Quality Manual of Spectral Irradiance Measurements

Transcription

1 Aalto University School of Electrical Engineering Metrology Research Institute Petri Kärhä Jari Hovila Maksim Shpak Tuomas Poikonen Hans Baumgartner Quality Manual of Spectral Version /12/2015

2 Page 2 (46) 1. Table of contents Quality Manual of Spectral Table of contents Definition Scope Object and field of application Features Principle of realisation Equipment Description of setups Calibration requirements Filter radiometer Spectroradiometers Measurement traceability Calibration and measurement procedures including validation methods Measurement ranges Measurement range using the filter radiometer Measurement range using the DM150 spectroradiometer Measuring a primary standard lamp Analysing the results Calculating colour co-ordinates and colour temperature Adjusting colour temperature Using the DM150 spectroradiometer Installing the setup Preparations Filter wheel and wavelength indicator Measurement Saving the files Temperature correction Calibration of the wavelength scale Calibration of the spectral irradiance responsivity Using the DTMc300 spectroradiometer Preparations Measurement Instrument settings Handling of calibration items Uncertainty budgets Uncertainty of the filter radiometer Uncertainty of the DM150 spectroradiometer measurements Uncertainty of colorimetric quantities Accommodation and environmental conditions... 41

3 Page 3 (46) 9. Field calibrations Control data Certificates Intercomparisons Publications... 46

4 Page 4 (46) 2. Definition 2.1. Scope This quality manual describes the principle and operation of the equipment used for spectral irradiance measurements. Procedures are defined to calibrate primary standard lamps of spectral irradiance using a filter radiometer. Operation and calibration of spectroradiometer is also described. Procedures are given to calculate colour coordinates (x,y) and correlated color temperature Tc. Calibrations can be performed on DXW and FEL lamps, and occasionally on Wi41/G lamps Object and field of application Filter radiometer: Primary standard for spectral irradiance measurements. Used for calibrating primary standard lamps. Spectroradiometer: Secondary transfer standard calibrated with the primary standard lamps and thereafter used for calibrating light sources of customers Features a) Filter radiometer The filter radiometer consists of a trap detector, a set of optical band-pass filters, a copper oven for temperature stabilisation of the filters, and a precision aperture. All of the components form a compact and robust filter radiometer. One filter at a time can be used. Change of the filter has been made straightforward and reliable. The filter radiometer can be used to calibrate a primary standard lamp of spectral irradiance in the wavelength region from 290 nm to 900 nm, with an uncertainty between 2.6 % and 0.7 %. Using the filters, 14 discrete spectral irradiance values may be obtained. These values may be interpolated to give a continuous spectrum throughout the region. Measurement of a lamp using all the filters takes approximately 2 hours, excluding the time needed for alignment. b) The spectroradiometer Five spectroradiometers are available for customer calibrations. All spectroradiometers are mostly automated, computer controlled, and convenient to use. They are calibrated with a primary standard lamp measured with the filter radiometer. The BENTHAM DM150 spectroradiometer can be used in low-accuracy measurements between 200 nm and 845 nm wavelengths. A lamp measurement with 1 nm intervals takes about 15 minutes. The results need to be corrected for temperature. The BENTHAM DTMc300 spectroradiometer can be used in higher-accuracy measurements over wide wavelength region between 250 nm and 2500 nm wavelengths. A typical lamp measurement over whole region takes about 30 minutes. The two detectors of the spectroradiometer are temperature stabilised.

5 Page 5 (46) Konica Minolta CS1000 and CS2000 are imaging spectroradiometers capable of measuring luminance and spectral radiance. They can be equipped with a diffuser to facilitate spectral irradiance measurement. Ocean Optics QE65 Pro is an array spectrograph capable for fast spectral irradiance measurements Principle of realisation The principle of the realisation of the spectral irradiance scale has been explained thoroughly in Refs. [1, 2, 3]. Therefore only a brief description is given here. The realisation is based on a known detector instead of the traditional method of using a known source. A simplified measurement setup is shown in Figure 1. A detector with known spectral responsivity is equipped with a narrow band optical filter with measured spectral transmittance. The field of view is limited using a precision aperture with accurately known area at a measured distance d from the lamp. By using several interchangeable filters, the spectral irradiance of the lamp may be derived at several discrete wavelengths. These discrete values may be interpolated using suitable physical models. Lamp Optional filter Detector d Aperture Figure 1. Principle of detector-based realisation of spectral irradiance. In practice, the aperture, the detector and the interchangeable filters have been combined as a practical filter radiometer. The mechanical construction of the filter radiometer is presented in Figure 2. The temperature of the filters is stabilised using a copper [1] P. Kärhä, P. Toivanen, F. Manoochehri, and E. Ikonen, Development of a detector-based absolute spectral irradiance scale in the nm spectral range, Appl. Opt. 36, (1997). [2] P. Kärhä, Trap detectors and their applications in the realisation of spectral responsivity, luminous intensity and spectral irradiance scales, Doctors thesis, Helsinki University of Technology, 1997, 92 p. [3] T. Kübarsepp, P. Kärhä, F. Manoocheri, S. Nevas, L. Ylianttila, and E. Ikonen, Spectral irradiance measurements of tungsten lamps with filter radiometers in the spectral range 290 nm to 900 nm, Metrologia 37, (2000).

