Yankee Environmental Systems Home Page

Made in USA

Optical Calibration Services


Most sensors must be characterized in terms of engineering units before they can be used. Our Optical Calibration Facility is a special purpose facility devoted entirely to the development of the characterization and calibration of optical radiation instrumentation. Our optical laboratory facilities give us the capability to fully characterize the performance of all YES manufactured optical instruments in terms of their spectral, absolute and cosine responses. In addition, we can also characterize the responses of many other optical instruments, such as total solar pyranometers and other radiometers. If you would like to discuss your custom optical characterization needs with us, please contact us. The YES optical laboratory includes four major facilities for the measurement of the spectral, cosine and absolute responses of optical detectors. These facilities are described below.

Cosine Response Facility (CRF)

The purpose of this facility is to measure the cosine response of a detector under test (DUT). Most instruments are designed to detect radiation incident on a flat surface. The output signal of an ideal instrument of this type is proportional to the cosine of the angle of incidence of the light (as measured with respect to the normal of the surface). This reflects the fact that the effective illuminated sensitive surface of the instrument is proportional to the cosine of the angle of incidence of the light. Deviations from an ideal cosine response can lead to measurement errors of the direct-normal beam for light sources, such as the sun, which illuminate the detector at range of incidence angles. Measured instrumental cosine response is traditionally expressed as the ratio of the measured response to the ideal cosine response of a Lambertian receiver. The MFR instrument family of field radiometers uses this measurement to perform automated cosine corrections of direct-normal data.

The cosine response measurement of the DUT is carried out on a precision rotary mechanical table and a custom designed motor controller. The DUT is illuminated by a uniform, parallel light beam during rotation over ± 90 degrees and its output signal is recorded. The ratios of the raw response of DUT vs. a true cosine, normalized to the output at 0° with respect to the beam, produces the "cosine response" of the instrument. The entire process is conducted under computer control. A schematic diagram of the YES Cosine Response Facility (CRF) is shown in Figure 1.

Figure 1. Cosine Response Facility for determining angular error function.

An Oriel 500 Watt Xe-Hg arc lamp serves as the CRF light source. The light output of the lamp is monitored and controlled by an Oriel Model 68850 photo-feedback system. A temperature-controlled silicon photodiode, which views the lamp cavity through a 1 mm diameter pinhole and a 1% transmission neutral density filter, is the feedback system light sensor. The output signal of the photodiode is used to adjust the electrical current supplied to the lamp in order to maintain a constant intensity of the light. This system helps to stabilize the intensity of the light beam such that it does not vary during the time of the scan.

The image of the CRF lamp arc is focused into a parallel beam down a 5 m long collimating tube, painted black on the inside and equipped with anti-scatter baffles. The downstream end of the drift tube is connected to a large enclosure, also painted black on the inside. Inside the enclosure is a custom designed, heavy duty computer-controlled turntable. The output signal of a detector undergoing testing is digitized using either a Data Translation DT-2829 16 bit data acquisition board, a 6½ digit Kiethley 2000 10 channel DVM or a 5½ digit Fluke 45 DVM, depending on the model of the DUT. When the CRF was initially developed, YES engineers verified beam spot uniformity at the DUT using a silicon photodiode detector with 1 mm diameter collimator. This detector was moved linearly across the output of the tube and the output signal was recorded as a function of position.

During a CRF test run, a DUT is then mounted in place and the rotary actuator moves it from -90° to +90° , in angular increments of 1 to 5 degrees (depending on the DUT), and its output signal is recorded. If necessary, the angular scans may be repeated several time for increased accuracy. Once the angular scans are completed, the cosine response of the DUT is computed by dividing the recorded values of the DUT output signals at the various angles by the DUT output signal value at 0° with respect to the beam. A typical cosine response plot, obtained using the YES CRF is shown in Figure 2.

Figure 2. Example of a typical measured MFR cosine response (error function).


Figure 3. Schematic diagram of the Spectral Response Facility.

Spectral Response Facility (SRF)

The schematic diagram of the YES Spectral Response Facility (SRF) is shown in Figure 3. The SRF characterizes the relative spectral response of an optical radiometer. The DUT is illuminated by a monochromatic beam and its output signal is recorded. A portion of the light beam is directed via a beam splitter onto a reference photodetector which monitors the beam intensity. The output signals of the DUT and the reference detector are recorded as the beam wavelength is varied. The spectral response of the DUT at each wavelength is determined by dividing its output signal at that wavelength by the beam intensity, as determined by the reference detector. The SRF can be configured to permit relative spectral response measurements in the range of 270 to 1100 nm, and has a wavelength resolution of 0.1 nm.

