Urban astrophotographers have a big problem: light pollution. Stray light from cars, houses, street lights, it all makes our skies glow and drowns out the faint deep sky objects that we want to photograph.
Certain targets like emission nebulae give out lots of light in narrowband wavelengths, and if you’ve got the right filters then you can basically image through the light pollution as if it’s not there. See the post on OSC vs Mono from a city for more details.
Other targets like galaxies and star clusters are broadband in nature, so our narrowband filters aren’t effective. Light from these broadband objects gets drowned out by light pollution, which makes imaging them difficult, but not impossible. Long integration times are key — I outline some tips on that topic here.
To help with this problem you can buy Light Pollution Suppression (LPS) filters. These work by blocking wavelengths of light associated with light pollution, for example sodium from old-style orange street lights. Modern LPS filters even claim they can help against LED lighting, which is the bane of many urban astrophotographers’ lives.
LPS filters aren’t perfect though. As well as blocking light pollution, they inevitably block some of the light from deep sky objects that we actually want to record. The aim is for a LPS filter to block far more of the “bad” light than the “good”, thereby improving your image’s signal-to-noise ratio.
Get on with the tests, I want to know which LPS filter to buy
I was recently in possession of three LPS filters from reputable company IDAS, and decided to have a shootout to see which was most effective for me. Throw “No Filter” into the mix as a control and we’ve got four options:
| No Filter | Our control in this experiment | Free! |
| IDAS-LPS D1 | Older generation filter but still well regarded | £169 |
| IDAS-LPS P3 | Latest generation | £199 |
| IDAS-LPS D3 | Latest generation, designed to help combat LED lighting | £181 |
Let the games begin!
For the first test I integrated three hours of data for each of the four options. I then edited each of these in the same way, following the same steps as when I properly edit photos, but keeping it simple just to produce images that could be compared visually.
Be honest: can you see much difference? I can’t. D1 looks a bit bluer, which is probably a colour cast that’s improved upon by the newer generation of LPS filters. None of them look dramatically better or worse than any of the others.
Round two, ding ding
Let’s crop in even tighter to M81. Maybe we can see some differences in the spiral arms.
Hmm, not really. Maybe if you squint a bit?
Round three, fight!
Last chance saloon for these filters. Surely they should provide some improvement over not using a filter at all. Let’s crop in really close to M82.
Maybe some very minor differences, but nothing that couldn’t be explained by the sky conditions being slightly different when the tests were made. And remember that we’re now cropped in really tight just in an effort to try and see any improvements.
Surprise bonus round!
It’s at this point in the tests that I sold the D1 and returned the P3. I still had possession of the D3, which in theory should be the best of the bunch — latest generation, and optimised for LED lights. Resident genius vlaiv from the Stargazers Lounge forum helped to run more detailed (and scientific) tests on No Filter versus D3. His results… drumroll…
The D3 helped to smooth out the background gradient of the image, but also degraded the light from the galaxies to such an extent that the all-important signal-to-noise ratio is actually worse when using this latest and greatest light pollution filter. I’m better off not using any light pollution filter at all!
Say what?!
Even if we’re generous and say that sky conditions were tipping the scales against these LPS filters, the benefits would be marginal at best. Certainly not worth spending almost £200 for.
I’m not saying that all LPS filters are pointless for everyone. Some people swear by them, and they can have extra benefits such as delaying skyglow washout and so allowing for longer exposure times. Plus, I’ve only tested three (albeit three with good reputations). Oh, and my simple processing steps may well have evened out any differences — although they’re the same first steps I take when processing images properly, so I feel it’s fair as a real-world test.
I think that the effectiveness of LPS filters is very dependent on your local sky conditions (how much light pollution, and what are the sources?) and also the imaging kit you’re using. My opinion is that LPS filter reviews from other astrophotographers are basically worthless unless they’re in the same location and using the same kit as you are. Do you live next door to an astroimaging YouTuber using the same made and model as your telescope and camera? Probably not.
You’ve got to test them for yourself. See if you can borrow some, or find a supplier with a good returns policy (thanks, First Light Optics). That’s a lot more work than just reading some online reviews, but no-one gets into astroimaging to have an easy life! You might even find that you’re better off without one.
Update, January 2024
Optolong sent me one of their new L-Quad Enhance light pollution filters to test. Is it any good? Check out my review here!
If this article has helped you, please consider making a donation.
That’s a really interesting comparison. Thank you for taking the time and effort to post this up. I found the link to your website on stargazerslounge btw. i’ve struggled with light pollution since I started imaging 10 years ago. I gave up about 5 years ago but now I’m back and wondered what the impact of street led lighting would have. Looking at your images I’ve decided to not bother with the IDAS and Astronomik LPF’s that I have. I’m using both OSC and DSLR.
