841287--Structural Design of Sandia 34m VAWT

SANDIA REPORT SAND84–1287. Unlimited Release UC–60

Printed April 1985

Structural Design of the Sandia 34-MeterVertical-Axis Wind Turbine

Dale E. Berg

Prepared bySandia National Laboratories

Albuquerque, New Mexico 87185 and Livermore, California 94550

for the United States Department of Energyunder Contract DE-AC04-76DPO0789

sf-29(1(1(11,881)

Issued by Sandia National Laboratories, operated for t.be United StatesDepartment of Energy by Sandia Corporation.NOTICE: This report was prepared as an account of work sponsored by anagency of the United Static Government. Neither tbe United StatesGovernment nor any agency thereof, nor any of their employees, nor any oftheir contractors, subcontractors, or their employees, makes any warranty,express or implied, or aeeumee any legal liability or reepopeibility for tbeaccuracy, completmiess, or usefulness of any information, apparatus, product,or process disclosed, or representa that ita use would not infringe privatslyowned rights. Reference herein to any specific commercial product, process,or service by trade name, trademark, manufacturer, OFotherwise, does notnecessarily constitute or imply ite endorsement, recommendation, or favoringby the United States Government, any agency thereof or any of theircontractors or subcontractors. The viewe and opinione expressed herein donot necessarily state or reflect those of the United Statee Government, anyagency thereof or any of their contractors or subcontractors.

Printed in the United S@@S of AmericaAvailable fromNational Technical Information SeniceU.S. Department of Commerce5285 Port Royal RoadSpringfield, VA 22161

NTIS price codeePrinted copy A02Microfiche copy AO1

AcknowledgmentsMany people contributed to the work described in thisreport. I will not attempt to name them all, but I wishto acknowledge a few who made significant contribu-tions to the test bed project: Robert Grover andTom Gardner did the design work; Tom Ashwill didthe structural and detailed design analyses; DebbyOscar and Tim Leonard contributed to the structuralanalysis; Tim Leonard also provided the computersupport for the project. Paul Klimas performed thework on aerodynamic and drivetrain dynamics, andRobert Nellums was responsible for the variable-speed generator. Emil Kadlec, as project leader, coor-dinated the entire effort, and Richard Braasch, super-visor of the Wind Energy Research Division, gave ushis wholehearted support and acquired the funding tomake this project possible. George Tennyson of theDepartment of Energy, Albuquerque, has been anenthusiastic supporter in our requests for DOE ap-proval and funding.

Introduction ................................................................................................................................................ ... 7

Design Tools ..................................................................................................................4............................. ... 8

ECON16 ............................................................................................................................................ .......... 8

FEVD ............................................................................................................................................ .............. 8

FFEVD ................................................................................................................................................ ........ 9

Design Process .....................................................................................O.........•.................O...............C............. 10Option 1 ............................................................................................................................................ .......... 10

Option 2 ................................................................................................................................................ ...... 11

Option 3 ................................................................................................................................................ ...... 12

Evaluation ................................................................................................................................................ ... 12

Detailed Design .............................................................................................................................................. 12

Summary .........?...................................................................................................................................... ......... 15

References ................................................................................................................................................ ....... 16

Figures123456789

10111213

Comparison of NACA 0018 Airfoil Profile with SAND 0018/50 ................................................... 7Finite Element Model of DOE 100-kW Turbine ............................................................................. 9Fan Plot of Rotor Frequencies for DOE 100-kW Turbine ............................................................. 9Fan Plot and System Driving Frequencies for DOE 100-kW Turbine ........................................ 10Fan Plot for Option 1 .......................................................................................................................... 11Fan Plot for Option 2 .......................................................................................................................... 12Fan Plot for Option 3 .......................................................................................................................... 12Blade-to-Tower Joint and Clamp ...................................................................................................... 13Blade-to-Blade Joint ............................................................................................................................ 13Construction of a 1.22-m Chord NACA 0021 Airfoil Section ........................................................ 14Upper Bearing Design .......................................................................................................................... 14Fan Plot for the Test Bed ................................................................................................................... 15Artist’s Conception of Sandia’s 34-m Test Bed ............................................................................... 15

5-6


IntroductionSandia National Laboratories (SNL), as the lead

DOE laboratory for Vertical-Axis Wind Turbine(VAWT) development, has developed state-of-the-artcomputer codes to model and analyze the structuraland aerodynamic performance of VAWTS. Data fromthe SNL 17-m research machine (built in 1976) andthe DOE 100-kW machines (designed in 1978 anderected in 1980-81) were essential in developing ouranalytic models. We continue to gain data from thesemachines but, in general, the data is well predicted byour current codes. This technology has been success-fully transferred to industry, as evidenced by thenumber of companies now building or planning tobuild VAWTS in the 17-m size range.

Over the past few years, SNL has developed andtested a family of natural laminar flow (NLF) airfoilsspecifically designed for use on Darrieus-type (curvedblade) VAWTS. These airfoils were designed to have avery low and wide drag bucket with abrupt stallcharacteristics to enable the VAWT to produce morepower at low wind speeds (compared to VAWTS withNACA 00XX airfoils) and to flatten off the powercurve at higher wind speeds. The family consists ofsymmetric airfoils 15 “i, 18 Y;, and 21p; thick (SAND0015/47, iSAND 0018/50, and SAND 0021/50, respec-tively), all with natural laminar flow (at zero angle ofattack) over nearly the entire forward half of theairfoil. A comparison of the NACA 0018 profile withthe Sandia-developed SAND 0018/50 laminar flowprofile is made in Figure 1. Additional information onthe design and testing of these airfoils maybe found inthe Klimas and Berg report.l

— NACh 0018

------ SAND 0018/60

Figure 1. Comparison of NACA 0018 Airfoil Profile withSAND 0019/50

These NLF airfoils perform best at Reynoldsnumbers between 1 million and 5 million. At lowReynolds numbers, such as those encountered nearthe tower, the performance of the NLF airfoils is lesssatisfactory than that of airfoils in the NACA 00XXseries. Thus, to make use of the NLF airfoils, we needto construct a blade with an NLF profile near theequator and a NACA 00XX profile near the tower.Aerodynamic and structural considerations show thatwe require a larger chord airfoil near the tower than atthe equator, so we also need to consider a nonuniformchord (tapered) blade. We want to continue to use anextrusion process to fabricate our blades, and becauseit is difficult to extrude a blade with a nonuniformchord, we decided to consider the use of step-taperedblades with uniform chord blade sections between thestep changes in chord length.

The potential economic advantages of using lami-nar flow airfoils and step-tapered blades have beeninvestigated by Kadlec.2 His work was based on ex-perimental data for NACA 00XX airfoils and a combi-nation of preliminary experimental and analyticaldata for the SAND 0018/50 and SAND 0021/50 NLFairfoils. Kadlec found that, with respect to a constant-chord NACA 0015 blade machine, a step-taperedblade machine using an NLF profile near the equatorand a NACA 00XX profile near the tower would resultin a significant decrease in the cost of energy (COE).

