ventral fin stops it but it doesn't appear to match the classic dutch roll motion.

[carholme]

Well after spending a considerable amount of time as a tech rep for Aerospatiale on the 332, I won't let my experience get in the way.

Have at er boys.

carholme

[[email protected]]

Carholme - there's no need to be like that, if you have more experience on a type than others here then say so, we are not mind-readers and most of us are always looking to learn and improve our helicopter knowledge.:ok:

[Shawn Coyle]

carlhome:
I was merely reporting what other UK test pilots who encountered the phenomena had told me.

[carholme]

crab;

What do you mean, "don't be like that"? All I did was offer my knowledge of the 330/332 with reference to D/R and you and Coyle subtly countered that to imply other reasons for the ventral fin.

If you feel those are the reasons for the fin, as I said, "have at er". I am not here to impose my experience or knowledge or impress anybody.

I am sure there are many pilots and engineers who have studied the autopilots in both aircraft and are familiar with their levels of capability and why the ventral fin was introduced on the 332.

carholme

[[email protected]]

Carholme - what I meant was that you made a statement about the ventral fin which to you was the absolute answer but, to people reading the post, was just the opinion of another guy on pprune who may or may not know what he is talking about.

If you want everyone to know you have extra inside knowledge of the Puma/332 when you make a post then mention it at the time.

As a tech rep then are you a pilot/test pilot?

As Shawn says, this phenomenon has been widely experienced by many pilots and test pilots and the Aerospatiale answer for the Puma was, from what I have been told, yaw/roll divergence aggravated by the particle separators, fixed on later models with the large ventral fin.

[carholme]

crab;

You are referencing helicopters equipped with large particle separators installed however the 330 had this problem long before the separators were installed as experienced with Schreiner and Sealand 330s, both without separators.

I am a licensed engineer with UK and Canadian licenses and do not feel for one minute that my comments require one to be a test pilot for confirmation.

carholme

[[email protected]]

Carholme - so did the 330 receive the ventral fin modification before the particle separators were fitted or did the separators aggravate the existing yaw instability enough that something had to be done about it?

Ref the test pilot point - I was just trying to ascertain if you had any practical flying experience on which to base your comments on handling qualities - this is a professional pilots rumour network after all:ok:

[carholme]

crab;


In the first place, the 330 in service did not have the ventral fin installed. It was installed on the AS332C and subsequent variants.

As for your other comment, I will leave that up to the mods to decide if engineers are allowed to post here.

carholme





Let's not go there again: please :=

Splot