39 replies to this topic

[Jethro7]

Hello CNers,

Some years ago I tried a 3nm or 3.5nm Ha filter and found that this filter caused too much scintellation with my PVS 14 it was a mess. All be it that Ha filter was a no name one that I purchased off Ebay. Who knows what band width that filter actually was and also the PVS 14 that was used trying out that cheap Ha filter was not near as good as the one I own now with one exception the EBI of my first tube was .2 and my present tube is .6 other than that all the other specs are much higher with my present tube. So I stayed with using a tried and true Baader 6nm Ha filter.  Last week Gavster, recommended a Antila 2.5nm Ha Narrowband filter to a CN member and we had a back and forth conversation of the merits between a 6nm Ha, 3nm and the Antila 2.5nm Ha. Any way I decided to take a $600.00 leap of faith in Gavsters opinion on the Antila 2.5nm Ha filter and purchased the filter. I recieved the filter on Saterday and had an opertunity to test the Antila 2.5nm Ha filter tonight. My ambient temp was 39°, Seeing was not very good, Transparency was excellent and of course my ever present Bortle 8+ skies. I was hoping for but not expecting good results. I decided to use my Tak TSA 120 F/7.5 scope because this scope is my most used scope. You know the drill, you set everything up turn it on and point your scope towards a Nebula and hope.  I chose M42 to start because it's awesome and there are alot of other Nebulae in the area. I was quite pleasantly surprised that I could tune out alot of the scintellation with the devices GAIN, presenting a very nice quiet and dark almost black contrast between the targets and the PVS 14's FOV. vs. the dark greyish contrast of a 6nm Ha filter. Of course M42 was as awesome as ever. I could definitely see a good bit more of the wispy nebulosity around the brightest part of M42 as well as much more detail in the Running man Nebula (NGC1975) From there, I moved on to the Horshead Nebula (IC 434) and the Flame Nebulae (NGC 2024) all were more prominent and presented more detail and this was true with being able to see more of Bernard's Loop (SH 2- 276) I used my TV 55/67 Plossl for most of tonight's viewing and tried out my newly aquired 26 Nagler with good results. Moving on from Orion, I went on and took in a Nebulae Tour and was delighted in what the Antila 2.5nm Ha filter allowed me to see. I did not know what or how much I have been missing till now. 

 

Thanks Gavster. 

 

KEEP LOOKING UP EVERYONE Jethro

 

 

Edited by Jethro7, 28 January 2024 - 11:45 PM.

[a__l]

Most likely it will be just as aggressive with how fast the telescope is. Therefore, the results will be mixed.

[Gavster]

Jethro, that’s great news!

[sixela]

Most likely it will be just as aggressive with how fast the telescope is. Therefore, the results will be mixed.

I’ve got one as well, currently one with a passband slightly blueshifted (from the 6 I have graphs of it’s the worst).

It works OK to f/4 but not below.

In another thread I already recommended that anyone with a scope close to f/4 or below contact their vendor to cherrypick one; they come with individual measurement reports so if they have a number in stock it’s not impossible.

If your scope is f/4.5 or slower I wouldn’t worry, if 6 samples are enough to guess what the tolerances on passband placement seem to be.

If your scope is faster than f/3.5 then even cherrypicking won’t work, the filter is just too narrow (the FWHM really is 2.5nm).

Edited by sixela, 29 January 2024 - 03:29 AM.

[Jethro7]

Most likely it will be just as aggressive with how fast the telescope is. Therefore, the results will be mixed.

Hello a__I, and sixela,

 

I’ve got one as well, currently one with a passband slightly blueshifted (from the 6 I have graphs of it’s the worst).

It works OK to f/4 but not below.

In another thread I already recommended that anyone with a scope close to f/4 or below contact their vendor to cherrypick one; they come with individual measurement reports so if they have a number in stock it’s not impossible.

If your scope is f/4.5 or slower I wouldn’t worry, if 6 samples are enough to guess what the tolerances on passband placement seem to be.

If your scope is faster than f/3.5 then even cherrypicking won’t work, the filter is just too narrow (the FWHM really is 2.5nm).