Preliminary work also indicated that the use ofNLF airfoils may result in a sizable decrease in themagnitude of the leadllag vibratory loads experiencedby the turbine blades. Certainly, the cyclic loads inhigh winds will be reduced significantly, for theseairfoils stall at a lower angle of attack, and the maxi-mum lift and drag acting on them are less than thecorresponding forces acting on a NACA 0015 blade. Ifthese blades do actually decrease the vibratory loads,the blade lifetimes may be extended, resulting in afurther decrease in the COE relative to a conventionalVAWT.

Further studies have shown that a 34-m diameterturbine would enable us to investigate the full benefitof the tailored blade concept, for the potential advan-tages increase quite rapidly with size up to 34-m and

7

then level off. This size machine would also provide uswith a new standard against which we could validateour structural and aerodynamic codes, a machine forwhich the gravitational and stochastic wind effectswould be much more important than any we have todate.

Considerable interest has been expressed in thepotential benefits of utilizing continuously variable-speed generators with wind turbines. With such agenerator, the rotational speed of the turbine could bevaried with wind speed to maximize energy capture.The only way to quantify any benefits that might berealized with variable-speed operation is to build amachine specifically designed to operate in such amode, since no existing machine has an operating rpmrange of any size that is free of resonances.

The Wind Energy Technology Division of theDepartment of Energy has funded Sandia NationalLaboratories to design and build a 34-m researchVAWT. This machine will incorporate the tailoredblade concept described above, be capable of continu-ously variable-speed operation, provide structural andaerodynamic data against which to validate our codes,and serve as a test bed for future VAWT research.

Design ToolsThroughout the design process we have main-

tained a strong interaction between the analyticaltools and our experienced hardware designers, pro-gressing through several sets of layout and conceptualdrawings as the design evolved. Our computer codeshelp us evaluate design concepts from structural andeconomic standpoints, but the experienced designer isstill indispensable.

The analytic design tools used in this effort com-prised three computer programs: an economic analysisknown as ECON16, a natural frequency analysisknown as FEVD (for Finite Element VAWT Dynam-ics), and a forced vibrational response analysis knowna~ FFEVD (for Forced Finite Element VAWT Dy-namics). A short summary of each program follows.

ECON16The design of a VAWT system must begin with

the specification of many system variables such asrotor diameter, rotor height-to-diameter ratio, bladechord(s), number of blades, rotor speed, blade shape,and blade cross-section profile and design. Bladecross-section design, in turn, includes the number,location and thickness of ribs, the thickness of theblade skin, and the amount of fill in the nose and tail

of the blade. The ECON16 code was developed atSandia to aid in selecting optimal combinations ofthese and other system components.3 It includes mod-els for estimating the costs (in 1978 dollars) of majorsystem elements and for estimating the total energyproduction of the system. ECON16 incorporates somemajor ground rules, including the following:

● Rotor to operate at constant rpm, controlled bythe utility grid through a synchronous or induc-tion generator

s Single rotating tower of tubular cross section,supported at the top by three guy cables

● Blade construction of constant cross section,thin-wall, hollow aluminum extrusions

Q Optimization based on minimizing annual sys-tem cost per unit of energy supplied

The model assumes that any turbine that it ana-lyzes has a 30-yr lifetime; it does not determine ananticipated lifetime or attribute an economic value toit.

FEVDFEVD is a finite element code based on NAS-

TRAN.4 The code is described in detail by Lobitz’ andCarrie et al.BIt requires a finite element model of thewind turbine including beam, rigid bar, spring, orconcentrated mass elements with the appropriatethree-dimensional properties. The model must accu-rately represent the tower constraints or supports(both top and bottom), the torsional stiffness of thedrivetrain, and the presence of bearings (modeled asconcentrated masses with appropriate element re-leases). In addition, the joint properties (blade-to-blade, tower-to-blade, and tower-to-tower) must becarefully determined. In some cases, one may need toconstruct a detailed three-dimensional model and an-alyze it extensively to determine appropriate jointstiffness for inclusion in the beam-type model. Ourfinite element model for the DOE 100-kW VAWT(Figure 2) is typical of the detail required. Moments ofinertia and material properties must be specified foreach element shown in the model, and the location ofeach grid point is required. A model may easily con-tain over 400 lines of variable specifications. TheFEVD code computes the rotating system effects forthis finite element model and generates the appropri-ate input data for the NASTRAN code. NASTRAN isthen used to compute the system’s natural frequenciesand mode shapes. The variation of the rotor responsewith rotor rpm is illustrated in a “fan plot” (Figure 3)for the DOE 100-kW machine. Only a few of the lowerfrequency, more important modes are plotted in Fig-ure 3, The name attached to each mode refers to its

8

sha~e at zero r~m; as the rotor rpm increases, the model) are quite conservative for the 17-m machine,modes assume v&y’ complex shape; and it is difficultto label them. lF is the first flatwise (symmetric andasymmetric) mode, lPr is the first propeller mode, lBis the first butterfly mode, lT is the first tower mode,2F~ is the second flatwise asymmetric mode, etc. Someturbines contain two distinct tower modes character-ized by the tower moving either in the plane of theblades (tower inplane, or TI) or perpendicular to theplane of the blades (tower out-of-plane, or TO). Illus-trations of these mode shapes may be found in theCarrie et al report.G

FFEVDFFE’VD is also a NASTRAN-based, finite element

code. In Jfact, it uses the same finite element model ofthe VAWT that FEVD uses. FFEVD calculates theeffects of’ the rotating system and estimates the rota-tionally resolved wind loading on each element of theturbine blades using a version of the double multiplestreamtube aerodynamic code known as CARDAA.YThe rotating system effects and wind loadings areadded to the finite element model to create a NAS-TRAN input deck, and NASTRAN then calculatesthe turbine response to wind loading. The NASTRANoutput is, plotted as the ‘mean and vibratory stresslevels for each element in the model. Further informa-tion on FFEVD can be found in the Lobitz works

The CARDAA code includes the Gormont dy-namic stall model,g modified with the Masse’ correc-tion.l(] Thlis empirical dynamic stall model is summa-rized by 13erg.11As implemented in FFEVD, this codeincludes the capability to analyze a machine whoseblades contain step changes in airfoil chord length (upto six different chords) and in section profile (up to sixdifferent profiles).

CARDAA assumes a steady incident wind with avertical g:radient to model the earth’s shear layer. Weare fully aware that the incident wind is actuallystochastic, with significant variations in both direc-tion and velocity. We also know that as machinesbecome larger, the relative size of typical atmosphericgusts or eddies becomes smaller than the size of theturbine, and the effect of a stochastic wind becomesgreater. IIn fact, Lobitz]z has recently shown that forthe Mod 2 Horizontal-Axis Wind Turbine (HAWT),the stochastic wind must be included in the forcedvibration model or a significant resonant problem ismissed entirely. Preliminary investigations (with asingle multiple streamtube code) indicate that theseeffects are less severe for VAWTS than they are forHAWTS, ibut they may still be important. A compari-son of FFEVD results with experimental data*;+indi-cates that the code results (without a stochastic wind

but that is not to say that we will not see significantstochastic wind effects on a larger machine. We arecurrently working on a new aerodynamic code thatincorporates a stochastic wind model, but completionis probably a couple years away.

