Power Rating of Multi-junction Solar Cells: Focus Thin Film

Transcription

1 Power Rating of Multi-junction Solar Cells: Focus Thin Film Content: 1. Basics of current matching a short review 2. Requirements on Solar Simulator 3. Status of Test labs concerning thin film tandem Prof. Dr. Franz Baumgartner, ZHAW, , Solarpraxis Berlin; bauf@zhaw.ch; 1

2 Measurement of the Current Voltage Characteristics of Standard PV Modules Single Junction Maximum Power P mp Swiss Mobile Flasher Bus comes to the plant site, high number of samples up to 100 a day F. Baumgartner, EUPVSEC, Hamburg Sept 2009 Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 2

3 Standard Method of PV Power Measurement I STC = I DUT, meas P P STC REF, meas IEC Mismatch correction STC Standard Test Conditions AM1.5 Spectra, 1000 W/m 2, 25 C SMMF Device under Test Flash light Typical today's Measurement Uncertainty of typ. 2 to 3% (k=2) at the best test labs possible for single junction cr-si Modules (costs about 100 to 500 / per test incl. logistics) Reference Cell Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 3

4 Stability +/-1%, Flasher Pulse IV Curve Measurement Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 4 Internal Data from Corina Droz, Pasan; 15. Dec 08

5 Power Tolerances of PV modules Standard cr-silicon Modules Manufactures PV Modules Datasheet: high eff. Si -5% -10% Data sheet 09/ W 205W 210W Nominal power 231W Manufactures PV Modules Datasheet: poly cr. Silicon Power tolerance +/-3% + Sorting: flasher report Power Warrenty 25 years -3% -3% P N [W] Nominal power Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 5

6 Law in action - To go to court? Manufacturers Datasheet: 210W (+/-3%) Tolerance Effective Tolerance -5% Independent Testlab Results: 201W (+/-2% uncertainty k=2) -u(t) +u(t) 205W > guaranteed upper limit 204W Justice decision: The independent test result can not prove that manufactures data specification failed! Result: OK!! Effective Tolerance (+/-5%) Testlab Results failed if below ( datasheet tolerance + testlab uncertainty) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 6

7 PV World Market and Uncertainty of nominal Power Turnover 2008 PV Module end customer Mio ; 2% uncertainty in Pn 300 Mio Only Thin-film Production Capac. Installations Europe 4800 MW A. Jäger-Waldau; Aug 2009; ISBN ; Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 7

8 Spectral Characteristics /PV Technologies of single junctions / single band gap cell H. Mulejans, JRC, EUPV Conf 2008, Valencia Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 8

9 4) Solar Spectra: Spectral Mismatch Factor MMF = SR SR DUT DUT ( λ) ( λ) E E estc ein ( λ) dλ ( λ) dλ SR SR REF REF ( λ) E ( λ) E ein estc ( λ) dλ ( λ) dλ Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 9 Quelle:

10 Singel cell s and tandem cell s energy lift Standard single bandgap Tandem cell red photon red photon blue photon Single cell E g1 blue photon Top cell E g2 - - Bottom cell E g1 Output 2e Voc ~2/3 E g1 /e - Energy level of the electron Output 1e - Voc ~2/3 E g1 /e + 2/3 E g2 /e higher voltage! - Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 10

11 Simple model of the tandem cell circuit schema / multi-junction TOP CELL LIMITIED SPECTRA at Vm=0V Im BOTTOM CELL LIMITIED SPECTRA at Vm=0V Im Isc_top Isc_top Im Im Isc_bot Isc_bot The photo generated currents of the bottom and the top cells are different in real tandem cells, due to each specific spectral cell response behaviour exposed to the given light spectra. One serial resistance and two parallel resistance complete the simple model Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 11

12 IV of tandem cell IV-curve tandem cell - serial connected bottom cell top cell tandem V_top+V_bot current [A] IV curve at Vm=0V limited by bottom cell V_bottom_cell V_bottom_cell Isc of the tandem cell is limited by the Isc of the bottom cell (or reverse IV characteristics) -1.0 voltage [V] Cell with the smallest Isc dominates terminal Isc (it no reverse break through). Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 12

13 1. Conclusion: Current Limit The current at maximum power of the tandem-cell is limited by the lowest short circuit current of the sub-cells. Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 13

14 Indoor have to adapted for outdoor spectra F.P. Baumgartner et. al.; EUPVSEC Dresden 2006 Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 14

