12 replies to this topic

[rmorein]

Many light pollution filters for imaging pass ir well into the  airglow region. This has little effect on cmos/ccd cameras, which have negligible response in this region. This extended passband, in combination with the accentuated long wave response of GaAs cathodes, could contribute to the difficulty of seeing galaxies.

 

Are there available filters that pass some ir, enough to enhance, without extending into the airglow region?

[TOMDEY]

I find that the GaAs responsivity is wonderful for globular clusters and galaxies. The red and NIR cut through haze like a knife and also penetrate the dust lanes in galaxies to see more of ~what's inside~. This is especially enhancing for the most remote "little globulars".    Tom

Edited by TOMDEY, 14 September 2024 - 09:23 PM.

[sixela]

Some of us use an Astronomik L1 filter for galaxies.

[Mauro Da Lio]

I made a comparison between a Harder Digital tube (GaAs) and a Photonis (multialkaly) tube in the IR. The GaAs is sensitive in the 800-1100 nm where the natural airglow is very strong. Thye Photonis should be less sensitive there.

 

I used two filters:

- Classical IR-pass (passes everything above 685 nm).

- Astronomik Proplanet (passes from 642 to 800 nm).

 

Findings:

- With the Photonis tube the IR-pass works better. I think the reason is because in the 642-685 band there is some residual light pollution.

- With the Harder Digital tube the Proplanet works better. I think the reason is because the Proplanet cuts the airglow.

 

The differences are small tough.

 

With GaAs and ideal filter might be the combination of the two, i.e., passing between 685 and 800.

Edited by Mauro Da Lio, 15 September 2024 - 03:15 AM.

[sixela]

The Astronomik passes to 720 nm.

The difference is not subtle: both the target and the background dim, but slightly less magnification restores the target surface brightness and you’re still left with a much darker background and more contrast.

Edited by sixela, 15 September 2024 - 03:30 AM.

[Mauro Da Lio]

Alexis, I think the OP was talking about filtering the artificial light pollution. So, L1 block above 720 nm, but passes the artificial light.

[sixela]

Quite. If you want to both filter light pollution but keep something in IR but not above 800 nm, then a Proplanet is pretty efficient on a Gen 3. Pretty dark though.

A “regular” light pollution filter like an IDAS LPS-D3 can also help, but it passes nothing above 700 nm.

Edited by sixela, 15 September 2024 - 11:20 AM.

[ButterFly]

Many light pollution filters for imaging pass ir well into the  airglow region. This has little effect on cmos/ccd cameras, which have negligible response in this region. This extended passband, in combination with the accentuated long wave response of GaAs cathodes, could contribute to the difficulty of seeing galaxies.

 

Are there available filters that pass some ir, enough to enhance, without extending into the airglow region?

Modern ccd cameras are in sharp decline.  Modern cmos cameras are very sensitive in near-IR up to a micron.  Here is the QE curve for my ASI220MM, for example.  If one adds a bayer matrix to the pixel array, the color filters' passbands multiply this base QE curve.

 

Most of the natural moonless airglow comes from molecular OH emissions.  Astronomik's 642 filter has a passband up to about 850nm, which span those bands.  There are OH rejection filters for you to explore as well, but not in 2" sizes.  There are no reports on people using them with NV.  During peaks of solar activity, not only do OH emissions increase, but there are also other bands as well.  Increased solar activity has effects on visual observing as well, so it's not just a near-IR phenomenon.  As with any other molecular emissions, the bands are rather wide.  If there is Moon out, the added skyglow is polarized in the same way as sunlight is.

 

Galaxies, as broadband emitters, will suffer with any filtering, unless the sky background is filtered out much more.  Molecular bands simply aren't narrow enough to reject them properly, without also rejecting a lot the galaxies' light as well.  If there is a strong polarization signal in the background, such as at ninety degrees from the moon, that will help.

 

Where NV helps with galaxies is where there is already inherent contrast.  Dust lanes are a clear example, where neighboring regions of the phosphor screen light up by very different amounts.  Making everything brighter there helps because there is already sufficient contrast there.  Smaller elliptical galaxies are another place where there is a clear gain.  Even sharply peaked spiral arms, like M51's, show very little gain in contrast over an eyepiece.  The disk is getting bright just as the peaks of the arms are, and the background as well.

 

Back to your original assertion:

 

Many light pollution filters for imaging pass ir well into the  airglow region. This has little effect on cmos/ccd cameras, ... .

Skyglow has less of an effect on imaging because of stacking.  One can stack and reduce uncorrelated background.  The longer the total integration time, the better the reduction of uncorrelated background.  Gradient removal is also available after stacking.  There is no stacking or gradient removal with NV.