6 Page 6 (46) oven whose temperature is measured by a temperature sensor and adjusted by a circular see-through Peltier element. Details about the construction and its performance may be found in [4]. When exposed to optical radiation, the filter radiometer will produce a photocurrent l ( l) t ( l) R( l) dl i = A E, (1) where A is the area of the aperture, is the wavelength, E( ) is the spectral irradiance to be determined at the aperture plane, t(l) is the known spectral transmittance of the filter and R(l) is the known spectral responsivity of the trap detector. The photocurrent is measured with a HP3458A digital voltmeter equipped with a sensitive VINCULUM current-to-voltage converter. Trap detector Annular Thermoelectric Element Heat Sink Filters Aperture Three photodiodes To Control Electronics Black Anodised Aluminum Copper / Brass Figure 2. Construction of the filter radiometer. The spectral irradiance of the tungsten filament lamps used can be modelled as ( ) Plastic Be ( l) [ ( c lt )-1] Temperature-Controlled Copper Oven Temperature Sensor E l =, (2) 5 l exp 2 where B is an auxiliary multiplication factor, e (l) represents the effective spectral emissivity of the lamp, c 2 = m K is the second radiation constant, and T is the temperature of the lamp filament in kelvins. The effective emissivity e (l) is modelled with an N th degree polynomial as [4] P. Kärhä, A. Haapalinna, P. Toivanen, F. Manoochehri, and E. Ikonen, Filter radiometry based on direct utilisation of trap detectors, Metrologia 35, (1999).

7 Page 7 (46) ( l) N b i i= 0 i e = l, (3) where b i s are free parameters, and maximum N is 7. Equations (1) and (2) are solved with a recursive iteration method so, that the spectral irradiance values at the effective wavelengths of the filters match. le ( lt ) ( l) R( l) dl l leff = (5) E l c c ( lt ) ( l) R( l) dl As the result, we obtain the discrete spectral irradiance values at the effective wavelengths and a function in the form of Eqs. (2) and (3) giving a continuous spectral irradiance curve. Selecting the values for N and for the integration limits depend on the number of filters in use. N has to be approximately 2-3 lower than the number of filters in use, in order to prevent unexpected behaviour of the interpolation function between the fixed values. The obtained interpolation function is valid only between the effective wave-lengths of the lowest and highest filters. In some cases it may be advantageous to model the spectral irradiance separately in UV and visible/nir. The calibration is stored in the filter radiometer. For short periods, the calibration of the lamps may be used. The calibrated lamp is used to calibrate the spectroradiometer, which may then be used in customer calibrations.

8 Page 8 (46) 3. Equipment 3.1. Description of setups Table 1. Equipment needed when calibrating a primary standard lamp. Description Quantity *) Remarks/Id Filter radiometer 1997 Trap detector 1(~10) None, used in transmittance Temperature controlled filter holder Temperature controller electronics 1 (black with brass ring) 1 (self-made) Precision aperture 1(2) (A10, 3 mm) Briefcase containing the items mentioned above Filter radiometer 2001 Trap detector 1(~10) UVFR-8 Temperature controlled filter holder Temperature controller electronics 1 1 (black with aluminium ring) 1 Thorlabs Precision aperture 1(1) NFRA1 (3 mm) Optical stand for trap detector Filter radiometer Trap detector 1(~10) MRI-9807 (blue with black front) Temperature controlled filter holder Temperature controller electronics 1 (blue with aluminium ring) 1 Thorlabs Precision aperture 1 (1) NFRA2 (3 mm) Optical stand for trap detector Filter radiometer accessories 1

9 Page 9 (46) Filter set 1 (14) 290, 300, 312, 330, 350, 365, 380, 440, 500, V(l), 600, 700, 800, 850, 900 Current-to-voltage converter 1 (2) Vinculum SP042 Digital voltmeter 1 (5) HP3458A Optical Bench **) Optical rail in light tight enclosure 1 Installed in Optical power laboratory **) Baffle **) 1 (2) Mounting plate for rail **) 2 One with electronic distance measurement unit Alignment laser **) 1 OMTec two-beam alignment laser Digital distance measurement electronics 1 **) Light Source Lamp with corresponding lamp holder **) 1 (12) OSRAM FEL-S.T6: Lamps and their nominated use are listed in a separate database SPIRR_Lamps.xlsx Alignment jig **) 1 Lamp specific Shunt resistor **) 1 (2) Guildline Current source **) 1 (2) Heinzinger PTN Digital voltmeter **) 2 HP3458A, for measurement of lamp current and voltage Light Source with spectral lines PSU for WL-lamp 1 Oriel Lamp PSU, 6061 Argon-mercury WL-lamp 1 Oriel Ar-Hg lamp, 6035 Krypton WL-lamp 1 Oriel Kr lamp, 6031 PSU for WL-lamp 1 UVP 0.21 A PSU for Neon lamps,

10 Page 10 (46) Software Neon WL-lamp 1 UVP Neon Pen Ray (10 ma), Neon WL-lamp 1 UVP Neon-Hg (18 ma), Mercury-lamp + external AC-adapter Excel model for analysing results 1 Ocean Optics HG-1, add s/n 1 Spirr30.xls (August 2014) *)The number in parentheses indicates the number of optional items available. The number not in parentheses indicates the minimum amount of the concerned item needed. **)In field calibrations, these items may be replaced with corresponding equipment of the customer. Table 2. Equipment needed for spectroradiometer measurements with DM150. Description Quantity Remarks/Id Spectroradiometer Electronics Software Monochromator 1 BENTHAM 150BC Input fibre and diffuser 1 BENTHAM D5, PTA/4, FOP/UV Photomultiplier tube 1 BENTHAM DH3 Rack + power supply 1 BENTHAM 217 Monochromator controller BENTHAM PMC DC current amplifier 1 BENTHAM 267 Integrating A/D-converter 1 BENTHAM 228A High voltage power supply 1 BENTHAM 215 Computer 1 IBM Thinkpad (M125) IEEE card 1 Ines PCMCIA 488 IEEE cable 1 GPIB-PMC-C Digital voltmeter 1 Agilent 34410A, for monitoring the PMT temperature 10 k NTC-thermistor 1 Control and data acquisition software 1 BENSOFT (bs.exe)

