Fish-tailing

[GrumpyGnome]

Any amateur or professional aerodynamicists out there ?

I have a HK Spirit. It's nice but seems to fishtail almost all the time - not hugely but there is a noticaeble left to right to left movement of the nose.

Anyone know the cause or cure ?

GG

[Frank Skilbeck]

Often referred to as a Dutch Roll a larger fin might cure it.

[Peter Christy]

Not familiar with that model, but its something that a lot of models suffer from! Phil Kraft's Kwik-Flis in the late 60s / early 70s were notorious for it, and they were World Championship level F3A machines! It was known as the "Kwik-Fli Dance"!

I know one pilot of my acquaintance who cured his Kwik-Fli by replacing the "flat plate" horizontal tail with a thin symmetrical one! Most of the aerobatic models I flew in that era had symmetrical section tails anyway (mostly Tauruses) and never suffered from the issue.

I have absolutely no idea why changing the section of the horizontal stab would affect an oscillation in *yaw* - perhaps something to do with the different drag pattern? But it did the trick on the Kwik-Fli!

--

Pete

 

Edited By Peter Christy on 14/11/2013 15:00:58

[John Lee]

Posted by Peter Christy on 14/11/2013 15:00:35:

Not familiar with that model, but its something that a lot of models suffer from! Phil Kraft's Kwik-Flis in the late 60s / early 70s were notorious for it, and they were World Championship level F3A machines! It was known as the "Kwik-Fli Dance"!

I well remember it Pete, the other cure at the time was to fit a fairing between the wing & fuselage.

Looking at the HK Spirit I strongly suspect that it's inherent with the short moment arm coupled with a large fin/rudder. The set up will have a small moment of inertia to help dampen the yaw/roll couple.

[Frank Skilbeck]

Apparently the Mustang had a similar problem when they changed to the bubble canopy version due to reduced fuselage side area behind the main wing and a fairing was fitted to the fin to cure this.

[Ian Jones]

My TN Vampire does it - a touch of throttle seems to "pull it" back in line. Full size vampires did it too and I suspect that's why later versions had the fins with a fairing sweeping forward onto the booms.

[Biggles' Elder Brother - Moderator]

Its usually caused by an interaction between the yaw inducing effect of the wing dihedral and the natural tendancy of the fin trying to hold the aircraft straight.

We all know that if we have a model with dihedral then the rudder will induce roll (this is how single channel models fly); well it works in reverse as well! If a model with dihedral rolls slightly it will naturally tend to yaw. The fin then acts like rudder applied in the opposite direction to the induced yaw and so tries to yaw the aircraft back straight. In doing so it can overcorrect and then, via the dihedral ,induces a roll in the opposite direction! This induces a yaw, which produces a roll and so the merry dance goes on and on.

Why is this more of problem with some models than others? Well if one of the mechanisms dominates then that stops the process. Its perpetuates when the two effects (dihedral induced roll to yaw and fin induced yaw to roll) are more or less balanced and, most importantly, slight out of step with each other - ie there is a lag in the effect.

Aircraft with swept back wings are particularly prone to Dutch Roll - which is why many of them have wings with anhedral rather than dihedral in order to weaken the wings' ability to induce yaw. Also many modern aircraft are fitted with a rudder damping system to stop Dutch Rolling - called yaw damping. These work by effectively slowing the fins impact in a small yaw by putting in opposite rudder.

How to fix it in a model? The usual trick is to make one of the effects dominant - prefereably the fin. You can do this by; lessening the amount of dihedral, lengthing the tail moment or making the fin bigger. Unfortunatly none of those are particularly easy to do to a finished model!

I've never seen this done - but it might be interesting to see if a gyro on the rudder would stop it? The problem is the gyro's gain would have to be set so that its reaction time was 3-5 times faster than the natural period of the Dutch roll - I have no experience of these gyros so I don't know if they could do that. But if they could they would basically act like a yaw damper on a full-size aircraft. Might be worth a shot?

BEB

Edited By Biggles' Elder Brother - Moderator on 14/11/2013 18:10:10

[Ultymate]

I had a 100cc Extra (H9) that fishtailed badly, it was easily cured by fitting a turbulator on the rear end of the rudder. I just cut a length of 1/4" internal diameter plastic tubing to length then split it lengthways opened the split and slid it onto the rudder and held it in place with 3 or 4 cable ties through the rudder and round said tube. Not very pretty but cure the issue and also served to protect the sharp edge of the rudder from hangar rash

[Bob Cotsford]

My small constant chord zero dihedral flying wing does it, adding extra fin area in the form of a rudder reduced the fishtailing but it still literally 'gets a wiggle on' at full chat.

[Biggles' Elder Brother - Moderator]

Ah, bringing flying wings in is opening a whole new box of tricks where this is concerned!

BEB

[Erfolg]

One of the mysteries to me, has been the reluctance of both modellers and full scale aircraft to have rudder/fin areas which are adequate.

In my distant past, I have visions of reading articles which warned modellers to keep the fin area small or "spiral instability" could be the penalty.

As has been written the number of full size aircraft where the fin/rudder area being initially inadequate has been legend, as has been registered on this thread. Another one to add is the prototype was the Flying Fortress, that needed and benefited from a substantial increase in fin/rudder area. Yet the list is almost endless.

