POWERENG 2007, April 12-14, 2007, Setubal, Portugal

Rotation-Elevation of Sun Tracking Mode to Gain

High Concentration Solar Energy

Omar Aliman, Ismail DautSchool of Electrical System Engineering

Kolej Universiti Kejuruteraan Utara Malaysia (KUKUM)02600 Jejawi, Perlis, Malaysia

E-mail: omaralimangkukum.edu.my

Abstract - Power conversion from solar thermal energy toelectrical energy is still very cost-intensive. Serious effort has to begiven in the development of the concentrator or heliostatstructure expenditure which contributing the most expensivecomponent in a central receiver solar power plant. With currentdevelopment to find alternatives and lower down the capital, analternative scheme of sun tracking has been developed andfeasibility tested. The proposed rotation-elevation mode of suntracking has significantly benefits use in high temperature andhigh concentration solar energy applications. Instead of offeringstationary location for large and heavy solar powered Stirlingengine application, the process of optical alignment works arereasonably easier and less time consuming. This paper at first willhighlight and elaborate its concept and follow with someexperimental results.

Keywords - Sun Tracking, Heliostat, High Concentration SolarEnergy.

I. INTRODUCTION

Central receiver system, so called power tower is a field ofheliostats that reflects the incident sunray to a receiver at thetop of the central tower. This concept [1]-[4] uses multiplenumbers of sun tracking as concentrator devices for biggerscale and higher power comparing to that of moving targetsolar collectors. However, the reflected images on the targethave relative large dimension and limiting the concentration[5]. Therefore, it cannot be directly used for a small receivingarea of some thermo-mechanical convector, e.g., the heat partof Stirling engine. In contrast, Solar furnace employs two-stagereflection which can produce very high concentration andfocus the solar rays at a small area [6]. However, this methodneeds high cost and long engineering time, particularly for theprecise optical alignments [6]-[8].On the next section, this paper will elaborate an alternative

to conventional azimuth-elevation sun tracking mode [9]-[12],that proposed rotation-elevation tracking mode [13] of suntracking.

II. TRACKING CONCEPT

In this technique, mirrors are arranged in rows and columns.A master mirror is fixed at the centre. Slave mirrors whichshare the same frame but have two extra moving freedomsabout their pivot points are surrounding the master mirror.Each slave mirror is angularly moved about its pivot point toreflect sun rays into the same fixed target of the master. The

result at the target is the total combination of individual mirrorimages. (See Fig. 1)

Incident Sun / ,

Target

s | : ! ! . . , 5<, ~~~A3x3 mirrors

\ ~~~~~~~~focusingOverlappedIm age

Fig. 1. Rotation-Elevation Mode concept. Mirror at the center is actingas master mirror.

A. Primary Tracking

The objective of primary tracking is to target the sun imageof master mirror onto a stationary target. Then, this mastermirror image acts as a reference for secondary tracking whereall the slave mirror images will be projected on it.

(a) Rotation Mode

In Fig. 2(a), it illustrates the rotation modes of the suntracking. ON is defined as the normal vector of the heliostatsurface; OS is the vector that points to the sun; OT is thevector that points to a fixed target.

In Fig. 2(b), it illustrates the rotation of the plane ofreflection. In the new reflection plane, the vector os points tothe new position of the sun and the vector ON-is the reflectornormal of the new orientation so that the sunlight is stillreflected towards the target.

Fig. 3 illustrates the mounting of sun tracking which has twotracking axes that are perpendicular to each other. Firstrotational axis is pointing toward the target and it is indicatedby TT' axis. The second axis, elevation axis is attached parallelto the reflector and it is shown as FF' axis.

The heliostat has to rotate about the TT' axis so that theplane can follow the rotation of the vector OS. As the sunmoves through the sky from the morning to solar noon, theplane will rotate starting from horizontal and turning tovertical. The angular movement about this rotation axis isdenoted as p.