NODE 13,

2526.8 cm (994.8 ln)–

NODE 202.ELEMENT 203—

“’%ssNODE 239NODE 240

0 ELEMENT 1

+.0UYN2ADW0

-s—ELEMENT 101

\

‘NODE 101

-NODE 6

-ELEMENT 5

-NODE 5

/’/t40DE 139/NODE 140

NODE 2‘NODE 1‘9~lVE TRAIN

Figure 2. Finite Element Model of DOE 100-kW Turbine

5

4

:

;3

vzill302uuL

1

0

OPERATING RPM -

2B

.

.

.‘2FA .

.

..

.

..

:..I I I 1 . I I

o 10 20 30 40 50 60

ROTOR RPM

Figure 3. Fan Plot of Rotor Frequencies for DOE 100-kWTurbine

9

Design ProcessSince this is to be a research machine, several

special considerations must be incorporated into thedesign. The fact that we intend to use this machine toevaluate other blade designs in the future imposes therequirement of a very stiff tower design. The vibrationmodes and frequencies of a stiff tower are less sensi-tive to the exact blade design than a soft tower wouldbe, and the design of new blades should be much easierwith a stiff tower. The tower must also be designed toallow the blade/tower joint geometry to be readilychanged, for new blades may well require differentmounting angles. In addition, we will be investigatingthe potential advantages of a continuously variable-speed generator on this machine, so we cannot designit for operation at a single rpm; we must have a rangeof operating rpm in which the machine is free ofresonant conditions.

Updated NLF experimental data have been in-cluded in the ECON16 model since it was used byKadlec for his work,z which was expanded signifi-cantly during the basic design stage of this project.The basic procedure was to identify attractive config-urations with the ECON16 program and then to eval-uate their structural feasibility with the FEVD andFFEVD programs.

FEVD was used to determine the machine modesand frequencies of vibration as functions of rotationalspeed. The results were plotted as fan plots (Figure 4).Presentation of this data in this manner, overlaid withthe harmonic lines of the rotation frequency, enablesone to readily spot potential resonance problems. Theintersection of these per-rev (P) frequencies and thenatural rotor frequencies show where resonance prob-lems could occur. Some of the modes are basically“odd” and will couple only with odd per-revs (1P, 3P,etc.), whereas others are “even” and will couple onlywith even per-revs (2P, 4P, etc.). For example, the firsttower ( lT) mode shown in the figure is an odd modeand will not be driven by 2P, 4P, etc. In addition, thehigher per-revs (5P or above) do not contain sufficientenergy, in general, to significantly drive a resonantcondition. We have found that some mode crossingsare very benign and may be safely ignored, whereasothers are potentially catastrophic. Examples of thelatter are the lB/lP, the lT/3P, and the lFs/2Pcrossings.

Once we located a potentially dangerous modecrossing, we used FFEVD to determine the antici-pated rotor response both near the mode crossing andwell away from it, where there should be no resonance.The mean and vibratory stress levels calculated byFFEVD indicated how badly the mode coupled with

the driving frequency near the mode crossing (howbad the resonance was). If the mode coupled toostrongly, we tried to alter the turbine design to shiftthe crossing out of the operating range. If the stresslevels away from resonance were too high, we modifiedthe design of turbine blade sections, or changed thebasic design of the turbine to attempt to reduce thoselevels. ,4fter the necessary structural modificationshad been incorporated into the model, we returned tothe ECON16 code to reoptimize it.

Three design options were pursued throughoutthe basic design process. Each option is discussed insome detail below, and an evaluation follows.

6

4

:

:3

0zw

<2Illau.

1

0

OPERATING RPM.“

20

k

IT .

lB .

.

lF

e:~s”~.: . . ...””””””””””-” :

.

5P

4P

3P

2P

1P

o 10 20 30 40 50 60

ROTOR RPM

Figure 4. Fan Plot and System Driving Frequencies forDOE 100-kW Turbine .

Option 1This was the first design to be pursued, and we

chose a conservative design with extremely rigidblades that would not need struts of any type. InOption 1, we wanted to design a machine that we wereconfident could be built, without paying too muchattention to the comparative COE. We would thenmove onto alternate designs that would address COE.The ECON16 code and structural considerations bothindicated that we should consider a step-taperedblade with a longer chord near the blade/tower sup-port and a shorter chord near the equator. We tried toavoid any blade flatwise resonance problems in Op- .tion 1 by driving those frequencies above the lowerper-rev driving frequencies. We also needed to keepthe first TI mode above the 3P line throughout theoperating range of -27 to 42 rpm.

The results of our efforts are shown in the fan plotin Figure 5. We were unable to avoid a lF crossing ofthe 3P line at 31 rpm, a lB crossing of the 5P line at 35

10

rpm, and a 2F~ crossing of the 5P line at 34 rpm— allexhibiting some degree of resonance. The lF/3P cross-ing defined the lower end of the operating range; theupper end was limited by the lTIK’ crossing at 47

rpm. The crossings at 34 and 35 rpm meant that wecould not operate the turbine close to these rates ofrpm; therefore we had two small operating windows(31-33 and 37-44 rpm) rather than one large one. Thisdesign evolved into a machine having a height-to-diameter (H/D) ratio of 1.4, a 3.05-m (10-ft) diameter,a 0.95-cm (0.375 -in.) wall steel tower, and a blade thathad a 1.83-m (72-in.) chord NACA 0021 profile nearthe tower, a 1.07-m (42-in.) chord SAND 0021/50profile in the intermediate region, and a 0.91-m (36-in.) chord SAND 0021/50 profile at the equator. Thefirst column in Table 1 summarizes the characteristicsof Option 1.

.—

Table 1. Option Summary

Option 1 Option ‘2 Option 3

Diameter, m (ft)

HID

Operating Range,rpm

Operating Mode

Annual Energy,YIWh

AirfoilsRootIntermediateCenter

Chord, m (in.)RootIntermediateCenter

Struts

Solidity, ?,

34 (110) 34 (110)

1.4 1.3

31-33 30-34

37-44 36-42

Variable VariableSpeed Speed

1.23 1.13

NACA0021 NACA0021SAND21/50 SAND 18/50SAND21/50 SAND18/50

1.83 (72) 1.22 (48)

1.07 (42) 1.07 (42)

0.91 (36) 0.91 (36)

None MiniRigid

15 12

34 (110)

1.2

29-32

36-42

Variable

Speed

0.96

NACA 0018

SAND 18/50

SAND 18/50

1.07 (42)

0.91 (36)

0.76 (30)

DeepCable

10

All Options have a steel tower, 3.05 m (10 ft) in diameter,with a 0.95-cm (0.38 -in.) thick wall.

