outdoor measurements and comparison of thin film solar ... · - pyranometer and c-si ref. cell ......

Durability of Thin Film Solar Cells

04.04.2012 Empa, Dübendorf

Hans-Dieter Mohring

Zentrum für Sonnenenergie- und Wasserstoff-Forschung (ZSW)Baden-Württemberg

Outdoor measurements and comparisonof thin film solar cell technologies

- 2 -

• Motivation

• Methods for field data evaluation

- STC power and temperature coefficients

- Self referencing

- Dependence on irradiance level

- Long term stability

• Summary

Content

- 3 -

Motivation

Higher uncertainties of module parameters for thin film technologies(CIGS, CdTe, a-Si/µ-Si) compared to c-Si.

Operational parameters, energy yield and long term stability can bedetermined from outdoor measurements.

- 4 -

More energy from thin film PV –Marketing or reality?

Quelle: Broschüre Q-Cells

Quelle: Schott Solar Thin Film, EUPVSEC 2010

Quelle: http://www.nexpw.com/techology_tfa.html

More energy from thin film PV in principlepossible by:

• lower temperature coefficient• good performance under low/diffuse light• annealing or light-soaking effects

- 5 -

Information from data sheet

• STC power/tolerance

• Temperature coeff.

• Dependence on irradiance

Source: http://www.avancis.de/fileadmin/media/portal/produkt/Datenblatt_PowerMax-STRONG.pdf

- 6 -

Information from simulation programs

Source: PVSYST

Data for

• STC power/efficiency

• Temperature coefficients

• Dependence on irradiance

Where are these data from??

- 7 -

Information from simulator measurements

Quelle: TÜV Rheinland

Measurement according to IEC 61853(PV Module Performance Testing and Energy Rating):

• Irradiance 100, 200, 400, 600, 800, 1000, 1100 W/m²• Temperature 15, 25, 50, 75 °C

Restrictions for thin film flasher measurements:

• Metastability effects• Crystalline Si reference cells• Current mismatch for tandem cells• Transient effects

- 8 -

Results from outdoor module characterization

• STC power:- data collection days to weeks- spectrum dependent on location and season- stability presupposed

• Energy generation, Performance Ratio (PR):- typically 1 year- comparison of different technologies (benchmark c-Si)- valid for selected site

• Parameter determination:- typically 1 year- power, Voc, Isc,.. under STC- temperature coefficients, operation temperature- dependence on irradiance, incidence angle, spectrum

• Long term stability- several years

- 9 -

Energy and Performance Ratio

• Energy Yield:

Y = Ecounter / Pnom (kWh/kWp)

• Performance Ratio (PR)

PR = Y / EPOA (%)

• Which Pnom ? Label? Flasher or outdoor measurement?

• Measured irradiation in module plane (EPOA) depends on sensor

- 10 -

Outdoor measurements at ZSW• Widderstall (since 1988)• Girona/Spain (since 2011)

• Module measurements- module kept at MPP- I-V characteristics (1 per minute)- Tmod- Pyranometer and c-Si ref. cell- Ta, GHI, DHI, DNI, TNI, vwind

• Generator measurements- 1kW, identical inverters- DC, AC voltage and current- Tmod, POA irradiance

• PID-test stand

- 11 -

Outdoor measurements at ZSW

0.0

0.5

1.0

1.5

2.0

2.5

0 10 20 30 40 50

Spannung (V)

Stro

m (A

)

Module I-V characteristics

- 12 -

Power and temperature coefficient:(1) Selection of irradiance interval

Calculation of P1000 (linear scaling with reference irradiance)

1000 W/m² +/- 10%2854 Data points

1000 W/m² +/- 2%492 Data points

1000 W/m² +/- 0,2%74 Data points

70

80

90

100

110

120

10 20 30 40 50 60 70

Tmod [°C]

P10

00 [

W]

70

80

90

100

110

120

10 20 30 40 50 60 70

Tmod [°C]

P10

00 [

W]

70

80

90

100

110

120

10 20 30 40 50 60 70

Tmod [°C]

P10

00 [

W]

Data collection September, Widderstall, Si-TF module

Reasonable range:• Irradiance interval +/-1 to 2 %• some 100 to 1000 data points

- 13 -

y = -0.2793x + 85.416

y = -0.2756x + 85.491

66

68

70

72

74

76

78

80

82

20 25 30 35 40 45 50 55 60 65

Modultemperatur (°C)

Mod

ulle

istu

ng (W

) AprilJuliLi (A il)

Power and temperature coefficient:(2) Different measurement periods

Measurement months April and July, Widderstall, CIGS module

• Higher module temperature in July• No change of STC module power

- 14 -

P1000 vs. Tmod

• Determination of Tcoeff. fromslope (-0,21 %/K)

