Autogyros - how do they work?

[Biggles' Elder Brother - Moderator]

HI Tony,

think about it this way; imagine a hand held child's windmill, drag the windmill through the air. The air makes the windmill turn - but you'd also feel a drag force on the windmill pushing it backwards.

So the prop at the front pulls the windmill (rotor) through the air. The windmill is so orientated that its almost horizontal but slightly tilted backwards, so it slightly faces the oncoming air - so the air turns the rotor, this creates a drag force. But becuase of the orientation of the windmill being only slightly into the oncoming air (being only slightly tilted back) the vertical component of that drag force is much larger than the horizontal component. If the vertical component of the drag force is bigger than the weight of aircraft - up you go! That's were the lift comes from.

Going back to hand held windmill analogy, imagine you had such a child's windmill on a stick. If you tilted it so it was nearly, but not quite, horizontal and then move it quickly forward you would feel a strong upward force as the vertical component of the drag - again this is the autogyro's "lift".

In a sense all lift is caused by drag. Aerodynamists can argue until the cows come home on whether drag is a by-product of lift or if lift and drag are basically the same thing just viewed from a different perspective! In one, perfectly reasonable interepretation, lift is just upward directed drag!

BEB

Edited By Biggles' Elder Brother - Moderator on 04/11/2012 13:27:06

[Erfolg]

Lat night I was thinking about the rotor as a flywheel. It seems that you have two servos on the mast. Although I am guessing you could have one on the elevator (instead) to control pitch/AoA of the rotor. The second controlling the roll plane of the rotor.

Looking on the top of the rotor, which is the conventional rotation, clockwise?

Stemming from this I like all my servos, to have the safe movement, to put the linkage to the control surface in tension, rather than in compression, where the tendency to crippling occurs.

So know looking at our rotor in plan, which sides do you place the servos to achieve this.

I personally have always have had issues in understanding flywheels/gyros, when attempting to move them from their stationary position. Are the loads on the servos high, indicating that something with more force than a 9g servo is required to move the device from the static position, when spinning?

[Tom Wright 2]

Erflog

Autogyros with rotor disc diameters of up to 40" and with AUW of under 2 lbs seem to fine in terms of control system reliability when 9gm metal geared servos like the MG 90 are used .Plastic gear servos are only suitable for very small light machines .

Rotor blades can be configured to rotate in either direction, in basic terms the direction determines the requirement for left or right roll trim .

Pitch and roll involve both the servos working together via the delta TX mix or a separate component .

I f roll control only is used on the rotor ,(pitch controlled by elevator) two servos are desirable ,but the delta mix would not be required.

Just going back to one of you earlier questions on rotor chord ...I flew the Crane Fly Major today in near perfectly calm conditions This model is heavy (around 4lbs) and has the same rotor disc diameter as the standard model ,but with the chord increased by 30% . Two things were very noticeable in that rotor spin up was achieved easily and the slow speed performance was remarkable .

Need to be away now for the rest of the evening but will come back with more on this subject soon.

 

Tom.

 

Edited By Tom Wright 2 on 04/11/2012 17:58:24

[Tom Wright 2]

Continued from above......The flights with broad blades were successful but with insufficient time to explore the extremes of the flight envelope. I am now hoping to fly the model again today in an effort to establish any serious downsides associated with low aspect blades.

Just as a matter of general interest the blades consist of a 6mm depron core sheeted top and bottom with 1/16" balsa sheet. Each blade( three in total ) is 467 mm span. 80 mm chord . with a max section thickness of 10 mm . This results in an aspect ratio of 5.8 which is considerably lower than seen on most popular Autogyros. So I expect to encounter some interesting aspects of performance that will lead to further information on the subject of broader blades.

Tom.

[Tom Wright 2]

Much more wind today ,good for checking performance in less than ideal conditions .

All I can say at this stage, is that increasing the rotor chord results in substantially more lift .Keeping in mind, this is the major effect noted so far on my experimental model .

Tom.

Edited By Tom Wright 2 on 05/11/2012 15:05:42

[Erfolg]

The problem with problems, is that gnaw away at you. Again I slowly woke up this morning still thinking of rotors. Unfortunately not with coherent set of answers, or questions, just amazement that autogyros actually work.

Although most have no issue that the rotor spins by being edge on to the prevailing air current, I am surprised how well it works. I found I was thinking what made a cup anemometer work. Both cross sectional areas being identical, just the shape, depending on cd difference to drive the rotoer around. This made me think how fast would it be possible to get the rotor to spin, all dependant on cd difference.