We will see how this goes with my other scopes when the opertunity arrives. My fastest scope is my Discovery 12.5" Dob at F/5, or the Altair Starwave 152 F/5.9 for a refractor.  As far as cherry picking filters go, I did not know that this was an option, it's a little too late for that but by my results at first light with the 2.5nm Ha filter are very encouraging. At the least the filter works well with my Tak TSA 120 F/7.5 that I like to use far more often than any of my other scopes. 

 

HAPPY SKIES AND KEEP LOOKING UP Jethro

[sixela]

No worries, as I said on the 6 test reports I saw the centre of the passband was never more than 0.4nm from the emission line. And the filter has a really flat top so with your scopes all 6 filters for which I have seen graphs will work flawlessly.

f/5 has 0.7nm shift at the extreme edge of the aperture. The FWHM edge of a perfectly centred passband is at 1.3nm of the middle. 0.7+0.4<1.3. So it’s likely the aperture is completely in-band on the flat top of the transmission.

For all your scopes that filter is a no-brainer.

 

At f/3.5, though, only 2 of the 6 filters I saw will work really well (if the graphs are accurate). For 2 others there'd already be some signal loss, but the more narrow bandwith will still lean to increased contrast, so they'll likely still be better than a 6nm or even 3.5nm filter. For the two last ones you'd really rather have a less narrow filter (or the much more expensive 3nm Chroma filter, which has less bandpass variation and some slight preshit of the passband to the red).

Edited by sixela, 29 January 2024 - 02:23 PM.

[Jethro7]

No worries,

For all your scopes that filter is a no-brainer.

 

 

Hello sixela,

Thank you for the filter education and encouragement. We will see how it goes over the next few days. Looking at my weather forcast it appears that I just might get a few nights of good skies. These good skies have been 

Kind of rare lately.

 

HAPPY SKIES AND KEEP LOOKING UP Jethro

[Deadlake]

Jethro, that’s great news!

How do you rate the Antlia 2.5 compared with the Chroma 3 nm?

 

About the same without the cost?

[sixela]

I wouldn't say "the same" if you have a fast scope (close to f/4 or below), unless you cherry-pick from a vendor.

 

The good thing about Chroma is that they really nail the passband location and flat top with very little variation. Now if you have an f/5 scope that doesn't matter, but not everyone has an f/5.

 

[I got the most blue-shifted bandpass Antlia 2.5nm H-alpha of all 6 originally in stock at one vendor, but they're letting me swap it for the most red-shifted one instead; should have asked first instead, given my scopes are f/4 and f/3.72.] 

 

The good thing about Antlia is that at least they provide an individual test report with each filter. With the Optolong L-Ultimate, you know some filters will do (the one Jim Thompson tested is preshifted to red in quite a perfect way) and some will not and no one can actually cherry-pick, it's just a lottery with a little wiggle in the bandpass centring. Which is fine at f/5 but not f/4 and below.

Edited by sixela, 29 January 2024 - 04:12 PM.

[WheezyGod]

Yea I’ve been waiting to see how the Baader 3.5nm HS filter does. I’m expecting more scintillation but so far it hasn’t bothered me enough. If I like it for handheld use then I’ll plan to buy my own and know I could go more narrow than the 4.5nm 2in I have from Antila.

I would think that nebula that are already very bright (the top 10 or so) might be better with a less aggressive filter but depending on how dark someone’s skies are there’s other faint nebula in the area that could become detectable with a more narrow filter.

For those of you who dislike the more aggressive filters is it because there’s more scintillation, less stars, the sky background looks unrealistic or a combination of the 3?

[sixela]

I still have to confirm the second Antlia filter I'm getting really has a bandpass matching the individual test report. It's evidently from another batch, so I have only one graph to go by. The batch characteristic does match that of  those tested on astropolis, so my guess is that it is from that batch, but I only have one graph so I don't know if there is indeed the spread you would expect on the individual graphs (the two filters tested and reported on astropolis show the same wiggle in bandpass centring).

 

The other 5 test reports are evidently individual *and* honest, and all from the same later batch, and the angular dependence on whether H-alpha emission is in-band or out of band does match the graph that came with my filter. Perhaps I should have picked the best of those instead, instead of the one of the older batch ;-).