Moderate degree of correlation

Power and temperature coefficient: (3) Evaluation

y = -0,0006x + 99,172R2 = 2E-06

70

80

90

100

110

120

0 10 20 30 40 50 60 70

Tmod

P100

0 Tk

orrP1000 T-corr. vs. Tmod

• STC power 99,2 W

• Deviation by time constants of - pyranometer- module back temperature

y = -0,2085x + 104,36R2 = 0,2107

70

80

90

100

110

120

0 10 20 30 40 50 60 70

Tmod [°C]

P100

0 [W

]

- 15 -

Voc1000 vs. Tmod

• Determination of STC-Voc

• Determination Tcoeff Voc

• Correction of Tmod usingVoc1000 Teff

Power and temperature coefficient :(4) Improvement of accuracy

100

110

120

130

140

0 10 20 30 40 50 60 70

Tmod [°C]

Voc1

000

[V]

y = 0,003x + 1,386R2 = 0,161

1,1

1,2

1,3

1,4

1,5

1,6

1,7

1,8

0 10 20 30 40 50 60 70

Teff [°C]

Isc1

000

[A

]

Isc1000 vs. Teff

• Determination of Isc@STC

• Definition: Eeff = Isc/ Isc@STC

Module used as irradiance sensor

=> “Self referencing”

y = 0,003x + 1,386R2 = 0,161

1,1

1,2

1,3

1,4

1,5

1,6

1,7

1,8

0 10 20 30 40 50 60 70

Teff [°C]

Isc1

000

[A

]

- 16 -

020406080

100120140160180200

20 30 40 50 60 70Modultemperatur (°C)

Mod

ulle

istu

ng P

1000

(W)

c-Si

a-Siµc-Si

DSV

CdTea-Si/µc-Sic-Si

-0.29-0.2-0.42Tk Pmpp (%/K)

59.1122.4185.2Pmpp @ 25 °C (W)

Power and temperature coefficient :Example c-Si, a-Si/µc-Si, CdTe

- 17 -

Parameter determination under low irradianceJune 2011, irradiance interval (Pyranometer) (98…102) W/m²

Linear Correction of measured power to100 W/m²

Plot vs. back temperature

y = -0,097x + 17,764R2 = 0,166

1011121314151617181920

5 10 15 20 25Tback (°C)

P100

(W)

Spectral- and incidance angle influences

No accurate evaluation possible!

- 18 -

Parameter determination under low irradianceJune 2011, Self referencing Eeff = (0.098…0.102) suns

Linear Correction of measured power to 0.1 suns

Plot vs. back temperature

Accurate evaluation possible! low irradiance degradation

y = -0,066x + 16,226R2 = 0,992

1011121314151617181920

5 10 15 20 25Tback (°C)

P100

eff (

W)

- 19 -

Self referencing of irradiance

ZSW Testfeld WIdderstall

- 20 -

Reference cell or pyranometer?

0

0.2

0.4

0.6

0.8

1

1.2

06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

G R

efze

lle/G

Pyr

anom

eter

8.7.

22.7.

Pyranometer

ISET Sensors

Ratio of c-Si sensor to pyranometersunny (8.7.) und cloudy (22.7.) day

• Moderate deviation under direct irradiation around noon time and under diffuse light • strong deviation under direct light morning/evening (incidence angle, spectrum)

Quelle: Kipp & Zonen

- 21 -

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 200 400 600 800 1000 1200

G Pyranometer (W/m²)

Wirk

ungs

grad

bedecktklar

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 200 400 600 800 1000 1200

G Pyranometer (W/m²)

Wirk

ungs

grad

bedecktklar

Efficiency vs. pyranometer irradiance

Data are as measured:

• Cloudy day: high efficiency

• Sunny day:strong influence of spectrumand incidence angle morning/eveningTemperature influence at noon

• NO UNIQUE FUNCTION!

a-Si/µc-Si: Efficiency vs. G

c-Si: Efficiency vs. G

- 22 -

c-Si type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2Eff. irradiance (suns)

norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

CIGS type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00

Time (CET)

Nor

m. e

ffici

ency

cloudy clear

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear

Efficieny vs. effective irradiance (1)

c-Si CIGS

Normalized efficiency vs. effective irradiance with parameter Tmod (top) and daily profile cloudy/clear (bottom)

Similar to c-Si, different at low

light level

- 23 -

c-Si type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear


c-Si CdTe

Normalized efficiency vs. effective irradiance and daily profile cloudy/clear

CdTe type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear

smaller T-coeff.