I then thought, just going round generates no lift, so the shape and how it is angled to the airstream matters a lot. I was now thinking i bet the difference in revs from climbing to sinking is very little, or is it. I wonder has any one has run a tachometer on one? I was by now thinking that normal wings have to operate over quite a varied airspeed. Does the rotor, possibly, the hub static, the tips significantly faster. It was the issue of if a highly cambered section, would be better than a low camber section? For most models the shallow drag bucket of highly cambered sections, being a nuisance, but would it be true for a rotor?

It also struck me that the blades on my helicopters taken on quite a catenery or a half parabola when held by the tips. Of course the rotor operates more like a UDL than a point load, although I guess the lift is anything but uniform (perhaps scimitar or Hoener tips would help). Where I was really heading, the centrifugal/centrifugal forces putting the blade essentially into tension, rather than top in compression and the bottom tension, or does it. Which made me think, tip weights could help with strength or could it, after all if it was a good idea, full size helicopters would have tip weights. I then thought, perhaps they do, you just do not know enough about them.

When I fully woke up, I was no nearer an understanding, realising some basic information would put me nearer an understanding.

[Tom Wright 2]

Erfolg.

You may find this sheds some light on your deliberations.

Tom.

Edited By Tom Wright 2 on 05/11/2012 19:30:26

[Tom Wright 2]

And this for the mathematically minded.

Tom.

[Erfolg]

I have read the links, it does seem that there are two major issues the first is the Delta plate hinges.

It seems at model size the use of a flexible plate is one of experience in having an idea what will work and a bit of hit or miss. This process probably is very much experience dependant, in that you now need to know what to look for. The use of mechanical hinges is probably much the same.

With blades it has been suggested that the C30 blades provide a difficult to control model. Which indicates that blade design/build is not necessarily straight forward. It has been suggested that biasing the weight to the leading edge, and tip weighting.

[Tom Wright 2]

Erfolg.

Welcome back .The model autogyro delta plate is simply a flexible material with sufficient compliance to allow limited blade up and down movement and resist undue sag.

The material most commonly used is smooth both sides GF sheet ,this material is very tough and resists loads applied in tension compression and torsion very well.

The material thickness and dimensions are selected to suit the weight and span of the rotors .If the delta plate is to thin or to large then blade sag and unruly behaviour in the air will be apparent even to the point were one blade can be seen to displaced vertically by a large amount causing temporary loss of control or worse.

On the other hand if the plate is to stiff undesirable effects in pitch and roll will be apparent, with severe pitch up tendances often occurring

I do not have or need a mathematical way of determining plate design although a formula were model weight and rotor disc area could be inserted to give plate dimensions and thickness would certainly be handy.

My experiments and observations over nine months of intensive autogyo design and many hours flying suggests that model autogyros of between 0.75 and 1 m rotor disc diameter and auw's between 1.25 and 2 lbs will perform well on a plates that are between 0.8 and 1 mm thick having sides between 4 and 4.5 inches.

Blade design for models within the parameters mentioned can indeed be very simple starting with some med - hard even grained balsa 2" wide and shaped to an Aquilla section 5-6mm at max thickness. Of course there are plenty of other options but it is in fact quite easy to build a model that flies well by using some very basic information and materials.A great deal can be learnt from such models thus paving the way to more sophisticated designs.In view of your extensive modelling and flying experience I feel you could produce your own autogyro based on one A3 page of general information.

Tom.

 

Edited By Tom Wright 2 on 07/11/2012 21:25:58

[Erfolg]

I would think the idea of being able to calculate the thickness and other criteria of the delta plate is some way of. The reason being that I suspect little reliable data is available on existing systems and that it may not be possible to accurately evaluate existing components.

With respect to airfoil sections, i do not know anything about Aquila wing section. I think it is possible to talk about general features of some commonly used wing sections. Such as those of Eppler 205, which is not a million miles different to Clark Y, there are some differences. The first being is that E205 has maximum thickness further back, than when compared to Clark Y. If memory serves me well, I think that E205 is slightly thinner, with lower camber, although i do not think the difference is significant. If a modern section like MH 32 is considered then the camber is low, with about 8% thickness, again with  arearwards maximium thickness, so quite thin.

Are the same considerations viable, with a Autogryro. Some how I think the issues are different enough, so that comparison is meaningless. On model aircraft, it is possible to ensure that the wing is operating in a region that is predictable. By going to to 10" chord, the section is almost certainly turbulent being always in the  super critical area. Rather than at least theoretically jumping from laminar to supercritical Reynolds numbers, enhancing the hysteris effect, On such narrow chords, as is the norm on model autogyros. the calcs I am sure will indicate laminar flow (although tripping into turbulent earlier than envisaged).

So what does matter? is it camber, is it percentage of maximum thickness, where maximum thiskness occurs, or something else?

It does strike me for hand made blades, it should be easier to produce a section which is nearer to that intended, with broder chords, than is possible with narrow chords.