11 Page 11 (46) Optical Bench Light source (as above) (as above)

12 Page 12 (46) Table 3. Equipment needed for spectroradiometer measurements with DTMc300. Description Quantity Remarks/Id S/N Spectroradiometer Electronics Monochromator 1 Bentham DTMc Input fiber and diffuser 1 Bentham D7/D3, 1-m or 2- m fiber 8176/1 Photomultiplier tube 1 Bentham DH-50-Te SA5573 Lead sulfide detector 1 Bentham DH-PbS-Te 7223 Rack + power supply 1 Bentham DC current pre-amplifier 1 Bentham 277 Lock-in amplifier 1 Bentham 225 Integrating A/D-converter 1 Bentham 228A High voltage power supply 1 Bentham 215 PMT cooler power supply 1 Bentham CPS Detector cooler power supply 1 Bentham CPS1M Optical chopper 1 Bentham 218H Chopper controller 1 Bentham Software Computer 1 Fujitsu Siemens Amilo IEEE GPIB USB adapter 1 National Instruments GPIB-USB-HS IEEE cable 1 IEEE488 GPIB cable Control and data acquisition software Optical Bench Light Source 1 BenWin+ (BenWin+.exe) (as above) (as above)

13 Page 13 (46) 3.2. Calibration requirements Filter radiometer The multimeters used are auto-calibrated every day before the start of the calibration measurements. Before autocalibration, the temperature of the DVM has to be allowed to stabilise for at least 1 h (longer time recommended). Calibration of the multimeters is done according to the MRI calibration schedule. The Vinculum current to voltage converter is calibrated according to the MRI calibration schedule. The trap detector is calibrated once every two years against the cryogenic absolute radiometer. The reflectance is measured, and the spectral responsivity is extrapolated. The methods used are described in [5]. Filter transmittances are calibrated twice a year using methods described in [6]. If stability measurements of certain filters indicate that shorter calibration intervals are needed, then be so. The calibration intervals are re-estimated with each measurement. Before filter radiometer measurements, the filters are inspected visually for dust or degradation. The filters that degrade appear foggy and need to be replaced. Filter transmittances are calibrated in the wavelength region 0 ±50 nm, where 0 is the peak wavelength of the filter. Out-of-band leakage s are studied in the wavelength region [ nm nm] once for each new filter. If Deuterium lamps are to be measured, also possible leakage s below 0 need to be considered. Reflectances of the filters have to be measured once for each filter when taking into use. These reflectances are needed to take into account the interreflections between the filter and the trap detector. If reflectances are not measured, a value (10±10) % is used. The uncertainty is taken into account in calculations. The trap detectors are stored with the protective cap closed whenever they are not in use. This is to keep dust away from the surfaces of the photodiodes. Before measurements, it is visually checked that there is no dust inside the detector. If there is visible dust, it may be carefully cleaned from the first photodiode with Kodak lens cleaning tissue. If other tissues are used, make sure that there is no silicon in the tissue. The two inner diodes cannot be cleaned without disassembling. [5] Quality manual of optical power laboratory. [6] Quality manual of reference spectrometer laboratory.

14 Page 14 (46) The areas of the precision apertures are measured once every four years by VTT- MIKES length section. Alternately, they may also be calibrated using the direct optical method, DOM [7,8]. The shunt resistor is calibrated according to the MRI calibration schedule Spectroradiometers The spectroradiometers are calibrated for spectral irradiance responsivity before calibration measurements if the previous calibration is more than six months old. The wavelength scales of the spectroradiometers are checked once a year. Incandescent lamps may be used for storing the spectral irradiance calibrations for short periods. Calibration interval for OSRAM FEL T6 lamps is 2 years or 15 hours of burning time, whichever exceeds first. Change in lamp voltage (by more than 0.3 V) indicates need for recalibration. [7] A. Lassila, P. Toivanen, and E. Ikonen, An optical method for direct determination of the radiometric aperture area at high accuracy, Meas. Sci. Technol. 8, (1997). [8] E. Ikonen, P. Toivanen, and A. Lassila, A new optical method for high-accuracy determination of aperture area, Metrologia 35, (1999).

15 Page 15 (46) 4. Measurement traceability Figure 3. Traceability scheme of the spectral irradiance measurements.

16 Page 16 (46) 5. Calibration and measurement procedures including validation methods 5.1. Measurement ranges Measurement range using the filter radiometer Lowest current that can be measured reliably with the trap detector is 1 na. The highest measurable current is 1 ma due to nonlinearity [9]. The lowest and highest measurable spectral irradiance values resulting in these limiting current values are presented in Table 4 for all filters presently in use. The values have been calculated assuming a light source of T c = 3130 K, and an aperture area of mm 2 (3 mm diameter) for high end and diameter of 4 mm for low end. Table 4. Minimum and maximum allowed spectral irradiance values for the filters in use with the filter radiometer. Filter E min [pw mm -2 nm -1 ] E max [mw mm -2 nm -1 ] Based on the values, we can define our typical measurement range as 10 pw mm -2 nm mw mm -2 nm -1 using 3-mm and 4-mm apertures. [9] T. Kübarsepp, A. Haapalinna, P. Kärhä, and E. Ikonen, Nonlinearity measurements of silicon photodetectors, Appl. Opt. 37, (1998).

