Skip to main content

Cell Measurements

NREL researchers measure: 1) current vs voltage (I-V) of both non-concentrator and concentrator solar cells and multijunction devices; 2) spectral responsivity of solar devices; and 3) linearity of short-circuit current and total irradiance for reference cells. Below are the key capabilities related to cell measurements.

Cell Current vs Voltage

For non-concentrator solar cells and multijunction devices, we use the following:

  • Abet 11048 Solar Simulator. This test bed is used to measure the 1-sun I-V characteristics of all cells we evaluate. The test bed measures I-V characteristics of PV cells as large as 30 cm in diameter. It uses a 3-kW xenon arc lamp filtered to provide a standard AM1.5 Global reference spectrum (IEC 60904-3 edition 2). The irradiance is adjustable from 0.2 to 2 suns (for smaller areas) and has a spatial nonuniformity of ±2%. The test device and monitor cell temperatures are controlled separately. The test bed can accommodate any sample that can fit under the beam.

    Related reading:

    K. Emery, Tutorial "Rating PV Power and Energy — Cell, Module, and System Measurements," Proc. 37th IEEE Photovoltaic Spec. Conf., Seattle, WA, June 19–24, 2011, NREL report NREL/PR-5J00-65976, https://www.nrel.gov/docs/fy16osti/65976.pdf.

  • One-Sun Multi-Source Solar Simulator (OSMSS). This test bed is used to measure the I-V characteristics of multijunction cells. The system uses two tungsten and two Xe lamps along with filters to produce light in nine wavelength regions in fiber-optic bundles corresponding to the spectral responsivities of potential multijunction subcells. The system combines the nine channels plus a 10th fiber-optic bundle into a single bundle that goes through a beam homogenizer to produce a uniform, spectrally adjustable beam that is 9 cm x 9 cm in size at 1-sun. Each channel can be adjusted with automated variable apertures from 0% to 100%. The spectrum is automatically adjusted so that each subcell is within 1% of its target photocurrent. This has been demonstrated for a high-efficiency multijunction cell with six subcells. The system is capable of measuring the energy that a cell would produce in a year based on a set of spectral irradiances, total irradiance, and temperatures derived from the typical meteorological year.

    Related reading:

    T.E. Moriarty, R.M. France, and M.A. Steiner, "Rapid, Enhanced IV Characterization of Multi-junction PV Devices under One Sun at NREL," Proc. 42nd IEEE Photovoltaic Spec. Conf. New Orleans, LA, June 14–19, 2015.

    T. Moriarty, J. Jablonski, and K. Emery, "Algorithm for Building a Spectrum for NREL One-Sun Multi-Source Simulator," Proc. 38th IEEE Photovoltaic Spec. Conf., Austin, TX, June 3–8, 2012, pp. 1291–1295.

For concentrator solar cells, we use the following:

  • Tunable High-Intensity Pulsed Solar Simulator (T-HIPSS). The system is used to measure I-V characteristics of high-efficiency multijunction concentrator solar cells. The T-HIPSS is a commercial system from Spectrolab with a temperature-controlled vacuum plate that has an electrically isolated voltage contact. It has six separate spectral adjustment ranges plus lamp voltage and a filter plate to adjust the spectrum, along with a dedicated 300–1700-nm spectroradiometer. The 1-ms pulses of light with an intensity of up to ˜1000-suns concentration illuminate a 4 cm × 5 cm area. Linearity is measured with a set of mesh screens covering two decades of attenuation. The system measures the spectrum as it is adjusted so that it can be easily tuned for state-of-the-art multijunction concentrating PV cells.

    Related reading:

    C.R. Osterwald, M.W. Wanlass, T. Moriarty, M.A. Steiner, and K.A. Emery, "Effects of Spectral Error in Efficiency Measurements of GaInAs-Based Concentrator Solar Cells," NREL Technical Report NREL/TP-5200-60748, March 2014.

    C.R. Osterwald, M.W. Wanlass, T. Moriarty, M.A. Steiner, and K.A. Emery, "Concentrator Cell Efficiency Measurement Errors Caused by Unfiltered Xenon Flash Solar Simulators," Proc. CPV-10, AIP Proceedings volume 1616, ISBN: 978-0-7354-1253-8, DOI: /10.1063/1.4897049, pp. 149–153, Albuquerque, New Mexico, April 7–9, 2014.