5 I I 1 1 1 5P

,/’

/

4

1

//

>4P

/ /’

//

/0

la\

*7O

~--1P, ----

_---—

o~0 20 40 60

ROTOR RPM

Figure 5. Fan Plot for Option 1

Option 2With Option 2 we took a far less conservative

approach than with option 1. We did not like the largechord that evolved to meet the requirements of Option1, so we decided Option 2 would use a smaller chordblade near the root and shallow rigid struts, if neces-sary. We sought to keep the struts near the blade-to-tower joints to minimize their effect on the aerody-namic performance of the machine. Our experiencewith deep rigid struts on our 17-m research turbinehas made us very aware of how detrimental they canbe. With a shallow strut and an arbitrarily imposedconstraint that we would limit the chord of the bladeat the root to 1.22 m (48 in.), we found we could notdrive the blade lF frequencies” high enough to avoidthe 3P frequency throughout the operating window.FFEVD indicated that the lFA mode would be drivenby a 3P crossing, and the lFS mode would be driven bya 2P crossing. We elected to tailor the blade responseto lie between the 2P and 3P lines throughout theoperating window. We also shortened the tower tokeep the lTI mode from crossing the 3P line in thedesired operating range.

Figure 6 shows the fan plot that resulted from ourefforts on this design. As you can see, we ended upwith a IF crossing of the 3P line at 28 rpm, a lBcrossing of the 3P line at 35 rpm, and a lTI crossing ofthe 3P line at 43 rpm. The lF/3P crossing effectivelydefined the low end of the operating range; the lTIcrossing defined the upper end of that range. FFEVDindicated that the lB/3P crossing was not a severeone, but we still would not want to operate right on topof it. Again, it effectively split our operating range andwe had two small operating windows rather than onelarge one. This design evolved into a machine with anH/D ratio of 1.3 and a blade having a 1.22-m (48-in.)chord NACA 0021 profile at the root, a 1,07-m (42-in. )

11

chord SAND 0018/50 profile in the intermediate sec-tion, and a 0.91-m (36-in.) chord SAND 0018/50 pro-file at the equator. It used the same tower as Option 1and shallow (0.9 m or 2.9 ft toward the equator fromthe blade-to-tower joints) rigid struts. Option 2 issummarized in the second column of Table 1.

5 I 1 I I 5P1

,/’

/

4 /-

//

2171

●“ 31To

vzu

: 2FA 2F8:2 / . /

\ / /lB

1lFA. <F, .-=

)Pr--” ---

01 1 1 I 1 I I0 20 40 60

ROTOR RPM

Figure 6. Fan Plot for Option 2

Option 3Our philosophy in the design of Option 3 was to

use blades that were soft in flatwise bending and toincorporate deep cable struts to boost the blade IFfrequencies above the 3P line. Cable struts seem to befar less detrimental than deep rigid struts to theaerodynamic performance of a turbine and have beenused successfully in the past. We again had to shortenthe tower to get the lTI/3P crossing out of the desiredoperating range.

The Option 3 fan plot is shown in Figure 7. Thecable struts did push the lF modes up high enoughthat they did not cross the 2P or 3P lines in theoperating range. The lB crossing of the 3P line, how-ever, was still a problem. FFEVD indicated that thiscrossing was much more severe for Option 3 than forOption 2, probably because the blades for Option 3 areless rigid in the lead/lag direction than the blades forOption 2. Once more, we ended up with two smalloperating windows rather than one large one. Option 3was a turbine with a 1.2 H/D ratio and a blade having

a 1.07-m (42-in.) chord NACA 0018 profile at the root,a 0.91-m (36-in. ) chord SAND 0018/50 profile in theintermediate area, and a 0.76-m (30-in.) chord SAND0018/50 profile at the equator. It also used the sametower as Option 1. Option 3 is summarized in the thirdcolumn of Table 1.

6 1 I 1 1 5P1

/’/

4

i!2s.

110>“ 3v~ \a 111$

a21FA,9FsIL

30

,/”1 -/ . 1P

1 Pr~-=

------

-----\

o~

0 20 40 60

ROTOR RPM

Figure 7. Fan Plot for option 3

EvaluationOnce these conceptual designs were completed, we

evaluated the three options we had developed in orderto select one configuration that we would continue towork on. We compared the stress levels throughoutthe 25- to 42-rpm range, especially at the mode cross-ings within the desired operating window. We deter-mined the operating windows that existed with thosemode crossings and estimated how difficult it wouldbe to eliminate the resonance problems, either byeliminating the crossing or by changing the structuraldesign so the mode would not be so readily driven. Wecalculated the COE for each design, examined ourmodeling confidence for each of the three options, andestimated the difficulty of fabricating each of them.

Option 1 was eliminated rather quickly. Its COEwas much higher than for the other two options, thelarge chord blade near the root would require at leastthree and possibly four extrusions and would be diffi-cult to fabricate, and it would be difficult to move orcontrol the resonant mode crossings within the de-sired operating window.

Options 2 and 3 had very comparable COE fig-ures, and the fabrication difficulty appeared to beabout the same. We felt, however, that the more severelB/3P resonance problem of Option 3 would be moredifficult to mitigate, and we had greater confidence inthe modeling of Option 2 with the rigid struts than wehad in the modeling of Option 3 with the cable struts.Therefore, we selected the Option 2 configuration forfurther detailed design work.

Detailed DesignOnce we selected Option 2 as the basic design with

which we would continue, we got down to the realdetails of how, precisely, we would build this machine.

12

Exactly how would we attach the blade to the tower?How about blade-to-blade joints? How would we con-struct the blade from multiple extrusions? Whatwould be the best tower-to-blade angles? These and ahundred more questions must be answered before aturbine is actually built. I will mention here some ofthe major features that we have established at thistime.

After further design studies of Option 2, we de-cided to eliminate the struts completely, using morerigid tower-to-blade joints and longer blade clamps tocontrol the blade flatwise frequencies. Sketches of thetower-to-blade joint are presented in Figure 8. Thesame de:sign will be used for the upper and lowerjoints. These joints actually serve two functions: astransitions from the large torque tube to the smallertorque shafts that pass through the upper and lowerbearings, and as mounts for the blade clamps. Theblade clamps are attached to the transition flanges byplates on the leading and trailing ends of the clamps.The blade’s mounting angle may be changed by sim-ply replacing the existing plates with plates havingclamp-mounting holes drilled for the desired angle. Inaddition, the use of other blade profiles or chords willrequire orily the replacement of the blade clamps andmounting plates. The blade end of each blade clamp istapered to decrease the stress concentration that willoccur in the blade at the end of the clamp. By elimi-nating the struts and reworking this joint, we wereable to drop the lF/3P crossing from 28 to 26 rpm,thus expanding our operating window slightly.

