The Laser-DSR Method

29.10.2015

New Developments in the Field of Primary Calibrations

of Reference Solar Cells

Overview

o Demand for high accuracy solar cell calibrations

o Standard Test Conditions

o DSR Method

o The new PTB setup: Advantages and disadvantages

o Validation by International Comparisons => WPVS

Economic Impact of Measurement Uncertainty for Photovoltaics

Source of data: EPIA (European Photovoltaic Industry Association) and for 2015 according to IHS market research institute

• Financial uncertainty = Global annual installation Price Uncertainty• 2012: Financial uncertainty = 30 GW/year 1.7 €/W 1 % = 500 M€/year

A measurement uncertainty of 1% leads to a financial uncertainty of 500 M€/year High demand for high accuracy solar cell calibrations

Solar parks are financed from banks, who add the financial uncertainty arising from measurement uncertainty to the total amount to be financed. A low uncertainty leads to competitive advantage.

Source of data: www.solarwirtschaft.de/preisindex

2006

2007

2008

2009

2010

2011

2012

0

1

2

3

4

5

Syste

m p

rices for

end c

usto

mers

in

Eu

ro / W

att

Prices in Euro/Watt

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015*0

10

20

30

40

50

Yearly w

orldw

ide n

ew

lyin

sta

lled P

V P

ow

er

/ G

W

New PV installations

Economic Impact of Measurement Uncertainty for Photovoltaics

Source of data: EPIA (European Photovoltaic Industry Association) and for 2015 according to IHS market research institute

To achieve lower uncertainties, in addition more realistic standards are needed

Energy rating instead of Maximum Power for PV classification (EMRP-Project PhotoClass) Energy rating approach needs much more parameters

• Temperature dependence• Non-Linearity• Angular dependency

Source of data: www.solarwirtschaft.de/preisindex

2006

2007

2008

2009

2010

2011

2012

0

1

2

3

4

5

Syste

m p

rices for

end c

usto

mers

in

Eu

ro / W

att

Prices in Euro/Watt

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015*0

10

20

30

40

50

Yearly w

orldw

ide n

ew

lyin

sta

lled P

V P

ow

er

/ G

W

New PV installations

Demand for a multifunctional calibration facilities

Standard Test Conditions

• Reference solar spectrum AM1,5

• Irradiance ESTC = 1000 W/m²

• Cell-Temperature (25°C)

• Angular distribution important, but not defined

The calibration procedure must take into account these STC according to IEC 60904-3.

Metrological background

,NormPhotocurrent: ( ) d( )E sI

Reference solar spectrum AM1.5 according to IEC 60904-3:2008

Spectral responsivity of different solar cells

200 400 600 800 1000 1200 1400 1600 1800Wavelength / nm

2468 135Photon energy / eV

0,0

0,2

0,4

0,6

0,8

Abso

lute

spec

tral

res

ponsi

vit

y, s

/ A

W-1

= 100%c-Si

GaInAs

CdTe

Poly-Si

CIS

a-SiGaInP

Ge Component Cell

400 600 800 1000 1200Wavelength, / nm

0

500

1000

1500

2000

2500

Sp

ectr

al i

rrad

ian

ce E

/

mW

m-2

nm

-1

AM1.5 (IEC 60904-3:2008)

Metrological background

,NormPhotocurrent: ( ) d( )E sI

Reference solar spectrum AM1.5 according to IEC 60904-3:2008

Spectral responsivity ofdifferent solar cells

200 400 600 800 1000 1200 1400 1600 1800Wavelength / nm

2468 135Photon energy / eV

0,0

0,2

0,4

0,6

0,8

Abso

lute

spec

tral

res

ponsi

vit

y, s

/ A

W-1

= 100%c-Si

GaInAs

CdTe

Poly-Si

CIS

a-SiGaInP

Ge Component Cell

400 600 800 1000 1200Wavelength, / nm

0

500

1000

1500

2000

2500

Sp

ectr

al i

rrad

ian

ce E

/

mW

m-2

nm

-1

Sun simulator with AM1.5 filterfrom its data sheet

AM1.5 (IEC 60904-3:2008)

Comparison of integral and spectral measurements

Integral Measurement Spectral Measurement

Spectrum can not be interpolated+ Spectral responsivity can be

interpolated

Uncertainty due to the measurement of the spectrum, e.g. spectral straylightwithin the spectroradiometer