 

But not everyone need obsess: that's all for us maniacs who want to combine f/4 scopes with super-tight filters. All the graphs I've seen and the individual tests by an independent tester are all fine at f/4.5 and above.

 

BTW, @Jethro7, care to share the test report that came with your filter? The back page only, because the front page is generic.

Edited by sixela, 29 January 2024 - 07:17 PM.

[WheezyGod]

I still have to confirm the second Antlia filter I'm getting really has a bandpass matching the individual test report. It's evidently from another batch, so I have only one graph to go by. The batch characteristic does match that of those tested on astropolis, so my guess is that it is from that batch, but I only have one graph so I don't know if there is indeed the spread you would expect on the individual graphs (the two filters tested and reported on astropolis show the same wiggle in bandpass centring).

The other 5 test reports are evidently individual *and* honest, and all from the same later batch, and the angular dependence on whether H-alpha emission is in-band or out of band does match the graph that came with my filter. Perhaps I should have picked the best of those instead, instead of the one of the older batch ;-).

But not everyone need obsess: that's all for us maniacs who want to combine f/4 scopes with super-tight filters. All the graphs I've seen and the individual tests by an independent tester are all fine at f/4.5 and above.

BTW, @Jethro7, care to share the test report that came with your filter? The back page only, because the front page is generic.

I’m quite surprised they even have individual test reports. Everything I read explained that most brands provide batch level stats at best.

[sixela]

No, the filters I've seen from the November tests indeed have individual test reports, and show the small variation in bandpass centring that you expect given Antlia's history. Mine is the worst of the lot for fast scopes and should deliver only 54% transmission on my f/3.72, and unfortunately when using it front-mounted on a 1x or 3x objective it's fairly easy to confirm that indeed at angles over 5° H-alpha goes off-band, which agrees perfectly with the test graph (for an effective refractive index of roughly 1.85).

 

It'd be perfectly fine for an f/4.5 scope, though.

 

From the earlier batch (July 2023) I only have seen only one test report (since the vendor has sold all the others from that batch). If that graph indeed belongs with that filter I'll be glad, with roughy 80% transmission even on my f/3.72. The snag is that on astropolis.pl something that looks like from the same batch seemed to be more blueshifted than what the test graph implies, which casts more doubts on the accuracy of the graph supplied.

 

But at least they do seem to be individual test reports. Mind you, only the back side is individualised. The front is just boilerplate from one representative sample, but it's not what is really informational anyway.

 

The most funny graph I have is the one from my ALP-T (which, BTW, performs quite well, and astropolis.pl tests of those indeed showed very little individual variations). It's clearly an individual report at the back as well, but since it shows both OIII and H-alpha passbands on one single graph there's no way you can eke out any useful information out of it with respect to passband centring. They might as well show a generic test because you can't read any small differences at that scale anyway.

Edited by sixela, 30 January 2024 - 03:51 AM.

[Jethro7]

No, the filters I've seen from the November tests indeed have individual test reports, and show the small variation in bandpass centring that you expect given Antlia's history. Mine is the worst of the lot for fast scopes and should deliver only 54% transmission on my f/3.72, and unfortunately when using it front-mounted on a 1x or 3x objective it's fairly easy to confirm that indeed at angles over 5° H-alpha goes off-band, which agrees perfectly with the test graph (for an effective refractive index of roughly 1.85).

 

It'd be perfectly fine for an f/4.5 scope, though.

 

From the earlier batch (July 2023) I only have seen only one test report (since the vendor has sold all the others from that batch). If that graph indeed belongs with that filter I'll be glad, with roughy 80% transmission even on my f/3.72. The snag is that on astropolis.pl something that looks like from the same batch seemed to be more blueshifted than what the test graph implies, which casts more doubts on the accuracy of the graph supplied.

 

But at least they do seem to be individual test reports. Mind you, only the back side is individualised. The front is just boilerplate from one representative sample, but it's not what is really informational anyway.

 

The most funny graph I have is the one from my ALP-T (which, BTW, performs quite well, and astropolis.pl tests of those indeed showed very little individual variations). It's clearly an individual report at the back as well, but since it shows both OIII and H-alpha passbands on one single graph there's no way you can eke out any useful information out of it with respect to passband centring. They might as well show a generic test because you can't read any small differences at that scale anyway.

