IE010310A

Chapter 2
Drawing The Line

As an exercise of some curiosity it was decided to look through a few selected published works within the body of the literature involving wind generator blade aerodynamics and see if the words "airflow deflection" or just "deflection" occurred anywhere within, not that any particular importance can be ascribed to these terms at this point in this presentation. Here is what was found.

Albuquerque

The Sandia National Laboratories Report SAND90-1615/ UC-261 for unlimited release of February 1992 entitled "Selected Papers on Wind Energy Technology, January 1989-January 1990" by the Sandia National Laboratories Staff of Albuquerque, New Mexico includes several relevant papers, all in reference to the 34-meter Darrieus vertical axis machine that was placed in operation near Bushland, Texas. In every case where the documentation approached the question of understanding the blade force and power production quantities, the following sentences, or similar phraseology, almost word-for-word, appeared:

"Turbine performance predictions have been made with the double-multiple streamtube computer code known as SLICEIT. This code is based on the CARDAA code developed by Paraschivoiu(5) and uses the Gormont dynamic stall model(6) as modified by Masse(7). It accounts for local Reynolds number effects, and is capable of analyzing the performance of rotors with multiple section profiles and step changes in chord length.......
References:
5. Paraschivoiu, I., "Aerodynamic Loads and Performance of the Darrieus Rotor," Journal of Energy, 6(6):406 (1982).
6. Gormont, R.E., "A Mathematical Model of Unsteady Aerodynamics and Radial Flow for Application to Helicopter Rotors," US Army Air Mobility R&D Laboratory, Vertol Division, Philadelphia, Pennsylvania, Report on Boeing-Vertol Contract DAAJ02-71-C-0045, May 1973.
7. Masse, B., "Description de Deux Programmes d'Ordinateur pour le Calcul des Performances et des Charges Aerodynamiques pour des Eoliennes A'Axe Vertical," Institut de Recherche de L'Hydro-Quebec, Report IREQ 2379, Varennes, Quebec, July 1981."

Not to betray any thoughts bordering on dissatisfaction with this, but it is clear that recourse was invariably taken to previous work codified into the computer. Test results, a subject frequently discussed, often coincided with this number-crunched information and discussions elaborated on cases in which it deviated somewhat. In other paragraphs of this treatment, mention was made of the "angle of attack" of the blades and the fact that its characteristic cyclic variation in this style of machine complicated the analysis to some degree. Other theoretical material was presented, especially the academically de rigueur contribution to the accumulating three-dimensional analysis on the subject, but all to less practical effect and of a second order nature.

What was most noteworthy overall is that the machine actually performed to the predicted specifications, obtained using these methods. It produced 500 KW of power in reasonable wind conditions and approximated the performance of horizontal-style machines of similar blade swept area. One of the graphs displaying these results is inserted below:

But no one, based on a superficial reading of these reports and trying to give them the benefit of the doubt, could explain how, for example, energy transfer occurs at the blades on a more theoretical level. The details of airflow and momentum transfer were buried within previous computer work and notions that are not easily explained, similar to the treatment of aircraft wing lift theory widely published and referred to previously. Things like the pitch angles of the blades (clearly an angle of, or very close to, zero degrees) never were reviewed as a subject worth exploring in any detail. A few cursory remarks were made about the various airfoil shapes chosen from a list of standards but without the benefit of any justification.

Rather, much scrutiny and many of the papers centered on the fatigue strength performance of the blades and their structural dynamics. The words "airflow deflection" or "deflection" were never found, subject, as the effort to do so remains, to our correction, if anyone cares to assist us in this minor item of research. Some anecdotal information was provided that was left unexplained as follows:

"Later is the spring of 1990 the blades became contaminated with bug residue. A set of data was collected at 28 rpm with the dirty, faired blades (4/90) and is compared to the clean, faired blade data..... Here we observe that the dirty blades exhibit lower performance in winds to 11 m/s, but then significantly outperform the clean blades in winds greater that 11 m/s. This increased performance in high winds by the dirty blades is the opposite of anything observed before on either HAWTs or VAWTs. This behavior is not clearly understood, but it is speculated that the bugs are acting like tiny vortex generators. Further study is underway to understand this phenomenon.
(SANDI91-2228 Unlimited Release of July 1992, "Measured Data For The Sandia 34-Meter Vertical Axis Wind Turbine" by Thomas D. Ashwill, Wind Energy Research Division, Sandia National Laboratories, Albuquerque, New Mexico, par. 3.3.3, pg.32)"

Other discussions on matters of the effects of paint flaking from the blade leading edge surfaces and irregular, discontinuous flow interruptions such as bolt heads and nuts at blade joint connections were provided but with results of no great substance. The paint flaking story appears several times and in the tone of a mea culpa but no parallels were referenced to other instances of such effects elsewhere in the field.

