measuring the throughput in spectrographs...fiber linked echelle spectrograph. caos efficiency of an...

By Gerardo Avila & Carlos Guirao

CAOS (https://spectroscopy.wordpress.com/)

10-11 June 2017St. Niklausen - ASpekt 20171

Measuring the throughput in

spectrographs

CAOS

CAOS group

10-11 June 2017St. Niklausen - ASpekt 20172

Gerardo Avila

Vadim Burwitz

Carlos Guirao

Jesús Rodriguez

CAOS

Purposes of our presentation

10-11 June 2017St. Niklausen – ASpekt 20173

The spectrograph throughput is one of the most important features for astronomy observations. However its measurement is difficult and usually is calculated by extrapolation of individual component efficiencies. We propose a simple method of measuring the overall efficiency at single wavelength by using a laser based device. We demonstrate it on a fiber linked Echelle spectrograph.

CAOS

Efficiency of an optical system

10-11 June 2017St. Niklausen – ASpekt 20174

The optical efficiency of an optical system is defined as the rate of optical energy through the system, divided by the energy coming directly from the source:

h= 𝐸𝑖

𝐸𝑜100%

CAOS

Basic Echelle design

10-11 June 2017St. Niklausen – ASpekt 20175

Full parabola as collimator

Output fibre end

CCD Tele-objective X-disperser Échelle

9º

CAOS

Echelle spectrograph efficiency

10-11 June 2017St. Niklausen – ASpekt 20176

The efficiency includes: Fibre link: telescope-fibre coupling, fibre

transmission and fibre-collimator coupling Collimator Echelle Cross-disperser (Prism or Grating) Camera CCD

CAOS

Fibre link efficiency (1)

10-11 June 2017St. Niklausen – ASpekt 20177

Pinhole – seeing ratio

Pinhole – lens/fibre misalignment

F/# matching and misalignment

Polishing quality

Fresnel, input and output lenses

Lenses misalignment errors and aberrations

Internal fibre transmission

Focal ratio degradation (FRD)

CAOS

Pinhole - seeing ratio

10-11 June 2017St. Niklausen – ASpekt 20178

CAOS

Fibre link efficiency (2)

Typical values

10-11 June 2017St. Niklausen – ASpekt 20179

Pinhole – seeing (when seeing = ø fibre) 50%

Polishing quality 95%

Fresnel, input and output lenses 92%

Lenses misalignment errors 98%

Internal fibre transmission (length) 96%

Focal ratio degradation (FRD) 60-95%

Total (not including seeing) 49-78%

CAOS

Telescope – fibre coupling - FRD

10-11 June 2017St. Niklausen – ASpekt 201710

Fibre Optics

Output

Collimator

Telescope

aperture

Input beam

aperture

beam

CAOS

Telescope – fibre input coupling

10-11 June 2017St. Niklausen – ASpekt 201711

F F'

Fibre

Telescope pupilTelescope Image

Fibre

Telescope pupil

StarStar

CAOS

Fibre coupling

10-11 June 2017St. Niklausen – ASpekt 201712

With 1x grin-lens With 2x grin-lenses

CAOS

Fibre output – collimator coupling

10-11 June 2017St. Niklausen – ASpekt 201713

F/3 f 5mm F/20

CAOS

Collimator mirror reflectivity

10-11 June 2017St. Niklausen – ASpekt 201714

Edmund-Optics coatings for mirrors:

CAOS

Echelle grating efficiency

10-11 June 2017St. Niklausen – ASpekt 201715

The maximum efficiency of echelles is maintained near the Littrow condition (when angle of incidence equals angle of diffraction)

Within each order, the efficiency will be maximum at the middle of the order. Typically reaching 50 to 75%. It drops to about one-half these values at the crossover points (free spectral range)

Out of Littrow configuration, the echelle reflectivity reduces rapidly (shadowing effect)

CAOS

Echelle orders efficiency

10-11 June 2017St. Niklausen – ASpekt 201716

Peak efficiency (100%)

Green laser 532nm (94%)

Correction coefficient: 1.064

CAOS

Shadowing effect

10-11 June 2017St. Niklausen – ASpekt 201717

BN

70

75

80

85

90

95

100

0 1 2 3 4 5 6

Rel

ativ

e ef

fici

ecny

Grating angle

Shadowing in Echelles

CAOS

Échelle out of Littrow

10-11 June 2017St. Niklausen – ASpekt 201718

The “grating angle” should be as small as possible in order to approach the Littrow configuration where the efficiency is the maximum. In our FLECHAS design, we found that 9° was the best compromise between the size of the camera objective and the optical table.