    C.R. Osterwald, M.W. Wanlass, T. Moriarty, M.A. Steiner, and K.A. Emery, "Empirical Procedure to Correct Concentrator Cell Efficiency Measurement Errors Caused by Unfiltered Xenon Flash Solar Simulators," Proc. 40th IEEE Photovoltaic Spec. Conf., June 8–13, 2014, Denver, CO.

  • High-Intensity Pulsed Solar Simulator (HIPSS). The HIPSS is a commercial system from Spectrolab with a temperature-controlled vacuum plate that has an electrically isolated voltage contact. It can accommodate 10 cm × 10 cm cells. The system is used to measure I-V characteristics of both concentrator solar cells and thermophotovoltaic (TPV) cells. Its light source is two low-pressure xenon arc lamps that are adjusted between 1,200 and 3,200 V. They deliver 1-ms pulses of light with an intensity of up to 2 × 106 Wm-2 and a spatial nonuniformity of ±3% over the area of 17 cm × 3 cm. The beam is adjustable to provide concentrations of 1 to 2,000 suns. The spectrum can be tuned for two junctions by setting the lamp voltage and using apertures to set the intensity level.

    Related reading:

    J. Kiehl, K. Emery, and A. Andreas, "Testing Concentrator Cells: Spectral Considerations of a Flash Lamp System," Proc. 19th European Photovoltaic Solar Energy Conference and Exhibition, Paris France, June 7–11 2004, pp. 2463–2465.

  • Continuous Illumination Concentrator Simulation System. For concentrator cells, this system uses a 1-kW short-arc xenon lamp. The light from the xenon source is reflected off a mirror onto a concentrator lens mounted on a translation stage. The system can be adjusted to achieve concentration ratios of 0.1 to 200 suns over an area that ranges from 4 cm2 to less than 0.1 cm2. This system is useful for measuring slow-responding cells under concentrated light. A 1-ms shutter is used to measure open-circuit voltage (Voc) prior to heating, allowing the stage to be cooled to obtain the I-V at a known temperature.

The following table provides a condensed list of characteristics for cell I-V measurement test beds.

Major Instrumentation for Cell I-V Measurements
System Typical Applications Special Features Light Source Test Bed Voltage Resolution/Limit Current Resolution/Limit
Abet 11048 solar simulator 1-sun I-V measurements for cells and small modules Wide current and voltage ranges Filtered 3-kW Xe; 0.1 to ˜10 suns 30 cm diameter; 15° to 50°C 5 µV / ±50 V ±10 pA to ±16 A
One-sun multisource simulator (OSMSS) Multijunction cells, I-V vs. irradiance and temperature 10 separate adjustable wavelength bands; dedicated spectroradiometer 2 Xe; 2 tungsten lamps; fiber optic 9 cm × 9 cm; 10° to 80°C 0.1 µV / ±40 V 1 nA to ±5 A
Tunable high-intensity pulsed solar simulator (T-HIPSS) I-V measurements for concentrator and TPV cells Spectrally adjustable; ˜1-ms flash; minimal heating; dedicated spectroradiometer 2 Xe flash lamps 30 cm long with mirror; 1 to ˜1000 suns 4 cm × 4 cm, <1 cm2 typical); 15° to 80°C 0.1 mV / 20 V 100 µA to 20 A
High-intensity pulsed solar simulator (HIPPS) I-V measurements for concentrator and TPV cells Spectrally adjustable; ˜1-ms flash; minimal heating 2 Xe flash lamps 30 cm long with mirror; 1 to 2000 suns 10 cm × 10 cm; 5° to 80°C 0.1 mV / 100 V 500 µA to 50 A
Continuous illumination concentrator simulation system I-V measurements for concentrator cells Spectrally adjustable; user-controlled bias conditions 1-kW Xe or 3-kW tungsten; 0.1 to 200 suns ˜1-cm diameter for Xe; 5° to 80°C 5 µV / ±10 V ±1 µA to ±10 A

Spectral Responsivity

Spectral responsivity (SR) measurement is an important part of the NREL PV device performance assessment process. Spectral responsivity systems measure how a device responds to selected narrow (spectral) bands of irradiance. Responsivity is measured in units of amps per watt vs wavelength and reported in terms of quantum efficiency (QE) — a measure of how efficiently a device converts incoming photons to charge carriers in an external circuit. SR systems measure the spectral response at different temperatures (10° to 80°C) voltages (±15 V), light levels (0 to few suns), and different chopping frequencies (<0.2 to 400 Hz).