The SRF light source is a Oriel 150 Watt Xe arc lamp. The light beam is focused onto the entrance slits of an Acton Research Corp. SpectraPro 275 monochromator equipped with a 1200 line/mm holographic grating. Focusing is performed to match the monochromator input optic f number (f3.5). The monochromatic beam passes through the exit slits and strikes a beam splitter element, mounted at an angle of 45° with respect to the beam. This element transmits approximately half of the light and reflects the remainder. The transmitted fraction is directed to a Gamma-Scientific DR-2550-2BNC NIST-traceable silicon photodiode which serves as the beam intensity monitor. The reflected portion of the monochromatic beam strikes the DUT. This technique normalizes the detector response to eliminate the effects of uneven spectral distribution that is characteristic of all high pressure Xe arc lamps. The major components of the automatic test system (ATE) include the following components:

  1. An ARC 275 monochromator that is connected via an RS-232 port to a PC running system software that controls the wavelength setting,
  2. A DR-2550-2BNC reference detector connected to an EG&G PhotoComp electrometer which digitizes the photocurrent. The output is provided via a second RS-232 port permits the systems software to record the detector photocurrent,
  3. A DUT whose output is connected one of several digitizing instruments, a Data Translation 16 bit ADC DT-2829 data acquisition board, a 5½ digit Fluke 45 DVM or a 6½ digit HP 34401 DVM. The choice of digitizer depends on the type of DUT.

The normal operation of the SRF during a wavelength scan proceeds in the following way:

  1. The system software sets the monochromator to a desired wavelength, following directions in a "scan list" file,
  2. The reference detector ouput photocurrent is measured and the measured value is stored by the software,
  3. The DUT output signal is measured and stored by the system software,
  4. Steps 1-3 are repeated for all wavelengths of interest until the end of the scan list,

Once the spectral scan is complete, the relative spectral response of the DUT is calculated from the stored data. The measured reference detector photocurrent at each wavelength is converted to an irradiance using its NIST-traceable spectral calibration. This irradiance is then used to normalize the DUT output signal to the beam intensity at each measurement wavelength. The set of normalized DUT output signals gives the relative spectral response of the instrument. If the DUT is an MFR-7, UVMFR or SPUV the scan output is later used to create "calibration information" file by the ARF as well as a .SPN report file. If the DUT is a UVB-1 or UVA-1 the systems oftware creates an ASCII relative spectral response scan file that can be used by our UV_Calc UV modeling software. A typical normalized relative spectral response plot is shown in Figure 4 for a visible/NIR MFR-7 instrument with six narrow band filter channels.

Figure 4. Examples of measured MFR relative spectral passband responses.

If the DUT is an imaging CCD detector (such as used in our TSI or RSS), its spectral response can also be determined using a second SRF configuration, where both the reference detector and the DUT are mounted on the exit ports of a specialized 12", SpectralonÔ lined integrating sphere that was custom designed for YES, Inc. by LabSphere Inc. One side advantage of this scheme is that, the knowledge of the port geometry can theoretically permit the measurement of absolute spectral response of the DUT under limited circumstances. Another more practical advantage is that it is a diffuse source and is therefore the only configuration suitable for far field illumination of CCD image sensors. A disadvantage to this configuration is that the dynamic range of the measurement is significantly less than in the standard SRF configuration due to the inherently low optical throughput (efficiency) of the integrating sphere geometry.

Absolute Response Facility (ARF)

The purpose of this facility is to measure the absolute response of a narrow bandwidth detector or radiometer to incident irradiance. This response is sometimes referred to as the instrument or detector voltage sensitivity transfer function. A schematic diagram of the YES Absolute Response Facility (ARF) is shown in Figure 5. A key component of the system is a 1,000 Watt calibrated, NIST-traceable FEL lamp, that acts as a spectral irradiance transfer standard. The calibration of the FEL lamp is an accurate determination of the lamp's absolute spectral irradiance at a distance of 50 cm from the lamp over the working spectrum of the bulb. Measurement of the response of the DUT while exposed to the known lamp irradiance convolved with the DUT's relative spectral response determines the absolute voltage transfer function of the DUT.

Figure 5. Schematic diagram of the Absolute Response Facility.

The FEL lamp and the DUT are mounted precisely 50 cm apart on mechanical fixtures inside a light-tight enclosure with black walls. The black walls and two light baffles help ensure that only direct radiation from the lamp reaches the DUT. Extra care was taken to reduce infrared reflecting off the rear wall of the enclosure. Since the FEL lamp calibration is only valid only when the electrical current through the lamp is 8 A, the ARF utilizes precision equipment to ensure that the lamp current is held constant with an accuracy of better than 0.8 mA (1 part in 10,000). The ARF monitors the FEL lamp current continuously to verify that it is within the allowable current range, and the operator can dither the current a small amount to make adjustments (we do not rely on the built-in regulation of the HP 6030A FEL supply.) The HP 6030A current output is controlled via a digital-to-analog converter output by a Data Translation DT2829 data acquisition card, under software control. The lamp current passes through a precision high power shunt resistor (Julie Research Corp. Model CS-1R-10-01A, 0.1 W , 0.01% accuracy), and the voltage drop across the shunt is measured by a NIST-traceable, HP 34401 6½ digit DVM. The output signal of the DUT, while illuminated by the lamp, is digitized by a ten input 6½ Kiethley 2000 DVM and recorded.