The blade-to-blade joint is illustrated in Figure 9.It is an external joint, similar to those we are currentlyusing on our 17-m research turbine NLF blade. Thesleeve portions of the joint will be extruded, bent tothe proper shapes for each airfoil, and bolted securelyto those airfoils. The two sleeves will then be attachedto either side of a 7.6-cm (3.O-in.) interface plate.Although it is not shown in Figure 9, the area immedi-ately on each side of that plate will be aerodynamic-ally faired with nonstructural material. This type ofjoint is relatively cheap to fabricate, is easy to install,and will allow us to interchange blade sections readily.Wind tunnel and 17-m turbine tests indicate thatexternal joints such as these do not seriously degradethe aerodynamic performance of the machine.

6+.++...,...+...:.+++...+.+....++..+++..

/SLEEVE

. ++.....++. .+. . . . . ..+.

++..++.+. ++++++++

++..++,++

++, .++,+ +++.+.+,

.+ ...+..+

+++...++ :+ .++++.+

I +.++..+++

.. +++ +.+. . ., ++++.,++

.+ .++.+

+.++..++ .+ ..+.+.+..,.+

,, .+.,....+.. ..0

tINTERFACE PLATE

Figure 9. Blade-to-Blade Joint

The blade sections will be constructed of multipleextrusions (Figure 10). The maximum size of eachextrusion is limited by the size of the extruding pressavailable; the largest single extrusion blade that canbe fabricated today is -0.74 m (29 in.) in chord. The1.22-m (48-in.) chord NACA 0021 blade profile shownin this figure will be fabricated from three extrusions,all with 0.79-cm (0.31 -in.) wall and rib thicknesses.The 1.07-m (42-in.) and 0.91-m (36-in.) chord SAND0018/50 blade sections will use two extrusions each,with 0.64-cm (0.25 -in.) wall and rib thicknesses. Thenose of each blade section is 1.27 cm (0.50 in, ) thick,

B.Jand the tail is filled, as illustrated in the drawing.Current plans call for the extrusions to be boltedtogether, but we are not irrevocably committed to that

Figure 8. Bl[ade-to-Tower Joint and Clamp method.

13

“a‘--...>*-1‘i: 3E-”—.-.—1’

Figure 10. Construction of a 1.22-m Chord NACA 0021Airfoil Section

We have elected to use an aluminum tower 3.05 m(10 ft) in diameter, with a 1.27-cm (0.50 -in.) wall,rather than the steel tower we developed for theconceptual design. The aluminum tower will allow usto reduce the stiffness of the guy cable by 25% andstill keep the first tower inplane mode at the desiredlevel. In addition, the difference in the coefficient ofthermal expansion for aluminum and steel, combinedwith the 2-to-1 difference in lengths of the steel cablesand aluminum tower, will result in much smallertemperature-induced cable tension changes than wewould see with a steel tower.

To obtain the necessary stiffness at the upperbearing of 78810 N/cm (45 000 lb/in.), we will usethree 8.9-cm (3.5 -in.) diameter guy cables -90 m (295ft) long. The size of these cables mandates the use ofhydraulic tensioning devices for each cable, but onceproper tension is obtained it will be maintained by amechanical device. We must keep the natural guycable frequency above the 2P turbine driving fre-quency at all rotational speeds to avoid cable reso-nance. The cable tension necessary to obtain suchfrequencies for cables of this size is excessive (595 000lb per cable). To reduce the cable tension required, weintend to use cable anchors to force a cable node one-third of the way up the cable and to utilize active cabledamping. These details will be worked out as thedesign progresses.

The upper bearing design is shown in Figure 11.The upper plate on the bearing housing is designed toaccept 3, 4, 6, or 12 guy cables, although the currentconfiguration calls for the use of 3. The vertical loadcaused by cable tension is carried by the thrust bear-ing at the bottom of the bearing assembly, whereas theradial loads are distributed between the lower bearingand the upper radial bearing.

The rotor base is a simple design with four angledlegs and single braces on each side. The top plate is a10.2-cm (4.O-in.) thick steel plate upon which thelower rotor bearing and the disk brake calipers aremounted. The entire weight of the rotor and thevertical load caused by guy cable tension is borne bythe lower rotor bearing. The ground-level, right-angle

transmission does not bear any vertical load. Thecurrent base design places the lower rotor bearing 4.8m (15.8 ft) above ground level, which provides clear-ance between it and the transmission for two sets ofslip rings (a total of 100 channels), a torque sensor,and two elastomeric couplings to protect the torquesensor. The instrumentation lines will be run from theslip rings inside the blades and will be brought to theoutside at the blade joint near where the lines arerequired.

Figure 12 is a fan plot of the current design. It isonly slightly changed from the one in Option 2 (Figure6), with the IF and 2F lines a little lower as a result ofthe changes in the towerlblade joint design. ThelF/3P crossing, which defines the lower end of theoperating range, is now at 26 rpm, the 1B13P crossingis at 34 rpm, and the lTI/3P crossing is at 45 rpm.NASTRAN does not predict a severe resonance at 34rpm, but unless we see a very light resonance duringactual operation we will be restricted to the use of twosmall operating windows (28-33 and 35-43 rpm) ratherthan the one large window that we would like.

Our current design has a rated power of 480 kW ina 12-m/s (27-mph) wind at 37.5 rpm and an estimated

annual energy output of 1.15 x 10G kWh (based on amean wind speed of 6 mls and a Rayleigh wind speed

distribution). The turbine cutout wind speed is 20 mls(45 mph). An artist’s conception of the 34-m test bedis presented in Figure 13.

M‘ “1” “,/+.

●\, ‘

1’1 — LIQHTNINO ROO

L-JFigure 11. Upper Bearing Design

14

s

r

[ { I5P

, 1

I I,/’

I

0 l-~.0 20 40

ROTOR RPM

Figure 12. Fan Plot for the Test Bed

2 IcEJ==-

—/’””

——

Figure 13. Artist’s Conception of Sandia’s 34-m Test Bed

Although our test plan has not been worked out indetail at this time, the general procedure has beenestablished. We will instrument the blades, joints, andtower with approximately 75 strain gages to allow usto determine the local and global stress distributionsfor the turbine. Some gages will be placed to detectmaximum stress levels and to determine stress con-centration details. These will be used to spot potentialproblem areas. Other gages will be placed to yield datathat can be directly compared with our code results. Atorque sensor on the low-speed shaft will be used tomeasure the rotor torque and to determine the aerody-namic performance of the machine. We will conductmodal tests of various sections of the machine duringconstruction and of the full machine prior to firstturn. The results of these tests will be used to fine-tune our finite element model to match the actualmachine. We will start operating the turbine atselected fixed speeds to characterize the machineresponses. At each speed we will seek overall perfor-mance data as well as stress data from all of our gages,We will pay particular attention to operating speeds inthe vicinity of the lB/3P mode crossing, where weanticipate a resonance condition. Once we have ade-quately characterized the machine in the fixed-speedmode, we will develop operating algorithms to operatein a continuously variable-speed mode in each of theanticipated operating windows. Finally, we will de-velop an algorithm to allow us to operate in a continu-ously variable-speed mode over the entire operatingrange, excluding only the rpm band around the lB/3Pcrossing. If lB/3P resonance is ~ mild one, we may nothave to exclude that speed from our operating range.

