Narrow Band Filters - 3nm vs 7nm etc

[Midnight_lightning]

I am going to upgrade my camera (separate post) and will need larger filters. Scope 135mm (FL 686mm/915mm depending on Reducer/Flattener)

I currently use a Baader NB set of 7nm filters (Ha OIII SII) but thinking I may splash out on some Astrodon or Chroma filters and wondering about the merits of 3nm - especially for OIII. 

We have Bortle 4 skies here but the rate of building in the area means this will deteriorate.

I have seen images taken with 3nm filters and they do seem to have more depth/contrast and particularly with OIII handle large bright stars well.

But presumably, if they are letting less light through, I will need longer exposures (?)

I'm out of my depth knowledge wise when it comes to filters so would be grateful for any  thoughts about the relative merits and disadvantages of the various bandwidths.

 

 

 

[Skipper Billy]

Following.

[DaveS]

3 nm filters won't let less of the wanted light through unless their peak transmission is lower, Astrodon and Chroma both quote above 90%. Myself, unless you really want [NII], which is really a PN emission line, then I would go for Chroma as they are significantly less expensive while being just as good.

[newbie alert]

Anyone use Antlia? 3.5nm ?

[Ken82]

I think it depends really on how bad your light pollution is ?? Narrower the filter bandwidth the better the SNR. 

From what I learnt before purchasing, The 5nm chroma filters are generally Considered the best option for most circumstances. Unless of course you have quite  bad light pollution like me then the 3nm filters are better. 
 

Absolutely the  biggest improvement for me between baader 7nm and chroma 3nm was the lack of halos. This is also true of the rgb filter set I have. 
 

Just be careful with f ratio as I think they are only recommended for f4 and slower. 

[Skipper Billy]

So what is the downside to choosing the 3nm ones as the 3nm and the 5nm are a very similar price ??  I dont suffer from any significant light pollution.

I am just trying to be absolutely certain before spending 2.5k on three tiny bits of glass !!! 🙂

 

[DaveS]

Disadvantage would be that the 3nm H-alpha will exclude the [NII] line which otherwise would add signal to the HII. Of course, you may want to separate the two in order to get a purer HII signal.

Chroma don't manufacture a 3nm [NII] filter, though Astrodon do, and should work with the Chroma. But unless you are devoted to imaging Planetary Nebulae I have found [NII] to be of questionable value.

Chroma also spec their 3nm filters for F ratios above 4, below that they recommend the 5 nm due to spectral shift.

 

Edit: 3nm is more resistant to that other source of LP that we all have to deal with, namely the Moon, and could give you a few more night imaging with the moon up.

Edited August 28, 2020 by DaveS

[Skipper Billy]

Thanks for that information - sounds like the 3nm are the ones for me (I am imaging at f5).

 

 

[Ken82]

15 minutes ago, Skipper Billy said:

Thanks for that information - sounds like the 3nm are the ones for me (I am imaging at f5).

 

 

I also made the same decision but based more on my light pollution. 
 

I believe the chroma guys are on record as saying “in most circumstances the 5nm are the right choice” so this made me really think about it. 
 

 

[Midnight_lightning]

6 hours ago, Ken82 said:

I also made the same decision but based more on my light pollution. 
 

I believe the chroma guys are on record as saying “in most circumstances the 5nm are the right choice” so this made me really think about it. 
 

 

I'm tempted to say the Chroma guys would say that given they don't make a 3nm but maybe that's a bit unfair - perhaps that's why they dont make one. 

That said for the relatively small increase in price it would be good to be able to image when there is a moon - probably double the number of imaging nights I get a year.

 

[Ken82]

21 minutes ago, Midnight_lightning said:

I'm tempted to say the Chroma guys would say that given they don't make a 3nm but maybe that's a bit unfair - perhaps that's why they dont make one. 

That said for the relatively small increase in price it would be good to be able to image when there is a moon - probably double the number of imaging nights I get a year.

 

Chroma make both 3nm and 5nm and they are the same price. 👍

[kirkster501]

The 3nm Astrodon Ha is the best filter I have.  I have it in both 1.25" and 31mm.  It is a spectacular filter on Ha objects.  You could just image with this one filter alone.