17 Page 17 (46) Measurement range using the DM150 spectroradiometer The lowest measurable spectral irradiance of the DM150 spectroradiometer was studied by measuring the dark current of the photomultiplier tube. A measurement was made with the spectroradiometer, with its diffuser completely blocked and no calibration set for the spectroradiometer. The standard deviation of the measured signal was 2.9 pa. It appears that the dark current cancellation does not completely cancel the dark current. In this measurement, the recorded signal was on the average 8.3 pa above the zero level. Assuming this value as maximum, we may conclude that the standard uncertainty of the photocurrent measurement is 2 2 u = 3 + (8/ 3) pa= 5,5pA. c The calibration factor used to convert the current values to the corresponding spectral irradiance values is presented in Figure 4 for the wavelength region from 350 to 800 nm. The minimum value is W mm -2 nm -1 na -1 at the wavelength of 523 nm. The lowest spectral irradiance values may be measured at this wavelength. The noise equivalent spectral irradiance value is W mm -2 nm -1. This sets the lowest measurable value. If we require e.g. 5 % accuracy (k = 1), the lowest measurable spectral irradiance value is 4 nw mm -2 nm -1. Calibration factor [W mm -2 nm -1 na -1 ] 1E-05 1E-06 1E-07 1E Wavelength [nm] Figure 4. Calibration factor of the DM150 spectroradiometer. The highest measurable value depends on the nonlinearity of the photomultiplier tube, which has not been studied yet.

18 Page 18 (46) 5.2. Measuring a primary standard lamp Build the setup as described in [10]. The reference also gives instructions on operating the lamps. The lamps to be used as spectral irradiance standards include 1 kw FEL and DXW type lamps. With HUT FEL lamps: 1) Measurement distance is 500 mm measured from the outermost surface of the lamp base 2) Operating current is A 3) Lamp is aligned using the targets in the briefcase of the lamps. The grooved surface of the alignment jig must be pointed away from the lamp (towards the filter radiometer). Lamp numbers have been written in the jigs. With DXW lamps typically: 1) Measurement distance is 500 mm or 700 mm measured from the outermost surface of the alignment jig. 2) Operating current is A 3) Lamp is aligned using the target provided by the customer. Operating procedures of customer lamps must be agreed with the customer because the alignment procedures vary. Alignment with all lamps is done using the 900c filter in the filter radiometer. OMTec alignment laser between the lamp and the filter radiometer is to be used in the alignment unless there is special need for other methods. If calibrating customer lamps in the field, measure the distances to the appropriate reference planes using the equipment of customers. Measure the lamp using each filter. Before measurement, record the dark current, and subtract it from the measured signal. With each measurement, check/adjust the lamp current, and record the lamp voltage. Start with 900 nm filter and measure it in the beginning and at the end of the measurement sequence. When changing the filters, allow the temperature of the new filter to stabilise for 5 minutes before measuring. During this time, light must be blocked. The time may be shortened if the temperature of the laboratory is close to 25 C. Record the temperature and humidity of the laboratory to be later added to the calibration certificate Analysing the results Analyse the results using the excel model, as described in [1, 3]. [10] Instruction manual for operating standard lamps.

19 Page 19 (46) 5.4. Calculating colour co-ordinates and colour temperature Colour co-ordinates and colour temperature are calculated using the interpolated spectrum (1 nm interval) of the lamp with equations and colour-matching functions defined in [11]. Spectral irradiance values are multiplied with colour-matching functions x, y, z and the products are summed to obtain tristimulus values X, Y, Z. Chromaticity co-ordinates x, y are calculated as x = X/(X+Y+Z) and y = Y/(X+Y+Z). Colour co-ordinates u, v (1960 UCS diagram) are calculated as u = 4X/(X+15Y+3Z), v = 6Y/(X+15Y+3Z). Correlated colour temperature T c is obtained as the temperature of a black body radiator, whose color difference in 1960 UCS diagram is minimised with respect to the colour co-ordinates u, v of the lamp. For analysing the T c of a measured spectrum, an excel file MRI_CCT_Calc_2014.xlsm can be used. The latest version is stored in folder T405/MIKES-Aalto/Quality/radiom/. The file has three calculation sheets for 1 nm, 2 nm and 5 nm spectral data between 360 nm and 830 nm. Copy the measured spectral data and paste it to the corresponding column in the file and then use the built-in iterative function Excel solver to calculate the CCT of the measured spectrum. The solver can be used with the pre-defined settings of the file that automatically refer to the correct cells. In the analysis, the Euclidean distance (u,v) between the chromaticity coordinates of the measured source and a Planckian radiator is minimised by changing the temperature of the Planckian radiator Adjusting colour temperature Colour temperatures of lamps are adjusted to certain values (e.g K for CIE standard illuminant A) using a spectroradiometer. It is preferred to use the Konica Minolta CS2000A spectroradiometer with the dedicated diffuser head for adjusting the T c of lamps. The CS2000A allows quick and repeatable measurement of the spectrum of the source and reduces the total burn time of the lamp significantly compared to using a scanning spectroradiometer for adjustment of the lamp T c. The CS2000A calculates the T c of the lamp using an approximative method that results in a T c value 1.5 K higher than analysed using the iterative method, when targeting at T c of 2856 K. This should be taken into account when adjusting the CCT of a lamp. In practice, adjusting a lamp to T c of 2856 K requires the CS2000A to display K. The final adjusted T c of a lamp should be verified with the MRI_CCT_Calc2014.xlsm file using the iterative method. In addition, it must be noted that the difference between the approximate method and the iterative method is nonlinear in nature and the difference of 1.5 K applies only for T c of 2856 K. Adjusting the lamp to any other T c than 2856 K needs to be checked with the iterative calculation method. [11] Commission Internationale de l Éclairage (CIE), Technical Report, Colorimetry, Second Edition, CIE

20 Page 20 (46) Procedure for adjusting the T c of a lamp: 1. Measure the spectrum of the lamp within WL range of nm (CS2000A). 2. Adjust the lamp current, wait for 1 minute and make a new measurement. 3. Repeat steps 1 and 2 until the colour temperature is within the specifications given by the customer, typically within 0.5 K from the target T c. 4. Repeat the measurement a few times within a period of min to verify that the adjusted T c is stable and within the specifications. Adjust current, if necessary. 5. Verify the final T c of the lamp with the iterative calculation method using the file MRI_CCT_Calc2014.xlsm 5.6. Using the DM150 spectroradiometer The Bentham DM150 spectroradiometer (Registry code M124) is controlled by IBM Thinkpad laptop computer (Registry code M125) via IEEE488 interface bus. The measurement setup is placed on a carriage which can be moved. MS-DOS 6.2 boot disk labeled Spectro-boot should be used for starting up the computer. This way more conventional RAM is available for the measurement program. The location of the measurement program is C:\bentham\bs.exe Installing the setup The setup is similar to the setup used with the filter radiometer measurements (Section 5.2) but the filter radiometer is replaced with the measurement head (equipped with a Teflon diffuser) of the spectroradiometer. Alignment is carried out as with the filter radiometer. A mirror is placed in contact with the diffuser to generate reflection for angular alignment. Distance is measured with the outermost surface of the diffuser as the reference plane Preparations In order to reach a stable operating temperature, the controller unit of the spectroradiometer and its high voltage module should be turned on the day before the measurements take place. Check that the optical fiber is connected properly to the monochromator and that it cannot move during the measurement. Check that all cables are connected. Figure 5 shows the controller unit of the spectroradiometer and its front panel connections. The signal cable from the photomultiplier tube (PMT) must be connected to the input 1 of the current amplifier.

21 Page 21 (46) Figure 5. The controller unit of the DM150 spectroradiometer and its connections. The temperature of the PMT is monitored using a 10 k NTC-thermistor which is attached on the surface of the PMT. A multimeter of type Agilent 34410A is used for converting the resistance of the thermistor to a temperature reading. Check that the cables are connected to the thermistor and turn the multimeter on. Press Shift + Config to enter the temperature measurement mode. Press Config again and select Thermistor 2W using the arrow keys. Press Enter and select 10k. Press Enter so many times that you get out of the menu. Finally press Shift + Trigger to start the temperature measurement Filter wheel and wavelength indicator The filter wheel position and the wavelength indicator of the spectroradiometer must be checked before measuring with the device because they can sometimes get misaligned. In the measurement program, go to the Scan menu Select wavelength. Type 400 in the New wavelength -field and press Enter. This drives the monochromator to 400 nm and the filter wheel to position 2. The correct position of the filter wheel is shown in Figure 6. If the filter wheel position is wrong, exit the program and switch the controller unit off. Carefully turn the filter wheel knob that it points at position 2. Switch the equipment on again and launch the program. If the knob of the filter wheel is turned manually while the device power is switched on, a mechanical failure will occur.

22 Page 22 (46) Figure 6. Filter wheel knob position. Correct position (left) and wrong position (right). Check that the wavelength indicator shows , as shown in Figure 7. This compensates for the typical wavelength offset of the device. The instructions for calibrating the wavelength scale can be found in chapter If the wavelength indicator does not show , go to Hardware Monochromator, and adjust the dial reading to obtain the offset. After changing the dial reading value, drive the monochromator to for example 380 nm and back to 400 nm to verify the change. Figure 7. Correct reading of the DM150 wavelength indicator after driving the monochromator to 400 nm Measurement The latest calibration file should be used when measuring with the spectroradiometer. The calibration file in use defines the wavelength step and the maximum wavelength range of the measurement. Typically the step is 1 nm and the range is 250 nm 830 nm. The calibration file can be manually loaded and viewed in the Scan menu Calibra on file.

23 Page 23 (46) The monochromator can be used with three different measurement bandwidths by changing the slits manually before the measurement. 1 nm. 2 nm and 5 nm bandwidths are obtained by using slit sizes of 0.56 mm mm and 2.78 mm. respectively. Higher bandwidths can be used in measurements of low signal levels to obtain better signal to noise ratio. Two slits of each size are kept in a plastic bag which can be found in the carriage of the spectroradiometer. The monochromator has 3 slots for slit installation. The two slots that are in use are labelled Entrance and Exit and are located at the optical fiber port and the PMT port. The slits are installed in pairs and must be of the same size. The entrance slit must be facing the optical fiber whereas the output slit must be facing the PMT. Open the first slit installation slot by unscrewing the two screws and pull the slit up gently by using pliers. Be careful not to drop any dust into the slot. Install the new slit carefully by pushing it in as far as it goes by using for example a screwdriver and close the slot door immediately to avoid any unnecessary dust from entering the slot. A calibration file corresponding to the new slit sizes must be loaded before the measurements. More information about the calibration files can be found in Chapters and The measurement settings can be changed in the Scan menu Scan op ons. The following settings should be selected for all typical measurements: Scan type ( ) Equidistant wl Pre-scan [x] Confirm spec. id [x] Zero calibration Scan [x] Graphics mode [x] Data correction Post scan [x] Return to start [x] Close shutter The comments for the measurement file can be typed in the Scan menu Next spec options Comment. A er typing the comments, press Escape and select Ok. The measurement range can be set in the Scan menu Equidistant wavelength. Give the start and stop wavelengths and the measurement step. The measurement step must be the same as in the calibration file. The start and stop wavelengths cannot exceed the range of the calibration file. The program uses automatic indexing for the filenames. The default filename is spec***.dat, where the *-symbols form the indexing part. If three *-symbols are used, the file names will be spec001.dat. spec002.dat etc. The file name can be changed in the Scan menu Next spec op ons ID. A total of 8 characters can be used for the name and the extension must be.dat. At least one *-symbol must be included in the file name that the indexing works. The indexing can be disabled by deselecting the Auto increment sequence in the same menu. To start the measurement, select the Scan menu Start measurement. The program suggests a file name defined in the ID menu. Confirm or change the file name and press Enter. During the measurement, the program draws the spectral irradiance graph on the