We currently plan to let contracts on the turbineblades and transmission (the long lead time items)early in calender year 1985; contracts for the rest ofthe equipment will follow throughout 1985. Erectionand first turn of the test bed should occur sometime in1986 at a site to be selected by the DOE Wind EnergyTechnology Division.

SummarySandia National Laboratories is currently

performing detailed design studies for a research-oriented, 34-m diameter, Darrieus-type Vertical-AxisWind Turbine. This work is a continuation of theconceptual design stage completed in May 1984, inwhich we looked at three potential VAWT configura-tions. Our primary analytic tools during these designstudies have been the Sandia-developed ECON16, ,FEVD, and FFEVD codes. All three VAWT configu-rations investigated during the conceptual design

15

stage incorporated a 3.05-m (10-ft) diameter steeltower and step-tapered blades with NLF sections nearthe turbine equator. The main features of the threeoptions are summarized in Table 1.

After evaluating these options, we elected to con-tinue our detailed design phase with Option 2.

A detailed design of the turbine is currently un-derway, and the machine will continue to evolve as thedesign progresses. The current design includes the

following features:

●

●

●

●

●

●

●

✎

A 3.05-m (10-ft) diameter aluminum tower witha 1.27-cm (0.50 -in.) wall

An H/D ratio of 1.3

Continuous variable-speed operation

No struts

External blade-to-blade joints

SAND 0018/50 blade section near the equator

NACA 0021 blade section near the tower

Multiple extrusion blade profiles

We anticipate the completion of fabrication andthe start of testing sometime in calendar year 1986.

References1P. C. Klimas and D. E. Berg, “Aerodynamic Design of a

Midsized Vertical Axis Wind Turbine Using Natural Lami-nar Flow Blade Elements,” Proceedings of the Sixth Bien-nial Wind Energy Conference and Workshop (AmericanSolar Energy Society, Inc., June 1983) pp 239-48.

‘E. G. Kadlec, “The Potential of Advanced DarrieusWind Turbines,” Proceedings of the Sixth Biennial WindEnergy Conference and Workshop (American Solar EnergySociety, Inc., June 1983) pp 213-18.

‘W. N. Sullivan, Economic Analysis of Darrieus Verti-cal Axis Wind Turbine Systems for the Generation ofUtility Grid Electrical Power, Volume II—The EconomicOptimization Model, SAND78-0962 (Albuquerque, NM:Sandia National Laboratories, August 1979).

4MSCINASTRAN User’s Manual, Vol I and II (LosAngeles, CA: MacNea}-Schwendler Corp., 1981).

5D.W. Lobitz, Dynamic Analysis of Darrieus VerticalAxis Wind Turbine Rotors, SAND80-2820 (Albuquerque,NM: Sandia National Laboratories, May 1981).

‘T. G. Carrie et al, Finite Element Analysis and ModalTesting of a Rotating Wind Turbine, SAND82-0345 (Albu-querque, NM: Sandia National Laboratories, October 1982).

71.Paraschivoiu, “Aerodynamic Loads and Performanceof the Darrieus Rotor,” Journal of Energy, 6(6):406 (1982).

‘D. W. Lobitz, “Forced Vibration Analysis of RotatingStructures with Application to Vertical Axis Wind Tur-bines: Proceedings of the Fifth Biennial Wind EnergyConference and Workshop, SERI/CP-635-1340, Vol III(Golden, CO: Solar Energy Research Institute, October1981).

‘R. E. Gormont, “A Mathematical Model of UnsteadyAerodynamics and Radial Flow for Application to Helicop-ter Rotors,” USAAMRDL Technical Report 72-67 (FortEustis, VA: US Army Air Mobility R&D Laboratory, May1973).

1°B.Masse’, Description de Deux Programmed d’Ordin-ateur Pour le Caleul des Performance et des Charges AxeVertical, IREQ-2379 (Varennes, Quebec, Canada: Institutde Recherche d’Hydro-Quebec, July 1981).

“D. E. Berg, “Recent Improvements to the VDART3VAWT Code,” Proceedings of the 1983 Wind and SolarEnergy Technology Conference (Columbia, MO: Wind/Solar Proceedings Engineering Extension, College of Engi-neering, University of Missouri—Columbia, April 1983) pp31-41.

‘2D. W. Lobitz, “A NASTRAN-Based Computer Pro-gram for Structural Dynamic Analysis of Horizontal AxisWind Turbines” (presented at the NASA/DOE Workshopon Horizontal Axis Wind Turbine, Technology, Cleveland,OH, 8-10 May 1984).

lSD. W. Lobitz and W. N. Sullivan, A Comparison ofFinite Element Predictions and Experimental Data far theForced Response of the DOE 100-k W Vertical Axis WindTurbine, SAND82-2534 (Albuquerque, NM: SandiaNational Laboratories, February 1984).

16

DISTFUBUTION:

Aerolite, Inc.Attn: IR. K. St. Aubin550 Russells Mills Rd.South Dartmouth, MA 02748

Aluminum Company of America (5)Technical CenterAttn: D. K. Ai

J. T. HuangJ. R. JombockMarshall KlingensmithJ. L. Prohaska

Alcoa Center, PA 15069

Alternative Sources of EnergyAttn: Larry StoiakenMilaca, MN 56353

Amarillo CollegeAttn: E. GilmoreAmarillo, TX 79100

Americim Wind Energy Assn.1516 King St.Alexanchia, VA 22314

Arizona State UniversityUniversity LibraryAttn: M. E. BeecherTempe, AZ 85281

R. G. RichardsAtlantic Wind Test SitePO BOX 189Tignish P. E. I., COB 2B0CANADA

Battelle -Pacific Northwest LaboratoryAttn: Li~rry WendellPO Box 999Richland, WA 99352

National Technical UniversityDept. of Mechanical EngineeringAttn: Dir. G. Bergeles42, Patission StreetAthens, GREECE

Bonneville Power AdministrationAttn: N. ButlerPO BOX 3621Portland, OR 97225

Burns & Roe, Inc.Attn: G. A. Fontana800 Kinderkamack Rd.Oradell, NJ 07649

Tom Watson

Canadian Standards Assn.

178 Rexdale Blvd.Rexdale, Ontario, M9W 1R3

CANADA

University of AucklandSchool of EngineeringAttn: V. A. L. Chasteau, Prof.Private BagAuckland, NEW ZEALAND

Colorado State UniversityDept. of Civil EngineeringAttn: R. N. MeroneyFort Collins, CO 80521

Commonwealth Electric Co.Attn: D. W. DunhamBOX 368Vineyard Haven, MA 02568

Gale B. CurtisCurtis Associates3089 Oro Blanco Dr.Colorado Springs, CO 80917 .

