Noise and electronic gain evaluation of a sample of astronomical CCD cameras

Method

 

The algorithm is implemented in ISIS software (see tutorial and application here).

Result

Model

Detector

Pixel size (microns)

Gain (e-/ADU)

RON (e-)

Dark signal per pixel (e-/s)

Abs. quantum efficiency @ 650 nm

Limit magnitude
(see note)

Atik460EX

ICX694AL

4.54 x 4.54

0.260

5.1

0.0013 @ -10°C

66%

15.6

Atik460EX

ICX694AL

9.08 x 9.08 (internal bin)

0.260

5.2

0.0046 @ -10°C

66%

16.1

Starlight Xpress
SXVR-H694  

ICX694AL

4.54 x 4.54

0.319

6.1

0..0110 @ -10°C

66%

15.5

Atik314L+ #1

ICX285AL

6.45 x 6.45

0.255

4.5

0.0002 @ -10°C

53%

15.6

Atik314L+ #2

ICX285AL

6.45 x 6.45

0.254

6.7

0.0002 @ -10°C

53%

15.2

Atik314L+ #3

ICX285AL

6.45 x 6.45

0.261

7.7

0.0004 @ -10°C

53%

15.0

Atik314L+ #4

ICX285AL

6.45 x 6.45

0.259

5.4

-

53%

15.2

Atik314L+ #5

ICX285AL

6.45 x 6.45

0.255

4.7

-

53%

15.6

Atik314L+ #6

ICX285AL

6.45 x 6.45

0.273

3.9

-

53%

15.5

Atik314L+ #7

ICX285AL

6.45 x 6.45

0.253

4.0

-

53%

15.5

Atik314L+ #8

ICX285AL

6.45 x 6.45

0.265

3.9

-

53%

15.6

Atik314L+ #9

ICX285AL

6.45 x 6.45

0.258

3.8

-

53%

15.6

Atik Titan

ICX424AL

7.4 x 7.4

0.334

7.4

0.014 @ 0°C

27%

14.3

Starlight Xpress SXVR-H9

ICX285AL

6.45 x 6.45

0.360

8.6

0.0002 @ -10°C

41%

14.8

Starlight Xpress MX916

ICX083AL

11.6 x 11.2

2.469

32.3

0.71e-/s @ -5°C

 -

-

QSI-583

KAF-8300

5.40 x 5.40

0.486

9.5

0.002 @ -10°C

44%

14.7

QSI-683

KAF-8300

5.40 x 5.40

0.494

8.6

0.0039 @ -10°C

44%

 

QSI-532 #1

KAF-3200ME

6.80 x 6.80

1.340

11.9

0.002 @ -10°C

79%

15.2

QSI-532 #2

KAF-3200ME

6.80 x 6.80

1.439

11.0

0.0008 @ -15°C

79%

15.2

QSI-520i

KAI-2020M

7.4 x 7.4

0.706

6.9

0.016 @ -10°C

32%

14.7

QSI-512

KAF-1603ME

9 x 9

2.828

17.8

0.022 @ -10°C

72%

 -

QHY11

KAF-11000M

9 x 9

0.695

10.49

0.039 @ -17°C

31%

-

SBIG STX-16803

KAF-16803

9 x 9

1.250

17.7

0.160  @ -10°C

40%

14.2

SBIG ST-402ME

KAF-402ME

9 x 9

1.127

16.7

0.062 @ -5°C

72%

14.8

SBIG ST-8XME

KAF-1603ME

9 x 9

1.534

15.9

0.042 @ -5°C

72%

14.9

Audine KAF-402ME

KAF-402ME

9 x 9

4.120

21.8

0.155 @ -5°C

72%

14.5


Note about limit magnitude computation: for a signal to noise ratio equal to 10 per resolution element (R=800) with a LISA spectrograph during observation of a A type star. The exposure time is 6 x 600 seconds (add of 6 elementary exposures). The telescope diameter is 28 cm for a f/ratio of 5.6. The slit width is 32 microns and the seeing of 3 arcsec. Sea altitude observation and star angular elevation of 45°.