The numbers are so significant that exhibited the need to larger areas has been so great, you have to ask the question of why? What is the downside of a larger area. Why minimisation seems so important, that the issue of dutch rolling and fishtailing are more important, than so called spiral instability.

As one who built a Tomboy, that was so unsuited to real RC operation, due to Dutch Rolling, as a consequence of to a great dihedral and smallish rudder and fin combination. Which was only solved by getting rid of the dihedral. Although i did initially try masking tape to attach a temporary strake to achieve a greater fin area. Which did improve things, The elimination of dihedral was the real solution, and the easier option,

From observing some FF models going in, the real issue seemed to be the marginal stability around all axis that is more of an issue, than selecting the fin/rudder for special attention to minimise the area. I guess that the marginal stability is an attempt to improve duration?

But why have full size designers been so reluctant in the past to provide adequate fin/rudder area? Just think of those early WW1 aircraft, Sopwiths and Fokkers, tiny rudders, often no fin.

[Stephen Jones]

This Model also Fish tails at high speed or when there is a side wind hitting it so why ?

[Simon Chaddock]

The only comment I can add is that I have planes with folding props that fishtail under power but glide arrow straight when the prop is folded.

Rotating props add effective side area so a nose mounted prop tends to reduce directional stability where as a pusher tends to add to it.

[GrumpyGnome]

Gosh - that's a lot of options !

Think I'll take a handful of batteries out on the weekend, plus some double sided tape and bits n bobs to try increasing the fin or rudder, and adding a turbulator.

Will also check if it's better/worse with power on; and look to see if the tailplane is a flat plate (I think it's pretty much symmetrical but will check).

Thanks fellas!

GG

Edited By GrumpyGnome on 15/11/2013 08:14:35

[YakMad]

I have two large Extras 300's which have fairly large fins and rudders, both fish tail badly, my solution was to fit cheap gyros on the rudder servos with the gyro rate controlled by a switch, I set zero percent rate on position 1 and 6 to 10 percent on position 2 which allows me to turn the gyro off in case it induces yaw instead of reducing it because the gain is set too high. Once set-up correctly not only does it stop the "fish tailing" but both models fly much smoother which I've proved to other club members by switching the gyro off (0% gain). The gyro has a secondary advantage as the tail will swing less on take-off and landing, particulary in cross wind as the gyro is applying rudder faster than you can react.

[Bob Cotsford]

Posted by Erfolg on 14/11/2013 22:29:22:

In my distant past, I have visions of reading articles which warned modellers to keep the fin area small or "spiral instability" could be the penalty.

Wasn't this the case that with the massively oversize fins used on some early designs when the model reached a certain angle of bank it staterd to sideslip. At this point the rudder is effectively acting as in-turn rudder tightening the turn, increasing the angle of bank, increasing the sideslip, increasing the in-turn rudder effect ad infinitum or ad terra.

[Ultymate]

A pictures worth a thousand words, here's the turbulator I fitted to my Extra, I did eventually fit one the full length of the rudder's trailing edge, it worked a treat. Gyros are banned in competitive IMAC flying and I suspect may well be in F3A

[Bob Cotsford]

Ulty, does leaving the rudder TE thick and square edged (eg 1/4" thick) have a similar effect?

[Ultymate]

Posted by Bob Cotsford on 15/11/2013 11:00:58:

Ulty, does leaving the rudder TE thick and square edged (eg 1/4" thick) have a similar effect?

Yes absolutely, some of the F3A guys are flying with extremely thick T/E sections actually flared outwards if that makes sense

Edited By Ultymate on 15/11/2013 11:57:53

[Erfolg]

The phenomena suggested seems to point towards Von Karman Vortex Shedding.

If a blunt trailing edge cures the problem great. It never entered my head that the issue could possibly affect models. Seems you live and learn.

[Stevo]

Indeed, loads of options. I had this on an old Wild Card, eventually changed the rudder servo for one much quicker and tightened up the linkages, - it cured it immediately!

[Simon Chaddock]

I would have thought Karmen vortex shedding would tend to make the fish tailing worse.

My own guess is it has more to do with making the rudder aerodynamically stable so it does not oscillate in the straight ahead position against the 'give' in the linkage. Mass balancing should have the same effect.

On the other hand my Super Endurance tends to fish tail and that only has a fixed fin with no rudder so the complete airframe aerodynamics must also play a part.

[Erfolg]

Simon

Vortex shedding happens from memory at all speeds, although again from memory, there is for any structure, a point where a critical frequency is reached, where the integrity of the structure is compomised being the cause potentially of a structural failure, as the loads are changing with respect to time and increasing.

The turbulator idea would have the effect of breaking up the flow. of any vortexes forming.

I am not sure about the thick trailing edge. Conventional wisdom is this fills with stagnant air, and effectively increases the length of the chord. I am not sure hoe or if that helps, unless in some way it were to either dampen eddies or perhaps stimulates random turbulence either way preventing a stable oscillating pattern within the air flow.

I do not know it is von Karman shedding, just threw it in, Although if the effect is just along the longitudinal axis. It would seem to be a simple aerodynamic effect.

I suspect that flutter is related to instability in fluid flow and the characteristics of the surface which is undergoing the phenomena. Which is not dissimilar to the Von Karman vortex street idea.

[Ultymate]

Here's a shot of a rudder under current use in F3A

[Simon Chaddock]

Fascinating!

A 'fish tail' rudder?