M. M. Curvin11169 Loop Rd.Soddy Daisy, TN 37379

Department of Economic Planning and DevelopmentAttn: G. N. MonssonBarrett BuildingCheyenne, WY 82002

Otto de VriesNational Aerospace LaboratoryAnthony Fokkerweg 2Amsterdam 1017THE NETHERLANDS

Department of EnergyAttn: G. P. TennysonAlbuquerque, NM 87115

17

DISTRIBUTION (cent):

Department of EnergyAttn: Capt. J. L. Hanson, USAFEnergy Technology Liaison OfficeNGDAlbuquerque, NM 87115

Department of Energy (20)Attn: D. F. Ancona (19)

P. Goldman1000 Independence Ave.Washington, DC 20585

Dominion Aluminum Fabricating, Ltd. (2)Attn: L. Schienbein

C. Wood357o Hawkestone Rd.Mississauga, Ontario, L5C 2V8CANADA

J. B. DragtNederlands Energy Research Foundation(E. C.N.)Physics Dept.Westerduinweg 3 Petten ,(nh)THE NETHERLANDS

Dynergy Systems Corp.Attn: C. Fagundes821 West L St.Los Banes, CA 93635

Dr. Norman E. Farb10705 Providence Dr.Villa Park, CA 92667

Electric Power Research Institute (2)Attn: E. Demeo

F. Goodman3412 Hillview Ave.Palo Alto, CA 94304

Alcir de Faro OrlandoPontificia Universidade Catolica-PUC/RjMechanical Engineering Dept.R. Marques de S. Vicente 225Rio de Janeiro. BRAZIL

FloWind Corp. (4)Attn: H. M. Drees

S. TremouletI. E. VasR. Watson

214 68th Ave. SouthKent, WA 98031

Gates LearjetMid-Continent AirportAttn: G. D. ParkPO Box 7707Wichita, KS 67277

H. GerardinMechanical Engineering Dept.Faculty of Sciences and EngineeringUniversity Lavel-Quebec, GIK 7P4CANADA

R. T. GriffithsUniversity College of SwanseaDept. of Mechanical EngineeringSingleton ParkSwansea, SA2 8PPUNITED KINGDOM

Helion, Inc.Attn: J. Park, PresidentBox 445Brownsville, CA 95919 “

Institut de Recherche d’Hydro-Quebec (2)Attn: Gaston Beaulieu

Bernard Masse’1800, Monte6 Ste-JulieVarennes, Quebec, JOL 2P0CANADA -

Iowa State UniversityAgricultural Engineering,Attn: L. H. SoderholmAmes, IA 50010

Rm. 213

Kaiser Aluminum and Chemical Sales, Inc.Attn: A. A. Hagman14200 Cottage Grove Ave.Dolton, IL 60419

18


Kaiser ,41uminum and Chemical Sales, Inc.Attn: D. D. Doerr6177 Su.nol Blvd.PO BOX 877Pleasanton, CA 94566

Kaman Aerospace Corp.Attn: W. BatesolOld Windsor Rd.Bloomfield, CT 06002

Kansas State UniversityElectrical Engineering Dept.Attn: G. L. JohnsonManhattan, KS 66506

R. E. KellandThe College of Trades and TechnologyPO BOX 1693Prince F’hilip Dr.St. John’s, Newfoundland, AIC 5P7CANADA

KW Control Systems, Inc.Attn: R. H. KleinRD#4, Box 914CSouth Plank Rd.Middletown, NY 10940

Kalman Nagy LehoczkyCort Adelers GT. 30Oslo 2, NORWAY

L. Liljidi~hlBldg. 005, Rm. 304Bare-West~eltsville, MD 20705

Ole LjungstromFFA, The Aeronautical Research InstituteBox 1102!1S-16111 Bromma, SWEDEN

R. Lynette & Assoc., Inc.Attn: R. Lynette15921 SE 46th WayBellevue, WA 98006

Massachusetts Institute of Technology (2)Attn: N. D. Ham, Prof.

W. L. Harris, Aero/Astro Dept.77 Massachusetts Ave.Cambridge, MA 02139

H. S. MatsudaComposite Materials LaboratoryPioneering R&D LaboratoriesToray Industries, Inc.Sonoyama, Otsu, ShigaJAPAN 520

Michigan State UniversityDiv. of Engineering ResearchAttn: O. KraussEast Lansing, MI 48825

Napier College of Commerce and TechnologyTutor Librarian, Technology FacultyColinton RoadEdinburgh, EH1O 5DTENGLAND

NASA Lewis Research Center (2)Attn: D. Baldwin

J. Savino21000 Brookpark Rd.Cleveland, OH 44135

.National Rural Electric Cooperative Assn.Attn: W. Prichett 1111800 Massachusetts Ave. NWWashington, DC 20036

Natural Power, Inc.Attn: L. NicholsNew Boston, NH 03070

Northwestern UniversityDept. of Civil EngineeringAttn: R. A. ParmaleeEvanston, IL 60201

Ohio State UniversityAeronautical and Astronautical Dept.Attn: G. Gregorek, Prof.2070 Neil Ave.Columbus, OH 43210

19


Oklahoma State UniversityMechanical Engineering Dept.Attn: D. K. McLaughlinStillwater, OK 76074

Oregon State University (2)Mechanical Engineering Dept.Attn: R. W. Thresher

R. E. WilsonCorvallis, OR 97331

Pacific Gas & ElectricAttn: T. Hillesland34oO Crow Canyon Rd.San Ramon, CA 94583

Ion ParaschivoiuDept. of Mechanical EngineeringEcole PolytecniqueCP 6079Succursale EMontreal H3C 3A7CANADA

Riso National Laboratory (2)Attn: Troels Friis Pedersen

Helge PetersenPostbox 49DK-4000 RoskildeDENMARK

The Power Company, Inc.Attm A. A. NeddPO Box 221Genesee Depot, WI 53217

Power Technologies Inc.Attn: Eric N. HinrichsenPO BOX 1058Schenectady, NY 12301-1058

Public Service Co. of New HampshireAttn: D. L. C. Frederick1000 Elm St.Manchester, NH 03105

Public Service Co. of New MexicoAttn. M. LechnerPO BOX 2267Albuquerque, NM 87103

RANN, Inc.Attn: Alfred J. Eggers, Jr.260 Sheridan Ave., Suite 414Palo Alto, CA 94306

Renewable Energy VenturesAttn: G. W. Stricker190 South King St., Sta. 2460Honolulu, HI 96813

The Resources AgencyDept. of Water Resources Energy DivisionAttn: R. G. FerreiraPO BOX 388Sacramento, CA 95802

Reynolds Metals Co.Mill Products Div.Attn: G. E. Lennox6601 West Broad St.Richmond, VA 23261

Memorial University of NewfoundlandFaculty of Engineering and Applied SciencesAttn: A. RobbSt. John’s, Newfoundland, AIC 5S7CANADA

Rockwell International (2)Rocky Flats Plant .Attn: A. TrenkaPO BOX 464Golden, CO 80401

Dr. kg. Hans RuscheweyhInstitut fur LeichbauTechnische Hochschule AachenWullnerstrasse 7FEDERAL REPUBLIC OF GERMANY

Beatrice de Saint LouventEstablissement d’Etudes et de

Recherches Meteorologigues77 rue de Serves92106 Bouloigne-Billancourt CedexFRANCE