24 Page 24 (46) computer display. The graphs can be viewed and compared after the measurements in the Graphics menu Display specs. The program can be set to measure multiple scans if needed. Go to the Scan menu Scan options and select Multiple in addition to the settings listed earlier. Additional settings for the multiple scan measurement can be made in the Scan menu Mul ple scan setup. This measurement method is useful for averaging troubleshooting. All typical measurements should be performed as single scans. More detailed information about using the measurement program is found in [12] Saving the files The measurement files are stored in the computer s RAM and must be saved onto a floppy disk or the hard drive before quitting the program. Go to the File menu Diskmemory manager. Press Tabulator to move the cursor to the right side of the screen. Select the files which you want to save by using the Space bar and press e and select Save. Choose the destination directory and select Save. The files can be temporarily saved in the directory C:\bentham\temp\ Temperature correction Because of the temperature sensitivity of the PMT. its temperature must be written down when conducting the measurements. If the measurement temperature differs from the calibration temperature, a correction must be applied. To check the calibration temperature, go to the Scan menu Calibra on le Data info. An excel-file including the instructions for applying the correction can be found in the directory \\metrology\webpages\quality\radiom\srm temperature correction.xls Calibration of the wavelength scale The wavelength scale can be calibrated using a mercury lamp, or another lamp with wellknown spectral lines. Suitable wavelengths of a mercury lamp are presented in Table 5. The lamps suitable for wavelength calibration of spectroradiometers are listed in Table 1. Table 5. Mercury lamp lines used in the wavelength scale calibration. Wavelength in air [nm] Order of wavelength [12] BENSOFT Manual.

25 Page 25 (46) (doublet) These wavelengths are compared with the wavelengths measured with the spectroradiometer. The wavelength calibration is conducted in four parts with 0.1 nm steps and 1 nm bandwidth (see next section) using four different calibration files. These files ( cal, cal, cal and cal) are located in directory C:\bentham\calibras\wave\ Each spectral line is measured separately (using a range of approximately ±3 nm around the peak wavelength) and stored into the same directory with name Hgxxx.dat, where xxx is the wavelength (without decimals). Because of relatively wide measurement bandwidth, the measured peaks are broadened. Therefore the peak wavelengths of the measured lines have to be calculated using FWHM (Full Width Half Maximum) method. Resulting wavelength correction file bs.wcl is stored in directory C:\bentham\ In a typical measurement, it is enough to use the method of chapter , and verify that the wavelength indicator shows after driving the monochromator to 400 nm. In this case, the wavelength correction can be left unchecked in the Scan options menu Calibration of the spectral irradiance responsivity The spectral irradiance responsivity of the DM150 spectroradiometer is calibrated using the known spectral irradiance of a standard lamp Osram FEL T6. The measurement distance is 500 mm between the front surface of the diffuser and the front surface of the lamp. The measurement bandwidth depends on the width of the entrance and exit slits used in the monochromator. Usually 0.56 mm slits are used to obtain 1 nm bandwidth. For low power light sources 2.78 mm slits (5 nm bandwidth) can be used. The procedure to calibrate the DM150 spectroradiometer is listed below. 1. Align the standard lamp and the diffuser head on the optical rail.

26 Page 26 (46) 2. Reset the distance measurement display while holding the 50.8 mm calibrated length between the diffuser head and the front surface of the lamp. Move the rail carrier of the diffuser head further away from the lamp that the distance measuring display shows corresponding to the 500 mm measurement distance. 3. Install slits with correct size to obtain the desired measurement bandwidth. 4. Check the correct filter wheel and wavelength indicator positions by driving the monochromator to 400 nm and follow the instructions of chapter Turn on the light source and let it to stabilise for at least 20 minutes before the calibration measurements. 6. Set the wavelength range to be measured. Scan Equidistant wavelength: nm. 1 nm step 7. Set the integration time of the measurement. Hardware ADC Setup Samples per reading: Write comments for the new calibration file in Scan Next spec options Comment -menu. The following information should be included: Calibration of DM150 with FEL nm, step: 1 nm, slit: 1 nm (0.56 mm) Temperature: 25.0 C. humidity 38 % PMT temperature: 26.7 C 9. Make two measurements to warm up the PMT. 10. The DM150 is calibrated against a spectral irradiance standard lamp by performing three consecutive scans of the PMT current. The data correction must be disabled in the measurement. The average of the scans is called the system response. The device is able to make multiple successive scans automatically. Check that the options in Scan Scan Options -menu are as follows: Scan type [x] Multiple Pre-scan [x] Confirm spec id Scan [x] Graphics mode Post scan [x] Return to start ( ) Equidistant wl [x] Zero calibration [ ] Data correction [x] Close shutter Go to Scan Multiple scan options menu and configure the options for multiple successive scans.

