Camera Labs

Hopefully my converted camera isn't too far off. As written elsewhere, one of the lenses I'm planning on using I found has enough longitudinal CA that it requires separate channel imaging for best quality. And if I'm going to do that, I might as well go narrowband.

For starters, I think the H-alpha is a given since that covers a lot of "stuff" out there. Where I'm less clued up on, is what gives off what else? Is there a dummies guide to what objects give what spectrum?

I'm also debating if it's worth jumping into OIII, SII or others or should I stick to H-alpha+G+B for starters. I'm guessing for serious narrowband imaging I'd probably best ditch the SLR anyway, and so the cost goes ever upwards... sky's the limit!

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

As you know I'm just getting into this myself so my thoughts are hardly authoritative. I've just ordered my own "starter kit" of filters comprising a set of Astrodon Tru-Balance LRGB filters and a 3nm H-alpha filter, those being all the bank balance will stand at the moment because I need 50mm square ones!!!

But as you are working with a DSLR sensor why do you need RGB filters at all? I'd have thought a single IDAS light pollution filter would be all you'd need for RGB imaging as the camera's Bayer matrix will do the rest.

Will you be using something like Atronomik's clip filters for your narrowband imaging?

Check out Astrodon's narrowband FAQ page for some hints about which filters to get.

Oh, and if you focus to get the sharpest reds then Noel Carboni's Astronomy Tools has a PhotoShop action which helps reduce the effect of blue/violet halos around stars which works quite well.

Bob

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

On the filtering, I guess I should clarify. My intention was to take two types of images first. One is plain RGB through the astronomik CLS filter. The other is H-alpha, and drop that into the R channel. I wasn't looking at dedicated R+G+B filters.

3nm? I was looking at the astronomik filters last night and wondering what the practical difference between 12nm and 6nm versions where, other than the price difference! Also their clip filters are not rated for high speeds, varying by specific type, but even f/2.8 is pushing some of their design. I did eliminate the filter as a factor to the optical effects I was seeing last night though. I guess a lens front filter might do better, but do you know anyone with affordable 78mm circular filters? And would that have a higher risk of unwanted reflections?

The FAQ page in particular is making interesting reading although I'll have to finish looking at it at work.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

You are right, 3nm filters only make sense if the f-ratio isn't too fast. As I'll be operating at f/7 with the 'scope I'll have no issues but I might see some effects with my intended use of a Pentax MF lens which can go to f/2.8. If I do I'll just drop back to f/4 and double the exposure times. Astrodon make a good case for use of 3nm filters in their own right (they would, they sell 'em!) in terms of signal to noise but for the future I have a particular need in that I want to exclude as much as possible from my Hα images of the adjacent NII emission. If it wasn't for that I'd have opted for something in the 5 to 6nm range to save a few pennies.

As an aside, I did a few calculations a while back and it turns out that 3nm is still wide enough to avoid losing the signal because of Doppler shift for pretty much every target nebula in the Milky Way and for nearby galaxies.

I bought a 72mm IDAS light pollution filter (link in my previous post) a while back and my unscientific impression is that it performs better than a similar filter I used to use mounted directly to the body of my old 40D. My guess is that the IDAS filter mounted behind the lens wasn't particularly happy with the light cone from my 85mm lens wide open at f/1.2. At the time I bought the 72mm filter the EF 135mm f/2.0L USM was still on my radar and the expense seemed worthwhile as I could have used it on both that and my EF 85mm f/1.2L II USM. I've not tested other LPS filters but from what I've read on other forums the IDAS ones work best.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

I'm hesitant to get a different light pollution filter just yet, although strictly once I get my full spectrum modified camera back the filter I have isn't the ideal one. It passes IR which wont be blocked on the new camera so that may cause issues. Would using a light pollution filter help any more over just using a narrowband filter?

If I were to go lens front filters, it would be the narrowband filters I'd like to get for that and I haven't seen any yet. Then again, based on what even 2 inch eyepiece filters cost, I don't think I would want to pay for a 72/78mm one!

I hadn't even considered if any Doppler effects would be significant. You've definitely been taking this far beyond my current thinking level.

So back to my original thoughts, I'm still wondering if H-alpha + G&B from bayer would be a good enough starting point, or if it is worth jumping into more narrowband filters like SII and OIII.

On a related note, are there special de-bayer considerations that can be made when dealing with narrowband imaging? Since you're dealing with nominally a single wavelength, then could you not extract data from more points than in conventional visible colour photography? I'd have to look up the colour filter spectral characteristics before I continue this thought further.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

The only answer to the above I'm comfortable giving is that if you are using a narrowband filter then a light pollution suppression filter doesn't add anything to the party and is just a superfluous extra optical element. I'm not aware of any round narrowband filters of diameter greater than 50mm. Indeed, for the cost of one of those if they did exist I think it would be cheaper to buy a small dedicated CCD camera that can accept 1 1/4" filters!!!

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

If you want better result, you would need to remove the IR/Cut filter that is installed on your DSLR Camera. That would be my first upgrade.