The use of lead and the heatshrink seem to have merit. Would practical applications indicate the same?

Edited By Erfolg on 07/11/2012 23:16:11

Edited By Erfolg on 07/11/2012 23:18:37

[Tom Wright 2]

When considering the section better efficiency can be achieved by incorporating little or no negative incidence,but this leaves the question of how to achieve spin up .

Current thinking seems to suggest a sharp edged section of between 7-10% thickness ,with no under cut from the l/e to the flat bottom.is a good place to start as rotor run up can be achieved with very little or even no negative incidence but other features need to be added to get good run up at low air flow values.Building the blades light can aid easier run up but may need tip weights to damp roll sensitivity.

As Rich points out over on the Fire Fly thread placing the rotor pivot point behind the lateral blade c/g ,and scalloping the blade root ,further aids reliable run up.However for a very simple machine it is also perfectly possible to design for a very low rotor loading giving in effect spare lift that can be sacrificed to employ enough negative incidence to ensure easy run up.

Tom

[Erfolg]

Much of my thinking was about the HK blades. What is it about them that is lacking. Essentially comparing things that appear to work well, with those which do not.

I was thinking about where the blades are attached, along the chord. Then I thought, if they are bolted solid, it is the compliance of the mounting plate, and possibly its plan shape, which could be a determining factor. So your comments have now surprised me. I guess i do not understand how blades are attached, and move.

I am a little surprised at the sharp leading edge, as in model aircraft wing, the feature is a recipe for a very sharp stall. Yet in reality making no contribution to efficiency to the wing performance at any speed, as we do not enter the transonic or supersonic speed range, however fast the model is crashing.

My intrinsic thinking of broader chords, is that it is easier to be accurate, within the context of deviation from the wanted % contour points, when hand made.

I would have thought that there is broad correlation of requirements between autogyro and model helicopter requirements. So studying and comparing the blades could be useful.

I am not necessarily convinced that a smooth upper surface is necessarily a must, as a slightly rough surface on model wings can aid reattachment, even if the Reynolds numbers indicate laminar flow, particularly as this normally is achieved over a limited area of the upper front region. Having said that, I would have no issue with using it as an easy finishing method. A bit of steel wool polishing to the upper rear would do the trick, if thought necessary.

Until I feel happier in at least believing I have some idea of what is required, I will not be starting a model. Although I wonder how many plans are available for electric models.

[Tom Wright 2]

I am waiting delivery of four sets of HK blades ,and hoping to use them on two new autogyo projects . The problem with these blades seems to be the size as they are smaller in span than blades used on the trainer models seen on this forum,so I propose to build a light airframe to suit and see what happens.

As for blade attachment ...a single 4mm bolt is is passed though the blade root and then through the delta plate .The bolt is tightened sufficiently to hold station under low radial loads but the moderate friction between the rotor under surface and the delta plate allows the blade to move and find its natural radial position when rotating. The up and down ( flapping motion) is controlled by the delta plate compliance .

I have flown many blade types and they all work but the sharp l/e minimises blade weight and seems to be better at achieving spin up .

Airfoil section accuracy is of course desirable in the same way as all aspects of the airframe construction should be engineered and built to the best standard possible .but if a simple model is designed to be tolerant of inaccuracies then it is less likely to fail to fly when in the hands of of a first timer.

Many would say a smooth upper blade surface is essential ,but I have successfully flown blades that are just the opposite it all depends at what level these matters are applied .It must be made clear that all my comments are made in relation to simple light weight autogyros that I have built and flown.Heavyer high performance machines obey the same laws of physics but require special considerations rather like comparing an Impala glider with an Ava .

[Tom Wright 2]

Erfolg

Just picking up on your wing section comments the Eppler 205 as shown below is indeed similar to a Carke Y .

If the section below the horizontal line is removed then it becomes what is loosely known as the Aquilla ,which would be the section found on many modern model Autogyro rotors .Maybe a little thinner on average but close enough to illustrate the idea. 

Tom.

 

Edited By Tom Wright 2 on 08/11/2012 21:30:07

[Tom Wright 2]

If those looking in have ever taken a look at a variety of Autogyro designs ,then the generous fin areas used may have been noticed .

Unfortunately large fins conflict with the need for good back rotor clearance, so the the designer needs to consider practical ways of getting around this problem.

Employing a pusher motor and a short moment arm can make life easier but such models require a little more attention to other design features and set up.

Just mentioned this in case anyone is thinking of designing their own tractor Autogyro .........and I hope there is .......and may be this has provided a little "food for thought" .

Tom.

[simon burch]

On the issue of tail design, I've just recently changed the design of my Spyro with amazing results. After deviating from the original design, I've struggeld to get my machine flying the way I wanted. One thing autogyro flying teaches you is to experiment.

