Username:

Password:

Altair AstroDIO DehumidifiersAtik CamerasModern AstronomyDavid HindsNe3 Filters

Author Topic: Do you really need a LP filter?  (Read 2771 times)

0 Members and 1 Guest are viewing this topic.

Offline dciobota

  • Daniel
  • Galactic Poster
  • ******
  • Posts: 9722
Do you really need a LP filter?
« on: 00:33:10, 29 January, 2011 »
 Warning!!! Long and possibly boring post!

 I can't answer this question for everyone, but I've been mulling this question over for years, and in my case at least I've found convincing arguments that I do not.  I posted some of my thoughts in response to Steve's (starf) request for comparison shots on the M82 image he posted, and I realized maybe others can benefit from my experiences, or at least generate some discussion.  :)

 So, the background info.  I live in what I believe is a moderate light pollution area (see first image).  In addition to that, there is a bank across the street to the north which has 9 lamps that are on all night.  The light on the north side of my observatory is strong enough at night to almost read by (see image 2).  Btw, that image is a 30sec exposure at iso1600 and f4.  That camera is unmodded.

 The observatory is built so that the roof sections, when open, shield most of the immediate sources of light.  So the scope does not use a light or dew shield.  Also, the imaging scope has an f9 ratio, which also makes it somewhat resistant to off axis light pollution.

 So, essentially the light pollution I experience is mostly from ambient light.  That will vary widely with atmospheric transparency; on a warm, humid night it pretty much useless for me to image anything deep sky related.

 On relatively clear nights though, my site can offer decent deep sky views.  I judge a clear night by how well I see certain dimmer stars.  Right now, since I'm imaging north, I observe the Big Dipper as to how well I see Megrez, the dimmest one.  If I can see it clearly, without averted vision, then it's a good night to image.

 The above will give you a good way to compare how much LP you have relative to my site.

 I cannot stress enough though how important it is to first block all direct sources of light at your site, be it with shields, portable canvas walls, or even as part of an observatory construction.  When I first started imaging, my images were totally washed out with light pollution.  A lot of that went away when I started using dew shields, and later on the observatory.  It really is like a little cave of darkness in a sea of light when I image nowadays.

 I hope I haven't bored, but all this is really leading to the filter question.  ;)

 So, how bad is the lp at my site now?  Still pretty strong I think.  Image 3 is a side by side comparison of two 10 minute exposures at iso800 and f9 through my scope, taken on similar nights as far as transparency.  The camera is a modded 1000D, which is admittedly more sensitive to red.  The original factory filter was removed, but not replaced with any other filter.  The image on the left was taken with a clip on LP filter, the Astronomik CLS.  The image on the right was taken with just a plain luminance filter (uv/ir).  As you can see from the image on the right, there is still plenty of LP due to ambient light.  A rough calculation shows that the image on the right is about 45% brighter, a significant amount, although some of it is not due to LP but simply because the LP filter blocks some of the useful light as well.

 Well, looking at the images, one would be led to conclude that the left image has better contrast and better information than the one on the left.  Certainly the one on the left looks kinda washed out and has the dreaded orange background so difficult to remove. 

But is that really true? I took that very image into photoshop, selected just the right hand image, and adjusted the levels on each channel to balance the color and drop the background luminance to the same values as on the right... essentially, correcting for LP.  It's a very easy adjustment actually.  Image 4 shows the result.  The non filtered image has more background noise, true, but the depth of the exposure is the same as on the left, actually better detail in some of the fainter areas.  And the color on the left image is fiendishly hard to correct in my experience, I could never get both stars and object to achieve a good color balance.

 So, there are pros and cons to using an LP filter.  In my case, it's really a borderline call, and eventually I chose to do without.  The noise is not as bad an issue as shown here, probably because my processing of a jpeg image made it a lot worse.  On similar number of exposures, and after processing, the non filtered image stack had about the same level of noise as the filtered one.  In fact, the final image I posted not too long ago was entirely using unfiltered data, since it yielded the best results.

 Hope this helps.