We use two SR systems:

  • Filter System. This system is used for solar cells and modules and has much more power in the beam than the grating system. Typical bandwidth is 10 nm, but has a 20- to 50-nm wavelength interval from 290 to 1,900 nm.
  • Grating System. This system is used for applications that require a narrow bandwidth and wavelength interval is required. This system has a typical minimum beam size of 1 mm × 3 mm, which makes it ideal for absolute spectral responsivity measurements.

Although important differences exist between the two systems, the basic procedures are similar: a wide-spectrum light source is chopped and filtered or diffracted into a discrete succession of narrow spectral bands, each of which is directed onto the test device. The device current produced from the monochromatic light is converted to an alternating current (ac) voltage signal. A lock-in amplifier locks into the chopper frequency of the light signal and measures the corresponding ac voltage produced by the light. Using a time-periodic light signal and a lock-in amplifier allows us to distinguish signals produced by the relevant spectral band from those that may be produced by other light sources.

Each system is controlled by a computer. Once the operator sets the parameters, the computer does the rest: runs the procedure through the selected wavelength range, acquires the data, calculates QE, saves the data, and updates the directory in a standardized manner as a tab-delimited text file.

Related reading:

  • K. Emery, D. Dunlavy, H. Field, and T. Moriarty, "Photovoltaic Spectral Responsivity Measurements," Proc. 2nd World Conference and Exhibition on Photovoltaic Solar Energy Conversion, Vienna Austria July 6–10, 1998, Joint Research Center report EUR 18656, pp. 2298–2301.

  • K. Emery, Tutorial "Rating PV Power and Energy — Cell, Module, and System Measurements," Proc. 37th IEEE Photovoltaic Spec. Conf., Seattle, WA, June 19–24, 2011, NREL report NREL/PR-5J00-65976, https://www.nrel.gov/docs/fy16osti/65976.pdf.

The following table is a condensed list of major instrumentation characteristics for spectral response measurements.

Major Instrumentation for Spectral Response Measurements
System Typical Applications Special Features Light Source Wavelength Range Bandwidth Voltage Bias Light Bias
Filter spectral responsivity SR measurements for solar cells and modules; 15°C-80°C High flux density; variable beam size; 61 filters on four filter wheels; adjustable chopping frequency 1-kW Xe 280 to 1,900 nm 10 nm full width at half maximum (FWHM) ±40 V Up to 200 mA
Grating spectral responsivity SR measurements for small-area and multijunction cells; 15°C-80°C 3 gratings for visible and infrared; adjustable chopping frequency 75-W Xe 300 to 3,000 nm >1 nm FWHM ±5 V Up to 200 mA

Linearity

The linearity of the short-circuit current (Isc) with total irradiance is an important measurement for reference cells because the standards require the reference cell to be linear over its range of operation. NREL measures the linearity of Isc in the range of 0 to 2 suns using two lamps and neutral-density filters. If the sample is linear, then the Isc from one lamp plus the Isc from the other lamp must equal the Isc when both lamps are on. This value is expressed as a percentage deviation from linearity and is typically measured over the range of 0.1 to 1.1 sun. This method is insensitive to the spectrum and spatial nonuniformity of the light from the two lamps at varying light levels.

Related reading:

  • K. Emery, S. Winter, S. Pinegar, and D. Nalley, "Linearity Testing of Photovoltaic Cells," Proc. IEEE 4th World Conference on Photovoltaic Energy Conversion, pp. 2177–2180, May 7–12, 2006 Waikoloa, HI.

  • K. Emery, Tutorial "Rating PV Power and Energy — Cell, Module, and System Measurements," Proc. 37th IEEE Photovoltaic Spec. Conf., Seattle, WA, June 19–24, 2011, NREL report NREL/PR-5J00-65976, https://www.nrel.gov/docs/fy16osti/65976.pdf.