Prop design, modeling and carving

[Nikolash]

I'm considering making props from wood on a cnc mill.

I want to make props for glow engines, that will be ideal match for an engine and airframe (3d flying in mind) to squeeze every bit of hp from engine.
Actually I'm aware that wood wont make prop of ultimate performance, but it is cheap, readily available, easy to work with,and making your own prop will give personal touch to a plane


I'm aware that it will be trial and error, and step by step proces.
Most important would be:
-designing a prop based on relevant calculations, theory
-than making an airworthy prototype
-than goes improving

Today I was playing with auto cad to see how difficult is to model a prop.
It is fairly simple process: you have to align root, middle and tip blade airfoil to tangents of their respective helices, make hub fairing and then command loft.
You only need 3 curves(airfoils) and a rectangular (hub cross section) and their relative position to make 3d model of a prop. It takes less time than writing this post.

But big problem is which airfoils to use ? (on these 3dmodels i used "in situ" hand drafted airfoils - artistic approach ;) ) .......and many others...

If nothing else, reverse engineering would help to establish some parameters, like chord thickness, prop surface and shape .

You are wellcome to share your experience and ideas.

To draw a line-
For now i know how to model a prop of a given pitch and dia.
I have no clue wich airfoils to use and how changing the airfoils will affect prop's performance.
I have free acces to 3d cnc mill, but dont know which tool to use for wood?

[Nikolash]

http://i277.photobucket.com/albums/kk61/nikolash_photo/Corel2.jpg
http://i277.photobucket.com/albums/kk61/nikolash_photo/Corel22.jpg
http://i277.photobucket.com/albums/kk61/nikolash_photo/Corel21.jpg
http://i277.photobucket.com/albums/kk61/nikolash_photo/elisa5.jpg

[Brandano]

very hard to say what would be an ideal section for a model prop. They operate in a weird environment, odd reynolds range, airflow not always aligned with the blade chord, different airspeeds between the root and the tip. I suspect that most commercial prop sections are chosen mainly for mechanical reasons than for aerodynamic reasons.

[Nikolash]

I suspect that most commercial prop sections are chosen mainly for mechanical reasons than for aerodynamic reasons.
Brandano, thats true, especially for sections close to hub.
Every brand has each own way in blending prob blade to hub, and often it gives impression of aerodinamical crudeness like on apc's, or on TF wood -almost half of the prop is square, sharpie?!
http://graphics.hobbypeople.net/gallery/534600.jpg
But I gues that central part of propeler does not contribute to thrust as it might seem.
Let's say one third of dia. gives only 11% of disc surface, considering that relative airspeed of central section goes from 0 to only 1/3 of tip's speed, acording to this if we also consider Re numbers, I would estimate that central third of prop's diameter generates less than 3% of thrust.
http://i277.photobucket.com/albums/kk61/nikolash_photo/Corel2-2.jpg
So, statement that airfoil sections close to hub of a prop serve only for structural integrity is backed with facts, so performance depends on propeller shape, surface area of blades, and outer airfoils.
So now it is easier knowing that i have great margin for making strong props, without noticeable sacrificing performance an vice versa.

And here is small tutorial of modeling a prop in CAD, profile sections and shapes are fictonal,just for illustration purpose.
Blending could be done in many ways but this one seems close for what is expected from wooden prop
http://i277.photobucket.com/albums/kk61/nikolash_photo/KOMPLET.jpg

[vintage1]

I've been designing and making laser cut laminated props for some time now.

I even detailed the complete design process in 2D CAD. It's somewhere about..

Aha

http://www.rcgroups.com/forums/showthread.php?t=936534

Read that..


As far as sections go, for normal cruise flight a true helically cut prop is very near optimal.

Sections are tapered from a 'strong' - up to 30% thick - root to a 10-15% tip chord depth at the tips.

Thinner undercambered stuff does well at slow RPM, but i harder to carve and fragile.

[BMatthews]

Some years back when a buddy and I got into electric flying it was all about efficiency. We did a BUNCH of testing of various props including reshaping some of the wood props to thin them and both add and reduce the camber of the blade airfoils.

