VOL. 5, NO. 1, JAN.-FEB. 1981 J.

ENERGY 39

AIAA80-0655R

Aerodynamic Performance of the DOE/Sandia 17-m-Diameter

Vertical-Axis Wind Turbine
Mark H. Worstell*
Sandia National Laboratories, Albuquerque, New Mexico

The aerodynamic performance of the DOE/Sandia 17-m-diam Darrieus wind turbine is presented for two
rotor configurations: 1) two strutted composite structure blades using a NACA-0012 airfoil section with a 21-in.
(0.533-m) chord; and 2) two unstrutted extruded aluminum blades with a 24-in. (0.610-m) chord NACA-0015
section. Significant increases in overall performance were obtained for the extruded blades which represent a
currently viable low cost blade construction technique applicable to the Darrieus wind turbine.
Downloaded by UNIVERSITY OF QUEENSLAND on September 5, 2013 | http://arc.aiaa.org | DOI: 10.2514/3.62496

Introduction turbine rotor. This is corrected by a wind shear exponent of

T HE DOE/Sandia 17-m-diam vertical-axis wind turbine
(VAWT) has been in operation since March 1977. This
machine is being funded by the Department of Energy to
provide field-test data from a large Darrieus wind turbine to
augment ongoing analytical studies, and provide operational
experience. Data being collected from this wind turbine fall
into four basic categories: 1) aerodynamic performance; 2)
dynamic and static blade loading; 3) drive-train dynamics;
and 4) tiedown cable dynamics. This paper will address
aerodynamic performance.
To date, the DOE/Sandia 17-m VAWT has been tested in
three rotor configurations: two strutted blades, three strutted
blades, and two unstrutted blades. Figures 1 and 2 depict the
strutted and unstrutted two-bladed configurations. The struts
are structural support elements identical in both shape and
structure to the rest of the blade. They extend from the center
of the rotor tower to a point at approximately 67% of the
turbine maximum radius. The strutted configurations utilized
pin joints in all blade connections; the unstrutted extruded
blades used clamped end fittings.
Figure 3 illustrates the drive train of the 17-m turbine. The
basic specifications of the 17-m are found in Table 1. In
considering the nature of this paper, two distinct areas of
Table 1 standout:
1) The composite blades are a NACA-0012 with a 21-in.
(0.533-m) chord. The extruded blades are a NACA-0015 with
a 24-in. (0.610-m) chord.
2) The original speed increaser and right-angle drive used in
all strutted blade testing was replaced by a right-angle drive
parallel shaft gearbox used in the extruded blade testing. This
introduced slight changes in rotor operating speed.
When directly comparing the aerodynamic performance
data of the strutted and unstrutted blades, both 1) and 2) will
present some difficulties. However, since it is the overall goal
of national wind-energy programs to minimize the cost of
wind energy, a direct aerodynamic performance comparison
can be quite informative since the extruded blades represent
current low-cost technology that may reach $2/lbm (0.454 kg)
of blade. In the case of the 17-m VAWT, the extruded blades
cost significantly less than the original composite blades.
Aerodynamic performance testing of the DOE/Sandia 17-
m VAWT has been established through the method of
"Bins."1 Windspeed is measured by two anemometers
located at a 94.0-ft (28.7-m) height immediately above the
0.1 to correspond to the rotor centerline height of 44.0 ft (13.4
m). 2 Air density is calculated through measured air tem-
perature and barometric pressure.
Turbine rotor output is measured by a torque sensor located
immediately below the rotor as seen in Fig. 3. The tare torque
(friction torque) of the rotor thrust bearings is added to the
measured rotor output torque. This value is equal to 1.71 kW
at 42.0 rpm of the rotor. In all testing, the rotational speed of
the turbine rotor is essentially constant which is maintained by
the induction generator (the synchronous generator was not
used in these particular tests). It should also be noted that all
aerodynamic performance data presented are corrected to an
air density of 0.0625 lbm/ft 3 (1.00 kg/m 3 ). This corresponds
to a standard day of 60°F (15.6°C) temperature, 830 mb
barometric pressure in Albuquerque, New Mexico [elevation
5440 ft (1660 m)], the location of the test site. The per-
formance data obtained in the manner described above
represent the aerodynamic performance of the turbine rotor
and does not include drive train power losses.
The first rotor configuration to be tested was that of two
strutted blades. Testing covered a period of 10 months
beginning in March 1977. A summary of this test data is
presented in Table 2. Here, Cpmax denotes maximum rotor
efficiency and Xis the tip speed ratio, Ru/V^. Noticeable in
Table 2 are several basic trends:
(1) Breakeven windspeed increases with turbine rotational
speed.
(2) Cp tends to reach its highest level at 37 to 42 rpm.
(3) Pealc turbine output increases with turbine speed, oc-
curring at progressively higher windspeeds.
One noticeable discrepancy is Cp at 45.5 rpm. This was

