Please check the specifications of camera manufacturers before making a purchasing decision.
Financial fortune permitting, it is favorable to own two astrocams, a small sensitive model for planets, the moon and autoguiding and another with a larger sensor for deepsky targets, preferably a cooled version. If you don't mind fumbling with color filters, choose monochrome versions for higher sensitivity and resolution.
If you are owning a decent DSLR, then you may wish to stay with it for deepsky imaging without PC and cables, plus instant gratification. Anyway, there are options.
There is no explicit formula for choosing a CMOS camera for astrophotography. Any choice will be a compromise based on personal preference while depending on a few major factors all of which cannot team up ideally in all imaging situations and optical trains (primarily focal length dependent). The choice is often based on the following criteria:
The chip dimensions mated with a given focal length of an optical train (telescope) determines the field of view (FOV), or how much area of the sky is captured. It can be determined by:
fov = 2 * Atan(length / (2 * fl)) * 57.296
The diagonal size of a sensor can be determined by:
d = Sqrt(width * width + height * height)
The larger the chip area, the more computation power is required, meaning that smaller chips provide higher video frame rates. A "region of interest", ROI, can be set electronically to reduce the effective area for processing in order to achieve faster videos which are crucial for imaging the planets (and the moon) at high magnification. For this reason, "planetary cameras" contain small sensors. Note that you may need to crop stacking artifacts. Stacking software offers "drizzling" by which image size can be increased without interpolation (PC memory capacity permitting).
The larger a pixel the more light it can collect during a given duration hence requiring less exposure time, while also allowing longer exposure times since it takes more photons to saturation (full well capacity). Smaller pixels gather less light and are saturated faster which can result in blooming. Typical pixel sizes are from 1.45μm to 9μm. Smaller pixels provide higher image resolution per millimeter of sensor size and are best for short focal length telescopes. The resulting pixel (image) scale in arc seconds per pixel is obtained by:
206.265 * pixel size [μm] / fl [mm]
Atmospheric seeing largely determines how much pixels a star will cover, in that pixel (image) scale should be taken a theoretical and with a grain of salt. Most cameras allow 2x2 'binning' to double pixel size and quadruple pixel area electronically while rendering the image accordingly smaller sacrificing resolution. Binning enables higher sensitivity thus shorter exposure times which can be useful for focusing faint objects before imaging without binning. Another welcome advantage of 2 x 2 binning is the twice better signal-to-noise ratio, SNR:
Sqrt(4) = 2
Some cameras allow 2 x 2, 3 x 3 and 4 x 4 binning.
Quantum efficiency, QE, in percent is the fraction of received photons that the sensor is capable of converting to electrons and is measured against wavelength. The latest image sensors feature improved noise characteristics while performing at around 90% QE, therefore achieving high sensitivity and a wide dynamic range. The higher this "signal-to-noise ratio", SNR, the better the images.
An example of a response curve of a camera with a specified QE of 91%. The added 642nm and 742nm lines mark the points where often used IR-pass filters open. Here the near-infrared response lies at the high end.
Click on the curve to view the near-infrared RELATIVE response of the IMX462 and IMX464 which is truly excellent.
On average monochrome sensors are 10% more sensitive over the entire wavelength domain as compared with their color versions.
Sensors smaller than the 1" class are usually classified as Planetary (and Lunar) Cameras. The latest models housing sensors like the IMX585 or the IMX662 sport a QE response of 90% plus improved full-well capacity and readout noise. In addition come features such as "Amp Glow free", "Passive Cooling" and "Dead Pixel Removal". This qualifies the newest so equipped planetary cameras also for imaging deepsky objects where low noise and high sensitivity are crucial, such EAA (Electronically Assisted Astronomy).
The Analog to Digital Converter, ADC, is an electronic circuit inside the camera outputting a digital representation of the collected photons to the controlling PC via USB. The higher the bit resolution (usually from 10-bit to 16-bit) the more dynamic range (and noise, if present) can be transmitted. The lower the bit resolution, the less computation power is required thus resulting in faster videos (more image frames per second). Though camera dependent, the bit resolution, (also labelled "High Speed ON/OFF") can be selected in capture software.
