An exoplanet transit can dim its star by one percent or much less. To see that dip, a camera has to measure the same star, frame after frame, for hours, with errors well below the size of the signal. Variable-star monitoring and stellar occultations ask for the same thing in different forms: steady, repeatable measurements and accurate times.
This post covers what time-series photometry asks of a cooled scientific camera and the questions worth settling before you choose one. It applies to any camera, and we use Finger Lakes Instrumentation (FLI) models as examples.
What time-series photometry asks of a camera
Imaging and photometry reward different things. A beautiful image needs low noise and a wide field. A light curve needs stability: the same response to the same light, hour after hour, as the target moves across the sky and the air cools.
That means the camera's job is less about any one headline spec and more about consistency. Cooling that holds its setpoint, a predictable response across the range of signal you plan to use, calibration you can repeat every night, and timestamps you can trust.
The noise budget
Every photometric measurement carries several noise terms. The standard "CCD equation" in Howell's Handbook of CCD Astronomy adds them up: photon noise from the star, photon noise from the sky background, dark current, and read noise for every pixel inside the photometry aperture.
For bright targets, the star's own photon noise usually dominates, and on a ground-based telescope scintillation adds another floor. Scintillation is the rapid twinkling caused by turbulence high in the atmosphere. It shrinks with larger apertures and longer exposures, and it grows toward the horizon (Young 1967; Osborn et al. 2015). No camera removes it, but your exposure strategy can reduce its share.
For faint targets, sky background, dark current and read noise matter more. That's where cooling and low read noise pay off. Run the numbers for your own targets and aperture before you pick a camera, because the answer changes with what you observe.
Cooling and dark-current stability
Dark current falls steeply as a sensor gets colder, so cooling helps. For photometry, stability matters as much as depth. If the sensor temperature drifts during a night, the dark signal drifts with it, and that change can creep into a light curve.
Choose a setpoint the camera can hold on your warmest night, with margin, rather than the coldest temperature it can reach on a cool one. Then log the sensor temperature with every frame so you can check it later. For cooling hardware choices, see our post on cameras for remote and polar observatories.
Linearity, full well and defocusing
Photometry assumes that twice the light gives twice the signal. Near saturation, that stops being true. Keep the peak pixel of your brightest comparison star well below full well, and measure where your camera's response starts to bend. Our post on acceptance testing for gain, noise and linearity shows how.
Many transit observers defocus on purpose. Spreading each star over many pixels lets you collect more light per frame without saturating, and it averages out small differences between pixels and small tracking errors (Southworth et al. 2009). The cost is a larger aperture with more sky background and more read noise, so balance it against your faintest comparison stars.
Sampling still matters when you stay in focus. Our post on pixel size and sampling covers the arcseconds-per-pixel math.
Calibration frames without a shutter
Bias, dark and flat frames remove the camera's own signature from your data. For time-series work, take them the same way every night, at the same setpoint and with the same exposure times you use on the sky.
Not every camera has a mechanical shutter. FLI Aurora cameras are not offered with one. CMOS sensors don't need a shutter for exposure timing, and a shutter is a moving part. For darks, use a filter wheel with a blocked position, such as our CFW8-8RND, which holds seven 50 mm round filters plus a blank, or a third-party external shutter. Kepler cameras support internal shutters if you prefer one.
Flats deserve extra care, because any error in them follows the star as it drifts across the sensor. Good guiding that keeps the star on the same pixels all night helps more than almost anything else.
Timing and timestamps
Transit mid-times and occultation timings are only as good as the clock behind them. Know exactly what each timestamp means: the start of the exposure, the middle, or the moment of readout. Then convert times to a standard such as BJD_TDB, as Eastman, Siverd and Gaudi (2010) recommend for exoplanet work.
Readout architecture affects timing too. Kepler sensors use a rolling shutter with a global-reset mode, so ask how the start time is defined across rows before you design a timing-critical program.
A GPS timestamp option is listed for Kepler and Aurora cameras; contact FLI for current status on your model.
Filters and the filter wheel
Standard photometric filters let you compare results with other observers and models. Many transit programs also use a broad filter for more light, or several bands to check that a dip looks the same in each color.
The filter wheel sits in the light path, so plan its spacing and filter size with the camera. FLI makes filter wheels in a range of sizes, listed on our accessories page. For multi-band time series, the FLI High Speed Filter Wheel can store a sequence of up to 50 filter moves and step through it from a trigger input, without waiting on the host computer. Our post on back focus, flanges and filter wheels covers the spacing.
Choosing the camera line
- Long exposures on faint targets. Aurora cooled CMOS cameras use Sony back-illuminated sensors with very low read noise. The ProLine PL23042 CCD has lower dark current and is easier to calibrate for multi-minute exposures, but it reads out slowly.
- Fast cadence and continuous monitoring. Kepler CMOS cameras are designed for fast, continuous, short-exposure imaging, and exoplanet research is one of their listed applications. We don't recommend single Kepler exposures longer than about five minutes. Stack shorter frames instead.
For software, FLI provides an ASCOM driver for Kepler cameras, and our ASCOM driver supports FLI CCD cameras and filter wheels. INDI drivers are available for FLI CCD cameras and filter wheels. Drivers and manuals are on our support page. If you run a network of telescopes, our post on SDA sensor networks covers fleet planning.
Questions to settle before you choose
- How faint is your faintest target, and how deep is the signal you need to detect?
- What cadence do you need: seconds, tens of seconds or minutes?
- Will you defocus, and how large will your apertures be?
- What cooling setpoint can you hold on your warmest night?
- How will you take darks: a shutter, a blocked filter position or an external shutter?
- Which time standard do you report, and how accurate must the timestamps be?
- Which filters, and what size, does your optical path allow?
Data-sheet values are typical. A production test summary exists for each camera and can be provided by serial number, which helps when you build a noise model for your program.
Talk to our sales team
FLI cameras are Made in USA, and our sensor chambers are assembled in our 1,000+ sq ft ISO Class 7 clean room. See how on our in-house manufacturing page.
Send our sales team your telescope, targets, cadence and filters, and we'll help you choose a camera and filter wheel for your photometry program. Request a quote or email sales@flicamera.com.