- 24 -

c-Si type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear


c-Si a-Si/µc-Si


aSi-µcSi type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear

Smaller Tcoeff. different at low

light level

- 25 -

c-Si type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

.effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear

Efficieny v s. effective irradiance(4)

c-Si a-Si/a-Si


aSi-aSi type

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1


norm

. effi

cien

cy

Tmod = 10 °C 25 °C 40 °C 55 °C

0.5

0.6

0.7

0.8

0.9

1

1.1

4:00 8:00 12:00 16:00 20:00Time (CET)

Nor

m. e

ffic

ienc

y

cloudy clear

Best Tcoeff., good at low light

level

- 26 -

Long term stability

ZSW Test Field WIdderstall

- 27 -

Initial Light-Soaking of CIGS Module

P: +10% in 6 Wochen

50

55

60

65

70

75

80

85

Jun. 03 Jul. 03 Aug. 03 Aug. 03 Okt. 03

P (W

), U

oc (V

)P@25°C Uoc@25°C

- 28 -

Initial degradation of a-Si/a-Si Modul

Source: Schott Solar, 21st EU PVSC

Staebler-Wronski-Effect, SWE

- 29 -

Long term stability of CIGS Module

Long term stability CIS @ 1000 W/m²

8.0

8.5

9.0

9.5

10.0

10.5

11.0

Jan. 03 Jan. 04 Jan. 05 Jan. 06 Jan. 07 Jan. 08

Tem

p. c

orr.

effic

ienc

y (%

)

Measurement close to AM1.5 and normal incidence, then temperature correction

- 30 -

Long term stability of CIGS Module @ 800 W/m²

Continuous evaluation possible

0.9

0.95

1

1.05

1.1

0 0.5 1 1.5 2 2.5 3Jahre

P_M

PP_

800(

t)/P

_MPP

_800

(0)

- 31 -

Which degradation rate can be detected?

Here assumed 0,3 % per year

0.9

0.95

1

1.05

1.1

0 0.5 1 1.5 2 2.5 3Jahre

P_M

PP_

800(

t)/P_

MPP

_800

(0)

- 32 -

Effective efficiency, corrected to reference temperatureand normalized to Pnom

0,00

0,20

0,40

0,60

0,80

1,00

1,20

0,0 0,2 0,4 0,6 0,8 1,0 1,2

effective Irradiance [suns]

norm

. Per

form

ance

Meas. Period 1

Meas. Period 2

- 33 -

Low light level degradation / STC-Degradation

Outdoor measurements

0.94

0.96

0.98

1

1.02

1.04

Time ------->

eta(

t)/et

a(0)

1000 W/m² 100 W/m²

Different kinetics

- 34 -

Potential Induced Degradation

(1) Leakage currentindoor

35....85 °C50…95 % -1000 Vsome hours

1,E-07

1,E-06

1,E-05

1,E-04

1,E-03

00:00 12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00

Time (h)

Leak

age

curr

ent d

ensi

ty (

A/m

²)

Reihe1Reihe2

(3) Leakage currentoutdoor

Module temperatureHumidityGenerator Voltagemonths to years

(2) Power degrad. indoor

85 °C85 % -1000 V1000 h

Teils sonniger, trocken-kalter Tag

1,0E-08

1,0E-07

1,0E-06

1,0E-05

03.01.201206:00

03.01.201208:00

03.01.201210:00

03.01.201212:00

03.01.201214:00

03.01.201216:00

03.01.201218:00

Leck

stro

m

[A] Si-TF 2

Si TF 3CIGS 3CIGS 4CIGS 5CdTe 2c-Si 2

Estimation of Time to PID Failure

- 35 -

Time to PID Failure

> 5*E21,4> 300Si-TF 6

2,63533 Si-TF 27,97,823CdTe 2

0,820,6c-Si 2

> 5*E20,5> 87CIGS 4> 5*E20,19> 37CIGS 3

2,51,61,4CIGS 1

Time* to P90 t [a]

Daily* Qdfrom Outdoor

[mC/m²]

Q _ HV-DH for P90[C/m²]

ModuleType

• Time to PID Failure depends on climate

* Average values fall/winter at Widderstall

„PID-Failure“90 % of initial power (P90)

- 36 -

Summary

• Methods for evaluation of characteristic module parameters from fielddata have been presented

• Self referencing uses the module itself as a best matched irradiancesensor

• Module power is a unique function of effective irradiance, transformation to laboratory measurements under reference conditions

• From long term measurements stability trends can be extracted withhigh accuracy

• Time to PID Failure can be estimated

- 37 -

Thank you for your attention!

ZSW Testfeld WIdderstall

outdoor measurements and comparison of thin film solar ... · - pyranometer and c-si ref. cell ......

Documents