Suitability for exoplanet observations

[badgerchap]

I'm going to be doing some observation/detection of known exoplanets for my final project in my degree. I'm looking into an appropriate camera for the detection process. I'll be using the recently released OSCAAR software to process and compare the light curves. What I'm looking for though is some advice regarding a suitable camera for the image capture process.

Suitable target stars range from about 8th to 12th magnitude, with transit depths ranging from around 1/1000 to around 1/50 of a magnitude. 

My uni only has an 8-bit camera (i.e. 2⁸=256 brightness levels), which will be insufficient for all but the deepest transit depths.

I'm trying to work out whether or not my unmodded Canon 1100D will be sensitive enough to detect the faint variations in star brightness during the transits.

The Canon sensor appears to be 14 bit (2¹⁴=16,384 brightness levels) although 14 bit sounds mighty odd to me.

Further to this, as it is colour, the Bayer matrix means that the camera is less sensitive than a mono version of a comparable chip. Will this affect my ability to detect the variations though, or will this depend only on the bit-number of the chip? Also, it's a CMOS chip rather than a CCD.

I'm looking at taking 30 sec exposures through a 12" Newt, so I think the magnitude itself is realistic.

I would really like to get hold of a 16-bit dedicated CCD for this (would give me 4x the brightness levels), but I may encounter budgetary restrictions (my new name for the wife), so I'd like to have the option to use the Canon if possible.

I appreciate that people may not have experience of exoplanetary observation, but I know people know these cameras inside out and upside down!

Does anyone have any thoughts on this? Thanks!

[gkec]

Comments:

The bit depth is dependant on the ADC.  14 bits or even 12 bits is not unusual.  Always assume 2 bits of noise for a single reading.  So effectively remove 2 bits from the quoted bit depth.  These can be recovered by taking multiple images.

Why exo-planets?  Is it because it is "sexy" at the moment?

There is more original observation/research to be made in the field of variable stars such as the R R Lyra types. There are many variable stars and not enough observers.

[nmoushon]

I'm not techy enough to really help you with this one much but I remember reading an article about Canon producing an ultra-high sensitive CMOS chip. Not sure the price but I would think since its a CMOS it would be cheaper than a CCD. Here's the Canon site that has small write up on it. I've always wanted to try this so will be following with interest to see what you'll be able to get with a students budgets. As, so many of us on here, have a budgetary restriction. If you get me haha. 

[badgerchap]

Mainly because I want to do something outside the usual remit of the department. We're supplied with a long list of available, predefined projects which, whilst interesting, have all been done by previous students. I want to do something which hasn't been done there before. I think it'll help me get a better mark when it comes to it. Showing initiative and a bit of creativity should hopefully be a feather-in-cap. Plus, yeah, it's sexy - and it's one of the reasons I got into Astro in the first place - the whole 'x billion stars in the galaxy, x billion galaxies..." statistic. Pre 1995 I loved the mystery of the potential habitability of other solar systems. I think it's dead exciting to be alive right now.

[badgerchap]

I'm not techy enough to really help you with this one much but I remember reading an article about Canon producing an ultra-high sensitive CMOS chip. Not sure the price but I would think since its a CMOS it would be cheaper than a CCD. Here's the Canon site that has small write up on it. I've always wanted to try this so will be following with interest to see what you'll be able to get with a students budgets. As, so many of us on here, have a budgetary restriction. If you get me haha. 

Looking of the size of that thing, looks like it could be quite pricey!

[gkec]

I'm not techy enough to really help you with this one much but I remember reading an article about Canon producing an ultra-high sensitive CMOS chip. Not sure the price but I would think since its a CMOS it would be cheaper than a CCD. Here's the Canon site that has small write up on it. I've always wanted to try this so will be following with interest to see what you'll be able to get with a students budgets. As, so many of us on here, have a budgetary restriction. If you get me haha. 

A 20cm x 20cm sensor in a commercially available camera?  Although they don't mention it the device is probably cooled.  CCDs have always been able to go low light when cooled.  What is interesting about this device is the size which is approaching the size of the photographic plates astronomers used to use until electronic detectors became more effective, and the ability to readout such a device at video rate.  Nice but unfortunately not cheap.  Sorry .

[Paul81]

You will also need a sensor with a linear response over a large range of the well depth of the camera - at some point most sensors exhibit a non linear response past a certain count. The linear region is the only part you can use for photometry.

This value can be hard to find in some data sheets so has to be determined experimentally. I read a great paper on amateur exoplanet hunting and this was one of the first steps!

Good luck!

[dph1nm]

Try reading this paper

http://adsabs.harvard.edu/abs/2010JAVSO..38..212L

Also, googling "exoplanets DSLR" throws up quite a few links to people who have attempted this. I think the answer is it is quite feasible.

NigelM

[badgerchap]

Thanks Nigel

[FraserClarke]

Made sure you consider the whole system, not just the camera. 8-bit isn't necessarily a problem, though at 12/16-bit camera will certainly make it easier. You certainly want a linear detector, as it will make calibrating a heck of a lot easier. You can probably find very suitable detectors second-hand, as you don't really need huge fields of view. As it's for a University project, can you get some funding from the department to buy a camera? A good argument here is often that they'll be able to use it for projects in future years -- developing enough final year projects is usually a limiting factor in University departments...

Key points for exoplanet transit work, in order of importance;

+ Detect enough photons to do better than 0.1% photometry

+ Have enough field of view to make sure you have enough comparison stars

+ Be able to calibrate the detector/telescope/observations well enough that you are (as close as possible to) photon-noise limited.

+ Make sure you can polar align/guide accurately enough to keep the target in the same position as closely as possible (ideally on the same pixel)

+ Try to control the focus as well as you (and ideally de-focus the image to improve photometry)

+ Understand what filter response you have, so that you can physically interpret the results better (depends on what you're trying to achieve)

As it's your final year project -- I'm not going to give you the answers on how to do above

[badgerchap]

Thanks for all the advice guys. Some new info there and some old stuff confirmed. I'm going to do a series of practice runs this year (I'm still in Year 2) and see how they go. I'll post on here regarding results, failures etc. Thanks again!