Are dedicated osc astro cams more sensitive?

Hi,

EDIT: After writing a lot and reading your question a second time, I noticed your DSLR is *not* astromodified. Long story short: a OSC astro camera is more sensitive in Ha (reason see below).

I haven't done a quantitative comparison for DSLR and dedicated OSC CMOS cameras for astro imaging but I'd say: the sensors have the same sensitivity. Of course you'll find differences within DSLRs and within astro cameras, so you could compare the worst DSLR with the best OSC cam but I guess that's not your question. I'm not aware that CMOS sensors are developed with dedication for astro cameras. I checked a few astro cameras and I always found the sensors being used in industrial or consumer applications. In fact, the fact that CMOS cameras are everywhere made them being the now standard for astro cameras. CCDs will be phased out as some companies will stop or have stopped production. Although from a technological point of view, w.r.t to astrophotography, I'd say a CCD is preferable.

I assume you are aware that the astromod of a DSLR is simply due to the fact that they contain an IR cut filter for daylight photography, where nobody cares about only 6% of sensitivity at the Ha line. Another difference is the cooling. DSLRs have the environment temperature or a bit above, depending on much the internal electronics heats the sensor. Therefore, the control of thermal noise isn't there. For dedicated cooled astrocameras the noise level is lowered through cooling and kept at constant values which allows easier image calibration and noise reduction.

In my opinion, switching to mono makes sense once you decide to shoot narrowband. Before you do a mod of your DSLR, go for a dedicated OSC.

CS!

Björn

Thanks for the detailed response!   I guess im wondering if the cost to benefit works out with getting a dedicated osc vs the unmodded dslr in terms of ha sensitivity since they both have similar filters over the sensor to inable that rgb color spectrum.   
a tantalizing feature of mono as i was thinking about it more this morning is i could be collecting data now with the almost full moon out. At least that is my understanding.   A lot to be said for that!

You’ve received some good answers here so only I’ll add a few comments.  First, if you want to stick with a DSLR, the best thing you can do is to have the camera modified.  The difference in sensitivity to signals in the red, including Ha, is significant.  Going from an unmodified to a modified DSLR makes a very significant difference in the results you can achieve.

Having said that, a DSLR is not well suited to narrow band imaging for one big reason.  The Baeyer filter that sits on top of the sensor only allows the light passed by the NB filter to pass through to only one out of four pixels.  With a mono camera, every pixel produces signal.  

Like Arun, I started with an unmodified 7D and then quickly moved to a modified 6D and that camera was fantastic.  Then I moved to a FLI-ML16803 mono sensor and that produced another leap in capability.  High quality mono cameras with relatively large sensors are priced in the same range as a DSLR so it’s a lot less expensive to make the leap but you still have to buy filters and a FW so it’s more pricy to get into mono imaging.  If you aren’t sure about how far you want to go with it, my advice is to get a modified DSLR and stick with RGB.  Get that process dialed in and you’ll learn pretty much everything you need to know to make the jump to mono down the road.

John

Doug Summers:
Where OSC begins to fall apart (rapidly) is in high bortle.   Bortle 4 is borderline.

It is not. I routinely shoot under Bortle 6/7 skies and feel none of the limitation you suggest. Just add a decent LP filter and off you go...

Then I would recommend going for an OSC. It will show you the Ha regions with much more ease than your DSLR. If you gear up one day with a mono, you‘ll have the option to use the OSC to capture the RGB data and the mono to capture additional data (Luminance and/or narrowband).

Hi,

In my opinion the extra sensitivity in Ha is overrated. Certain cameras cut all of it but many others, your Canon 6D included, don't. It is practically the same as an OSC astrocam.  

Now, to answer whether it's worth the money, the pros are:
- Ergonomics: It is designed to connect to a telescope, it is lighter, smaller, sensor is accessible to clean. You can simply swap the IR cut filter with an IR pass if you want to image in near IR and then put the original filter back, it is not a $200 modification.
- Flexibility when controlling the sensor: change gain, offset, binning, bit depth, region of interest at a whim.
- More accurate calibration and less noisy long exposures to begin with (if cooled)
- Truly raw image to work with: no star eaters, no in-camera preprocessing. Although this is not really an issue with Canons. 

And the cons are:
- Probably less capable/lower quality sensor than the DSLR
- Needs a computer
- Needs a power source (if cooled)

Extra sensitivity in red is not really a factor, especially for giant bright nebulae such as the Lagoon or Rosette. If you do not see red with the Rosette, chances are you need to look at your post processing or increase integration time, either way I doubt it is a matter of 30% less sensitivity in red or so. 

Narrowband imaging of faint emission targets is something else and you need a mono camera, filters and quite long exposure times for that. 

Cheers,
Dimitris

In my opinion the extra sensitivity in Ha is overrated. Certain cameras cut all of it but many others, your Canon 6D included, don't. It is practically the same as an OSC astrocam.