+ The AM1.5 spectrum of the standard can be used directly

=> No uncertainty component

+ High signal Low signal,NormPhotocurrent: ( ) d( )E sI

200 400 600 800 1000 1200 1400 1600 1800Wavelength / nm

2468 135Photon energy / eV

0,0

0,2

0,4

0,6

0,8

Abso

lute

spec

tral

res

pon

sivit

y, s

/ A

W-1

= 100%c-Si

GaInAs

CdTe

Poly-Si

CIS

a-SiGaInP

Ge Component Cell

400 600 800 1000 1200Wavelength, / nm

0

500

1000

1500

2000

2500

Sp

ectr

al i

rrad

ian

ce E

/

mW

m-2

nm

-1

Sun simulator with AM1.5 filterfrom its data sheet

AM1.5 (IEC 60904-3:2008)

Solar cell

Bias radiationEb

Modulated monochromatic

radiation, measured with Ref.-PD

DE()

Ib + DIsc(, Eb)

DC-Multimeter + Lock-In-Amplifier

DSR-Method Differential Spectral Responsivity

0 20 40 60Time t

ms0

2

4

6

8

10

Irra

dia

nce

E

: Monochromatic + bias radiation

: Bias radiation

𝑠 𝐸𝑏 =∆𝐼(𝜆, 𝐸𝑏)

∆𝐸

=d𝐼

d𝐸

∆𝐸 ≪ 𝐸𝑏

Result of DSR measurement

Absolute differential spectral responsivity ~s (, I(E)) of polycrystalline Si solar cell

400 600 800 1000 1200

Wavelength, in nm

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

~ sab

s i

n

mA

W /

m²

0.01 0.1 1 10 100I in mA

0.1 1 10 100 1000E in W / m²

75

80

85

90

95

~ s i

n

mA

10

00 W

/ m

²

-20%

-15%

-10%

-5%

0%

5%

Polycrystalline Si

AM1.5( ) ( ) dE s s =

𝐼STC

0𝐼STC 1

𝑠(𝐼)d𝐼

Calibration objects

Reference solar cells

200 400 600 800 1000 1200 1400 1600 1800Wavelength / nm

2468 135Photon energy / eV

0,0

0,2

0,4

0,6

0,8

Abso

lute

spec

tral

res

ponsi

vit

y, s

/ A

W-1

= 100%c-Si

GaInAs

CdTe

Poly-Si

CIS

a-SiGaInP

Ge Component Cell

Component solar cells for Space Application

Laboratory method instead of balloon flight

Calibration objects

Industry solar cells

absolute differenzielle spektrale Empfindlichkeiten ~s (, I(E)) von ISFH Referenz

400 600 800 1000 1200

Wellenlänge, in nm

0

2

4

6

8

10

12

14

16

~ sab

s in

m

A

W /

m²

0%

1%

2%

3%

4%

5%

rel.

Un

sich

erhei

tk=

2

: I = 27,4 mA 30

: I = 191,6 mA 32

: I = 1010,1 mA 34

: I = 2887,4 mA 35

: I = 4481,8 mA 36

: s(, 1000 W/m²)

: Ableitung

Validation of the DSR method for non Si responsivities

• Comparison of component solar cells with “extraterrestrial” calibration by ESA

• The results agree within 0.4% and 0.6%.

300 400 500 600 700 800 900Wavelength, / nm

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

~ sab

s /

mA

W /

m²

C. Baur, G. Siefer, R. Kern, S. Winter: “Primary solar cell standards – comparison of extraterrestrial and synthetic calibration”, in Proceedings of “9th European Space Power Conference (ESPC)”, (Saint Raphael, Juni 2011)

Realisation und maintainance of the„World Photovoltaic Scale“

NREL

PTB

TIPS

AIST

EURAMET supplementary comparison S.5

o Started in 2013, Draft A this year

o Star-like order: Participant -> PTB -> Participant

o Calibration of Isc under STC

o Participants

o PTB (Coordinator)

o VNIIOFI

o KRISS

o A*STAR

o ITRI

o NIM

o Asked for participation in the next round: NIST

How to measure the Differential Spectral Responsivity

o Xenon or quartz halogen lamp based system (DSR, SCF)(or laser-driven xenon lamps or supercontinuum systems)

