Mono or OSC, your suggestions

[Dr_Ju_ju]

…… leading on from the colour transmission system, the starting point of the signal, i.e. the television camera, especially in the early days, was based around a 3 monochrome camera system, i.e. RGB, where each camera being fed by the appropriate colour light, that had been split by a dichroic mirror, into its RGB components....  

now there's a thought for a full bandwidth 3 way system, an OCS with 3 mono cameras....

[vlaiv]

I think next simple advancement would be OSC camera with slightly different bayer matrix, instead of RGGB, one can imagine something like LRGB - one pixel being full spectrum instead of having filter.

[gorann]

23 minutes ago, vlaiv said:

I think next simple advancement would be OSC camera with slightly different bayer matrix, instead of RGGB, one can imagine something like LRGB - one pixel being full spectrum instead of having filter.

Great idea but I doubt our small amateur AP-camera manufacturers would ever talk Sony or Panasonic into making such a chip. The LRG (or LRB or LGB) suggestion by Stub Mandrel with calculated third colour is also very interesting. Only problem is that it would not save much time (at most 1/4th) but it could be a way out for someone lacking one colour after the clouds moved in one night (as they often do). Strategy should then be to shoot Lum and then one colour at a time and be happy to get at least two.

This thread is really interesting and creative!

[Stub Mandrel]

11 hours ago, Dr_Ju_ju said:

…… leading on from the colour transmission system, the starting point of the signal, i.e. the television camera, especially in the early days, was based around a 3 monochrome camera system, i.e. RGB, where each camera being fed by the appropriate colour light, that had been split by a dichroic mirror, into its RGB components....  

now there's a thought for a full bandwidth 3 way system, an OCS with 3 mono cameras....

Not diffcult to arrange, the splitter prisms come up as surplus items regularly. With stacking you don't even need the precise alignment of all three sensors, as long as they are focused.

[Dr_Ju_ju]

Yea, I know, I have a few from my Beeb days.... its just another thing on my to-do list...

[alexbb]

12 hours ago, vlaiv said:

I think next simple advancement would be OSC camera with slightly different bayer matrix, instead of RGGB, one can imagine something like LRGB - one pixel being full spectrum instead of having filter.

There's already some work done for sensors in this area: https://www.anandtech.com/show/7149/aptina-announces-ar1331cp-13-mp-cmos-with-clarity-plus-we-take-a-look

Having one on an astronomical camera? Perhaps not so soon.

[vlaiv]

7 minutes ago, alexbb said:

There's already some work done for sensors in this area: https://www.anandtech.com/show/7149/aptina-announces-ar1331cp-13-mp-cmos-with-clarity-plus-we-take-a-look

Having one on an astronomical camera? Perhaps not so soon.

That is actually quite clever as it combines above approach with color subtraction. Two clear pixels and red and blue one! It looks like best option for OSC astronomy. This sort of sensor will be a bonus for EEVA as well.

[vlaiv]

Just seen that quoted article is from 2013 - if they wanted to market such sensor, they would have done it by now. I don't think it's going to happen any time soon.

[andrew s]

Clearly the right way to go is two dichroic mirrors separating red and blue with 3 mono cameras one for each beam ( red and blue side arms green straight through) with colour optimised field flateners / correctors for each beam.

I seem to recall some  upmarket Sony video cameras used such an arrangement.

Regards Andrew 

Could always add in adaptive optics for fun.

[vlaiv]

Not sure why would we go thru the trouble of splitting beam when we can have multi scope setup, each having its own sensor and filter.

Quad scope, each equipped with L, R, G, and B filter respectively, aimed on the same part of the sky would provide further benefits - like increased total aperture (or if you want to look at it like that - capturing LRGB at the same time at full resolution and full aperture of single scope).

[andrew s]

12 minutes ago, vlaiv said:

Not sure why would we go thru the trouble of splitting beam when we can have multi scope setup, each having its own sensor and filter.

Quad scope, each equipped with L, R, G, and B filter respectively, aimed on the same part of the sky would provide further benefits - like increased total aperture (or if you want to look at it like that - capturing LRGB at the same time at full resolution and full aperture of single scope).

Cost, size, weight. It could be a retro fit single imaging package i.e. integrated imaging if all 3 cameras controlled as one with one image file. 

The technology exists in current Sony TV cameras.

Regards Andrew 

[davew]

14 hours ago, vlaiv said:

I think next simple advancement would be OSC camera with slightly different bayer matrix, instead of RGGB, one can imagine something like LRGB - one pixel being full spectrum instead of having filter.

These ideas come and go. Some from quite a long way back too.

https://www.dpreview.com/articles/8732455081/kodakhighsens

Dave.

[smr]

With regards to a modified DSLR vs my unmodified Canon 80D...

If i were to buy a Canon 600D modified for instance, would I be giving anything up in terms of quality?

My Canon 80D sensor is quite a bit newer than the 600D sensor.

The 80D has on chip ADC and is purported to be ISO-less meaning that there is no benefit in shooting higher than ISO 200.

With the 600D I think I would have to shoot at ISO 800 or 1600.

Would this be a negative thing compared to the ISO-less sensor on the 80D in terms of noise?

Obviously the advantage of the modified DSLR is that it is a lot more sensitive to Ha and I could have a go at using Ha filters to combine with broadband subs.

I could get a Canon 600D modified for about £250. And that would mean I am not straining my 80D (with long exposures) which I also use for landscape photography.

If I do go for a dedicated OSC I think it would be the well enamoured ASI 294MC Pro which is £1000. 

The biggest difference between the 294 and a modified 600D would be just the cooling and noise reduction / proper calibration or is there more to it than that?

