New type of dualband filter: OIII+SII

[Lee_P]

Here's something interesting: Askar are releasing a new type of dualband filter. It's OIII+SII, designed to be used in conjunction with the regular H-a+OIII. Askar are sending me one, so I'll give it a good test and will report back in the Members' Reviews section.

 

[The Admiral]

Makes perfect sense.

Ian

[vlaiv]

That poster is a bit misleading.

OSC + dual narrowband filters will never be able to do proper separation of signal like mono + regular NB filters.

This has nothing to do with dual narrowband filters - but rather with OSC sensor themselves.

If you look at QE chart for any OSC sensor - you will see that every pixel has sensitivity all over 400-700nm range.

here is QE graph of ASI2600.

Whenever you capture Ha or SII signal - it will be picked up by both green and blue pixels. If you are imaging OIII at the same time (which is similarly picked up by red pixels) - three will be some "crosstalk".

Some of Ha/SII signal will be imprinted in OIII signal and vice verse. You won't be able to record pure Ha/SII or pure OIII.

On the other hand - there is nothing stopping one from using regular NB filters with OSC camera as well. With careful processing one can even get better than often assumed sensitivities - 1/4 for Ha/SII and 1/2 for OIII - precisely because there is this "crosstalk" and all the pixels are sensitive to all the wavelengths (although not with the same QE).

Just to be clear - I'm not saying that dual band filters are not good (or tri/quad for that matter) - I'm just saying that comparing them to Mono+NB or dedicating them exclusively to use with OSC sensors is misleading (why not use dual band with mono as well?).

[The Admiral]

9 minutes ago, vlaiv said:

why not use dual band with mono as well?

Presumably because you won't be able to separate the signals at all with mono? I hear what you are saying though, there will be some degree of inability to separate signals on a OSC.

Ian

[Lee_P]

I think for me the interesting thing about the filter is that it moves away from what I consider to be the benefits of OSC -- namely simplicity of data acquisition, and avoiding having to buy lots of filters. But if the end results are good that it could be an appealing halfway-house between OSC and Mono in those regards. I'm really curious to see how well it works under real-life conditions.

[The Admiral]

Two things are going through my mind. The first is the need to expose though 2 filters, which is only 1 away from doing 3 filters with a mono camera, with its associated sensor efficiency benefits. So using mono becomes a more realistic option. Secondly, I wonder if the algorithm used to extract the separate channels can, to a first approximation, allow for the overlap using the published spectral response? But now we are adding complexity.

Personally, I'm all for producing an attractive image rather than one with scientific accuracy, so if you find that using the the Askar set can come up with the goods, then great! I look forward to seeing what you get.

Ian

Edited October 28, 2022 by The Admiral
Correct ASIAir to Askar

[Lee_P]

I thought that Askar were the first to try this but I've just seen that IDAS have already released an OIII+SII dualband filter -- the NB3, although I gather it's been discontinued. I wonder why...

[vlaiv]

9 minutes ago, The Admiral said:

Secondly, I wonder if the algorithm used to extract the separate channels can, to a first approximation, allow for the overlap using the published spectral response? But now we are adding complexity.

In principle - yes.

If we for the moment ignore that there is actual width to each of those bands and that they response is not uniform across these bands (not constant but rather more like bell shaped) - we actually have very simple system of equations.

Red_pixel = SII_signal * SII_QE_red + OIII_signal * OIII_QE_red

Green_pixel = SII_signal * SII_QE_green + OIII_signal * OIII_QE_green

We have pixel values and all QE values - so we can reconstruct SII and OIII signal that are unknowns in above equations.

In fact - we will have 3 equations with 2 unknowns.

In reality - things are not that simple as we have LP that is unknown and also - we don't have uniform response at each of these wavelengths (or rather wavelength ranges that filters pass), so above will be just approximation.

[licho52]

The IDAS wasn't too viable as it had a very wide bandpass, to the tune of 30nm IIRC.

I use both 2600MM and MC and I know very well what OSC is capable of, as well as its limitations in narrowband domain.  This should be a very good product in my opinion, and long overdue - it will however be very confusing to people who process these straight out of debayer without channel separation.  I expect some hilarity in "beginner" sections of AP internet.

[900SL]

On 27/10/2022 at 13:42, Lee_P said:

Here's something interesting: Askar are releasing a new type of dualband filter. It's OIII+SII, designed to be used in conjunction with the regular H-a+OIII. Askar are sending me one, so I'll give it a good test and will report back in the Members' Reviews section.

 

Looking forward to seeing your results. I have the IDAS NBZ dual band Ha Oiii OSC filter and am thinking of getting the Askar Oiii Sii as well. I haven't had great experiences with Askar QA on a scope so looking forward to any third party reviews 

[Lee_P]

9 minutes ago, 900SL said:

Looking forward to seeing your results. I have the IDAS NBZ dual band Ha Oiii OSC filter and am thinking of getting the Askar Oiii Sii as well. I haven't had great experiences with Askar QA on a scope so looking forward to any third party reviews 

I just need a few more clear nights and then I'll have enough data to produce two images using these Askar filters. I'll post a link to my tests in the Members' Reviews section when it's all done

 

[900SL]

On 01/12/2022 at 11:53, Lee_P said:

I just need a few more clear nights and then I'll have enough data to produce two images using these Askar filters. I'll post a link to my tests in the Members' Reviews section when it's all done

 

Any update on this filter Lee?

 

[Lee_P]

1 hour ago, 900SL said:

Any update on this filter Lee?

 

Yep, the review is all written but Askar have asked to see it prior to publication, as a courtesy to them. They said they hope to read it on 28 Dec - tomorrow! Then I'll be able to share the review.