1-4244-0895-4/07/$20.00 C 2007 IEEE 551

(a) (b)Fig. 2. (a). Rotation modes of sun tracking. (b). Rotation plane of suntracking rotation.

Fig. 4. Coordinate system attached to earth reference frame.Horizontal-axis

Slave mirror

Vertical

Rotation-axis

Elevation-axis_/F

Master mirror

I&Targe

F'

Heliostat

Fig. 3. Sun tracking mounting.

The coordinate system attached to earth reference frame isillustrated in Fig. 4. The vector CS that points towards the suncan be described in terms of hour angle ca and the declinationangle 6. The observer Q is located at the latitude q. Bydefining a coordinate system with the origin, C, set at thecentre of the earth, the CM axis is a line from the origin to theintersection point between the equator and the meridian of theobserver at Q. The CE (east) axis in the equatorial plane isperpendicular to CM axis. The third orthogonal axis, CP is therotation axis of the earth.

The coordinate system attached to the local heliostatreference frame is illustrated in Fig. 5. The vector OS is in thefunction of the angle,/ and the angle p. The relationshipbetween the incidence angle o and the angle ,8 is fz/2 = 20 +,.The local heliostat reference frame is referred to only when theheliostat frame is oriented in such a way that the normal ofmaster mirror becomes parallel with the rotation axis. Theorigin of the coordinate system is defined at the centre of themaster mirror and is denoted as 0. The OR axis is parallel withthe array of mirrors arranged in the vertical direction of theheliostat frame. The second axis, OU axis is parallel with thearray of mirrors arranged in the horizontal direction. The thirdorthogonal axis, OT axis is a line pointing out from the origintowards the target direction.

Fig. 5. Coordinate system attached to the local heliostat reference frame.

(b) Elevation Mode

The rotation of the heliostat about FF' axis (perpendicular tothe plane) will adjust the heliostat normal position within theplane until it bisects the angle between OS and OT. As aresult, the sunlight will be reflected onto the target. Thisangular movement depends on the incident angle of the sunrelative to the heliostat normal and it is denoted as 0.

The formulas for p and 0 can be derived by transformationstudy of two different coordinate systems: one attaches to thecentre of the earth and the other attaches to the heliostat.Detailed derivation ofthese formulas can be found in [13].

With the movement of the sun tracking of central mastermirror, the images of the slave mirrors will be inevitablyaberrant. Therefore, in order to achieve a high concentration,the slave mirrors need to be adjusted accordingly to overcomethis aberration. This adjustment is rather minor, particularly ifthe target is far away from the heliostat. The tracking conceptof this secondary order tracking is described as follow.

B. Secondary Tracking

The proposed rotation-elevation mode encourages thearrangement of the slave mirrors to be grouped into rows andcolumns. Under this mode, the mirrors in the same row orcolumn will have the same movement. Fig. 6 shows the side

552

-

It g| lt.01

view of a 25-mirror heliostat with P representing the heliostatframe and the central row (row 3) contains master mirror.Slave mirrors of row 1,2,4 and 5 are attached to the heliostatframe in such a way that they can turn about their own pivotpoint P1, P2, P4 and P5 respectively. To overlap 4 rows of sunimages onto the central master image, each row of slavemirrors has to rotate through an angle, (.

u". 4higt

I Pi.,

Fig. 8. Daily tracking and recorded images.

u .fr ..... Hz' - 'M

N

ii:placcitiont devict

Fig. 6. Side view of a 25-mirror heliostat.

III. EXPERIMENTAL RESULT

In order to test the feasibility of the proposed sun trackingconcept, two generations of the prototype are constructed. Thepractical problems, such as tracking error and opticalalignment work, are analysed and solved through the proposedmethods. The detailed analysis of the tracking error is reportedin [15] while the methods proposed for optical alignment isreported in [16].

Tests have been conducted over years since 1997 inUniversiti Teknologi Malaysia by two different scales of suntracking prototypes. A typical structure of sun tracking of thisnew technique is illustrated in Fig. 7, which has successfullyfocused images into one fixed target and maintained itthroughout the day (See Fig. 8).