PLEASE REMOVE ME

Offline Scotty H

  • Full Member
  • ***
  • Posts: 243
Re: Do you really need a LP filter?
« Reply #1 on: 02:03:38, 29 January, 2011 »
I'm obviously only a newbie to astrophotography, yet this is a very intruiging experiment. You say the images were taken on similar nights transparency wise? I use a CLS filter as well but have never tried taking the same DSO picture with and without. I did try some fixed tripod widefield images images with and without the filter and found that it made alot of difference on one night and minimal another, then again the exposure times are very different. I am a pilot by trade and find that visually atmospherics can play a big part in what you see at different altitudes. (Ice crystals, high/low level cloud, humidity etc) I know from my site just recently that the light polution has been pretty bad just because it has been relatively damp, cold and foggy... Getting big halos around the moon, orange glow from the nearest town (14 miles away) Yet other nights very little LP, maybe the atmosphere is that bit more dry and calm and maybe little bit of luck is involved. Fortunately i live in a small village so i dont get street lamp problems as you do. I see in the two right hand images a red/pink star in the middle of the image and thought maybe a hot pixel yet in the two left hand images they're not there. Am i right in thinking the left hand images are with the CLS filter (Sorry if i'm wrong) if so the filter is cutting out alot more than i was expecting. Will definately have to experiment lots more myself. :D   
Mounts: 10 Micron GM1000HPS, Celestron CGEM
Scopes: Boren-Simon 8"CF Powernewt, Meade 5000 80mm Apo, Meade 8" ACF SCT
Cameras/Filters: SBIG STF8300, SX CFW7 LRGB, Ha, SII, OIII, Canon 1100Da, Astronomik Ha, Cls, QHY5
Processing: Pixinsight, PS CS6, DSS,
Capture: APT especially now it runs SBIG cameras, BYEOS

Offline dciobota

  • Daniel
  • Galactic Poster
  • ******
  • Posts: 9722
Re: Do you really need a LP filter?
« Reply #2 on: 05:07:48, 29 January, 2011 »
 Hi Scotty,

 Yes, there is one red hot pixel in the right hand side images that does not appear on the left side.  Due to framing differences, M82 is not exactly in the same place on the frame in the two images.  Hopefully the CLS wouldn't cut out stars like that, I've never seen it.

 Yes, the left hand side images are with the CLS.

 Interesting that you mention different kind of atmospherics, ice crystals, I haven't thought of that.  The last couple of nights here have been clear but "foggy", although I think those are ice crystals, since the temps are still well below freezing. 

 It would indeed be interesting to make this filter/no filter comparison on a more humid night.  I would suspect the results would confirm your findings, that the cls filter helps a lot more on these nights.  But a more important question would be, would the cls filter in that case result in an image of similar depth to a clear dry night?  I have become a lot more picky with my nights over the years, on clear humid nights I hardly ever image anymore.  Of course, if those nights are few and far between, then the CLS is definitely an option.
PLEASE REMOVE ME

Offline starf

  • Poster God
  • *****
  • Posts: 1888
Re: Do you really need a LP filter?
« Reply #3 on: 10:42:30, 30 January, 2011 »
thanks for the comparison daniel. The one without the CLS after processing contains much better colour information and detail. Its only the raw images that have more contrast if a CLS filter is used, which gives a totally false impression of what this filter is capable of.

i dont think ill ever go back to using a light pollution filter with a DSLR. They increase exposure times and mess with star colours. I thought maybe wideband filters may have a use for mono luminance though - just like using a UHC filter visually on bright nebula or a yellow filter on the moon.

some objects seem to like shine at very specific wavelengths. I thought the jets in M82 may be the same but from what ive seen they are extraordinarly present in the x-ray range as seen by the chandra observatory. I read somewhere that the halo and jets are thought to be synchrotron radiation caused by the galactic magnetic maelstrom.

It struck me that most people image M82 in H-alpha (there's a classic example in Burnham's
guidebooks)- but they should be visible in possibly O[II] and S[III] and certainly in wideband filters.

It could of course, be that the features are so faint when compared to the very bright ellipse of the galaxy that they would be completely washed out in anything other than narrowband.

ps. nice observatory too and no snow?

Offline dciobota

  • Daniel
  • Galactic Poster
  • ******
  • Posts: 9722
Re: Do you really need a LP filter?
« Reply #4 on: 18:39:14, 30 January, 2011 »
Lol Steve, that obsy pic was taken last summer, but here's one taken about two weeks ago.