The predetermined FEL absolute lamp spectral irradiance is mathematically convoluted with the relative spectral response of the DUT (measured earlier in the SRF) to arrive at the effective power incident on the DUT over its passband. The measured value of the DUT output signal is then divided by the effective power to calculate the absolute response coefficient of each channel of the DUT. A typical intermediate irradiance QC report is shown below in Figure 6. If the DUT is an MFR-7 or UVMFR the system processes the voltage response and previously measured normalized spectral response info a "calibration information" file with summary spectral scan data appended to the end of this file.

As routine maintenance, the voltage transfer function of the various ATE digitizing systems used in the ARF are periodically calibrated either by a NIST traceable standards lab or via a precision voltage source and a NIST-traceable 6½ digit HP 34401 DVM. The three precision shunt resistor is measured yearly by the manufacturer to verify its stability, and lamps are recharacterized after 50 hours of run time, which is logged by the system software.


This file: D436B.IRR   Calibrating MFRSR head and datalogger.
Date:  12/21/1998
Head ID: $C38E      System ID: $0000       FEL lamp distance =  50.00 cm
Narrowband channel information: 
Chan  L_m     FWHM    L_c     EBW     SpIrr       Sens         Offset    Sig
      (nm)    (nm)    (nm)    (nm)    uW/cm2-nm   V/W/m^2_nm   volts dc   (mV)
 1   413.9     9.9   414.5     9.7     2.4871    -0.2063     0.0029     -5.13
 2   497.8     9.3   497.6     9.0     6.6263    -0.3488     0.0023    -23.11
 3   614.8     9.9   614.4    10.0    13.7303    -0.7290     0.0025   -100.09
 4   672.3     9.8   672.6     9.7    16.7822    -1.0561     0.0022   -177.24
 5   869.8    11.0   869.4    11.2    22.0465    -1.9476     0.0026   -429.37
 6   938.1    10.7   937.6    10.9    22.2078    -2.5111     0.0014   -557.67

Figure 6. Example of a spectral irradiance QC report

Outdoor Test Facility (OTF)

In addition to the indoor optical laboratory YES maintains a fully instrumented field test station for monitoring solar radiation, evaluating new sensor designs and performing instrument calibrations using the sun as a source of irradiance. The test stand is equipped with two independent PC based data acquisition systems that allow up to 48 instruments be placed under test concurrently. The outdoor facility instrumentation includes:

  • YES TSP-700 WMO secondary standard total solar radiometer for measuring total global visible/NIR solar radiation (300-3000 nm),
  • EPLAB NIP Normal Incidence Pyrheliometer for measuring broadband direct-normal visible/NIR solar radiation (300-3000 nm),
  • YES PVT-3030 NIST-traceable, shielded and aspirated platinum resistance thermometer for monitoring ambient temperature,
  • YES UVB-1 and UVA-1 broadband reference instruments covering 280-320 nm and 280-400 nm respectively, and
  • YES MET-2010 Optical chilled mirror dew point hygrometer for monitoring atmospheric water vapor.

Unlike narrowband filter radiometers such as the MFR-7 and UVMFR and SPUV, broadband TSP, UVA and UVB-1 instruments are not absolute calibrated in the ARF since the shape and distribution of the FEL lamp spectrum differs vastly from that of the sun, (especially in the UV portion of the spectrum). Broadband radiometric instruments intended to measure atmospheric radiation are calibrated using the sun as a source of stable spectral irradiance against carefully monitored and controlled transfer detectors. Broadband visible/NIR Total Solar Pyranometers are calibrated against co-located pyranometers who are ultimately traceable back to WMO standards located at the World Radiation Center in Davos, Switzerland.

OTF broadband UV radiometer transfer detectors are characterized using co-located narrowband spectroradiometers, that are calibrated with NIST-traceable FEL lamps. By plotting the integrated wavelength scan of the spectroradiometer against the co-located broadband detector output over a range of solar zenith angles (e.g. a clear day in the summer), the slope of the X/Y plot determines the broadband radiometer's absolute calibration factor.


Copyright © 1998-2006 Yankee Environmental Systems, Inc., All Rights Reserved. Privacy Notice
This page was last updated on Monday, September 11, 2006 .