 

If you are looking for better views of galaxies over an eyepiece, get yourself a cmos camera, preferably cooled, and stack images to reduce uncorrelated background.  NV is not the right generalist tool for the job.
 

[rmorein]

I have an ASI533MC, With apologies to the serious astronomers who have favored their replies, my quest is more whimsical, varying from night to night. Your advice is surely good for some of those nights.

 

Please suggest some OH rejection filters.

Edited by rmorein, 15 September 2024 - 08:08 PM.

[ButterFly]

I have an ASI533MC, With apologies to the serious astronomers who have favored their replies, my quest is more whimsical, varying from night to night. Your advice is surely good for some of those nights.

 

Please suggest some OH rejection filters.

You'll have to call manufacturers for a custom notch filter design.  Omega, ThorLabs, Chroma, and Edmund will get you what you need.  They will not be cheap.

 

For the near-IR response of your camera, take images of the same star field, at the same exposure and gain, with and without a UVIR cut filter.  The added bloat is the near-IR response.  Here are the charts for your 553MC.  R, G, and B all have the same response above about 850!  When comparing your imaging setup to NV, compare a single frame, rather than a stack.  If you wish the make the comparison more "fair" (very loose here), look at just the red channel.

[Mauro Da Lio]

There are many bands for airglow emissions. https://sites.bu.edu...k-hart/airglow/

The longer the wavelength, the stronbger the bands. For example the bands above 800 nm and up to 1100 are much stronger than below 800 nm (figure 1).

Conversely artificial light is concentrated above 650-700 nm https://onlinelibrar....1111/gcb.13927 (figure 2).

 

 

Edited by Mauro Da Lio, 16 September 2024 - 03:14 AM.

[rmorein]

Modern ccd cameras are in sharp decline.  Modern cmos cameras are very sensitive in near-IR up to a micron.  Here is the QE curve for my ASI220MM, for example.  If one adds a bayer matrix to the pixel array, the color filters' passbands multiply this base QE curve.

 

Most of the natural moonless airglow comes from molecular OH emissions.  Astronomik's 642 filter has a passband up to about 850nm, which span those bands.  There are OH rejection filters for you to explore as well, but not in 2" sizes.  There are no reports on people using them with NV.  During peaks of solar activity, not only do OH emissions increase, but there are also other bands as well.  Increased solar activity has effects on visual observing as well, so it's not just a near-IR phenomenon.  As with any other molecular emissions, the bands are rather wide.  If there is Moon out, the added skyglow is polarized in the same way as sunlight is.

 

Galaxies, as broadband emitters, will suffer with any filtering, unless the sky background is filtered out much more.  Molecular bands simply aren't narrow enough to reject them properly, without also rejecting a lot the galaxies' light as well.  If there is a strong polarization signal in the background, such as at ninety degrees from the moon, that will help.

 

Where NV helps with galaxies is where there is already inherent contrast.  Dust lanes are a clear example, where neighboring regions of the phosphor screen light up by very different amounts.  Making everything brighter there helps because there is already sufficient contrast there.  Smaller elliptical galaxies are another place where there is a clear gain.  Even sharply peaked spiral arms, like M51's, show very little gain in contrast over an eyepiece.  The disk is getting bright just as the peaks of the arms are, and the background as well.

 

Back to your original assertion:

 

Skyglow has less of an effect on imaging because of stacking.  One can stack and reduce uncorrelated background.  The longer the total integration time, the better the reduction of uncorrelated background.  Gradient removal is also available after stacking.  There is no stacking or gradient removal with NV.

 

If you are looking for better views of galaxies over an eyepiece, get yourself a cmos camera, preferably cooled, and stack images to reduce uncorrelated background.  NV is not the right generalist tool for the job.
 

The above makes perfect sense. However, in my location, SE PA, Bortle 7, atmospheric conditions are between a rock and a  hard place. When NWS predicts "clear", they appear to have contextualized this to mean light haze, which varies minute to minute, rendering flats useless.   It has gotten worse in the past 20 years. Years pass without a single day with a western sky. Those of you who enjoy western skies, which I have experienced, might wonder why I am bothering at all. I will never manage an image that reaches "mediocre", so I have defined my interest as EAA, with immediate or near immediate gratification a substitute for quality. It is a personal, religious experience.

 

 Now I ask a question which some might say belongs in EAA, except that the comparison is with image intensifier technology. Given that the background varies rapidly in my location, which makes it impossible to methodically build contrast with patient stacking, are there tips and tricks in the form of real time processing, AI, etc., that allow a modern CMOS camera  (Starvis2, no amp glow) to overlap or supplement the immediacy of night vision?

[ButterFly]

There's always this: