DSLR as spectrophotometer calibration examples#

Introduction#

This is a repository showing the results of some tests performed to calibrate a DSLR camera (Canon ESOS 550D) with a difraction rating on the front of the lens as a spectrophotometer. The setup was mounted at Villaverde del Ducado. Different lamps were used for spectral calibration.

Test of 2022/12/30#

For the test of 2022/12/30 we used several LED of different color temperatures. The wavelength calibration was made with the aid of a CFL.

../_images/test_20221230.png

Fig. 94 Spectra and the lamps used in the test of 2022/12/30#

../_images/wl_ident_CFL_blue.png

Fig. 95 Identification of the emission lines of the CFL spectrum. Blue region.#

../_images/wl_ident_CFL_red.png

Fig. 96 Identification of the emission lines of the CFL spectrum. Red region.#

../_images/wl_fit_CFL.png

Fig. 97 Fit position versus wavelength for the wavelength calibration. On the right the residuals of the fit.#

../_images/spectra_test_20221230.png

Fig. 98 Wavelength calibrated final observed spectra of the LED lamps used in the 2022/12/30 test.#

Fig. 98 shows the comparison of the spectra pf LEDs of different color temperatura. Note that the peaks in the continuum spectra are not located in exactly the same wavelegth region.

The LED lamps used in this test are not available at the UCM-LICA database. To determine their spectra we have used a Sekonik C-3000 spectrophotometer.

../_images/spectra_test_20221230_sekonik.png

Fig. 99 Spectra of the LED lamps used in the 2022/12/30 test obtained with a Sekonik-C3000 spectrophotometer.#

It is interesting to note that (a) repeated observations of the same lamp provides different color temperature and (b) this color temperature is lower than the nominal temperature, i.e. that displayed in the box.

nominal

1

2

3

2700K

2466K

2504K

2534K

3200K

2860K

2891K

2852K

4000K

3800K

3816K

3708K

6500K

5375K

5576K

5622K

Other interesting result is the lack of enough signal in the blue region (even for the 6500K LED) for calibration purpouses (see Fig. 99 and Fig. 100). A bluer light source is needed for comparison.

../_images/mean_all_leds.png

Fig. 100 Individual and mean spectra for the different LED lamps used in the 2022/12/30 test obtained with a Sekonik-C3000 spectrophotometer.#

After the comparison of the observed spectrum and those measured with the spectrophotometer as a reference we can find the response curve.

../_images/response_LED6500_20221230.png

Fig. 101 Observed and expected response using a LED 6500K.#

The result is still not satisfactory in the blue part of the spectrum.

Test of 2023/01/02#

For the test of 2023/01/02 we used nothe white LED of 6500K (nominal color temperature) and a very white CFL that can be used for wavelength calibration but also for photometric calibration.

../_images/test_20230102_LED6500.png

Fig. 102 CFL lamp and its spectrum as recorded in the test and with Sekonik C-3000 spectrophotometer for the 2023/01/02 test.#

../_images/test_20230102_CFL.png

Fig. 103 CFL lamp and its spectrum as recorded in the test and with Sekonik C-3000 spectrophotometer.#

Note the emission of the CFL lamp in the 380-420 nm wavelength range.

../_images/test_20230102_2d_spectra.png

Fig. 104 Bidimensional spectra for the whole series of observation 2023/01/02. Not reproduced with the same scale.#

../_images/test_20230102_1d_LED.png

Fig. 105 Monodimensional spectra for the LED 6500K observation of 2023/01/02.#

../_images/test_20230102_1d_CFL.png

Fig. 106 Monodimensional spectra for the CFL 6500K observation of 2023/01/02.#

The CFL spectra is rather defocused but it is enought to perform wavelength calibration.

../_images/wl_ident_CFL_20230102.png

Fig. 107 Mean CFL 6500K spectrum with lines identified.#

../_images/observed_sekonik_CFL6500.png

Fig. 108 Mean CFL 6500K spectrum observed and measured with Sekonik C-3000 spectrophotometer.#

Warning

The test should be repeated with a better focus to get a proper response curve

Test of 2023/01/10#

This test was performed by Ignacio Hilera. He used several LED of different color temperatures and two CFL.

../_images/test_20230110_DSCF7313.jpg

Fig. 109 Lamps used in the 2023/01/10 test.#

../_images/DSCF7310.JPG

Fig. 110 Clip of picture DSC7310 with two CFL lamps of different temperature.#

../_images/test_20230110_DSCF311_recorte.png

Fig. 111 Clip of picture DSC7311 with two CFL lamps of different temperature.#

../_images/test_20230110_all.png

Fig. 112 Clips of pictures.#

../_images/test_20230110_ima_CFL.png

Fig. 113 Clip of DSCF73100 with two CFL spectra.#

../_images/test_20230110_mono_CFL.png

Fig. 114 Monodimensional CFL spectra of picture DSCF7310.#

../_images/CFL_7311.png

Fig. 115 Mono dimensional CFL spectrum of picture DSCF7311 with lines identified.#

../_images/CFL_7310_wl_ident.png

Fig. 116 Mono dimensional CFL spectrum of picture DSC7310 with lines identified.#

../_images/graph_Lexman_LED2700K.png

Fig. 117 Mono dimensional Lexman LED2700K spectrum of picture DSCF7306.#

../_images/graph_Lexman_LED4000K.png

Fig. 118 Mono dimensional Lexman LED4000K spectrum of picture DSCF7305.#

../_images/graph_Lexman_LED6500K.png

Fig. 119 Mono dimensional Lexman LED6500K spectrum of picture DSCF7304.#