 As far as wideband filters for luminance, that was my thought too, but they cut so much light I'd just as soon use a plain luma filter.  As I was writing the article though I did retry combining the CLS filter results with the non filtered ones, but just the red channel from the CLS.  I was somewhat successful improving the red channel info, to where more of the ejecta were visible.  But, as before, it affected the star colors as well, although not as bad.  So, in my opinion, the CLS filter may make a decent Ha filter in a pinch...maybe.  I don't have either a cls or Ha filter to do a comparison.

 As far as imaging M82 in narrowband, interesting point.  I love the Chandra/Hubble/Spitzer composite here:
http://www.nasa.gov/mission_pages/chandra/multimedia/photo10-038.html

 It may well be there would be more than just Ha ejecta, although it seems quite a bit of it shines nicely in IR.  I don't have an IR pass filter, but I guess I could try taking off the uv/ir filter at some point and see what results I get.  That may give some interesting results in the red channel.  Which can mean that some really hot galaxies like M82 could benefit from IR data... another interesting idea.  :)
PLEASE REMOVE ME

Offline disraeli

  • Poster God
  • *****
  • Posts: 1772
Re: Do you really need a LP filter?
« Reply #5 on: 09:32:33, 31 January, 2011 »
Dan, all of what you say is spot on particularly about local ambient light.  I am surrounded by mountains and no sodium light visible from my site except for a farm security light about 1 mile away.  I was finding what I thought was amp glow when stretching images. It was not! I have been using a 5W fluorescent light to see my laptops and yes you've guessed it, that was the problem coupled with the 250 watt yard light about 30 yards from the telescope which would come on if the cat passed by or my dear wife called me in for supper.   Fred.
Fred.----------------------------------------------------------------------------
HEQ5 pro, TMB 92 Triplet with Televue   TRF-2008 x0.8 flattener, ATIK 4000LE OSC. 300mm telephoto lens with Meade DSI IIIC as Guide Camera,  Guiding with PHD, Phillips SPC900NC webcam. Various camera lenses all wit

Offline dciobota

  • Daniel
  • Galactic Poster
  • ******
  • Posts: 9722
Re: Do you really need a LP filter?
« Reply #6 on: 18:39:09, 31 January, 2011 »
 Lol I feel your pain.  If you notice in my night pic of the obsy several solar lights, white ones at that.  Thank goodness the roof sections block them, I've had a good mind to tear them out never mind what the wife says!
PLEASE REMOVE ME

Offline dph1nm

  • Full Member
  • ***
  • Posts: 106
    • Nigel's Astrophotography Page
Re: Do you really need a LP filter?
« Reply #7 on: 18:35:45, 02 February, 2011 »
I wondered about this for a long time, as I suspect that for most broad spectrum objects LP filters do reduce the signal-to-noise, although there does seem to be a great variation between makes in how much light they actually block. However, I had a lot of trouble without an LP filter getting colour right, because my red channel was saturating much quicker than green or blue.  If you then subtract off the light pollution (which is the correct procedure) you end up with bright objects  being blue/green. To compensate for this you then have to stretch the red channel, but then the colours vary with intensity as you have a different stretch from the other two channels (and the red noise can become more prominent). So I bought an SkyWatcher LP filter and  find colours come out much better.

(I guess you should really clip the green and blue channels to the same range as the red at the RAW stage straight out of the camera, then the colours might work).

NigelM
8.5" F6 Newtonian + HEQ5 Goto / Celestron Nexstar SLT102 / afocal imaging with Olympus 2020Z P&S digital camera.

Offline dciobota

  • Daniel
  • Galactic Poster
  • ******
  • Posts: 9722
Re: Do you really need a LP filter?
« Reply #8 on: 19:48:25, 02 February, 2011 »
 That's interesting Nigel, my experience has been the opposite with colors. 

The last set of images (below the raw ones) show the effect of simple levels adjustment.  The image on the right is the result of the orangey unfiltered image with just levels applied to move each channel histogram to a base value of zero (where data actually starts).  Maybe with a worse affected image colors may become an issue, but in my experience, as long as you don't blow your highlights too severely, colors are retained in stars and objects as well.

 If you have an example of a hard to correct raw image, I would love to take a look at it and see how it can be corrected.  I always believe in considering more datapoints, especially ones that differ from mine.
PLEASE REMOVE ME

 

ukbuysellRemote Imaging from AustraliaSharpSkyblank APTUKAI on Facebook
TinyPortal © 2005-2012