_________________
Canon T2i (550D)- Tamron 17-50 F2.8 Non-VC, Sigma 70-200mm F2.8 EX DG OS HSM, Canon EF-S 55-250mm F4.5-F5.6
http://www.astronomyphotography.com

Given the non-trivial cost and relatively niche application of narrowband filters, I think I'll just get a H-alpha and see where I go from there.

On combining narrowband with general light pollution filters, I guess that's day job me speaking. Filters are not perfect. While the manufacturers show nice charts with 0 to 100%, I doubt it is actually 0% in the filtered zone, the question then is to what degree of attenuation are we looking at? The question then be will it remove even more unwanted light than any further loss in wanted light. This is speaking from a light polluted area point of view. If you don't have much unwanted light to get rid of, that would alter the balance a bit.

Symtex, you're a few threads out of date, but I did mention again in this thread I have a camera out being converted to full spectrum right now. Should be back in less than a week now.

Elsewhere: today I've been poking around my broken polar finder scope again. It has now been dropped twice. I've found and corrected the main previous issue, which is the optical axis being non-perpendicular to the mount. Next step is to get the infinity focus set then adjust the alignment screen, and it'll almost be good as new. Now with dust inside at no extra cost!

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

I've been meaning to see what I could find out about combining LPS and Hα filters but I've been a bit pre-occupied setting up my new smart-phone over the last few days!

I'll be interested to see whether you get any improvement in S/N ratio by adding an LPS filter in front of (or behind) a Hα filter. It seems quite difficult to get figures on out of band rejection for narrowband filters but Christian Buil has come to the rescue yet again. His page Astronomical Filter Curves contains a huge amount of information though it's still difficult to see just how good the rejection is out of band for most of the filters he measured.

There might indeed be a theoretical advantage in combining filters but a very quick Google search didn't throw up any significant results. For example, if one Hα filter is good you'd think two Hα filters would be twice as good, even if the second one was a cheaper 12nm version, but using that setup doesn't seem to be common. In asking myself why this should be the only answer I can come up with is that the second filter doesn't significantly improve the S/N ratio, not because it doesn't cut out more of the unwanted photons but because other noise sources within the sensor are already of a similar magnitude to the residual flux out of band using a single filter.

But hey, I'm rambling a bit. As I say, I'll be interested to see whether you get any improvement after optimal post-processing if you add an LPS filter to a Hα one.

Bob.

P.S. I'd assume that for such a dual filter setup you'd have to make sure the surfaces of the filters aren't in physical contact.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Thanks for the link, that's kinda answered some other questions I had in my mind regarding astronomik filter characteristics beyond their datasheet statements. I will bookmark that site as I remember it from before but can never find it when I need it...

On combining light pollution and narrowband filters, I'm not likely to get to test it anyway. The filters I would get will be EOS clip so only one can be fitted at a time. Lens front filters, if I can even find them, would be rather astronomical in cost...

On a related note, I have been bouncing the odd e-mail with astronomik on the 12nm vs 6nm question, and the recommendation seems to be go for the 12nm if I want to later use their SII and OIII for image reconstruction also. Reasoning was along the lines that the 6nm would only reduce star brightness and wouldn't significantly help with light pollution.

On the plus side, I questioned the use of fast optics with their filters, since the website typically states an optimal range for f/3.x with a usable range to f/2.8. They are usable to f/2 which is nice, and this will be updated on the website in a refresh some time.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

Some interesting information from Astronomik there. Not totally sure I buy into their 12nm recommendation though. From this Astrodon page (and yes, they have a vested interest!):

It is often asked whether the benefits of the narrower 3 nm filters are worth the extra cost over wider filters.
Besides minimizing the effect of moonlight and light pollution, a primary goal for selecting the narrower filter
is to increase the contrast in the shortest possible imaging time.

They have a link at the bottom of that page to this very nice Flash illustration comparing 6 and 3nm OIII images.

Speaking with all the (non) authority of one who has yet to expose a single narrowband photograph I think a major reason for opting for a narrower bandwidth is precisely to cut down on the star brightness. My own experiments in another thread have convinced me of the huge benefit during post-processing of being able to derive "just stars" and "no stars" versions from the same data allowing optimisation of each and then combining the two towards the end of the process. The less stars and the less bright those stars are in the narrowband image the easier it should be to do a really good job of removing the stars that are present and, of course, one can take longer exposures as star bloat should be reduced. I'm not sure yet where my own path will take me but it's even possible that I might not re-introduce the Hα "just stars" version back into the final image if I've also taken RGB images (the Hα and RGB lights would have to be registered before star removal in the Hα ones).

Bob.

P.S. I suppose I should also declare a vested interest here as I've already ordered a 3nm Hα filter.

P.P.S. Enough of this armchair philosophising. I think a cup of tea is called for, together with a spot of P.G. Wodehouse (book) followed by an evening of American football. I'll let the other Mods keep an eye on things for the rest of the day.

Computer "OFF"

Oh, I forgot: I don't have voice command installed yet...

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Star size/brightness aside, the argument for using a set of similar bandwidth filters was ease of direct recombination at the end. Star subtraction is a level beyond anything I'm considering at the moment.

I'm don't think astronomik have any financial interest in promoting the 12nm over their own 6nm, since the 6nm is more expensive anyway.

I've also verbally directed various computer to go "off" before too. Usually with more colourful prefixes.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.