9º

Échelle

Output fibre end

Full parabola as collimator

CCD Tele-Objective X-disperser

CAOS

FLECHAS Spectrograph

10-11 June 2017St. Niklausen – ASpekt 201719

Parabola f 444 mm, ø 75 mm, Edmund Optics

Collimator beam F/18

Pupil 25 mm

Échelle 79 li/mm 63º 25×50×9, Thorlabs

X-disperser Prism N-F2 60º 60x60x60mm

Objective f 200 F/2.8, Canon

CCD Atik 11000 4008×2672×9 µm (24×35)

CAOS

Fibre link

10-11 June 2017St. Niklausen – ASpekt 201720

Fibre ø 50 µm and 10 m long, Polymicro FIP

Fibre Injection F# F/3

Fibre Output F# F/3

Output lens Doublet f = 5 mm, ø 3 mm, Linos

Collimator F# F/18.5

Beam projection at the collimator (pupil)

ø 24 mm

Fibre image at the collimator (slit equivalent)

ø 308 µm

CAOS

Optical efficiencyaccording to manufacturer

10-11 June 2017St. Niklausen – Aspekt 201721

Parabolic Mirror 95 % Échelle 55 % Echelle vignetting 95 % Prism 92 % Canon objective 90 %

Fibre efficiency 49%-78 % Spectrograph only 41% Spectrograph + Fibre 20%-32%

CAOS

10-11 June 2017St. Niklausen – ASpekt 201722

0

10

20

30

40

50

60

70

80

90

100

400 500 600 700 800 900

Inte

rnal

tra

nsm

issi

on

(%

)

Wavelength (nm)

20 m

FBP FVP

Fibre transmission (internal) in 20 m. FLECHAS uses the FBP type

Fibre efficiency

CAOS

10-11 June 2017St. Niklausen – ASpekt 201723

88

90

92

94

96

98

100

350 400 450 500 550 600 650 700

Tra

nsm

issi

on (

%)

Wavelength (nm)

Reflectivity Enhanced Aluminium

Parabolic mirror reflectivityLab measurements

CAOS

X-disperser. Transmission grating efficiencyLab measurements

10-11 June 2017St. Niklausen – ASpekt 201724

0

10

20

30

40

50

60

70

80

90

100

350 400 450 500 550 600 650 700 750 800

Tra

nsm

issio

n

(%)

Wavelength (nm)

X-disperser Newport

CAOS

X-disperser. Prism efficiencyLab measurements

10-11 June 2017St. Niklausen – ASpekt 201725

80

82

84

86

88

90

92

94

96

98

100

350 400 450 500 550 600 650 700

Tra

nsm

issi

on (

%)

Wavelength (nm)

Prism N-F2 60 60x60x60mm

CAOS

Measuring prism efficiency

10-11 June 2017St. Niklausen – ASpekt 201726

CAOS

Camera efficiencyLab measurements

10-11 June 2017St. Niklausen – ASpekt 201727

40

50

60

70

80

90

100

350 400 450 500 550 600 650 700 750 800

Tra

nsm

issio

n(%

)

Wavelength (nm)

Canon 200mm F/2.8 EF

CAOS

Measurement of the reference flux (100%)

10-11 June 2017St. Niklausen – ASpekt 201728

Laser

3 X objective

0.5 mmpinhole

Iris

16 X objective

F/3

Integrating sphere

Photo diode

I / Vtransducer

CAOS

Measurement of the flux through fibre

10-11 June 2017St. Niklausen – ASpekt 201729

Laser

3 X objective

0.5 mmpinhole

Iris

16 X objective

Fibre

Iris

Integrating sphere

Photo diode

I / VtransducerF/3

F/3

CAOS

Integration sphere

10-11 June 2017St. Niklausen – ASpekt 201730

CAOS

10-11 June 2017St. Niklausen – Aspekt 201731

Measurements

CAOS

Measurements

10-11 June 2017St. Niklausen – ASpekt 201732

Injection in fibre F/3 Laser wavelength 532 nm Peak correction coefficient 1.064

h= 𝐸𝑖

𝐸𝑜100 (%)

Fibre efficiency 82.4% Spectrograph only 39.7% 42.2% (*)

Spectrograph + fibre 32.7% 34.8% (*)

(*) After applying peak correction coefficient

CAOS

Conclusions

10-11 June 2017St. Niklausen – ASpekt 201733

The efficiency of an instrument is a feature as important as any other (resolution, wavelength range, etc..)

Most critical component: the fibre. A bad FDR could cause you loosing 30% of the light (imagine you need a telescope 30% bigger!).

Efficiency measurements can be easily performed with lasers (blue, green and red).