The data transfer rate, or frame rate measured in frames per second, fps, is typically specified for a sensor's maximum resolution, 8-bit color space and 10-bit ADC selection. The larger the sensor, the more data has to be transferred per frame and the (s)lower the frame rate. For lunar imaging with poor tracking and clouds persistently crossing, a fast frame rate is desirable. When imaging planets only a small area of the sensor is required. The smaller this "region of interest, ROI", the faster the frame rate. For instance, a camera specified with a maximum frame rate of about 50fps can easily work at over 200fps when the ROI is set to, say, 640 x 480 pixels, assuming the camera is connected to an USB 3.x port and SSD for storage.
It lies in the nature of image sensors to produce more noise at higher ambient temperatures in that a cooled camera is warmly recommended for deepsky images with long exposure times, say, over a minute per frame. The noise characteristics of sensors vary while impressive progress has been made, but generally cooling is preferable. The drawback, apart from notably higher cost, is need for an external 12VDC power source. Planetary cameras come without cooling since exposure times are typically a fraction of a second. Also, smaller sensors cool down faster than large sensor during exposure pauses.
Due to absense of a color filter matrix (Bayer), monochrome cameras are more sensitive than their color versions, and provide higher resolution, but require the hassle of three color filters and triple or longer total integration time plus extra post-processing for a color image. It is a matter of preference and budget. Small autoguiding cameras are best in monochrome given their superior sensitivity and resolution both of which are crucial for fixing on guide stars. Monochrome cameras are more expensive since image sensor manufacturers produce more color sensors.
While monochrome sensors receive light photons straightforwardly into their pixel cells thus achieving best possible resolution, color sensors add a filter matrix. The pattern of this matrix is composed of two green, one red and one blue filter. This explains why color sensors provide less resolution than monochrome sensors. Full-color images are generated by an algorithm which interpolates a set of red, green and blue values for each pixel. Algorithms also estimate a value for a given pixel by reading surrounding pixels. Naturally, the algorithm requires computing power.
Pixel [µm] → Size [inch] ↓ |
1.45µm | 2.00µm | 2.40µm | 2.90µm | 3.75µm | 4.60µm | 5.86µm | 9.00µm |
1/3" | IMX224 | |||||||
1/2.8" | IMX715 | IMX290 IMX462 IMX662 | ||||||
1/1.9" | IMX385 | |||||||
1/1.8" | IMX678 | IMX178 | IMX464 IMX664 | |||||
1/1.2" | IMX585 IMX485 | IMX482 IMX174 | ||||||
2/3" | IMX429 | |||||||
1.0" | IMX183 | IMX533 | ||||||
1.1" | IMX432 | |||||||
4/3" | IMX492 | MN34230 | IMX294 | |||||
APS-C | IMX571 | IMX071 | ||||||
Full Frame | IMX455 | IMX410 |
Lunar & planetary & autoguiding cameras, Deepsky cameras
To avoid vignetting the APS-C sensor format requires a telescope with an image circle of at least Ø29mm, the Full Frame format requires Ø43mm minimum.
Click on a sensor part number to view details.
