Replies to This Discussion

Permalink Reply by Brad Hughey on April 22, 2012 at 7:52am

I am resolved to hold an admittedly minority opinion when I say that tilt-rotor designs are a solution looking for a problem.  It seems so many aeronautical engineers are fixated on the notion that the main problem with the current embodiment of the helicopter is that it doesn't fly fast enough.  Unfortunately, all the added complexity and non-fault-tolerant mechanical points of failure added by the tilt-anything designs exacerbate what I see as the primary issue with VTOL flight - expensive operating overhead.

That said, I do agree it is patently true that the ultimate performance of any propeller requires variable pitch to accommodate different inflow velocities (advance ratios).  To those that follow my random musings, note that this gentleman speaks of slower, symmetrical (both blades together) pitch changes on the order of 2 degrees per second or so, and therefore the propeller design would not be saddled with the pitch and inertial moment issues inherent in cyclic-pitch-control helicopters.

(hold on, I am going to answer the question.  :-))

I know of no better resource on the web than Martin Hepperle's site, including his exemplary programs JavaFoil and JavaProp.  All the tools are there, except there's not much focus on bridging the theoretical model gap between static and V/nD thrust.  Heck, nobody seems to cover that much anyway, making me think there's an excellent PhD-candidate research paper project which awaits the intrepid academic experimenter.  Anyway, here's the link:

http://www.mh-aerotools.de/airfoils/index.htm

If you're looking for an excellent technical foundation of the issues to get you started, this paper by Dr. Paul Pounds, now of Yale University, is a fairly complete treatise (see attached file).

Dr. Michael Selig of the University of Illinois and his team have done more actual investigation of low-Re propellers than most anyone.  His papers and data can be had here:

http://www.ae.illinois.edu/m-selig/

Enjoy!

Attachments:

• Pounds Modelling and Control of a Quad-Rotor Robot 33767.pdf, 2.8 MB

• ▶ Reply

Permalink Reply by Joshua Geating on April 23, 2012 at 3:57pm

Thanks for the reply. This is my first post on DIY Drones and I'm really glad to see there are professional, knowledgeable responses.

Although I have little experience in aeronautical engineering as a mechanical engineer (to my dismay), I can definitely believe you about the overhead associated with tilt-rotors, but I can't help but wonder if micro-UAVs somewhat inherently circumvent that problem because air frame loading is so low. I imagine even some servos could function entirely as the tilt mechanism when the entire aircraft may weigh 2-4 kg or less.

I am still confused about correct nomenclature. When I said variable pitch, I intended to imply changing the pitch rapidly and independently using servos for each propeller (although they would have independent motors as well).

Thanks for the resources! I'll be looking them over quite thoroughly.

• ▶ Reply

Permalink Reply by R. D. Starwalt on April 23, 2012 at 5:12pm

Are you familiar with the work shown in this video?

MIT ACL

I would hope your university connections allow some access to their information.

• ▶ Reply

Permalink Reply by Joshua Geating on April 23, 2012 at 8:09pm

I've seen that video several times, but haven't spent exorbitant amount of time looking it over. I was under the impression that the intent was more so for increased controllability than for efficiency. Their propellers appear to just be tapered flat rotors I've seen several times on hobbyking and similar hobby superstores. I am interested in propellers that actually have twist - could still possibly go in reverse (very inefficiently) if they needed to - but are primarily designed to operate over a wide velocity range. I would also like a much larger diameter, and possibly three blades instead of two. I can't help but notice how the osprey's rotors/propellers are much larger than a typical propeller, but smaller than a helicopter's rotors.

Does anyone know if R/C propellers come anywhere near supersonic at the tips? What is the primary factor limiting diameters? I would be interested in getting a motor like this one, http://www.hobbyking.com/hobbyking/store/__17924__NX_4008_620kv_Bru..., and increasing the diameter greatly for better efficiency.

• ▶ Reply

Permalink Reply by Joel Venable on April 23, 2012 at 11:37pm

 

Awhile back I actually dug into their whitepaper for the variable pitch quad, and found out the model of motor.  I believe they were Graupners if memory serves.  They're hard to come by in the states.  You're correct in that the focus was on the controllability rather than the airfoil efficiency. 

In my feeble mind, though, a propeller is very similar to any other airfoil with the exception that the tip always has a higher speed than the middle.

 

For a great deal of information regarding RC props, check out eCalc.  The short answer to your question about props is yes, it's possible.  It's more likely with a high KV motor and a smaller prop.  This might be done on a real RC plane to achieve a high top speed for a light plane that doesn't need a lot of torque.

• ▶ Reply

Permalink Reply by R. D. Starwalt on April 24, 2012 at 4:23am

You pose good questions and some of the data you ask about is measurable with 'amateur' techniques.

Regarding tip speed, this simple calculator might be of use.. Culver Propellors.

A text oriented toward prop dynamics.. Propellor Dynamics where the reviewer states..

Normally the shape and airfoil of propeller tips is not of much interest, as they don't much affect performance. However, in some racing classes the propeller tip speeds approach that of sound. This introduces a whole new realm of aerodynamics and the profile of the airfoil section becomes very critical indeed.

The first time I saw an Osprey 'in the wild' it was amazing. It is an aircraft that cannot be ignored.

I have always wondered how much the material in a prop distorts as the speed/load increases.

With the distortion, in your case a built in twist, at what point(s) does the 'bite' of the prop change?

There must be sections of the rotating disc that exhibit more thrust than others.

At least now we have some moderately decent model software.. if you can get a function that acurately describes the data.

I have a PhD friend, theoretical physics - surface modelling of particle scattering, that is always tweaking his software model trying to fit it to the collected data.

Measuring how your proposed blade variables affect performance will be the difficult part.

My PhD friend says I need to work in the experimental side of things.

 

• ▶ Reply

Permalink Reply by Brad Hughey on April 24, 2012 at 6:21am

@R.D.: That the Osprey is an incredible engineering feat cannot be denied.  Whether or not it was worth the effort is the subject to much on-going debate.

There is likely more aerodynamic twisting of the blades if the pitch moment of the airfoil is pronounced and the blade material is not rigid enough.  Unfortunately, if you're using the highest efficiency airfoils, the negative pitch moments can be very high.

Your PhD friend is right; unlike much of physics, many of the stalwart formulas used in aeronautical engineering are empirically- derived.

@Joshua: Full-scale variable-pitch propellers were found to be worth the complexity due to two facts: engines produce their best efficiencies at particular RPMs, and airfoils have their best lift to drag ratios at specific angles of attack.  The AoA changes, of course, with inflow velocity, so varying the pitch gives the pilot a means to change it back to quasi-optimal.

I would recommend reviewing this thread for a good treatise on the basics:

http://www.diydrones.com/forum/topics/the-case-for-large-scale-elec...

You'll get no argument from me that changing the pitch can provide a faster control response than fighting rotational inertia.  The fundamental questions are, is the increased cost and complexity (negative reliability impact) provide a commensurate performance advantage (in other words, is it just "cool" or does it really do something useful?), and does it come at the expense of other, simpler things to reduce control response times?

We have yet to see innovations in the hobby-class market for multicopters, like custom carbon-fiber blades with optimal taper and twist (see Paul Pounds' blades in his white paper), low inherent inertia motors (aluminum wire windings, et.al.), regenerative (or non) braking for blade slowing, and dynamic headroom spiking (higher voltages applied for short durations) to speed up leading motors, etc.  All of these things can be (ought to be?) tried before resorting to grafting on extra mechanical contraptions if our basic premise is that multicopters don't respond fast enough.  Otherwise, if you want ultimate efficiency and performance regardless of forward speed (or higher attainable speeds), there is no option other than variable pitch.  Just be aware of the single points of failure you're adding to the system (one blade set gets stuck at negative pitch and you're toast).

Feel free to disagree with any of my assertions, as such discussions are the stuff of enlightenment. 

• ▶ Reply

Permalink Reply by R. D. Starwalt on April 25, 2012 at 4:26am

Joshua,

Did you read William Premerlani's input on the discussion of VP from February?

LINK

It is not directly related to propellor voodoo but he has thought much about VP and quads.

• ▶ Reply

Permalink Reply by Brad Hughey on April 25, 2012 at 6:19am

What I found most interesting about that thread was Mark Cutler's paper covering his experiments.  There is no doubt that there are advantages to incorporating variable pitch.  However as Mark states in his conclusion, "...in nominal (non-agile) flight maneuvers there is little difference between fixed- and variable-pitch actuation."

So, if aerobatics or ultimate forward speed efficiencies are your goal, it seems the implementation of variable-pitch would be essential.  However, do note there will be a compromise incurred for this luxury in the form of significantly increased mechanical complexity and concomitant reductions in reliability.

With regards to tip mach number, I recall the now somewhat ancient pop-culture metaphor, "doctor, it hurts when I do this...".

• ▶ Reply

Permalink Reply by Joshua Geating on April 25, 2012 at 3:02pm

I guess I haven't really stated my intentions clearly, and may have been misleading. I am not concerned with controllability beyond stability, which could be either RPM or pitch control (after reading that VP thread, I'm leaning towards pitch control :). Also, I am really not all that concerned with efficiency for efficiency's sake. My main goal is to bring about completely new capabilities for UAVs not possible without variable pitch rotors. Namely, to allow hovering with a sizeable payload, as well as fast forward flight.

With a fixed pitch tilt rotor, a plane with X payload at hover could only reach X mph in forward flight, and the two are related to each other by pitch (at least partially). A fast, high pitch plane that performs well at high advance ratios would hover very inefficiently, meaning a lower payload. Furthermore, even a plane with large or wide, efficient wings cannot fly very fast due to drag.  

A plane with large multi-blade variable pitch rotors could attain an efficient, high-payload hover, as well as support very low drag, swept wing, high velocity flight. Essentially, it could have not mediocre performance across all velocities, but excellent performance at both high and low velocities (haven't thought much about mid where the wing loading is too high).

• ▶ Reply