She now flies perfect. More challenging than some of the tractor designs, but the sound and look of the pusher models is pretty unique.

Cheers,

Simon

[Tom Wright 2]

Great to hear from you again Simon ,and glad to see you have continued afer the Crane Fly build .

Nice video and a very smooth controlled landing ,(newbies take note how Simon used the power on aproach) .

Tom.

[Tom Wright 2]

Erfolg

I think you may find this of interest.

Tom.

[Erfolg]

Tom

I have read the link.

Although I understand the general principle, the actual implementation of the science, is defeating me from the practical implications,as to where pull and push the rotor disc to control pitch and roll.

It is the control of the rotor which appears to be the most demanding.

I did consider some aspects of the rotor and as usual have no real answer.

I considered a very small rotor, which was vanishingly small. It did seem that the necessary revs to support the model would be high, although the response to change of revs would be lively.

Next I considered a very large dia rotor. This would require lower revs to support the model, the downside is that the response would slow.

Additionally speeding up or down would be related to rotor mass also, due to inertia effects.

There and are no conclusions other than, yes these are factors, but how do you deternmine a viable rotor diameter and chord length.

More questions than answers. It does seem I will have to build one, to start getting a handle on the issues.

[Tom Wright 2]

Erflog.

Building a simple Autogyro is more fun than the theory ,although I applaud your persistence to study the theory as it's obviously an intelligent way of tackling a new subject.

It is normal practise to try and get the head roll and pitch pivots in a vertical line to the rotor retaining bolt.The length of the head roll and pitch actuating bars is often equal ,with the length decided to give the required amount of rotor movement .

For trainer /sport Autogyros a rotor loading of between 2 and 5 oz per sq ft is the norm ,but keeping the loading on the lighter side is beneficial for first attemps.The rotor loading is calculated by working out the rotor disc area ,and then using the weight of the model to work out the loading in oz per sq ft ....just the same way as a fixed wing .

Simple experiments suggest that a rotor disc diameter of up to 1m ,and with a blade chord of between 2" and 2.5" work well for models up to 2 lbs AUW ....lighter the better to start with.

The rotor control response is effected by servo type ,clevis position on arm ,head control arm length , and the weight distribution along the blades ,adding weights at the tips can soften roll response on a twitchy model, but to much can result in control response lag .The height of the mast also effects roll response eg short = responsive long = tame .

The primary design considerations I used for a simple 36" rotor diameter tractor Autogyro are as follows......

Design the fus to achieve good clearance between the rotor and tail .And to achieve a hang angle of 12 -15 degrees at AUW.

The u/c should be as wide track as possible ,with out adding un due weight.

The mast height can be one rotor blade length x 0.7

Rotor section flat bottomed with sharp l/e ( no undercut )

Mount the blades at between 30 - 40 % of chord and balance the blades chord wise to get the c/g on or in front of the mount centre. Also balance the blades length wise .

Use .08 thick GF to made a delta plate that has equal sides of between 4 and 4.5".

Make sure the servo to head control arm push rods are strong and cannot bend under flight loads. Use metal gear servos.

Do not skimp on fin area positive directional stability is a must.Tail area not critical,but the better the design the less tail area can be tolerated.

Use a motor set up to give 200-250 W .5-6 degrees down thrust, 2 degrees right side thrust.

There is absolutely no guarantee that a model made to these guidelines will fly off the board but it's a start point that worked for me.

Take advice on trim set up,and launch techniques before flying.

And most of all be persistent don't give up ,and ask questions .......answers may or not help lol.

Tom.

 

 

Edited By Tom Wright 2 on 13/11/2012 20:13:24

[Tom Wright 2]

A couple of views of a typical control head with bearing and delta plate.

 

Tom.

 

 

Edited By Tom Wright 2 on 13/11/2012 21:36:41

[Tom Wright 2]

The two control head operating servos are plugged into the rx pitch and roll channels via a delta mixer, or a standard mix can be set up on the TX .This mix function enables the servos to work together ,to achieve any combination of pitch and roll rotor positions determined by the tx stick deflections.

Tom.

[Tom Wright 2]

This view shows the set up and lay out on my Crane Fly.

[Tom Wright 2]

Posted by Erfolg on 08/11/2012 12:10:28:

Much of my thinking was about the HK blades. What is it about them that is lacking. Essentially comparing things that appear to work well, with those which do not.

Erfolg

It would seem possible that the HK blade performance was assessed on a model that that was perhaps to heavy for the application?

Having flown them on my Crane Fly with lipo weights up to 1.8 A I found the performance to be very good indeed,with no deserable draw backs found. The AUW of the model with 1.8A lipo is 1lb 9oz so I conclude the HK blades carry this weight quite easily.

Tom.