Quantum efficiency of some detectors
from constructor datasheet (for Kodak) and web source evaluation (for Sony - but unknow measure protocal)

Wavelength

ICX285ALG
(Sony)

ICX694ALG
(Sony)

KAF-8300
(Kodak)

KAF-3200
(Kodak)

KAI-2020
(Kodak)

KAF-1603ME
(Kodak)

KAI11000M
(Kodak)

3500 A

18 %

31 %

18 %

36 %

20 %

30 %

30 %

4000 A

38 %

65 %

37 %

55 %

48 %

44 %

42 %

4500 A

55 %

72 %

40 %

58 %

57 %

47 %

48 %

5000 A

64 %

76 %

48 %

62 %

56 %

52 %

50 %

5500 A

63 %

76 %

54 %

78 %

46 %

67 %

44 %

6000 A

60 %

72 %

51 %

80 %

39 %

72 %

37 %

6500 A

53 %

66 %

44 %

79 %

31 %

71 %

31 %

7000 A

39 %

54 %

36 %

69 %

22 %

67 %

23 %

7500 A

29 %

42 %

31 %

60 %

14 %

52 %

17 %

8000 A

21 %

30 %

27 %

47 %

10 %

46 %

11 %

8500 A

14 %

20 %

20 %

38 %

6 %

36 %

7 %

9000 A

8 %

12 %

13 %

27 %

4 %

24 %

4 %

9500 A

5 %

6 %

7 %

13 %

2 %

13 %

2 %

10000 A

2 %

3 %

3 %

5 %

1 %

6 %

1 %




Quantum efficiency plots (datasheet + web)

Evaluation of quantum efficency for some sensors by measure

The quantum efficiency of Sony ICX285 and ICX694 CCD devices is calculated from comparison with observed signal on a KAF-3200ME CCD camera (QSI-532) and a KAF-8300 CCD camera (QSI-583). The KAF-3200ME and KAF-8300 QE are assumed to be known and exact. The measurement is made with a spectrograph LISA. The formula used for extract quantum efficiency is:

Note: the computed effciency is the "apparent QE" for a given camera (camera glass windows and detector coverglass optical transmission impact the result).

Here the result (quantum efficiency in percent):

Wavelength

KAF-3200ME
(on QSI-532 camera)

KAF-8300
(on QSI-583 camera)

ICX285
(on ATIK314L+ camera)

ICX694 (or ICX674)
(on ATIK460EXcamera)

4000 A

54 %

36 %

35 %

54 %

4500 A

57 %

45 %

57 %

70 %

5000 A

60 %

50 %

58 %

73 %

5500 A

75 %

57 %

55 %

71 %

6000 A

79 %

52 %

46 %

64 %

6500 A

75 %

47 %

38 %

57 %

7000 A

66 %

38 %

32 %

43 %

7500 A

58 %

32 %

26 %

34 %

8000 A

46 %

28 %

21 %

25 %

Pixel size, measured typical camera readout noise and measured/adopted inverse gain for this analysis:

 

QSI-532 camera

QSI-583 camera

ATI314L+ camera

ATIK460EX camera

Pixel size (p)

6.80 microns

5.40 microns

6.45 microns

4.54 microns

Reaout noise

12.1 electrons

10.4 electrons

3.7 electrons

5.4 electrons

Inverse gain (G)

1.380 e-/ADU

0.496 e-/ADU

0.274 e-/ADU

0.270 e-/ADU



Measured apparent quantum efficiency.

Call to contributors: transmit your proper measure (gain, read-out-noise, dark signal for a given detector temperature)  to christian.buil_at_wanadoo.fr for archiving and usefull (anonymous) comparison for amateur communauty. Process the data yourself or transmit necessary frames. Long exposure for dark frames are mandatory for precision (300 to 600 seconds integration time typicaly). Note also CCD temperature and work in total obscurity for dark and bias frames. Use binning 1x1 acquisition mode. If possible, indicate serial number.

Back