 

Edited August 23, 2019 by smr

[vlaiv]

1 hour ago, smr said:

The biggest difference between the 294 and a modified 600D would be just the cooling and noise reduction / proper calibration or is there more to it than that?

Quite a bit of difference, lets just list relevant stats so you can compare them:

600D at ISO800 (close to unity):

pixel size: 4.3um
peak QE: 41%
read noise: 3.2e
dark current : ~0.4e/px/s (at about 33C sensor temperature)

ASI294 at Gain 120 (close to unity):

pixel size: 4.63um
peak QE: ~75% (estimated by ZWO)
read noise:  ~1.8e
dark current: 0.0065e/px/s (at -15C)

So, larger pixels (not necessarily what you might need but means higher sensitivity and lower resolution on same focal length), higher QE (more than 80% higher) and lower read noise (almost twice). Dark current is also quite high with 600D - let's take 200 seconds exposure and see how much noise is there from dark current. Accumulated dark current in 200s exposure will be about 80e, and square root of that will be ~8.95e - that is almost three times higher than read noise, so it's not negligible at all. Dark current and read noise combined give 9.5e of noise vs ~2.13e noise with ASI294 (1.8e from read noise and 1.14e dark current noise from 200s exposure at -15C).

600D will have more active pixels due to smaller pixel size (both are APS-C type sensors) - 5184 x 3456 vs 4144 x 2822.

 

[Stub Mandrel]

3 hours ago, smr said:

And that would mean I am not straining my 80D (with long exposures)

I don't think long exposures cause any 'strain', its sheer number of mechanical actuations that matters. Longer exposures = fewer exposures.

[swanny]

I use OSC for a few reasons:

1.  I am time limited.  I don't have the time to put 30-40 hours into one subject.  10-15 yes.  

2.  CMOS is what the future is going to be.  CCD is a dying technology.  So I went that way rather than have to switch things up later.  There are mono CMOS I believe too.

3.  I am willing to sacrifice a little bit of 'quality' for the simplification and time savings

4.  I use the QHY367 on my fsq106 and it fits the image circle great.  Not many CCD, if any, would economically do that like the 367.

5.  I do Hubble with this camera and they look very good.  

If I were retired and had all the time in the world, or had a job where I was home every night, I might do the mono.  But the image quality with this camera is very good so I don't see the need to worry about mono at the present time.  

[ollypenrice]

33 minutes ago, swanny said:

I use OSC for a few reasons:

1.  I am time limited.  I don't have the time to put 30-40 hours into one subject.  10-15 yes.  

 

You can't shoot luminance so how can OSC be faster? You can never shoot more than about 1/3 visible spectrum per pixel. A mono shoots all the visible spectrum per pixel in luminance mode.

Olly

 

[ollypenrice]

21 hours ago, Stub Mandrel said:

Surely you already have LRGB data to play with? Can you use 'pixel math' in Pixinsight to do the work?

I wouldn't use pixel math, I'd just use 'Subtract' in Astro Art, but when I get a sec I'll give it a go out of curiosity.

Olly

Edit: After a moment's thought I realize it's not that simple since I don't shoot L and RGB at the same exposure or, usually, in the same camera. A new set of captures would indeed be needed. On the other hand, using the data I do have, I could extract a 'subtracted from luminance' version of each colour and combine these as RGB to see what they look like. My guess is that they'll look like a mess. We are going to need a long spell of monsoon rain before I get round to trying this! 

Edited August 23, 2019 by ollypenrice

[andrew s]

20 hours ago, Dr_Ju_ju said:

…… leading on from the colour transmission system, the starting point of the signal, i.e. the television camera, especially in the early days, was based around a 3 monochrome camera system, i.e. RGB, where each camera being fed by the appropriate colour light, that had been split by a dichroic mirror, into its RGB components....  

now there's a thought for a full bandwidth 3 way system, an OCS with 3 mono cameras....

Missed your post before proposing the same thing. Still very much in use.

Regards Andrew 

[swanny]

33 minutes ago, ollypenrice said:

You can't shoot luminance so how can OSC be faster? You can never shoot more than about 1/3 visible spectrum per pixel. A mono shoots all the visible spectrum per pixel in luminance mode.

Olly

 

I assume you shoot RGBL?  So you are saying you only do 1.25MN subs?  Because I do 3-5MN on my camera for color frames.  Different CMOS sensors have different sensitivity as well so they are all not the same.  

[Whistlin Bob]

6 hours ago, smr said:

I could get a Canon 600D modified for about £250. And that would mean I am not straining my 80D (with long exposures) which I also use for landscape photography.

If I do go for a dedicated OSC I think it would be the well enamoured ASI 294MC Pro which is £1000. 

 

I think @vlaiv has given a very comprehensive answer to this, but to put it in a real world context I use a modded 600d in f4.5 reflectors (130 and 200) in Bortle 5 skies, and for many situations it performs very well. I'm not disputing that a 294 would be significantly better, but it's 4x the price. For that money I have the camera, both scopes (1 was 2nd hand), a coma corrector, an IDAS D2 LP filter, an Ha filter and half a Pixinsight license. I think that's a good deal!

 

 

[newbie alert]

Mono...end of the story...

Enjoy....

[Alien 13]

I cant imagine anything worse than faffing around with filter wheels and filters, besides mono cant do meteors or comets unless yo want them well in mono...

Alan

[Tim]

What telescope were you looking to use it with btw?

 

[smr]

1 hour ago, Tim said:

What telescope were you looking to use it with btw?

 

Me? 

Zenithstar 73 apo refractor, 430mm.