I’ve used FLO’s Clear outside app and it looks like I’m in a bortle 5 area.
I’ll be reading up a lot more on your website.
Again many thanks
Kindest regards
Pete
Thanks Pete, I’m glad you’re enjoying the website. More content coming soon!
LED lighting certainly make things more difficult for us astroimagers. If you’ve already got some filters, I’d say give them a go and see if they make an improvement. That’s really the only way to know for sure. Maybe they’re worth it for you given your equipment and sky conditions; or maybe not, in which case you can sell them!
Great analysis. I’ve done a simpler OSC test comparing frames with and without Orion’s Skyglow filter. (which isn’t all that different than the new hot Optolong L-Pro if you compare the bandpass spectrum). Can’t find the test data but seem to recall that imaging time on M51 was tripled to achieve the same image with no improvement. Things may have worked out better on a nebulosity. Or not. But I’m getting only 2 or 3 hours of exposures and that eliminates the LP filter option.
In the past I’ve done LRGB work and had great success with narrow band filters. LRGB is generally a whole lot better than OSC but it’s very time consuming and post processing can be a career in itself.
Thanks for the comment, interesting to hear about your test. I wonder how many imagers out there are using filters and just assuming that they’re helping?
Thank you for taking the time to do this. Good ‘real world’ review
Your test results confirm the same conclusion that I have gradually come to, i.e. I am better off without filters. (maybe that should be, I am no worse off without them)
Maybe one day a light pollution filter will be made that actually helps us out! At least we’re heading into darker months now, with more narrowband targets well-suited to filters like the L-eXtreme.
Great post!
Quick question: is LRGB on a mono better for high bottle areas for broadband targets. (In know that SHO is great for nebula, but I assume doesn’t work for galaxy, other broadband target)
I know that both LRGB and the dual narrowband are great for nebulae, etc. But if I want to do broadband targets from high LP backyard, is mono worth it?
In principle, mono plus LRGB can produce higher quality results than OSC for broadband targets from high-Bortle areas; but the kit will be more expensive, and more complex to use. See my article called “OSC vs Mono from a city” for more details. It’s certainly possible to get good broadband images using an OSC camera from a city centre though. Check out my recent photo of the Pleiades as an example. Whether using mono or OSC, the trick is to aim for long integration times to achieve a high signal-to-noise ratio.
Very interesting results Lee – thanks for posting. I was looking at getting an IDAS clip-in for my EOS and just looking at the passband curves couldn’t see any difference between the D3 and P3 in terms of specifications let alone a field test like yours. I do have a real mix of LED and sodium floodlights in my area (edge of Greater London near a London underground depot) so was thinking the D3 but then again using Pixinsight DBE can do wonders
Ouch, sounds like your light pollution is similar to mine. Brute-forcing your way through with long integration times can work really well though, maybe it’ll work for you too. Good luck!
Lee – A great test and write up. Even a few years later, this remains useful information. I am imaging from San Diego coastal skies in a B7+ area with my DSLR. I started off with no light pollution filter, but the skyglow was killing me. I opted for a Haida ClearNight filter (a basic neodymium filter) for the front of the lens, and saw improvement in contrast and reduced light pollution gradients, which, as a total newbie, were overwhelming me. Fast forward a couple of years, and I am now starting to use a small refractor and am looking for an LP filter that will take in all of the major nebula emission lines and reject or reduce HP/LP sodium, mercury-arc, and LEDs. We are still mostly LP sodium here, with an increased use of LEDs. After spending hours looking at filter absorption spectra, I am leaning towards either the IDAS LPS D3 or the Optolong L-Pro. Your results, though, do give me pause. Have you been able to follow-up on this with additional real-world results, perhaps with a more optimal exposure time for each filter? It would be instructive to see the difference between 22 hours of filter and unfiltered data. One may have to expose more to get better results with the LP filters. Again, thanks for such a thorough and thought-provoking article.
Hi Martin, thanks for your comment. Since I conducted these tests I’ve continued to image RGB data with no filter. (My camera have an in-built UV/IR cut filter). I’d be keen to try some updated light pollution filters if any come on the market, but my approach of battling through light pollution using long integration times has served me well. If you’re interested in nebulae, why not something like an Optolong L-Ultimate or equivalent? That’s what I use for narrowband targets (nowadays often in conjunction with an Askar D2), and it works well.
I have enjoyed with your explanation and valuable information since I’m newbie in this field. I live in a suburban area (level 6-5 on the Bortle scale) and it can be a little frustrating to get these types of inquiries only through companies in the sector. Thank you very much for your honest work. On the other hand, I am looking for an operational website where I can sell products related to this hobby (posibly more specialized than wallapop!!). Thank you very much again for your time.