15 Spectra and Top/Bottom Cell Performance Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 15 K. Yamamoto, Kaneka, Osaka PV Conf (2003)

16 Current matching high eff. cells InGaP / GaAs Current matching Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 16 Patent US , Sharp 2004

17 2. Conclusion: Thickness design The effective optical thickness of the top and bottom cell is the most important design parameter in a-si/uc-si Tandem cell to reach current matching (Remark: light trapping has an important effect on current matching and my change by the angle of irradiance on the module plane) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 17

18 Output tandem cell: GaInP/GaInAs Ga 0.65 In 0.35 P/Ga 0.83 In 0.17 As Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 18 Next Generation Photovoltaics; High efficiency through full spectrum utilization; ISBN ; ( 2004) Antonio Mart ı and Antonio Luque;

19 Current Matching: a-si/uc-si Tandem Light-soaking Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 19 K. Yamamoto, Kaneka, Osaka PV Conf (2003)

20 Seasonal Spectral Change at Noon Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 20 K. Yamamoto, Kaneka, Osaka PV Conf (2003)

21 Changing Temperature at the a-si/uc-si Tandem Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 21 K. Yamamoto, Kaneka, Osaka PV Conf (2003)

22 3. conclusion: Real life! The current matching at STC conditions AM 1.5 will not be reached in real outdoor operation due to 1. Change in Temperature and the inherent different temperature coefficients of the sub-cells 2. Spectral changes of sun light due to sun heights (AM 1.5) and different clouds 3. Unstable sub-cell: light soaking of the a-si Top cell 4. Light trapping and thus current matching my change at different angle of irradiance IEC Energy Rating of PV Modules have to be modified Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 22

23 Requirements on nominal power measurement!! The goal is to have all the relevant data to predict the kwh electricity production after 25 years! (IEC norms not ready!) Pn Test at one spectra is not enough! - daily and seasonal changes of sun spectra - power characteristics as a function of air mass Pm=f(AMxx) this requires a sun simulator with adjustable spectra!! Temperature coefficient of different matching conditions at least for bottom or top cell limitation condition have to be given Pn test results just after module fabrication is not enough: the degradation / light soaking has to be taken into account The spectral characteristics of each sub-cell have to be known to reduce measurement uncertainty Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 23

24 The optimal light source and the optimal reference cells Solar Flasher Simulator have to fit excellent to AM 1.5 but have to have the potential to shift the spectra to the blue and to the red. This could by realised either continuously (additionally filter arrangement) or in discrete steps by the use of additional light sources like LED s SMMF Spectral mismatch factor have to be applied to the limiting sub-cell SRTOP ( λ) EeIN ( λ) dλ SRBOT ( λ) EeIN ( λ) dλ For each junction of the DUT a spectral fitted reference cell should be used; for example a-si/uc-si a-si: (1 filtered cr-si cell for the top cell i.e. KG3 filter) uc-si: (1 filtered cr-si cell which represent the spectral input of the bottom cell, extracting the photons absorbed by the top cell) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 24

25 High-Fidelity Solar Simulators AIST Japan National Institute of Advanced Industrial Science and Technology (AIST) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 25 Yoshihiro Hishikawa and Sanekazu Igari, EUPVSEC Paris, 2004

26 Tandem moduls: Status of the PV Test labs OUTDOOR Measurements are not different from the standard setup; High qualified labs believe mostly in outdoor technique to reach lowest uncertainties (i.e. JRC) measurement takes time!! OUTDOOR is no option for industrial production lines with high throughput A few labs in central Europe give uncertainty values for tandem cells in the range of 3 to 6% (up to 10% for triple cells) Select other light spectra than AM1.5 is tested in fewer labs Figure of merit of Tandem cells current-matching are not discussed up to now in detail ( AMxx at matching) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 26

27 Nominal Power and current matching this 3 tandem modules have the same STC Power η P N,STC AM x1 matching coeff AM x2 matching coeff AM x3 matching coeff blue rich AM1.5G red rich Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 27

28 Conclusion Up to know most of the most experienced test labs are working on special indoor test procedures to measure the power of tandem solar cells Measurement uncertainties below 5% are not expected today! (only in a few selected Labs) Nominal power at AM1.5 and the matching factor optimum at AM xx should be given To predict yearly energy performance the low irradiance characteristics of the tandem cell have to be known. (The current mismatching at the changing spectra relative to AM 1.5G at low irradiance have to be taken into account) Franz Baumgartner, ; ; Solarpraxis Berlin, Slide 28