I am in the market for a new camera for astronomical photometry and the vendor is pushing me very hard to buy a CMOS camera instead of a CCD camera. A few questions. Please feel free to address one or more of the below questions. I am totally new to CMOS cameras so anything you contribute would be valuable information to me. Thanks.
1) Would a CMOS camera be able to do absolute and differential photometry as well as a CCD?
2) What is the process to calibrate a CMOS image vs. calibrating a CCD image (darks, bias, flats, etc.)?
3) Is a "filter wheel" necessary with a CMOS camera?
3) do the Johnson-Cousins filters or the SDSS filters mean anything for CMOS images? If not, what is used instead of these filters?
4) What is the best CMOS camera available for photometry?
I won't get to all of your questions, but can address some:
1: AAVSONET has a number of instruments available worldwide to do photometry. A number of these scopes have CMOS imagers, and several are in the process of being upgraded to larger CMOS imagers. That should answer whether you can use either. See https://www.aavso.org/aavsonet
2: Yep, pretty much the same.
3: If you wish to do BVRI type photometry, yes. That implies a mono sensor. Some DSLR and color CMOS users avoid that and submit images from various color channels.
4: Check out https://www.aavso.org/observing-manuals for these details.
5: the best one is the one you have. I have long found that "perfect" is the enemy of "the good enough". And everyone's situation is somewhat unique.
I hope some of that helps!
Peter
BPEC
Thank you Peter. For someone like me who knows almost nothing about CMOS, the information you provided is extremely important.
Clear skies,
Ed
"What is the…
Regarding:
"What is the process to calibrate a CMOS image vs. calibrating a CCD image (darks, bias, flats, etc.)?"
It is similar, but from my experience and reading you cannot scale dark frames of CMOS cameras as some of the "signal" that is in the dark frame does not increase in a linear fashion with exposure. That means you need to maintain a library of dark frames at your standard exposures and use the appropriate one. Having said that, for short exposures (say less than a minute) the dark signal is so small (as many CMOS cameras are so "quiet") that there is no practical difference between, for example, 1 and 30 seconds.
Thanks LDJ, That is very important to know.
Ed
I didn't see any further…
ED,
I didn't see any further response to your question on calibration frames; if you already gotten the info you need, just ignore this.
I can't speak to the actual post processing after stacking to prepare the VS data as I'm a newbie. But I've spent a lot of time getting over the learning curve with OSC CMOS cameras for the last two years and can tell you what needed to properly calibrate your CMOS LIGHT images.
LDJ is correct about the extreme low ev noise of CMOS sensors. However, each camera/sensor has a sweet spot in gain that gives the greatest well depth vs EV noise level. You sometimes read /hear this referenced in on line discussions as the unity gain.
I can give you specific examples of my two ZWO OSC cameras.
The first one I had was the ASI294MC-P - its unity gain was at 120 with an offset (or sometimes referred to as pedestal) of 20. My current OSC is an ASI2600MC-P which has a unity gain of 100 and an offset of 50. The offset/pedestal is to give your image some 'elevation' above the dark level to help prevent clipping good data during post processing.
You need three types of files to have a good calibration file setup for CMOS sensors (and it is slightly different than CCD due to the nature of the sensor).
DARKS: Must match your LIGHT frame settings in all parameters to be effective: Exposure time, sensor temperature, Gain, and Offset. The number required - more is better, but I usually keep a minimum of 15 on file for each of my 'standard' OSC setups (it is the exposure time that I may change for specific targets). This establishes the dark level for the sensor for your post processing software. Sounds easy, but light leaks in your imaging train can cause real problems. As LDJ indicated, most of the good quality CMOS cameras have EV' noise at or below 1ev at unity gain so for the short exposures that seem typical of VS imaging the DARKS are going to be very dark assuming no light leaks.
FLATS are next and they must match the LIGHTS in GAIN, Sensor Temp, and Offset and FILTER. The goal here is to achieve an exposure approximating the 50% ADU level for your camera sensor at the imaging settings used for the LIGHT frames (excluding exposure time). From my reading, the guidance is that if you must err, err on the low side of the 50% ADU level. The purpose of the FLATS are to provide a reference of the entire imaging train of dust, smudges, etc and sensor gradient so that the stacking software can calibrate that out of the image frames. The recommended exposure time is less than 3 secs. I've gotten my best/most consistent flats using a doubled white T-shirt stretched over the front of the OTA near dusk (pointing to a clear point in the sky for even illumination). I've tried light panels, even made my own from an electroluminescent sheet to get the 'perfect' even light, but the double T-shirt still seems to provide the best result.
DARK FLATS are a CMOS sensor requirement, and they must match the FLATS in every particular, except they are taken with a covered/dark OTA. If you have a CMOS sensor with lamp glow, this is the 'secret sauce' that lets the stacking software calibrate the lamp glow out of the stacked image. The ASI294MC-P is rather notorious for the lamp glow in its sensor, while the ASI2600MC-P is backlighted and does not have any lamp glow.
FLAT and DARK FLATS don't take long, so the more is better mantra applies here as well. I typically shoot at least 20 of each. My imaging management software is N.I.N.A which has almost as many features/capabilities as Voyager, but is freeware. One of the things N.I.N.A. has that I use is a FLATS Wizard - it will automatically create a matched set of FLAT/DARKFLATS which further eases the entire process.
As an aside, you'll see BIAS frames mentioned but those do not apply to CMOS sensors. BIAS frames are specific to CCD sensors which (typically) don't have shutters, so they can get a true zero exposure time dark frame, which is essentially what a BIAS frame is. CMOS sensors are physically not capable of shooting a true BIAS frame since they typically have some form of shutter, built into the pixel sensors. You'll see this referred to as a rolling shutter by some makers. CMOS sensors do have some very short exposure times, but they most definitely do NOT have a zero time exposure.