What we found is that it's REALLY hard to improve on most of the APC prop shapes. In fact none of my wood props were able to be made to perform better than the APC props other than in a couple of cases. Both of those were in the use of the really big, high pitch slow turning sizes used for geared motors back then. My 14x8 and 15x10 mods to Zinger "cores" were better than the APC options. But then those APC's were intended for use on the chainsaw/weedeater/leafblower/quarterscale gasoline engines of the day. With the newer APC electric props out now I'm not sure I could do a wood conversion that would work better than they do.

So get some APC props in your size range and test them. Then try to make something that works better. It won't be easy.....

[Nikolash]

@ Vintage1

Great thread, simple, effective and affordable way to make your own scale props, I have found your thread when I was searching this forum.
My concern was how to make that laminated props enough strong to use it on 4t engine,
If I cut wood sheets first than laminate, I'm afraid that I can't apply enough pressure to make good bond for such (ab)use, without distorting layers, so i thought that i had to start with solid wood.
Maybe I will reconsider your method and make couple prototypes to test their strenght.
It takes more time to design laser cut layered props than a solid ones, but it takes less time to cut (you have to do sand and varnishing them anyway) .





My view on carving props has dramaticly changed, at first I thought that there is space technology involved, and I was just hoping to find 3d models on net and feed cnc machine with them.
The more I was into I've realized that making good prop would be fairly simple (more simple than I expected), and that making best propeler would be rocket science.

@ Bmatthews

It is really hard to beat performance-efficiency of APC props especially with wood. Great !
I have more courage now.



I have a question:
What does width, area of a blade to static thrust, or better - what "eats" more torque for same increase of thrust - making blades wider or increasing diameter of a prop?

P.S. Vinatge1, what reference on airfiol is considering when determining pitch? Is it MAC, or bottom of prop, or ???

[vintage1]

@ Vintage1

Great thread, simple, effective and affordable way to make your own scale props, I have found your thread when I was searching this forum.
My concern was how to make that laminated props enough strong to use it on 4t engine,
If I cut wood sheets first than laminate, I'm afraid that I can't apply enough pressure to make good bond for such (ab)use, without distorting layers, so i thought that i had to start with solid wood.



Full scale props were laminated. The glue only has to be strong in shear, to resist layers sliding over each other. i.e. cart spring rather than solid lump!

If you are really worried use something like foaming PU glue or epoxy.

I broke a prop on landing. Made simply using thin CA. No glue line failure. Just snapped the wood. Some people have used my props on 4 cycles..



Maybe I will reconsider your method and make couple prototypes to test their strength.

Do it.

The key to high power usage is to forget the root area aerodynamics, and make it a solid slightly faired lump!

The actual failure modes, flutter excepted, is thrust breaking the root. Centrifugal forces are more than catered for by a decent root section.

It takes more time to design laser cut layered props than a solid ones, but it takes less time to cut (you have to do sand and varnishing them anyway) .



It takes me about 2 hours to do a wuickie, on the design, and 4 hours for a quality job.

making them is a two hour job at most.

My view on carving props has dramaticly changed, at first I thought that there is space technology involved, and I was just hoping to find 3d models on net and feed cnc machine with them.
The more I was into I've realized that making good prop would be fairly simple (more simple than I expected), and that making best propeler would be rocket science.

I don't think it is..again the law of diminishing returns.

My very first 'slowflyer' prop outperformed an APC slowfly. Just.

I've used a mixture of premade and hand made and APC 'E' props and the woods are usually a shade less good in the premade, and a shade better when I did them myself. The secret seems to be making the prop no thicker than it needs to be. Thin sections seem to be low drag, but ultimately te props get fragile, and flexible. You need then a bit heftier for an IC engine.


I have a question:
What does width, area of a blade to static thrust, or better - what "eats" more torque for same increase of thrust - making blades wider or increasing diameter of a prop?



Double the area, double the thrust. Double the power.

also thrust is area times velocity and velocity is related to diameter, so more diameter is more area AND more velocity both. Area is proportional to square of diameter, velocity is proportional to diameter so thrust and power proportional to diameter cubed. Thrust also proportional to pitch speed so thrust and power varies as pitch.

