Low Speed Theodorsen-Betz Propeller Testing With Schmitz Airfoil Data

[smed]

I am a high school student and I plan to conduct an experiment on low-Reynolds number propellers this summer. I already posted on this formum about my project, but now I think that it is finally coming together.

This is where I am:

I plan to research MAV (micro air vehicle) propellers approximately 3-6" in size. Reynolds numbers will range probably from 35,000 to 60,000. Actual testing of propellers will be conducted at a University where a professer very generously offered to help me obtain data. The wind tunnel is already set up to test props, too.
To design propellers, I will be using a program which calculates the correct Theodorsen- Betz propeller for minimum induced drag, but also takes into account parasite drag. This program goes hand and hand with a book on Theodorsen props, and the authors will be publishing both in about 6 months.
Apparently, for the program to work at these low Re #s, accurate airfoil drag coefficients are necessary for different stations on the propeller. For this, I have obtained F.W. Schmitz's work on testing on several airfoils at low Reynolds number. I am not sure of the accuracy of this information, which goes as low as 60-21,000 Re and I was hoping someone on this forum would have some input.
Then, I plan to make propellers with the 'Theodorsen-Betz minimum induced drag' program that I have, according to the dimensions of MAVs, but then make several propellers, each with the drag information from the different airfoils (with Schmitz's work, and maybe even Michael Selig's of the U of Illinois).
Then, I could compare the efficiency of the different propellers, and determine which airfoil's drag data is best to use when making low Reynolds number propellers. I am also hoping to show that Theodorsen propellers are the most efficient, even at these Reynolds numbers.
Also of interest is that the professer mentioned to me that although they thought that they designed propellers well at the University, the real challenge for them was fabricating the propellers. They said that they had an injection molding machine but never were happy with the results.
I do not know how I will be making my props, but it probably will involve making a 3d model of the props on an engineering program. I have a friend who has a machine shop, and I will ask him tomorrow if his cnc machines could machine propellers out of plastic if the machine was connected to a computer with SolidWorks, with a 3d model of my propeller design. Making carbon props does not seem feasible for this project because I would probably need about 20 molds.

Thanks

[davidfee]

You're a High School student? Wow! And believe me, I mean no offense! This is a very ambitious project, and you seem very well prepared... at least from a theory standpoint. I want to wish you well on this project!

Keep up the good work!
-David

[BMatthews]

Yes, best of luck with your work. Perhaps after you do the MAV work you could do some prop development for some of our electric models. MAV research is ongoing and obviously important but some good electric model designs would be commercially saleable immediatley to a "hungry" audience.

Xfoil may be what you're looking for to do the drag coefficient work for the airfoils. Either the DOS one or the imbedded version available with Profili ver. 2 should do well for your needs. Contact Marc Drela for confirmation that the polars will be meaningful at these very low Re's. There's a fair degree of extrapolation at those numbers I believe but Marc can confirm that for you.

If it looks good for you then there's also a pretty technical discussion group on Yahoo. Marc or someone else here can offer up the link. If you do a google for xfoil I suspect you'll also find it in the results.

[Ollie]

Your main challenge at this point is to shape the test propellers accurately enough that the airfoil data corresponds to the actual propeller crossection which you construct. This will very likely be a major source of error. Unless you can find the facilities to actually measure the shape of the airfoil and pitch of the test propellers at various places, you will be working with assumptions rather than facts. Every measurement has a probable error associated with it and the size of the measurement error should be within acceptable limits.

You can use X-Foil to generate polars of the intended airfoils but with various size errors introduced into the profile to see how much various size errors produce what variations in the polars. This will give you an idea of acceptable construction tolerances.

[Dick Huang]

Adam,
Good luck on your propeller project. When you have the shapes defined I would suggest that you get in touch with Landing Products maker of APC props. They now sell electric props that are under 6" in dia.; they might be interesed in helping you.
Dick Huang