Presented as Paper 80-0655 at the AIAA/SERI Wind Energy

Conference, Boulder, Colo., April 9-11, 1980; submitted May 9, 1980;
revision received Aug. 15, 1980. Copyright © American Institute of
Aeronautics and Astronautics, Inc., 1981. All rights reserved.
*Member, Technical Staff. I DOE/Sandia 17-m VAWT, strutted blades.
 40 M.H. WORSTELL J. ENERGY

most likely due to some form of error in the instrumentation
or data acquisition. Reference 3 contains the entire test results
for this configuration and should be consulted for a more in-
depth presentation of the data.
In March 1979, the DOE/Sandia 17-m VAWT was fitted
with two extruded aluminum blades. It was at this time that
the original speed increaser was replaced. This was done in
order to test the parallel shaft gearbox which appears to be
attractive for future VAWTs. Testing of the two extruded
blades is still in progress. The results obtained to date are
shown in Table 3.
The extruded blades demonstrated the same basic trends as
described for the strutted blades with the exception that Cp
occurred for 33.7 and 38.7 rpm. In comparing the extrucfecl

THRUST BEARING

GUY CABLE

ROTOR
Downloaded by UNIVERSITY OF QUEENSLAND on September 5, 2013 | http://arc.aiaa.org | DOI: 10.2514/3.62496

THRUST BEARING

TORQUE SENSOR
FLEXIBLE COUPLING
SYNCHRONOUS GENERATOR
SPEED INCREASER CLUTCH

RIGHT ANGLE DRIVE

FLEXIBLE C O U P L I N G

Fig. 2 DOE/Sandia 17-m VAWT, unstrutted blades. Fig. 3 DOE/Sandia 17-m VAWT drive train.

Table 1 DOE/Sandia 17-m VAWT specifications

Original specifications Present specifications (if different)
Rotor diameter 54.9 ft (16.7 m)
Rotor height 55.8 ft (17.0 m)
Swept area 2014 ft 2 (187 m 2 )
Ground clearance 16.0 ft (4.88 m)
Overall height 110 ft (33.5 m) 94 ft (28.7 m)
Operating speed 29.8-52.5 rpm 29.8-54.8 rpm
Number of blades 2or3
Rotor solidity 0.14-0.21 0.16-0.24
Blade manufacturer Kaman Alcoa
Blade material Fiberglass/Honeycomb/Aluminum extrusion Aluminum extrusion
Airfoil section NACA-0012 NACA-0015
Chord length 21.0 in. (0.533m) 24.0 in. (0.610m)
Use of struts Yes No
Blade length 79 ft (24.1m)
Blade shape Straight/circular/straight
Blade joints Pinned Rigid
Blade weight (each) 7131bm(323kg) 1370 Ibm (621 kg)
Strut weight (per blade) 46 Ibm (202 kg)
Speed increaser 3-stage planetary 3-stage parallel shaft
Manufacturer Crichton Philadelphia gear
Speed increaser ratio 42.9:1 35.58:1
Belt drive ratio to motor 1.42:1 to 0.8:1 1.7:1 to 0.92:1
Motor/generator
(induction) 50 hp (37.3 kW) squirrel cage 75 hp (55.9 kW) squirrel cage
(synchronous) 75hp(55.9kW)
Brake Dual independent 30.0 in. (0.762-m) disc
Brake torque capacity (each) 53,000 ft-lb (7330 kg-m)
Tower O.D. 20 in. (0.508m)
Tower I.D. 18 in. (0.457m)
Number of guy cables 4
Cable angle (to horizontal) 35deg
Cable diameter 1 in. (2.54cm)
Cable pretension 12,000-18,000 Ibf (5440-8160 kg) (5440-8160 kg)
Cable length 129 ft (39.3 m)
 JAN.-FEB. 1981 17-M-DIAMETER VERTICAL-AXIS TURBINE 41

