Precision of a Low-Cost InGaAs Detector for Near Infrared Photometry

We have designed, constructed, and tested an InGaAs near-infrared camera to explore whether low-cost detectors can make small (<1 m) telescopes capable of precise (<1 mmag) infrared photometry of relatively bright targets. The camera is constructed around the 640x512 pixel APS640C sensor built by FLIR Electro-Optical Components. We designed custom analog-to-digital electronics for maximum stability and minimum noise. The InGaAs dark current halves with every 7 deg C of cooling, and we reduce it to 840 e-/s/pixel (with a pixel-to-pixel variation of +/-200 e-/s/pixel) by cooling the array to -20 deg C. Beyond this point, glow from the readout dominates. The single-sample read noise of 149 e- is reduced to 54 e- through up-the-ramp sampling. Laboratory testing with a star field generated by a lenslet array shows that 2-star differential photometry is possible to a precision of 631 +/-205 ppm (0.68 mmag) hr^-0.5 at a flux of 2.4E4 e-/s. Employing three comparison stars and de-correlating reference signals further improves the precision to 483 +/-161 ppm (0.52 mmag) hr^-0.5. Photometric observations of HD80606 and HD80607 (J=7.7 and 7.8) in the Y band shows that differential photometry to a precision of 415 ppm (0.45 mmag) hr^-0.5 is achieved with an effective telescope aperture of 0.25 m. Next-generation InGaAs detectors should indeed enable Poisson-limited photometry of brighter dwarfs with particular advantage for late-M and L types. In addition, one might acquire near-infrared photometry simultaneously with optical photometry or radial velocity measurements to maximize the return of exoplanet searches with small telescopes.