You have to swallow hard when pressing the buy button so it is a "are you in or are you out" moment....  I got both mine when Ian King was closing down and I got 20% off.

Edited August 28, 2020 by kirkster501

[billhinge]

45 minutes ago, kirkster501 said:

You have to swallow hard when pressing the buy button so it is a "are you in or are you out" moment....  

Glad  its not just  me that thinks that way 🙂

[x6gas]

I have the 3nm OIII Astrodon and the 5nm SII and Ha.  I must confess that's because that's what Ian King recommended and I trust his judgement.

[DaveS]

I have a full set of Astrodon 3nm in 1.25", bought at the end of 2014 when they were "0nly" £385 each, now they've gone beyond silly money into stratospheric insanity. Still brilliant though if you can afford them.

[Ken82]

Here is a halo comparison I did earlier. 
 

same magnitude star -

Baader 7nm and RGB filters - 120 pixel halo 

IDAS D1 and D2 - 160 pixel halo 

Chroma RGB and 3nm - zero halo 

 

For me the removal of the halo is enough to justify the price even if it doesn’t improve anything else. 

Edited August 28, 2020 by Ken82

[Jkulin]

4 hours ago, Midnight_lightning said:

I'm tempted to say the Chroma guys would say that given they don't make a 3nm but maybe that's a bit unfair -

Hmm, I own 2" Unmounted 3nm Chromas which are certified at 3nm so not sure where you heard that?

Edited August 28, 2020 by Jkulin
typos

[ollypenrice]

I've used the 3 and 5 Astrodons and the 7 Baader in Ha. For me the 3 Astrodon is the clear winner. It gives the most striking contrasts and the smallest stars. Would I pay the Astrodon price for the 5? I'm really not sure...

The Baader 3 isn't slow, it's actually faster than my Baader 7. As said earlier, it's 'wanted light' we're after and good throughput.

Halos afflict two Baader OIII's I have and an Astronomik (which is supposed to be a no-halo replacement for an earlier one. Sorry, it's hopeless.) The Baaders are 'less bad.' I agree that for this reason alone it's worth going up-market for OIII.

Olly

[Midnight_lightning]

14 hours ago, Ken82 said:

Here is a halo comparison I did earlier. 

That's really useful Ken. The price of these things is crazy but having spent thousands on a mount, cameras, telescope and goodness knows how much time learning the trade I'm not going to put up with halo's when I dont need to

 

 

[Midnight_lightning]

12 hours ago, Jkulin said:

Hmm, I own 2" Unmounted 3nm Chromas which are certified at 3nm so not sure where you heard that?

Someone posted it above, I should have checked it was accurate before propagating it - apologies

[vlaiv]

http://www.astrosurf.com/brego-sky/A_glimpse_to_the_night_sky/Articles_files/filters.pdf

Quote

For fast telescopes (faster than f/4), filters narrower than 13nm will be degraded. For slower systems a narrower filter is recommended as it enhances the effect of the filter.

This is equation:

For 3nm filters, we need to have maximum shift of 1,5nm at 656.3nm.

654.8/656.3 = ~0.99771446

Now we square that and subtract from 1 we get ~0.0045658566.

Refraction index of air is 1.000293 and that of glass is around 1.5, so we have 0.445 * sin^2 angle = 0.0045658566

From this, we have sin angle = ~0.101356 = 0.1

Angle is 5.74 degrees.

This gives approx ratio of ~4.975 or F/5.

Interesting exercise. I expected longer F/ratio needed for such narrow filter, but it turns out that 3nm filters are good for F/5 and higher. They can certainly be used on even faster systems with slight loss of aperture due to filter acting as aperture stop. I also calculated for on axis beam - edge of field beam will have additional angle added that depends on focal length and size of sensor (but is relatively small in comparison).

I started this post believing that there would be significant difference between 3nm and 7nm on fast optics, but it turns out for regular F/ratios that we use in imaging, 3nm is just fine.