27 Page 27 (46) a) Total number of scans: 3 b) Start number: 1 [x] Single specs 11. Start the measurement. Scan Start Measurement 12. When using the multiple scans option, the measured data are saved in a single file, by default on the drive A: (floppy disk). The data need to be loaded into the memory for calculating the calibration file. Go to File Disk-Memory manager and open the file. The file opens as three separate spectral files in the memory. The average of the three files can now be calculated in Process Mean and error The name of the resulting system response file should be of type stddmmyy.dat, where ddmmyy is the calibration date. e.g. st dat. Insert the name of the system response file into the MeAn: Spec-Id field and press Process. Go to File Disk-Memory manager and save the file from the memory to the directory (if 0.56 mm slits are used) C:\bentham\system\056slit\ 13. The calibration file is calculated by using the produced system response file stddmmyy.dat and the spectral irradiance data of the lamp. The data of the spectral irradiance lamps are stored in the directory C:\bentham\lamps\fel1000\ The name of the lamp data file is of type XXX_mmyy.dat, where XXX is the lamp number (the last three digits of the serial number) and mmyy indicates when the spectral irradiance of the lamp has been calibrated. The latest available data should be used. Load the spectral irradiance data into the memory. Go to Scan Calibration file Calculate -menu. Get the proper lamp and system files and press Compute. Go to Graphics Display specs and check that the graphs are OK. 14. Save the calibration file to the directory (if 0.56 mm slits are used) C:\bentham\calibras\056slit\ The name of the calibration file is of type bsddmmyy.cal, where ddmmyy is the calibration date. 15. While the lamp is still operating, make a test measurement by using the new calibration file. Set the data correction on in Scan Scan Options menu. Load the new calibration file in Scan Calibration file Load data menu. Compare the measured spectrum to the lamp file; spectra should converge.

28 Page 28 (46) Note that the software makes a copy of the loaded calibration file called bs.cal and uses that file every time the software is started Using the DTMc300 spectroradiometer The new spectroradiometer (Bentham DTMc300. registry code MIKES004664) is controlled by a portable computer (Fujitsu Siemens Amilo) via IEEE488 interface bus. The measurement setup is placed on a carriage which can be moved Preparations The PMT needs at least 12 hours to reach a stable operating temperature, so the spectroradiometer should be turned on the day before the measurements take place. Check that the optical fiber is connected properly to the monochromator and that it cannot move during the measurement. Check all connections. Figure 8 shows the front panel connections of the rack unit, the chopper controller and the PMT cooler power supply. The signal from the PMT must be connected to the input 1 of the current pre-amplifier and the output of the pre-amplifier to the input 1 of the lock-in amplifier. The signal from the lead sulphide (PbS) detector must be connected to the input 2 of the pre-amplifier. High impedance of PbS allows to use it with a current pre-amp. Figure 8. Rack unit. chopper controller and PMT cooler power supply and their front panel connections. The D-connector of the PMT cooler cable includes four additional banana sockets. They are meant for troubleshooting and do not need to be connected when doing typical measurements. The D-connector should be connected to the socket at the front panel

29 Page 29 (46) of the CPS50 PMT cooler power supply. The detector output of the CPS1M cooler power supply must be connected to the PbS detector. The back panel output of the chopper controller should be connected to the optical chopper. Connect the computer to the rack unit using a GPIB-USB adapter. The monochromator should be connected to the same GPIB-socket using a standard GPIB-cable. Connect the 215 high voltage supply to the PMT and the PbS detector using the HV1 and HV2 outputs. These outputs use special MHV (high voltage BNC) connectors. They look like ordinary BNC-connectors but have higher voltage ratings. It is very important to use the supplied MHV cables for these connections to avoid causing any damage to the equipment and for safety reasons. The PbS detector has a built-in volt-age divider with a ratio of 1/11. If the PMT is run at a typical level of 750 V. the bias voltage of the PbS detector will be approximately 75 V. The optical chopper uses an unprotected chopping disk with 10 slots. Avoid touching the disk and make sure that no parts are in contact with the disk before turning the equipment on. Turn on the PMT cooler power supply, chopper controller, monochromator and the rack unit. From the rack unit, turn on the CPS1M detector cooler power supply and the 215 high voltage supply. Adjust the optical chopper frequency to 175Hz. This chopping frequency should always be used in a 50Hz power system to achieve the best signal to noise ratio in the measurements. The frequency display shows the correct frequency for a 10 slot chopping disk. For other disks, multiply the dial reading by the factor no. of slots in disc f =. 10 The display of the CPS50 cooler power supply shows either the target temperature or the actual temperature of the PMT. depending on the position of the front panel switch. Use the Set Temp -knob to adjust the target temperature to C for the PMT and lock the knob position by pushing the knob in. The monochromator uses two motorized slits (entrance slit and middle slit) and two manual slits (both exit slits). The motorized slits can either be set manually or con-trolled automatically based on the measurement step size. In the latter case, the pro-gram calculates corresponding slit width, using the step size as the bandwidth, and the linear dispersion of the grating in use. The monochromator has three gratings, which are configured to be used in the different spectral ranges as following: Table 6. Grating Range Linear Dispersion Detector nm 1.8 nm/mm PMT nm 2.7 nm/mm PMT nm 5.4 nm/mm PbS