Hello sixela,

Interesting, just by visual, my Antila 2.5nmHa filter works great but this filter did not come with any kind of test report. 

 

HAPPY SKIES AND KEEP LOOKING UP Jethro

[sixela]

Weird. Should have come with the filter. 

[eyeoftexas]

Is the Antlia 2.5nm filter available in 1.25” size?  I only find it as 2”.

[sixela]

Got the new Antlia 2.5nm filter. Took the liberty of putting them both on a spectrometer with a decent resolution. It's not yet perfectly calibrated, so I'll post something on the new filter only once that's done (since even 0.2nm can make a huge difference for some of these).

 

But the shape, transmission and relative bandpass centring between different filters is already known robustly with what we did tonight.

 

So:

-there is indeed a 0.3nm bandpass centring difference between the two Antlia 2.5nm H-alpha filters, and for an f/3.72 this does make a difference: on the red slope of the bandpass such a small shift makes a huge difference in transmission once the bandpass sihfts enough to the blue.

-a 7° angle of incidence causes 1.2-1.25nm of bandpass shift, which corresponds to an n_eff of roughly 2 (similar to other Antlia filters, and better than what was reported on astropolis.pl)

-there's a huge flat top and max transmission is above 95% over the entire top.

 

Summary: apart from the wiggle room in the bandpass centring, which is important only for scopes faster than f/4.5, these are truly excellent, and I can see why Gavster thinks this performs like a Chroma but at a better price.

 

For fast scopes it's possible to 'accidentally' (or not) get a filter that's just as good as a Chroma, but under f/4 you're taking chances that you won't be taking with a 3nm Chroma (I bet there's a lot less than the [-0.3,+0.3] nm wiggle that there seems to be at Antlia).

 

Don't know if cherry-picking Antlias on the basis of the supplied individual test reports works. It worked for me and the filters did indeed differ in the way the reports would indicate, but that could be sheer luck.

 

If I had to do it again for an f/3.72 I'd probably pick a Chroma. For an f/4, probably the Antlia. For anyone who wants a really good 6nm filter at a lower price than a Chroma, the Astronomik MaxFR is it (see below).

 

Also measured:

-3.5nm Baader H-alpha. Bandpass centring favourable to f/3.6 to f/6 --slightly redshifted. No flat top, but still fairly decent (and the best transmission is on the red side of the emission line, so good for fastish scopes). For the price, excellent, and actually the most narrow one that gives me good transmission across the entire wavefront of my f/3.72. Don't know if I was lucky with the bandpass being slightly redshifted; if that's just me being lucky and it wiggles that much to the other side there are going to be really bad ones too.

-Antlia ALPT-T. 5.5nm FWHM, slightly blue-shifted (not that good) but with a more gentle slope on the red side (good for fast scopes and somewhat mitigates the bandpass centring 'error' ).

-Optolong L-Ultimate. 4nm FWHM, bandpass quite blue-shifted (but for f/infinity the emission line is still on the top); it's like the Antlia with 1.5nm of the red side lopped off, unfortunately that's the wrong side for a filter that should still work at f/4. Confirmed as not that good for my f/3.72 and even at f/4 it suffers a bit. Less good than a colleague's filter, so for fast scopes you're playing Chinese roulette. I'd hesitate to stick my neck out for it for anything faster than f/4.5. Also slightly less transmission, but that's fairly irrelevant.

-the 6nm Astronomik H-alpha MaxFR bandpass and bandpass centring is just like the graph on the marketing blurb and is superb for anyone who wants that kind of bandwith. And I know that's not only this spectrometer, I could confirm it by looking at a solar spectrum and seeing where the H-alpha absorption line was in the bandpass. Obviously, not the same contrast at all when used at 3x compared to a 2.5nm Antlia (which makes the background pitch black in comparison).

Edited by sixela, 02 February 2024 - 09:56 AM.

[Thierry Legault]

Just got the 2.5 too.