National Atomic MuseumAttn: G. Schreiner, LibrarianAlbuquerque, NM 87185

20


Technion-Israel Institute of TechnologyDepartment of Aeronautical EngineeringAttn: PL SeginerProfessor of AerodynamicsHaifa, ISRAEL

Queen Mary CollegeDept. of Aeronautical EngineeringAttn: D. SharpeMile End Rd.London, El 4NSUNITED KINGDOM

Instituto Technologico Costa RicoAttn: K.. SmithApartadlo 159 CartagoCOSTA RICA

Solar InitiativeCiticorp Plaza, Suite 900Attn: J. Yudelson180 Grand Ave.Oakland., CA 94612

Bent SorensonRoskilde University CenterEnergy Group, Bldg. 17.2lMFUFAPO BOX 260DK-400 RoskildeDENMARK

South Dakota School of Mines and TechnologyDept. of Mechanical EngineeringAttn: E. E. AndersonRapid City, SD 57701

Southern California EdisonResearch & Development Dept., Rm. 497Attn: R. L. SchefflerPO Box 800Rosemead, CA 91770

Southern Illinois UniversitySchool of EngineeringAttn: C. W. DoddCarbondale, IL 62901

The University of ReadingDept. of EngineeringAttn: G. StaceyWhiteknights, Reading, RG6 2AYENGLAND

Stanford UniversityDept. of Aeronautics and Astronautics

Mechanical EngineeringAttn: H. AshleyStanford, CA 94305

Low-Speed Aerodynamics Laboratory (3)NRC-National Aeronautical EstablishmentAttn: R. J. TemplinMontreal Rd.Ottawa, Ontario, KIA 0R6CANADA

Texas Tech University (2)Mechanical Engineering Dept.Attn: J. W. Oler

J. StricklandPO Box 4389Lubbock, TX 79409

Tulane UniversityDept. of Mechanical EngineeringAttn: R. G. WattsNew Orleans, LA 70018 ●

Tumac Industries, Inc.Attn: J. R. McConnell650 Ford St.Colorado Springs, CO 80915

Air Force Systems Command (AFSC)

Aerospace Power Division/Aero Propulsion LabAttn: J. M. TurnerTerrestrial Energy Technology Program OfficeEnergy Conversion BranchWright-Patterson Air Force Base, OH 45433

United Engineers and Constructors, Inc.Attn: A. J. KaralisPO Box 8223Philadelphia, PA 19101

21


University of AlaskaGeophysical InstituteAttn: T. Wentink, Jr.Fairbanks, AK 99701

University of CaliforniaInstitute of Geophysics and Planetary PhysicsAttn: P. J. BaumRiverside, CA 92521

University of ColoradoDept. of Aerospace Engineering SciencesAttn: J. D. Fock, Jr.Boulder, CO 80309

University of MassachusettsMechanical and Aerospace Engineering Dept.Attn: D. E. CromackAmherst, MA 01003

University of New MexicoNM Engineering Research InstituteAttn: G. G. LeighCampus PO Box 25Albuquerque, NM 87131

University of OklahomaAero Engineering Dept.Attn: K. BergeyNorman. OK 73069

University of SherbrookeFaculty of Applied ScienceAttn: R. CamereroSherbrooke, Quebec, JIK 2R1CANADA

University of TennesseeDept. of Electrical EngineeringAttn: T. W. ReddochKnoxville, TN 37916

USDA, Agricultural Research ServiceSouthwest Great Plains Research CenterAttn: R. N. ClarkBushland, TX 79012

Utah Power and Light Co.Attn: K. R. Rasmussen51 East Main St.PO Box 277American Fork, UT 84003

W. A. Vachon & AssociatesAttn: W. A. VachonPO Box 149Manchester, MA 01944

VAWTPOWER, Inc.Attn: P. N. Vosburgh134 Rio Rancho Dr.Rio Rancho, NM 87124

Washington State UniversityDept. of Electrical EngineeringAttn: F. K. BechtelPullman, WA 99163

West Texas State UniversityGovernment Depository LibraryNumber 613Canyon, TX 79015

West Texas State UniversityDept. of PhysicsAttn: V. NelsonPO BOX 248Canyon, TX 79016

West Virginia UniversityDept. of Aero EngineeringAttn: R. Walters .1062 Kountz Ave.Morgantown, WV 26505

Central Lincoln People’s Utility DistrictAttn: D. Westlind2129 North Coast HighwayNewport, OR 97365-1795

Wichita State University (2)Aero Engineering Dept.Attn: M. Snyder

W. WentzWichita, KS 67208

Wind Energy ReportAttn: F. S. SeilerBox 14102 S. Village Ave.Rockville Centre, NY 11571

22


Wind Power DigestAttn: Michael EvansPO Box 700Bascom, OH 44809

Wisconsin Div. of State EnergyAttn: Wind Program Manager101 South Webster St,, 8th FloorMadison, WI 53702

1520 D. J. McCloskey

1522 R. C. Reuter, Jr.

1523 Jr. H. Biffle1524 I). W. Lobitz1524 W. N. Sullivan1600 Ft. G. Clem1630 Ft. C. Maydew1636 J. K. Cole2525 R. P. Clark3160 J. E. Mitchell (15)3161 P. S. Wilson6000 E. H. Beckner6200 V. L. Dugan6220 D. G. Schueler6225 T. D. Ashwill6225 DI. E. Berg (10)6225 H. M. Dodd (50)6225 R. D. Grover6225 E, G. Kadlec

6225 P. C. Klimas6225 T. Leonard6225 M. T. Mattison6225 R. O. Nellums6225 D. S. Oscar6225 M. E. Ralph6225 M. H. Worstell7111 J. W. Reed7544 T. G. Carrie7544 J. Lauffer7544 D. R. Schafer8024 M. A. Pound3141 C. M. Ostrander (5)3151 W. L. Garner3154-3 C. H. Dalin (28)

For DOE/TIC (Unlimited Release)

.

23

SAND84– 1287Unlimited ReleasePrinted April 1985

DistributionCategory UC—60


Dale E. BergWind Energy Research Division

Sandia National LaboratoriesAlbuquerque, NM 87185

AbstractSandia National Laboratories, as the lead Department of Energy laboratory for Vertical-AxisWind Turbine development, is currently designing a 34-m diameter Darrieus-type vertical-axiswind turbine. This turbine will be a research test bed that provides a focus for advancingtechnology and validating design and fabrication techniques in a size range suitable for utility use.Structural data from this machine will allow structural modeling to be refined and verified for aturbine on which the gravity effects and stochastic wind loading are significant. Performance datafrom it will allow aerodynamic modeling to be refined and verified. This design effort incorporates

Sandia’s state-of-the-art analysis tools in the design of a complete machine. In this report Idescribe the analytic tools we are using, summarize the conceptual design procedure, and presentportions of our detailed design as it existed in September 1984.

841287--Structural Design of Sandia 34m VAWT

Documents

Transcript of 841287--Structural Design of Sandia 34m VAWT