30 Page 30 (46) The manual slits must be set accordingly to the range of the measurement and step size combination: Slit size (mm) = Step size (nm) Linear dispersion (nm/mm) The exit slit to the PMT is set at 20% larger than above value because of subtractive configuration of double monochromator. Because two gratings are used with the PMT as a detector, and it is impractical to change manual slits during the measurement, in practice single measurement is limited to the range of nm nm or nm. If it is necessary to scan the whole range in one measurement, a variable step size must be used to compensate for the difference in the linear dispersion between gratings. By default, the step size for the infrared is 10 nm, so the PbS slit must be set at mm. For UV and visible step size is 2 nm for nm range and 3 nm for nm range, so the PMT slit must be set at mm. Width of the exit window of the fibrebundle should be considered when planning a measurement, as it effectively sets an upper limit on the slit widths. Start the computer and launch the program C:\Program Files\Bentham\BenWin+.exe A shortcut for the program is located on the desktop of the computer. The equipment must be initialized before the measurement. Go to the Tools menu and select Initialize. Normally instrument settings from previous measurement are loaded automatically. All major settings are listed in chapter Measurement Comments for the measurement file can be typed in the Scan menu Set File Information. Double-click the text fields to enable typing. Press + to add more rows. The measurement settings can be changed in the Scan menu Scan setup. Click the Advanced -button to view all parameters. The number of measurements can be set using the field Number of Spectra. Give the start and end wavelengths and the measurement step. To transform raw measured values to the real irradiance values. two methods can be used. Manual data correction: Measure only within one wavelength range at a time. Ranges are: (2nm step) (3nm step) (10nm step). Use following scan settings: ( ) Use Custom Wavelengths ( ) Auto Range ( ) Zero Calibrate ( ) Data Correction ( ) Close Shutter ( ) Return to Start Wavelength

31 Page 31 (46) Results will be raw values from the lock-in amplifier. They can be transformed into the irradiance values with the Excel table that contains correction multipliers. The table is found at: \\Metrology\webpages\quality\radiom\DTMc300-corrections.xls Automatic data correction (untested!): Calibration files can be used when measuring with the spectroradiometer, to automatically transform the raw values from detectors into correct scale. Calibration files can be manually loaded in the Scan menu Data Correc on Load From File. If the data correction is turned on, the start and end wavelengths cannot exceed the range used in the calibration. If the measurement range includes a grating change while using the same detector, the issue of fixed exit slits (described in previous chapter) has to be addressed by using variable step size. To do so, check Use Custom Wavelengths checkbox and click Load Custom Wavelengths. Custom wavelengths file is a text file with.dat extension, which has a list of wave-lengths to be measured. A file custom_wavelengths_all.dat has a range of nm, with 2 nm step for nm range, 3 nm step for nm range, and 10 nm step for nm range. The following scan options should be selected for the measurements: ( ) Use Custom Wavelengths ( ) Auto Range ( ) Zero Calibrate ( ) Close Shutter ( ) Data Correction ( ) Return to Start Wavelength Data Correction option is selectable only if calibration file is loaded in Data Correction options window. To start the measurement, press the New Scan -button. The program draws the spectral irradiance graph(s) into a window. After the measurement, copy the data to clipboard and paste it into a program of your choice or export the data to Excel. It might be better to export to Excel from the File menu, rather than from the dialog provided by the program after the measurement is finished, that way multiple measured spectra are not separated into different tabs in the Excel, but are on the same tab. separated by columns. If you chose the Save to File option, data will be saved in a text file. Use.ben as the file extension. The data from a scan containing multiple spectra can be saved either to a number of files or to just one file. The parameter for this is Save multiple spectra to a single file and can be found in the Tools menu Op ons. If this op on is deselected, the program asks for only one file name when saving but creates a number of files with automatic indexing, each file containing a single spectrum data. The monochromator can be driven to a certain wavelength using the Scan menu Signal Setup. Type the target wavelength into the corresponding field and press Go to

32 Page 32 (46) Wavelength. Multiple measurements of a single wavelength can be made using the Scan menu Sta onary Scan Setup. Give the scan length, time interval and wavelength and press New Scan to start the measurement. This measurement method is useful for troubleshooting. Note that for the Signal Setup and the Stationary Setup bandwidth (= slit width) is carried over from the previous measurement. Slits can be set manually by clicking on the Instruments menu and Slits Instrument settings All major settings can be found in Instrument menu. For Lock-In Amplifier window following settings are used: Table 7. Setup number Input Wavelength Min. Range 1 (10^0 V/A) 1 (10^0 V/A) 1 (10^0 V/A) 1 (10^0 V/A) Max. Range 4 (10^-3 V/A) 4 (10^-3 V/A) 4 (10^-3 V/A) 4 (10^-3 V/A) Target Range 1 (10^0 V/A) 1 (10^0 V/A) 1 (10^0 V/A) 1 (10^0 V/A) Phase Variable Phase Offset Time Constant (s) Effectively, only setups 1. 2 and 3 are used. Maximum range is set at 10^-3 V/A. because otherwise the lock-in can get stuck, switching between the ranges continuously when the signal level is low. Settle delay is set at 4000 ms. The settings are described in Figures 9 11.

33 Page 33 (46) Figure 9. Bentham DTM-c300 pre-amplifier settings. Figure 10. Bentham DTM-c300 filter wheel settings.

34 Page 34 (46) Figure 11. Bentham DTM-c300 monochromator settings. The grating properties of the DTM-c300 spectroradiometer are listed in Table 8. Table 8. Properties of the Bentham DTM-c300 spectroradiometer gratings. Turret 1 Turret 2 Grating 1 Grating 2 Grating 3 Grating 1 Grating 2 Grating 3 Line Density Zero Order Alpha Max Wavelength Swing Away Mirrors can be adjusted by double clicking on each if them in the list. Swing Away Mirror 1 is deflected on wavelength up to 850nm (both monochromators are used), and not deflected from 851nm and up (only first monochromator is used). Mirrors 2 and 3 are always deflected. Settle delay is 1000 ms. In Slits configuration Auto mode is typically used, so the slit width is adjusted accordingly to the wavelength step size. In miscellaneous settings Lock-in Preamplifier Input should be set to 1.