All at constant RPM. With respect to RPM, I think thrust and power is square law.


P.S. Vinatge1, what reference on airfiol is considering when determining pitch? Is it MAC, or bottom of prop, or ???


Interesting point...No one really knows. If you are really smart, you can calculate the flying speed, and the pitch slip, and set the blades at a constant angle of attack to the incoming flow all along the blade.. But I generally set the bottom of the flat section to the helix, and then assume that the actual airfoil shape will give me the extra incidence at the flying speed.

That probably makes my props a shade coarser pitch than advertised.But they should fly at the *calculated pitch speed* and still provide thrust.

Let me say one thing in conclusion. It is not hard to make a wooden prop that is as good as, or better, than anything you can buy. However whether you think the 6 hours of design and manufacture and half an hour of laser cutting, is worth it, is a personal decision. :D

[BMatthews]

The width to diameter represents the aspect ratio of the blade. For high RPM use the width doesn't need to be that wide to be up into the higher reynolds numbers and to avoid a lot of the losses of low reynolds number operation. But slow turning props can benifit from wider blades. If it helps to think of it this way consider that a wider blade will bite into the air better at slower speeds where the air is more "flexible". Meanwhile a higher RPM prop is seeing the air as being more "solid" so it doesn't need to be as wide to lock into the air. Now this description really butchers the whole reynolds number issue but I think it's a reasonable way to think of why slower turning props tend to have lower aspect ratios by having wider blades. It also takes relatively more power to move a wider blades but mostly because they are then doing the work instead of "slipping" like a narrow blade will do at lower speeds.

Again this explanation is loaded with aerodynamicly poor descriptions for the sake of trying to explain how the air sees the prop blades in simple terms.

[vintage1]

It's a guess, but the problem with high RPM work is getting ENOUGH diameter to get clear of the fuselage etc. So you tend to slim the blade down to get the loading right. Also there is less induced drag on a slender blade for a given thrust.. The wetted area which dominates at ultra high RPM though, doesn't change for a given thrust.

There is another issue with 3D props as well. These are not operating near the pitch speed at all, and with those if you want a prop with a decent wide speed range, you don't want to spread the thrust over the whole blade..so it makes sense to make a paddle prop with the thrust concentrated across a small area of the blade.

[Nikolash]

I have exams, so there is no time to deal with props fore now, :o but I hope I 'll make and test some props during next week. I would like to make series of props by varying pitch, dia, and blade surface . ;)
Have to find wood blanks and sheets first. :(

[Nikolash]

http://i277.photobucket.com/albums/kk61/nikolash_photo/Corel-1.jpg
http://i277.photobucket.com/albums/kk61/nikolash_photo/slice.jpg

I found some spare time and energy and just grabed one prop file to try to slice it.
I guess cad needs some tuning- increasing number of isolines etc. as you see slices are solids 8easy to extract their slice surfaces), but still looks promising.

[vintage1]

Looks pretty good to me.

[Nikolash]

Thanks vintage,
but that wooden like structure is not intentional, it is weird polgon mesh show up, and at close ups you can see how crude model actually is.

[Stagger_Wing]

Hey guys, nice work!

I have started a project that requires me to design my own propeller blades and I was hoping one of you could help me out with a few questions.

First of all could you describe the helix in a little more detail? I understand the basic concept but I am unsure of exactly how it is defined both in CAD and mathematically. Also what books if any are you referencing for propeller design? Neither of my aerodynamics books cover propellers in detail.

From looking at Nikolash’s screen shots it appears that the center of lift, or similar point, has been used to align the airfoil profiles before lofting, where as V1 appears to have aligned the trailing edges along a horizontal reference line. Is there a there a best method of lining up the profiles or does it not make notable difference in performance as long as it kept consistent throughout the design?

Thanks in advance.

Scott

[vintage1]

Hey guys, nice work!

I have started a project that requires me to design my own propeller blades and I was hoping one of you could help me out with a few questions.