Fig. 7. Prototype of a heliostat focusing to a target.

A. Solar Engine.

Modifying an internal combustion or diesel engine is not anew idea and the publication of demonstrating such idea wasappeared as early as 1980s [17]. Believing that the commercialparts of better quality ensure a good performance of the engine,it was brought the adoption of modifying a commercial Hondaengine model 4S G150 IC engine into a Stirling engine for thisexperiment [18].

Solar Stirling Engine System - The merit of the system isthat the target is stationary, thus the Stirling engine can havebigger size and heavier weight, hence higher power. This is incontrast with the dish Stirling system where the engine ismounted and heated at the focal point of the parabolic dish.Thus, it requires a complicated lubrication system and limitedsize and weight. The major obstacle of the dish Stirling systemis the uncertainty of the engine design and maintenance. Eachengine failure or overhaul may require the engine to beuninstalled from the focal point, and the subsequent installationmay result in optical realignment. Such operation would belaborious and costly.

Experimental Stirling Engine - The low 0.5900 overallefficiency gained from the experimental Stirling engine cannotbe used to reflect the performance of the matured commercialStirling engine in such system. For low cost demonstration, theengine was not designed to operate at high pressure. Some ofthe design dimensions of the experimental engine have notbeen optimized especially the dead volume ratio. The designwas also limited by the commercial parts of the internalcombustion engine such as the bore stroke ratio of the piston.Instead of sophisticated heat exchanger design, the simplifiedversion was adopted in the design to avoid complexity andtrouble shooting in the demonstration.

Fig. 9 illustrates the solar Stirling engine experiment. It islearned that further development to the proposed Stirlingengine and other potential applications should be studied

553

Ntrow flne

;*-.

seriously to gain the effective result ofhigh concentration fromthis alternative technique of sun tracking.

Fig. 9. Solar Stirling engine being tested.

B. High Temperature Concentrated Solar Energy.

High temperature concentrated solar energy has beenconstructed consisting of the prototype heliostat and a secondstage parabolic concentrator with diameter of 60.96cm. Thedistance between the heliostat and the parabolic concentrator is1Om and the focal point of the parabolic concentrator is15.24cm. The achievable temperature of solar furnace systemcan be confirmed through melting various kinds of materialsincluding stainless steel, ceramic materials, tungsten etc. In thematerial melting experiment, the tungsten wire wassuccessfully melted and its melting point is about 3400°C [14].From Fig. 10, the yellow colour shows that the tungsten wirewas oxidised at which the temperature was greater than 500°C,the following yellow crystal indicates the tungsten oxide wasmelted and the temperature of 1470°C has achieved. Finally atthe end of the tungsten wire, it shows that the tungsten wasmelted and also there is no existing of the tungsten oxidecrystal because it was vaporised at temperature higher than1 827°C. Detailed setup of high temperature solar furnace usingproposed sun tracking mode is reported in [14].

IV. CONCLUSION

The proposed alternative technique of sun tracking methodhas to be explored to make the construction of the sun trackingwith many element mirrors cost effective while maintaining aprecise sun tracking. In this proposed alternative trackingmode, by introducing a rotational axis to the sun trackingframe, the slave mirrors of the same column or the same rowcan be arranged to share the same driving device. Thus, thenumber of controlling components can be significantly reducedto M+N. With this design, the focusing area can be as small asthat of an element mirror or even smaller if the element mirrorsare concave.

ACKNOWLEDGMENT

We acknowledge with many thanks to Prof. Y.T.Chen, andmembers of Solar Group which we formerly in the same teamand effort, together struggling for the success of this researchwhile we were in Universiti Teknologi Malaysia.

REFERENCES

[1] Sakurai Takemaro and Shibata Yukio, "Theoretical Concentration ofSolar Radiation by Central Receiver Systems," International Journal ofSolar Energy, 31(3), 1983, 261-270.