Part Number | Pixel | Resolution [pixels] | Size | QE | Full-well | ADC |
---|---|---|---|---|---|---|
Sony IMX071 | 4.78µm | 4944 x 3284 | 23.6 x 15.6mm | ~50% | 46k e- | 14-bit |
Sony IMX094 | 4.88µm | 7376 x 4928 | 36 x 24mm | ~56% | 52k e- | 14-bit |
Sony IMX128 | 5.97µm | 6032 x 4032 | 36 x 24mm | ~53% | 74k e- | 14-bit |
Sony IMX174 | 5.86µm | 1936 x 1216 | 11.3 x 7.1mm | ~77% | 24.8k e- | 12-bit |
Sony IMX178 | 2.40µm | 3096 x 2080 | 7.4 x 5mm | ~81% | 15k e- | 14-bit |
Sony IMX183 | 2.40µm | 5496 x 3672 | 13.2 x 8.8mm | ~84% | 15k e- | 12-bit |
Sony IMX185 | 3.75µm | 1920 x 1200 | 7.2 x 4.5mm | ~75% | 22.8k e- | 12-bit |
Sony IMX193 | 3.76µm | 6024 x 4024 | 22.6 x 15.2mm | ~74% | 36k e- | 14-bit |
Sony IMX224 | 3.75µm | 1304 x 976 | 4.8 x 3.6mm | ~75% | 19.2k e- | 12-bit |
Sony IMX225 | 3.75µm | 1280 x 960 | 4.8 x 3.6mm | ~75% | 13k e- | 12-bit |
Sony IMX249 | 5.86µm | 1920 x 1200 | 11.3 x 7.1mm | ~77% | 32k e- | 12-bit |
Sony IMX269 | 3.30µm | 5280 x 3956 | 17.3 x 13.1mm | ~84% | 23k e- | 12-bit |
Sony IMX287 | 6.90µm | 720 x 540 | 4.97 x 3.73mm | ~71% | 0k e- | 12-bit |
Sony IMX290 | 2.90µm | 1936 x 1096 | 5.6 x 3.2mm | ~80% | 14.6k e- | 12-bit |
Sony IMX294 | 4.63µm | 4144 x 2822 | 19.1 x 13mm | ~90% | 66k e- | 14-bit |
Sony IMX385 | 3.75µm | 1936 x 1096 | 7.4 x 4.1mm | ~80% | 18.7k e- | 12-bit |
Sony IMX410 | 5.94µm | 6064 x 4040 | 35 x 26mm | ~80% | 100k e- | 14-bit |
Sony IMX428 | 4.50µm | 3208 x 2200 | 14.5 x 9.9mm | ~78% | 20k e- | 12-bit |
Sony IMX429 | 4.50µm | 1944 x 1472 | 8.75 x 6.6mm | ~79% | 25k e- | 12-bit |
Sony IMX432 | 9.00µm | 1608 x 1104 | 14.5 x 9.9mm | ~79% | 97k e- | 12-bit |
Sony IMX455 | 3.76µm | 9568 x 6380 | 36 x 24mm | ~91% | 51.4k e- | 16-bit |
Sony IMX461 | 3.76µm | 11700 x 8896 | 44 x 33mm | ~90% | 50k e- | 16-bit |
Sony IMX462 | 2.90µm | 1936 x 1096 | 5.6 x 3.2mm | ~80% | 14.6k e- | 12-bit |
Sony IMX464 | 2.90µm | 2712 x 1538 | 7.9 x 4.5mm | ~90% | 12k e- | 12-bit |
Sony IMX482 | 5.80µm | 1920 x 1080 | 11.1 x 6.2mm | ~85% | 51.5k e- | 12-bit |
Sony IMX485 | 2.90µm | 3840 x 2160 | 11.1 x 6.2mm | ~85% | 13k e- | 12-bit |
Sony IMX492 | 2.32µm | 8240 x 5628 | 19.1 x 13mm | ~90% | 66k e- | 12-bit |
Sony IMX533 | 3.76µm | 3008 x 3008 | 11.31 x 11.31mm | ~91% | 73k e- | 14-bit |
Sony IMX571 | 3.76µm | 6248 x 4176 | 23.5 x 15.7mm | ~91% | 51k e- | 16-bit |
Sony IMX585 | 2.90µm | 3840 x 2160 | 11.2 x 6.3mm | ~91% | 38.8k e- | 12-bit |
Sony IMX662 | 2.90µm | 1936 x 1100 | 5.6 x 3.2mm | ~91% | 37.8k e- | 12-bit |
Sony IMX664 | 2.90µm | 2704 x 1540 | 7.8 x 4.5mm | ~91% | 38.5k e- | 12-bit |
Sony IMX676 | 2.00µm | 3552 x 3552 | 7.1 x 7.1mm | ~83% | 10.55k e- | 12-bit |
Sony IMX678 | 2.00µm | 3840 x 2160 | 7.7 x 4.3mm | ~83% | 9k e- | 12-bit |
Sony IMX715 | 1.45µm | 3830 x 2160 | 5.6 x 3.2mm | ~80% | 5.7k e- | 12-bit |
Part Number | Pixel | Resolution [pixels] | Size | QE | Full-well | ADC |
QE is approximate and given for monochrome sensors whereever both monochrome and color sensor versions are used in commercial astro-cameras, else QE applies to color sensors. Be sure to check back with the camera vendor of your choice.
In total 35 records.