Italy

Next is a paper presented at the American Wind Energy Association conference of 1994 in Minneapolis, Minnesota, entitled "Improvements of Aerodynamic Experimental Measurement on Airfoil Sections Fit For Wind Turbine Blades, Extended to High Angles of Attack", given by S. D'Angelo and C. Ligorio of 'Politecnico di Torino' University, of Torino, Italy. It starts on page 251 of the proceedings of the conference, which bears no copyright and states only that those wishing a complete copy may contact the AWEA in order to obtain one.

In this paper the authors make a sincere effort to look at something not addressed in sufficient detail in previous work, a sort of comeuppance on what was available from other, more wealthy nations. Italy, after all, was not playing a major role in wind energy technology development at the time and questions about the future of such endeavors there were clouded in doubt.

What this says is that someone needs to look at the larger "angles of attack" required for wind generator blades vice those more typical of aircraft wings. What they are referring to, of course, is the pitch angle of the blades and, ultimately, the airflow deflection arising therefrom. But to be noticed, again, is that a careful review of the paper discloses never any mention made of the word "deflection". It is, not to seem overly wrought by this detail, conspicuous by its absence.

Worldwide, it seems, aerodynamic theoreticians, as a body, never, ever own up to any effect that airfoils capable of generating lift have on the airflow. The flow proceeds on undisturbed at the trailing edge in the manner of thinking so supported and, hence, is not subject to anything that can be termed "deflection", even at high so-called "angles of attack". On a more macro scale, the wind impacting on the entire blade swept area does see a velocity change and hence wind generators are subject to wake effects due to spacing and "array" considerations. But nothing like this happens at the more detailed level of the blades themselves or, at least, nothing of this nature is ever considered sufficiently important to justify study and analysis.

That said, it must be admitted that this paper was progress. Notice was given the wind energy community that higher angles of attack are needed, in the authors' own words at the outset of the Abstract:

"Wind turbine blades often operate at angles of attack beyond those usual in aeronautics."

What this paper did comes as no surprise. The surprise is that this point needed to be made at such a late date in wind energy development. What "high angle of attack" means physically, in the somewhat topsy-turvy fashion of the geometry involved, is a reduction in the pitch angle of the blade from a more positive angle to a less positive angle.

The Netherlands

One more attempt is made to find the word "deflection" in the technical literature. By now the pattern is set. This word just never seems to appear and no one would be surprised if it did not do so in the work of perhaps one of the premier wind energy aerodynamicists of the field, the one whose name appears to have become more closely associated with the technology than any other. The reference here is to material that has been published in leading wind energy technical journals under the authorship of Herman Snel of the Netherlands Energy Research Foundation, Petten, the Netherlands.

The particular item is a 24 page detailed, technical article that appeared in "Wind Energy" by Mr. Snel (ISSN 1095-4244), the Pilot Issue of Spring 1998, published by Wiley InterScience of John Wiley and Sons, Ltd., Chichester, West Sussex, UK. It was entitled, "Review of the Present Status of Rotor Aerodynamics", and can be found starting on page 46 of this issue. The entire journal issue and the article itself are protected under copyright and the subscription price for the issues published quarterly each year is quite steep, reflective of the prestige attached to their availability.

It must be recognized that material such as this takes for granted many of the ideas that may seem important on an elementary level. Even the parameter identified in other work as the "angle of attack" is only briefly noted here and then only in the glossary as the angle, a, which is seen in the work as the argument of certain coefficients. The treatment is typified, as most work of some erudition inevitably is, by attention to exceptions, unusual situations, and anomolous behavior. Much, for example, is made of flow separation and the phenomenon known as "stall". An eight page Appendix is incorporated right into the end of the article as a means of providing the basic equation systems for more commonplace and routine cases and for a wider public but even this is not readily digestible by anyone without years of experience in doing such analyses.

The nub of the matter and the reason why the word "deflection" is so important to be considered is that, without it, any flow analysis must make use of alternative methods of dealing with momentum transfer. So in this treatment, what is called "vortex generation" becomes a sort-of substitution for something that deserves to be identified with a plainer, more approachable term. It's all part of a larger concept known as "circulation" and identified with the Greek letter, G. Years of work are evident here and those involved deserve their accolades and support but questions remain.

Deflection. The word seems to be not a part of the vocabulary of those associated with wind energy theoretical aerodynamics. If it did not appear in the above three instances of material treating the subject, in all likelihood it does not appear anywhere in the larger body of literature of the field as well. Many concepts and ideas are discussed in this sort of material and many words used but the word "deflection" is not one of them. This is, after all, an amazing thing when it comes down to a common sense approach to this technology, a sort of discovery, an unexpected revelation that among all that has been accomplished maybe some concepts have yet to be explored.