ENERGY

ENERGYa proposal for a Multi-channel

Cmos Camera

F. Pedichini, A. Di Paola, R. SpezialiINAF Oss. Astr. Roma

h

e-

What is multi-channel imaging?The most standard duty of an astronomer

simply : Multi Band Photometry

How to do it?

The 3 most used technologies:

• 1- an old, but good, FILTER WHEEL + 1 CCD imager

All of these items deserve a... deep comparative analisys !

•2- a MOSAIC FILTERED CCD imager•3- a DICROICH focal plane splitter + several CCD imagers

an old, but good, FILTER WHEEL + 1 CCD imager

Has been used since the first digital imager, with great results, low cost, simple mechanics, user selectable bands and thx. to

modern CCDs also a very good Q.E.

......but......

different bands are exposed and acquired at different times due to detector readout time and filters set up time

Imager...10101000111....

a MOSAIC FILTERED CCD imager

widely used in low cost tv cameras, digital photography apparatus and amateur astronomy color

imagers with all the bands taken during the same exposure

.....but.....

no user selectable bands, low Q.E. (quite impossible to be used with back illuminated detectors) and the

bad MOIRE effect...

....RGBRGB.... matrixfilter

the MOIRE effect...

the color of each pixel is function of the signals of the nearest ones producing nice on sharp edges....!

astronomers dont’LIKE!

a DICROICH focal plane splitter + several CCD imagers

sometime used in astronomy with the best Q.E. and time

sincronicity in each channel ....but....

complex, expensive and polarization & f# sensitive

Imager

Imager

Imager

...10101000111....

...10101000111....

...10101000111....

polarization & f# sensitive• dicroichs are made up by dielectric layers and

partially polarize the reflected and transmitted beams

• best performances about field uniformity need collimated beams so a collimator and several camera optics must be added in the imager optical train

• often the weight and size of a “dicroich” imager make it impossible to fit some telescope focal planes as in the case of Schmidts or fast prime foci

and now we go to a comparative table...

comparative table....

Type Q.E.% Sampling Syncro Complexity

FilterWhell

90 Good No Standard

MosaicFilter

20-30 Moire Yes Low

Dicroich 80-90 Good Yes High

........is the world of C-MOS helping us with a fourth chance?

http://www.foveon.com

in this detector each pixel outputs

R,G and B signals with all the nice features of a CMOS “on chip camera”

•about 10Mpix in three layers•low power•on chip 12bit ADC •windowing and binning•4 f.p.s.•less than 30e- RON

foveon phisycs:on silicon “dicroich”

• red photons are deep penetrating the silicon

• green photons make photoelectrons midway

• blue photons interact early.....

foveon bands:

• bands are not user selectable

• how do they compare to Johnson filters

• what can we do with them in astronomy?

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

0.3 0.4 0.5 0.6 0.7 0.8 0.9

BJ

VJ

RJ

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.3 0.4 0.5 0.6 0.7 0.8 0.9

Bfov

Vfov

Rfov

Black Bodies simulations:BB spectra from 2000 to 20000 K convolved with BVR Jhonson and BVR Foveon converted to magnitudes

Teff B V R Bf Vf Rf20000 7.3 7.57 7.74 7.47 7.61 7.7515000 7.99 8.18 8.29 8.11 8.2 8.310000 9.24 9.24 9.24 9.24 9.24 9.248000 10.13 9.98 9.89 10.02 9.95 9.896000 11.57 11.16 10.91 11.27 11.09 10.94000 14.4 13.47 12.87 13.64 13.24 12.822000 22.58 20.22 18.5 20.05 19.22 18.24

0

5

10

15

20

25

0 5000 10000 15000 20000 25000

Teff

B

V

R

Bf

Vf

Rf

reduction to Johnson....

R2 = 0.9998

R2 = 0.9992

R2 = 0.9989

-0.5

0

0.5

1

1.5

2

2.5

3

-0.2 0 0.2 0.4 0.6 0.8 1 1.2Vf-Rf

B-Bf

V-Vf

R-Rf

Poli. (B-Bf)

Poli. (V-Vf)

Poli. (R-Rf)

it is possible using instr. color index Vf-Rf or Bf-Vf and a second order polinomya. error is 0.01 mag.

V-Vf = 0.7818(Vf-Rf)^2 + 0.2571 (Vf-Rf)- 0.0051

Energy camera:

•Uses Foveon M10 x3 CMOS

•2200 x 1500, 9 micron pixels

•3 bands, 4fps

•about 25 arcmin2 @ 12 m focal lenght

•0.15 arcsec/pixel

•30 e-RON

•0.25 seconds read out time

•no shutter needed

LBC comparison on a mag 14 star:CCD:exp. 0.013 secQE 90%2e/aduRON 5 e-

Energy:exp. 0.04 secQE 30%5e/aduRON 30 e-

Time for 3 bands:exp. 3x0.013 secfilter 3x1 secreadout 3x1sec total ~3 sec

Time for 3 bands:exp. 1x0.04 secfilter 0 secreadout 1x0.25sectotal ~0.3 sec

Target:star field BVRfast photometryat S/N = 100 to detectlight transients SKY 1”seeing ~21mag/arcesc^2

Telescope:Mirror 8.2m. 1.4F#(prime focus camera)

Both cameras: 9 micron pixelsabout 4Mpixel

LBC comparison on mags @ S/N 100:

0

2

4

6

8

10

12

11 13 15 17 19 21 23Mag

CCD expEnergy exp

3.9Hz

3.5Hz

2Hz

0.5Hz

2.9Hz

1.1Hz

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

4 6 8 10 12 14 16 18

ENERGY @ 60/90/180 Schmidt:

On a smaller telescope at the resolution of 1”/pixel the field is 1180 arcmin^2.

Energy becomes a:multiband GRB Hunter on raw satellite triggers

Mag

Hz

S/N=100S/N=10

2nd comparative table....

Type Q.E.% Sampling Syncro Complexity

FilterWhell

90 Good No Standard

MosaicFilter

20-30 Moire Yes Low

Dycroich 80-90 Good Yes High

CMOSFoveon

~30 Good Yes Low

open discussions:

•1) Is it useful in astronomy?

•2) Can Foveon modify the shape of the bands?

•3) Will we have more than 3 on-chip bands in the future?

•4) Can Foveon technology used on back ill. detectors?

•5) Can Foveon technology join CCDs?

the end.....Energy

is a Multi channel camera

or...

Mc2