I will disagree a little bit here. Canon cameras specifically will cut out about 75% of the H-alpha wavelength, letting in 25%. That certainly was the case with my 5D Mark IV. This difference matters. If you take an image of the Eagle nebula or even the Horsehead, an astromodded DSLR or a dedicated astro OSC will capture a great deal more of the H-alpha detail. So too, if you take an image of the Ophiuchus region. The red is really hard to get with an unmodded DSLR which only lets in 25% of the H-alpha light.

and this speaks to my main concern, if the osc dedicated astro cam is only marginally more sensitive to Ha vs unmoded 6d then in just those terms it probably isn't worth purchasing the osc cam.  but if considering much improved noise performance of osc then ya, if you have lots of funds available it is a good choice.  i wish the manuf would publish that specification of Ha sensitivity for their osc.

For the OSCs the data is usually available. Check the website of ZWO for example. You‘ll find the peak quantum efficiency and also the relative response for each channel as a function of the wavelength. 

For DSLRs it seems more difficult but some web research should give you some information, e.g.
ResearchGate.

For this specific DSLR in the article the Ha sensitivity is roughly 5%.
Compare this to the relative response provided for the 294MC.
Roughly speaking, the DSLR is „blind“ to Ha if compared to the OSC.
Yet, one can still collect data, e.g. my horse head was captured with an unmodded EOS250D. It’s more difficult but still possible. An OSC would do that with ease.

NGC2024, Flame Nebula and Horse Head Nebula

CS,
Björn

Tom Marsala:
With the OSC's you get less noise if they are TEC cooled, but there are ways around the noise with the uncooled 6D; you just have to get creative!

Nowadays CMOS sensors have very low thermal noise even up to room temperature compared to older sensors or to CCD sensors. 
I recommend people trying to find the thermal noise data of their sensors before they cool it to -15 deg C. The major advantage of TEC cooling is the constant sensor temperature which makes generating dark frames much easier. 
To talk about some numbers: the IMX183 sensor in my camera has about 0.01 electrons/second of dark current at around 10C. That means to get in the order of read out noise, I need to expose for more than 300 seconds.

Agreed,  Roger.  In a similar way, I got more color data on a 10 second sub of the Ring Nebula with the modified 6D than I did with a 3-hr integration when it was unmodified! Pretty crazy!

Like I said in my original message there is a difference between using a sensor without the relatively restrictive IR cut window normally integrated in every general purpose camera and using a sensor with a Ha filter which is the recommended way to image Ha targets. You do get a lot of red when the sensor is without the filter but it is not just Ha, it is the entire spectrum of red things above 600nm or so (where the stock window starts cutting) plus a great deal of near infrared.

Will things look spectacularly redder? Absolutely.

Will that be Ha detail? Extremely doubtful.

Ha is but a tiny line, less than a percent of the extra red that you are now more sensitive to. The stuff that is 1nm to the right of Ha in the spectrum for instance and the stuff 1 nm to the right of that and pretty much everything else all the way up to 700nm? It is guaranteed that the camera's sensitivity to those increased more than it did to Ha. And they all look red.

I have taken two pictures of the Lagoon. One is with a stock DSLR:

https://www.astrobin.com/413954/?nc=user

The other is 4 hours of OSC to which I added 4 hours of L with a mono sensor. Both cameras have better response to red than the D7500 because the ZWO IR filters cut at 700nm, almost 50nm above the Ha line. 4 times better in the mono sensor's case.

https://www.astrobin.com/9y4s1j/?nc=user

Ignore the fact that the detail is better (different scopes,  different integration times, guiding). The thing is the clear separation between the blueish core of the Lagoon and the pink periphery which was captured perfectly by the stock DSLR, is almost completely gone in the second picture where all colours are shifted towards red and that was after very painful post processing to make the parts that are supposed to be blue appear at least a little purple. I was actually quite disappointed at the second picture from a chromatic point of view. 

if you remove the effect of the IR cut window then the difference between OSC and DSLR is a matter of chance, because OSC astrocam sensors do not have any magic properties. They are the same sensors that you see in mobile phones, security cameras, action cams and of course DSLR and mirrorless general purpose cameras. For example Sony A7 R IV and ZWO ASI 6200MC use the same sensor (IMX455) and so do ASI 2600MC and Pentax K-3 mark III (IMX 571).

Assuming the sensors have similar tech (i.e. are not separated by a decade) it is the window and only the window that has a serious impact on spectral response. And there is a good reason stock DSLRs have that window, namely it makes it easier to obtain a natural (to the human eye) color balance. 

Cheers,

Dimitris

Scott Badger:
is a a modified dslr no longer workable as a regular

You can use your modified DSLR for landscape by using a custom white balance.  This is done by shooting a white or gray card in even sunlight and saving the white balance as custom.  It works very well.