[Adam J]

I have AD Ha Sii and Oiii all in 5nm I used to have Baader filters, the 5nm are a significant upgrade just due to halos but in my case I would say the image was also brighter in Ha with better contrast leading me to wonder what the peak transmission on the Baader ha was. I am under class 5 sky and I dont feel the need to upgrade to 3nm, however I paid £180 for my 0iii used but as new, £250 for the Ha new, and £330 for the Sii new. I acquired them in that order, at the current prices I would give AD and Chroma a miss unless you have a tone of spair cash and nothing else to spend it on. 

Lol just worked out that at current prices I could sell them second hand and actually make a profit....

Adam

Edited August 29, 2020 by Adam J

[Midnight_lightning]

3 hours ago, vlaiv said:

I started this post believing that there would be significant difference between 3nm and 7nm on fast optics, but it turns out for regular F/ratios that we use in imaging, 3nm is just fine.

Really helpful, many thanks vlaiv

[Buzzard75]

On 29/08/2020 at 06:12, vlaiv said:

http://www.astrosurf.com/brego-sky/A_glimpse_to_the_night_sky/Articles_files/filters.pdf

This is equation:

For 3nm filters, we need to have maximum shift of 1,5nm at 656.3nm.

654.8/656.3 = ~0.99771446

Now we square that and subtract from 1 we get ~0.0045658566.

Refraction index of air is 1.000293 and that of glass is around 1.5, so we have 0.445 * sin^2 angle = 0.0045658566

From this, we have sin angle = ~0.101356 = 0.1

Angle is 5.74 degrees.

This gives approx ratio of ~4.975 or F/5.

Interesting exercise. I expected longer F/ratio needed for such narrow filter, but it turns out that 3nm filters are good for F/5 and higher. They can certainly be used on even faster systems with slight loss of aperture due to filter acting as aperture stop. I also calculated for on axis beam - edge of field beam will have additional angle added that depends on focal length and size of sensor (but is relatively small in comparison).

I started this post believing that there would be significant difference between 3nm and 7nm on fast optics, but it turns out for regular F/ratios that we use in imaging, 3nm is just fine.

Vlaiv, thanks for this. I followed your math up to this point: "Angle is 5.74 degrees. This gives approx ratio of ~4.975 or F/5."

Solving for sin angle = 0.1 gives me an angle of 174.26. I'm assuming to get the angle you have, you subtracted that from 180, giving you 5.74. While I understand how you got that number, I'm not quite sure why you did it other than the fact that 180degrees is coplanar to 0 and you're just looking for an angle of incidence to that plane. The real question I'm trying to understand is the ratio and where that comes from. I'm assuming its the ratio of 5.74 degrees to some other angle, but I'm not sure what other angle.

[vlaiv]

3 hours ago, Buzzard75 said:

Vlaiv, thanks for this. I followed your math up to this point: "Angle is 5.74 degrees. This gives approx ratio of ~4.975 or F/5."

Solving for sin angle = 0.1 gives me an angle of 174.26. I'm assuming to get the angle you have, you subtracted that from 180, giving you 5.74. While I understand how you got that number, I'm not quite sure why you did it other than the fact that 180degrees is coplanar to 0 and you're just looking for an angle of incidence to that plane. The real question I'm trying to understand is the ratio and where that comes from. I'm assuming its the ratio of 5.74 degrees to some other angle, but I'm not sure what other angle.

We have two things here - square / square root and sin

Square root of something is not only positive value - its both positive and negative value (it has two roots). Then, there are some identities:

This means that you can take small angle as correct answer even if you've got large answer from arc sin in calculator

In fact it is +/- 5.74 degrees.

Now if you want F/ratio - you need to use tan of that angle.

Tan(5.74 degrees) = 0.1005184.... and inverse of that is 9.948426789....

However this is "half F/number" as incident ray comes at 5.74 degrees on one side and 5.74 on the other side - so above 9.948.... is ratio of radius of aperture to focal length and not diameter of aperture to focal length. We need to divide that with two, so we get: 4.97421339... = ~ F/4.975 = ~F/5

Makes sense?

Adjacent is focal length and Opposite is radius of aperture in our case.