First of all could you describe the helix in a little more detail? I understand the basic concept but I am unsure of exactly how it is defined both in CAD and mathematically. Also what books if any are you referencing for propeller design? Neither of my aerodynamics books cover propellers in detail.



Well, the disapponting news is that most model props do not seem - as far as I can tell - to be in any way scientifically designed: the rapidity with which I was able to match APC and GWS slofly props, followed by equally successful props able to take more power suggests to me that actually it's all a matter of taste and guesswork, largely.

I will toss some thoughts in. I use sort of flat bottomed clark-Yish, or on thinner blades, slightly undercambered sections, and tend to align the flat bottoms with the 'helix'. That means at 'helical' pitch speed all parts of the blade are at a slight angle of attack that is the same across the blade. That should give optimal performance at or near that pitch speed.

Obviously, if you want a 3D hovering prop, you don't want a helix at all. You want the blades at a constant angle to the hub more or less. To give constant angle of incidence to a more or less static airstream coming in.. There is in fact no way to optimise a prop for different airspeeds and thrusts. Not even a variable pitch one. Because a coarse pitched helix when rotated to fine pitch will have the inner part of the blade at a negative angle of attack..

IN fact what seems to dominate the designs for me, is that for high power use you need a thick blade near the bub, and the prop is less efficient as a result. The props with thin blades do well at low RPM, but tend to snap at higher..or even on heavy landings!

Ad since I am generally chasing scale appearance rather than performance, I am not that bothered.

Finally, the blade thrust per unit area of blade varies pretty much with the diameter, which means that the inner part of the blade doesn't do much at all, except hold the outer bits in place structurally, and add drag.

So don't knock yourself out trying to make a prefect prop. One that is as good as one you can buy or better, is plenty good enough.

Thin blades seem to work better, but are structurally compromised. In order for a prop to work the first thing is it should not fall to pieces. Unless you are making carbon blades, that tends to impose limits on what you can do.



From looking at Nikolash’s screen shots it appears that the center of lift, or similar point, has been used to align the airfoil profiles before lofting, where as V1 appears to have aligned the trailing edges along a horizontal reference line. Is there a there a best method of lining up the profiles or does it not make notable difference in performance as long as it kept consistent throughout the design?

Thanks in advance.

Scott

I don't think it matters much really. I only do what makes the prop look 'right' from the side. Allied WWI props tapered from the hub to the tip substantially from the front to the back. That places the tips about on a plane with the hub rear. Later props tapered more symmetrically. I am thinking of the Watts props on the gladiator and MKI hurricane etc.

Also the Gyspy Major props I have looked at are like that.

I suspect it was more an issue of strength, fashion, or production methods than anything cute aerodynamically. WWI props on German and Allied aircraft look very different, and yet performance was remarkably similar.

[Stagger_Wing]

Thanks for the reply vintage1. Your post was very informative. I’ll give it a try and see what I can come up with.

[lincoln]

If you REALLY want to get scientific about it, I seem to recall that a Professor Larrabee at MIT came up with some code for optimized props that was very helpful to the human powered airplane guys. I don't know what's happened with the code since then. I seem to recall that it was base on theoretical work done in the '30s.

However, I wonder if you guys aren't going all too high tech. Seems like this design and build stuff is taking a long time. I wonder how fast you could make a good prop with small power tools and methods based on the rubber power guys' tricks. I think there are some instructions that you can find with a link from the Twin Pusher pages. I've made some from balsa by hand, and based on that I think with stuff like a small bandsaw, sander, etc. you ought to be able to put together a prop pretty quickly. Of course if you're going to do a bunch of the same design, maybe high tech is worth it.

I know that some of the commercially available props aren't so great. I took a Graupner 6-3 folder and used Mark Drela's mods. Noticeable improvement in climb.

[vintage1]

I wonder how fast you could make a good prop with small power tools and methods based on the rubber power guys' tricks?

About 2 hours to design, 10 minutes to cut, and two hours to carve. Using a laser cutter.

I doubt I could do better carving a blank.

This is one area where the design for lasering is actually faster than scratch building!

If I had 3D cad, it would probably be even faster.