 

Quick look between clouds with a 16" f/4.5 from Bortle sky 8 (Paris suburbs): on the horsehead+flame, the improvement in contrast against an Optolong L-ultimate 3nm (plus red filter to isolate the Halpha line) is not huge but significant. That's what I expected and I'm not disappointed.

Edited by Thierry Legault, 02 February 2024 - 04:39 AM.

[Jethro7]

Just got the 2.5 too.

 

Quick look between clouds with a 16" f/4.5 from Bortle sky 8 (Paris suburbs): on the horsehead+flame, the improvement in contrast against an Optolong L-ultimate 3nm (plus red filter to isolate the Halpha line) is not huge but significant. That's what I expected and I'm not disappointed.

Hello Thierry,

Interesting point about the red filter. 

 

HAPPY SKIES AND KEEP LOOKING UP Jethro

 

P.S. I just wanted to thank you for sharing  your knowledge and insights, you have been a big help to me over the years.

[sixela]

Here's my first attempt at calibrated data; still rough since my calibration light source wasn't very precise, but enough to illustrate. The X-axis could still be off by 0.2nm, though.

 

Since I knocked the light pipe for the spectroscope out of alignment after measuring the new filter and comparing it with the unfiltered source (the top transmission for the new Antlia filter is at 96-97%), the bandpasses are all calibrated to their own maximum.

 

 

X-axis in Angstrom.

 

Red and blue are two measurements of the new Antlia. Green is a measurement of the old one. The bandpass shape is well nigh identical, but the placement isn't. Doesn't matter on slow scopes, matters on scopes from f/4.5 down (but remember: green is the worst filter over a sample of 6, and if my calibration is off by even 0.1nm then it's still perfect for f/4.5).

 

The blue/red graphs give you good performance at f/3.72 (85% transmission) and still decent performance at f/3.5.

 

Yellow is the Astronomik 6nm Max FR (which indeed has a bandpass setting that sets the emission line at the blue side of the bandpass, and it has a truly colossal blue shift reserve for something that also works on slow scopes.)

 

The calibration corresponds well to other data points:

-the bandpass setting is indeed exactly as indicated on the graphs supplied by Antlia

-the bandpass setting on the Astronomik is as validated by seeing where the H-alpha absorption line appears when looking at sunlight through the filter with a spectrograph.

 

This is a temporary result, I do plan to have some better calibrated results.

Edited by sixela, 02 February 2024 - 02:28 PM.

[sixela]

Antlia 2.5nm vs. Baader 3.5nm.

 

 

Quite good bandpass setting for fast scopes (whether that's an accident or not is unknown).

Edited by sixela, 02 February 2024 - 02:29 PM.

[sixela]

Optolong L-utlimate (one of two, the other has fine passband setting) vs. Antlia 2.5nm (new one) around H-alpha:

 

 

 

 

Yes, the emission line is in the flat top for slow scopes, but there's almost no blue shift reserve for faster scopes (less than even on the two 2.5nm Antlias!) Not good, I'll be trying to exchange it in the hope the next one fares better (the other one that's fine is not mine).

 

In fact it's 1nm (not 2nm) less wide than the ALP-T, but only on the wrong side, making the bandpass centring much worse than on the ALP-T:

 

Edited by sixela, 02 February 2024 - 03:35 PM.

[sixela]

Of course not all Optolong L-Ultimates will be like that, but on scopes faster than f/4.5 you're really playing Chinese roulette.

[Thierry Legault]

Of course not all Optolong L-Ultimates will be like that, but on scopes faster than f/4.5 you're really playing Chinese roulette.

I agree. I have coupled three filters with a Coronado double stack solar filter (0.5A) at F/5 and obtained the transmissions on the Ha line:

 

- Antlia 2.5nm: 91%

- Triad Radian Quad band (4nm): 91%

- Optolong L-Ultimate (3nm): 68%

 

I suspect that the Optolong bandpass is shifted! 

Edited by Thierry Legault, 02 February 2024 - 04:28 PM.

[sixela]

Mhh -- that makes the one out of two that I saw not such an outlier, then, there's even worse :-(. There's also better (Jim Thompson's filter that he tested some time ago has the bandshift to the other side).

Edited by sixela, 02 February 2024 - 05:09 PM.