Throughput Measurements
for various Gecko configurations

Before taking a look at the numbers in the following table, please remember a few things:

Data for these measurements have generally been obtained under photometric conditions. However, for some nights, there might have been some hardly detectable small cirrus clouds, which would then affect the throughput values.
Vega was used for some measurements because its magnitude is zero. When stars other than Vega have been used, we applied a correction for the magnitude using the UBVRI magnitude closest to the observed wavelength. However, this doesn't mean that the magnitude of the star for the wavelength observed is equal to the broad-band one. This can also affect the final throughput measurement.
Imperfect guiding or declination oscillations can lower the throughput.
The choice of filter or grism can significantly change the throughput.
The theoretical response of the different orders must be taken into account when using a value in the table to find the throughput value for another (close) wavelength.
The CCD QE must also be considered.
Using the UV optics for Gecko can change the throughput.
The throughput numbers below are the average number of electrons (not ADUs) you can expect to get per second when you sum all the pixels of your spectra in the spatial dimension, for a Magnitude=0 star.

There are throughput measurements for the following configurations:

Throughput measurements for Gecko + CAFE + EEV1, and the "main" optical fiber.

Wavelength (Å)	Order	Gecko Optics	Filter or Grism	THROUGHPUT (e-/col/sec)	Notes
8660	6	RED	1814	17840	Vega
7773	7	RED	1704	39400	Omi Her and Vega
7700	7	RED	1704	36810	Vega
6707	8	RED	1626	76000-90000	Omi Her and Vega
6565	9	RED	1617	14750	Vega
6428	9	RED	1617	36030	Vega
6286	9	RED	RED	47390	Vega
6150	9	RED	RED	76200	Omi Her and Vega
5957	9	RED	1521	97800	Omi Her and Vega
5890	9	RED	1519	78360
5890	9	RED	1521	94533	Vega
5786	10	RED	1521	42000	Omi Her and Vega
5657	10	RED	RED	64400	Vega
5143	11	RED	Blue	78000	Vega
4860	12	RED	Blue	17030	Vega
4352	13	RED	Blue	29410	Vega
4040	14	UV	UV	6500-7200	July 2001; 2 stars, 2 nights, 2 measurements
3950	15	RED	UV	23630	Vega
3947	14	UV	UV	22500	July 2001; 2 stars, 4 measurements, 3 nights
3930	14	UV	UV	10000-14000	July 2001; 2 stars, 2 measurements, 2 nights
3772	15	UV	UV	5264	HR39, Sep 14 2000
3635	16	UV	UV	350-600	July 2001; 3 stars, 2 nights, 3 measurements
3536	16	UV	UV	2242	HR39, Sep 14 2000
3328	17	UV	UV	949	HR39, Sep 14 2000
3130	18	UV	UV	168	HR39, Sep 14 2000

Wavelength (Å)	Order	Gecko Optics	Filter or Grism	THROUGHPUT (e-/col/sec)	Notes
8082	center of 7	RED	RED	11000	2 stars
7281	8	RED	1711	150002000	3 stars
7281	8	RED	filter	15000-30000?	2 stars, 2 results?
7072	center of 8	RED	RED	20900	2 stars
6707	8	RED	RED	75000	2 stars
6675	8	RED	RED	42000	1 star + cirrus
6645	8	RED	1617	33000-45000	2 stars
6600	8	RED	RED	40000	1 star + cirrus?
6562	8	RED	RED	35000
6335	9	RED	RED	70400	2 stars
6330	9	RED	1604	40000	4 stars
6330	9	RED	1604	34777	Xi Ori
6330	9	RED	RED	29201	Xi Ori
6286	center of 9	RED	RED	29500	3 stars
6150	9	RED	RED	35000-45000	3 stars
5657	center of 10	RED	BLUE	37900	3 stars
5143	center of 11	RED	BLUE	44000	1 star
4861	12	RED	BLUE	14600	2 stars
4714	center of 12	RED	UV	19700	2 stars
4455	12	RED	BLUE	23000
4352	center of 13	RED	UV	18100	2 stars
4041	center of 14	RED	UV	19400	3 stars
3950	14	RED	UV	11000	3 stars

Wavelength (Å)	Order	Gecko Optics	Filter or Grism	THROUGHPUT (e-/col/sec)	Notes
4100	14	RED	UV	3100	2 stars
5142	11	RED	BLUE	30000	3 stars
6565	8	RED	1617	43300	1 star
7281	8	RED	1711	19000	2 stars
8082	7	RED	RED	17000	2 stars

Wavelength (Å)	Order	Gecko Optics	Filter or Grism	THROUGHPUT (e-/col/sec)	Notes
3078	18	UV	UV	400	4 stars + cirrus

Take spectral data and biases (or darks)
Combine the biases or darks
Subtract the average bias or dark from the object frame
Do NOT flat field the data
Extract the spectra; use apall and carefully select the aperture to include all the slices
Get the mean signal in the ADU for the whole spectra; use imstat on the main portion of the spectra (cut the edges if there is vignetting)
Get the mean signal in e-/col/sec by multiplying the number of ADU by the gain, and dividing by the exposure time
Correct for the magnitude of the star to normalize to magnitude 0
The airmass correction is not computed; data was taken with reasonable airmasses.