[2] Bartel J.J. and Skvarna P.E., "10-MWe Solar Thermal Central ReceiverPilot Plant," Journal ofSolar Engineering, 106, 1984, 50-58.

[3] Hirono T. and Horigome T.A., "1-MWe Central Receiver Type SolarThermal Electric Power Pilot Plant," Journal ofSolar Engineering, 106,1984, 90-97.

[4] Holmes J.T., "Heliostat Operation at the Central Receiver Test Facility,1978-1980," Journal ofSolar Engineering 104, 1982, 133-138.

[5] Baranov V.K., "Heliostats and Concentrators with Variable ReflectingSurface Geometry," Journal ofGeliotekhnika, 16(2), 1980, 28-33.

[6] Azimov S.A., "Solntse Scientific Production Complex. Double-MirrorPolyheliostat Solar Furnace with Thermal Capacity of 1000kW,"Journal ofGeliotekhnika, 23(6), 1987, 3-9.

[7] Riskiev T.T. and Suleimanov K.H., "Double Mirror Polyheliostat SolarFurnace of 1000kW Thermal Power," Journal of Solar EnergyMaterials, 24, 1991, 625-632.

[8] Lewandowki Allan, Bingham Carl, O'Gallagher Joseph, WinstonRoland and Sagie Dan, "Performance Characterization of the SERIHigh-Flux Solar Furnace," Journal ofSolar Energy Materials, 24, 1991,550-563.

[9] Alpert D. J. et al., "Solar Concentrator Development in the UnitedStates," Journal ofSolar Energy Materials, 24, 1991, 307-319.

[10] Winter C.-J., Sizmann R.L and Vant-Hull L.L. (1991), "Solar PowerPlants Fundamentals, Technology, Systems, Economics," (Springer-Verlag, NY, 1966, 237-250).

[11] Fourakis E. and Severson A.M., "Low-Profile Heliostat Design for SolarCentral Receiver Systems," International Journal of Solar Energy, 19,1977, 349-356.

[12] Murphy L.M., "Stretched-Membrane Heliostat Technology,"International Journal ofSolar Energy Engineering, 108, 1986, 230-238.

[13] Y. T. Chen, K. K. Chong, Omar Aliman , T. P. Blighl, L. C. Chen,Jasmy Yunus, K. S. Kannan, B.H. Lim, C. S. Lim, M. A. Alias, NoriahBidin, Sahar Salehan, Shk. Abd. Rezan S.A.H., C. M. Tam And K. K.Tan, "Non-Imaging Focusing Heliostat," International Journal of SolarEnergy, 71 (3), 2001, 155-164.

[14] Chen, Y. T., Chong, K. K., Omar Aliman, Lim, C. S., Lim, B. H., Tan,K. K., Bligh, T. P., Tan, B. K., and Ghazally, I., "Report Of The FirstPrototype Of Non-Imaging Focusing Heliostat And Its Application InHigh Temperature Solar Furnace," International Journal of SolarEnergy, 72 (6), 2002, 531-544.

Fig. 10. Melted tungsten.

554

[15] Omar Aliman, Ismail Daut, Muzamir Isa, Mohd Rafi Adzman, TrackingError Analysis ofRotation-Elevation Mode Heliostat, American JournalofApplied Sciences, 4(3) (2007), pp. 172-176.

[16] Omar Aliman, Ismail Daut, Rosnazri Ali, Muzamir Isa, Mohd RafiAdzman, Optical Alignment ofSun Tracking, Proceeding Of MalaysianTechnical Universities Conference On Engineering & Technology(MUCET) 2006, Dec. 19-20, KUITTHO, Batu Pahat (2006)

[17] G. Walker, Future Coal-Burning Stirling Engine, InternationalMechanical Engineering Publication, 2, 1982.

[18] Omar Aliman , L.C Chen, C.S Lim, Ismail Daut, Muzamir Isa, MohdRafi Adzman, Solar Engine Application By Using Concentrated SolarEnergy, International Energy Journal, Vol. 8 (2) 2007, RERIC Press.

555