Table 2 17-m aerodynamic performance test summarya

A ^2.2 rpm, Unstrutted
Breakeven V, Pmax atV
O ^2.0 rpm, Strutted
rpm mph C atX kw mph
/>max
29.6 8.2 0.342 5.04 7.4 15.9
33.6 8.2 0.361 5.27 12.3 17.8
37.0 8.2 0.376 5.37 18.5 22.6
42.0 9.1 0.370 5.67 28.9 25.5
45.5 10.1 0.303 4.57 37.0 27.4 A
48.4 10.1 0.341 6.12 47.9 30.3 O
52.5 10.1 0.340 5.02 57.3b 31.3
a
Test site elevation 5440 ft (1660 m); 30-ft (9.14-m) wind reference height;
two blades with struts. b Higher winds are needed here to ascertain peak value.

Fig. 5 Two-bladed rotor efficiency vs tip speed ratio at 42.2 rpm,

Table 3 17-m aerodynamic performance8 unstrutted and 42 rpm, strutted.

Breakeven V,
rpm mph Cpaf t atJT kw mph
Downloaded by UNIVERSITY OF QUEENSLAND on September 5, 2013 | http://arc.aiaa.org | DOI: 10.2514/3.62496

33.7 6.3 0.453 5.74 18.2 19.7

38.7 6.3 0.454 6.07 27.6 23.6 A 33.7 rpm, Unstrutted
42.2 8.2 0.407 5.33 39.3 26.5 O 33.6 rpm, Strutted
46.7 9.2 0.409 5.54 Incomplete
50.6 9.2 0.409 5.66 Incomplete
a
Test site elevation 5440 ft (1660 m); 30-ft (9.14-m) wind reference height;
two unstrutted blades.

°°oooooo
A
o
~
A 33.7 rpm, Unstrutted
O 33.6 rpm, Strutted A A £ A
AO
A0

bo°
8 16 24 32 40 48 56
Km/hr

Fig. 6 Rotor output vs centerline windspeed at 33.7 rpm, unstrutted

and 33.6 rpm, strutted.

Fig. 4 Two-bladed rotor efficiency vs tip speed ratio at 33.7 rpm,

unstrutted and 33.6 rpm, strutted.

blades to the strutted composite blades, several trends become ~)OOOOO

noticeable:
1) Breakeven windspeed for the unstrutted blades is lower
than for the strutted.
2) Cpmax for the unstrutted blades is significantly higher.
3) TrTex unstrutted blades produce a higher peak power
output.
To provide insight into these trends, Figs. 4-7 are presented. ^, A ©

These comparisons between the strutted composite blades and

the unstrutted extruded blades were performed at essentially
the same rotational speed. Figures 4 and 5 indicate that the
extruded blades provided a higher Cp at all tipspeed ratios for
the rotational speeds presented. The direct implication of this
is better shown in Figs. 6 and 7 which plot turbine rotor
output power as a function of windspeed. This type of plot is
more reliable than a Cp plot because of the latter in-
corporating a V3 calculation which will clearly magnify any
windspeed measurement errors.
In examining Figs. 6 and 7, the most highly noticeable
characteristic is the higher power output of the unstrutted
extruded blades in medium to high winds. Another distinct
difference is that the windspeed for power regulation (that
windspeed where the turbine power output begins to level off)
is moderately higher for the extruded blades. In low wind-
speeds, the differences become less noticeable.
Fig. 7 Rotor output vs centerline windspeed at 42.2 rpm, unstrutted
and 42 rpm, strutted.
The differences exhibited between the strutted composite
blades and the unstrutted blades can be attributed to several
causes. First of all, the struts themselves operate in the wake
of those blade sections extracting the major portion of power
and may literally become drag devices. Of particular interest
is the joint between the struts and the blades where it is
suspected a substantial amount of drag could be generated.
The aerodynamic performance of the Darrieus VAWT, like
the majority of wind turbines, is quite adversely affected by
 42 M.H. WORSTELL J. ENERGY