[vintage1]

Thanks for the reply vintage1. Your post was very informative. I’ll give it a try and see what I can come up with.

Oh there is one other thing that's worth mentioning. I did linear thickness taper of the blade section as well, from around 35-40% at the root to around 10% at the tip.

In fact one could probably taper faster than that, but the maths was then not a guesstimate, it started to get too horrid.

[Nikolash]

How to define helix in aoutocad...you have generic comand in draw menu, but it in that shape is not much usable for design there i used it as visual reference.
But for actual reference for determining aoa of an airfoil you need simple trigonometric formula. I'll try to derive it here.


2*(distance of section from rotation axis)*Pi * tan(pitchAngle)= PITCH (in inches)

so :


PitchAngle= arctan (PITCH / (2*(distance of section from rotation axis)*Pi )


...about aligning prop sections...

A prop could be considered as wing, to certain point, so sweeping blades, making dihedral, like tappering ticknest toward front or rear face of hub, making scimitar blades(aligning sections along curve) and other extravagances are possible
BUT those changes are structurally critical due to generated centrifugal, and lift forces which are not axial any more.
If they are to much out of axis, they can induce unwanted bending , and shear forces and, oscillations which could easily result in prop failure.
These are downsides of not aligning airfoils along neutral point.

As wintage1 mentioned that some full scale wooden props had thickness tapered toward rear surface of hub, reason for that could be to minimize tension at rear face and introduce a bit of pressure at front face of a prop which is allready stressed by centrifugal forces.

[vintage1]

As wintage1 mentioned that some full scale wooden props had thickness tapered toward rear surface of hub, reason for that could be to minimize tension at rear face and introduce a bit of pressure at front face of a prop which is allready stressed by centrifugal forces.

I think its so that carpenters could lay them flat on a bench when making them, personally :D

Flexible scimitar blades like slofly props are interesting, in that they almost certainly flatten in pitch under high thrust conditions, making a crude sort of variable pitch prop.

Oh and it's Vintage1, axshully.. ;)

But for actual reference for determining aoa of an airfoil you need simple trigonometric formula. I'll try to derive it here.


2*(distance of section from rotation axis)*Pi * tan(pitchAngle)= PITCH (in inches)

so :


PitchAngle= arctan (PITCH / (2*(distance of section from rotation axis)*Pi )


Well is it though?

What you want to do is set your helix for the actual airspeed and RPM, and then add a (constant?) angle of attack to that.

So basically fix the flying speed you want the prop to be 'perfect' for and work out what RPM the motor is working best at, in terms of power, and set the helix off that..then you want to add the angle of incidence that nets you the best L/D ratio at the airspeed of the model vectored with the rotational airspeed... and bearing in mind the best angle will have to account for teh varying section thickness as well as speed, and so on..and the result in terms of thrust per unit area have to be integrated back to determine the force along the blade, and so dictated the structural thickness at that point.

"The maths gets horrid" is all I have to say. I'd probably write a program that simply kept iterating until the thing wasn't getting any better with each iteration..

However it seems from various calculations that an average prop is about 75% efficient in cruise. That's pretty good. Its not like we can make simple changes that will result in several times more thrust being produced. The single biggest benefit I have found is using a larger coarser pitched prop at slower RPM, for efficiency at cruise., The downside seems to be that it's very much more a 'one speed' prop than a smaller one a higher RPM.

A typical prop with pitch to diameter in the range of 0.7-1 is a very good all rounder. 0.5:1 is a good slow speed and hover prop, and 1:2 is good speed prop or warbird prop, but these seem to be approaching the limits of what is practicable..

Its hard to find real data, but RPM figures and top speeds suggest that light aircraft with fixed pitch props tendto be aroud the 0.7:1 ratio, WWI fighters about 1:1 and WWII fighters before the advent of variable pitch around 1:2

You need coarser pitch with fast aircraft because otherwise the tips go transonic. Tip speed is pitch speed times diameter over pitch times Pi..so a 1:1 prop goes transonic at pitch speed = mach 1/PI..under 200mph.To reach 400mph you need at least a 1:2 diameter to pitch.

WWI props were coarse pitched because engine RPM was low..