+ Easy to use

Low Power (100 µW) with subsequent problems

Uniformity

Large bandwidth

Bad Signal-To-Noise especially at high bias levels

Rel. & abs. measurement needed

Size of the solar cells is limited

S. Winter, T. Wittchen, J. Metzdorf in Proc. 16th EU PVSEC, (Glasgow 2000)

The new Laser-DSR setup

o Tunable laser substitutes lamps

Power increases up to a factor 100 – 10000

+ Good uniformity

+ Low bandwidth possible

+ No interpolation between relative and absolute measurement

monitorphotodiode

Bias lamps with dichroic mirrors for absolute calibration

Absolute spectral responsivity

ModelockedTi:Sapphire

laser

Optical parametricOscillator (OPO),SHG, THG, FHG

Pulse-to-CWconverter

Bandpasslimitation

(monochromator)

Pulsed with 80 MHz CW signal

680 nm – 1080 nm 190 nm – 4000 nm

Many further improvements in detail

x,y,z-Table

Monochromator

Optics (Lenses,

Apertures,

Monitor phodiode)

Biasradiator

Φ-θ-Goniometer

Pulse-to-CW

Convertor

(Fiber)

Chopper

Climate Chamber

Lasersystem

(Ti:Saphir Laser,

OPO, SHG, THG,

FHG)

Electronic

Compact Array

Spectroradiometer

Scientist

PhD-Student

Design of the new Facility

x-y-z-Table + GoniometerLaser systemBias RadiationOptics

Laser-DSR- Facility: From Design to Realisation

200 400 600 800 1000 1200 1400 1600

wavelength, / nm

1234567 1,5

photon energy / eV

10-4

10-3

0,01

0,1

1

10

100

1000

104

105

106

107

op

tica

l pow

er /

µW

GW: switch of monochromator grating

: DSR Xenon - lamp, incident on solar cell, measured with Si-PD

: DSR FEL-lamp, incident on solar cell, measured with thermocolumn

: LDSR, incident on solar cell, measured with powermeter

: LDSR, direct laser output, measured with powermeter

• Factor 100-10 000 higheroptical power than with theold facility

• Spectral range 200-1600nm

Measurement results: Radiation power

Conclusion

o DSR method is well suited for the primary calibration

of different types of solar cells

o The method is validated for many different

stable single junction solar cells

o PTB develops the next-generation of the DSR facility:

LASER-DSR

o The high power improves the signal to noise ratio and

enables higher resolution and better uniformity

o It is a multipurpose spectral comparison facility.

s = s(, E, fChopper, T, x, y, z, , )

Conclusion

Thank you for your attention

A part of the work was funded by EMRP.

The EMRP is jointly funded by the EMRP participating

countries within EURAMET and the European Union.

“Towards an energy-based parameter for photovoltaic classification”

The new Laser-DSR setup

o Tunable laser substitutes lamps

Power increases up to a factor 100 – 10000

+ Good uniformity

+ Low bandwidth possible

+ No interpolation between relative and absolute measurement

monitorphotodiode

Bias lamps with dichroic mirrors for absolute calibration

Absolute spectral responsivity

ModelockedTi:Sapphire

laser

Optical parametricOscillator (OPO),SHG, THG, FHG

Pulse-to-CWconverter

Bandpasslimitation

(monochromator)

Pulsed with 80 MHz CW signal

680 nm – 1080 nm 190 nm – 4000 nm

Why do we need a monochromator behind the laser?

The new Laser-DSR setup

o Tunable laser substitutes lamps

Power increases up to a factor 100 – 10000

+ Good uniformity

+ Low bandwidth possible

+ No interpolation between relative and absolute measurement

monitorphotodiode

Bias lamps with dichroic mirrors for absolute calibration

Absolute spectral responsivity

ModelockedTi:Sapphire

Laser

Optical parametricOscillator (OPO),SHG, THG, FHG

Pulse-to-CWConverter

Bandpasslimitation

(Monochromator)

Pulsed with 80 MHz CW signal

680 nm – 1080 nm 190 nm – 4000 nm

Pulse to cw converter

L0 = l0L1 = l0 + 2.5 cmL2 = l0 + 5.0 cmL3 = l0 + 7.5 cm…

L99 = l0 + 247.5 cm

t0 = t0

t1 = t0 + 0.125 nst2 = t0 + 0.250 nst3 = t0 + 0.375 ns…

t99 = t0 + 12.375 ns

And after 12.5 ns the next pulse from the laser appears.

Fiber bundle with 100 multimode fibers , each with an individual length

12.5 25 37.5 4

Time / ns

0.0

0.2

0.4

0.6

0.8

1.0

Sig

nal

: Laser Signal

: Signal with

"Pulsed in

cw-Converter"

Pulse to cw converter

Fiber 1

Fiber 1 andFiber 50

Fiber 1, 33 and 66

Fiber 1, 25, 50 and 75

Pulse to cw converter

All 100 fibers