the presence of drag. This would account for the overall
increase in the performance of the unstrutted blades when
compared to the strutted.
A second possible cause is the airfoil sections themselves.
The extruded blades utilize a NACA-0015 as compared to the
NACA-0012 of the composite blades. This introduces a
change in maximum lift coefficients. Coupled with this, the
larger chord of the extruded blades constrains the blades to
operate at a higher Reynolds number for the same rotor speed
as to the strutted composite blades. These effects would
account for the turbine regulation point moving toward
higher windspeeds for the unstrutted blades and the peak
power output being higher.
In conclusion, there is a clear difference exhibited between
the aerodynamic performance of the DOE/Sandia 17-m
VAWT fitted with unstrutted extruded aluminum blades
when compared to the strutted composite blades. The in-
creased aerodynamic performance of the extruded blades is
Downloaded by UNIVERSITY OF QUEENSLAND on September 5, 2013 | http://arc.aiaa.org | DOI: 10.2514/3.62496
quite encouraging, particularly in light of them representing a
viable low-cost means of blade construction for the Darrieus
VAWT.
Acknowledgment
This work was supported by the U.S. Department of
Energy under Contract DE-AC04-76DP00789.
References
^kins, R.E., "Performance Evaluation of Wind Energy Con-
version Systems Using the Method of Bins—Current Status," Sandia
National Laboratories, Rept. SAND77-1375, March 1978.
2
Akins, R.E., "Wind Characteristics at the VAWT Test Facility,"
Sandia National Laboratories, Rept. SAND78-0760, September 1978.
3
Worstell, M.H., "Aerodynamic Performance of the 17 Meter
Diameter Darrieus Wind Turbine," Sandia National Laboratories,
Rept. SAND78-1737, January 1979.

From theAIAA Progress in Astronautics and Aeronautics Series...

RADIATION ENERGY CONVERSION IN SPACE—v. 61

Edited by Kenneth W. Bill man, NASA Ames Research Center, Moffett Field, California

The principal theme of this volume is the analysis of potential methods for the effective utilization of solar energy for the
generation and transmission of large amounts of power from satellite power stations down to Earth for terrestrial purposes.
During the past decade, NASA has been sponsoring a wide variety of studies aimed at this goal, some directed at the physics
of solar energy conversion, some directed at the engineering problems involved, and some directed at the economic values
and side effects relative to other possible solutions to the much-discussed problems of energy supply on Earth. This volume
constitutes a progress report on these and other studies of SPS (space power satellite systems), but more than that the volume
contains a number of important papers that go beyond the concept of using the obvious stream of visible solar energy avail-
able in space. There are other radiations, particle streams, for example, whose energies can be trapped and converted by spe-
cial laser systems. The book contains scientific analyses of the feasibility of using such energy sources for useful power gen-
eration. In addition, there are papers addressed to the problems of developing smaller amounts of power from such radia-
tion sources, by novel means, for use on spacecraft themselves.
Physicists interested in the basic processes of the interaction of space radiations and matter in various forms, engineers
concerned with solutions to the terrestrial energy supply dilemma, spacecraft specialists involved in satellite power systems,
and economists and environmentalists concerned with energy will find in this volume many stimulating concepts deserving
of careful study.

690pp., 6x9, illus., $24.00Mem. $45.00List

TOORDER W R I T E : Publications Dept., AIAA, 1290 Avenue of the Americas, New York, N. Y. 10019