Pixel scale adequate?

[SCO]

Is a pixel scale of 0.9 arcsec/pixel capable of doing the highest quality exoplanet work?

[karenacollins]

Yes

[SCO]

Thanks Karen.

I probably should have explained why I asked this.  

I am buying a new camera specifically for exoplanet work.  Camera option 1 has a pixel size of 15 microns and a full-well size of 150,000 electrons.  Camera option 2 has a pixel size of 13.5 microns and a full-well size of 100,000 electrons.

The exoplanet manual I am reading says:

6.2.1. Choice of Binning The FWHM of stars in the target field should be spread out over 3-5 pixels. If the imaging camera is normally setup to bin more than 1x1, binning might have to be reduced to achieve this.  This then would result in an acceptable image scale (sometimes called “plate scale” or “pixel scale”). Image scale is expressed in arc-seconds per pixel and is a function of the focal length of the telescope and the size of the CCD photosite. For example, assume that the FWHM is 3.0 arcseconds, and the unbinned image scale of the observer’s imaging system is 0.5 arc-seconds/pixel.  This means that, without binning, the FWHM would be spread over 6 pixels and that 2x2 binning would mean that that the FWHM would be spread over 3 pixels, still within the acceptable, albeit the lower, part of the range.

Assume a FWHM is 3.0 arcseconds:

Camera 1 has an unbinned image scale of 0.9 arcseconds/pixel
Camera 2 has an unbinned image scale of 0.8 arcseconds/pixel

FWHM for camera 1 would spread out over 3.33 pixels
FWHM for camera 2 would spread out over 3.75 pixels

Both are technically between 3 and 5 pixels and are therefore acceptable (assuming a FWHM of 3.0 arcseconds), however camera 2 is more inline with the 3-5 pixel recommendation.  Therefore camera 2 would seem to be the better choice.

However camera 1 has a full-well capacity of 150,000 electrons vs. camera 2 which has a full-well capacity of 100,000 electrons.  Therefore camera 1 would be far less likely to saturate as the airmass changes across the sky.  In that way, camera 1 would be the better choice.

[Rick_N.]

 Yesterday I was finding a magnitude for a SN using a DECaLS FITS image file. This has a scale of approximately 2" for a 4px radius aperture, or 8px diameter. This is OK and occasionally I use a 3pixel radius (6px diameter) aperture. Below that, say for a 2px radius, you have an amount of pixels that might not tell you much anyway as it is a square made up of 9 pixels.

This is using the 4m DECAM CTIO telescope. PanSTARRS is much the same. I guess what I'm saying is that the
highest number of pixels per arcsecond is what is required, so Camera 2 seems a good choice.

[SCO]

That is good information.  Thanks.

[Rick_N.]

OK I've tried some different FITS files and found that for a 4px radius aperture, PanSTARRS has a 1.99" coverage, much like DECaM 2.01". Interestingly, SDSS with a 4px radius aperture has a 3.17" coverage.

[SCO]

Thanks Rick.  I have already called FLI and requested to change the chip from a 15 micron CCD230-42 to a 13.5 micron CCD42-40 (both made by Teledyne e2v).  I will lose some flexibility in terms of avoiding saturation, but I am more comfortable with the FWHM covering more pixels.  I greatly appreciate your input on this question.  It has allowed me to get off the fence and make a final decision.

By the way, my dream would be to get access to a 4 meter telescope.  Right now all I have is my little 0.5 meter scope.  And I feel very lucky to have even that.. :-)

Ed

[Rick_N.]

Well good luck with your searching. I would guess that if you are looking for exoplanets you should invest in a very good clock for your observatory so that the telescope tracks very accurately. Perhaps an exposure time is limited to how well everything tracks over say 1, 5, 10mins etc. but I don't know.
 
Most of the surveys mentioned have exposures of 2 or 3 minutes, but their main purpose is to cover as much sky as possible but still maintain quality.