3nm Ha versus 5nm Ha

[gregbradley]

I have been quite happy with my 5nm Ha Astrodon filter. But it is slightly damaged and I need to replace it.

I am wondering if I should look at getting the 3nm Ha filter in its place.

I will be using it at F3.8 so that is factor. Do these 3nm Ha filters work ok at fast F ratios?

Is there much of a difference between 3nm and 5nm Ha?

Is the difference more for when the moon is up filtering out more moonlight?

My main concern is whether its suitable at F3.8.

Greg.

[strongmanmike]

Ah huh??? you got an RHA on its way or something...? :)

Mike

[LewisM]

Peter sells AP R-H...

Greg sold AP Starfire and TEC...

Join the dots Mike :)

[Slawomir]

One question remains: will Rick sell his guide-scope -FSQ106... ;)

As for the original question, I have 3nm Astrodon Halpha and using it at F5.6. There is a HUGE difference between 12nm Astronomik and 3nm Astrodon (also cost-wise). After subtracting bias there is literally hardly any unwanted signal left in 5-minute subs from my horribly light polluted site (yellow city's night-sky).

Unfortuantely I cannot comment on a difference between 3 and 5 nm...nor on above-mentioned filter's effectiveness at F3.8... :(

[clive milne]

Don Goldman intimates that 3nm filters are useful and represent an improvement over 5nm filters down to f3.5.

That is different from saying that the efficiency of 3nm filters is not unchanged at focal ratios as fast as f3.5

Bandwidth versus f ratio (from Alluxa):
http://www.alluxa.com/white_papers/Ultra-Narrowband-Optical-Bandpass-Filters-With-Large-Format-and-Improved-Temperature-Stability-fig5.jpg

[SpaceNoob]

No doubt you will lose NII? I went with 5nm to keep it wide enough for Ha and NII. My OIII and SII are 3nm though and they work very well.

[gregbradley]

]


Yes a new one. Roland just finished making it and its on its way. Perhaps in a week. I got notified back in January letting me know my name is close on the notification list and double checking I still wanted one. That it was no guarantee you'll be offered one.

I said yes and then nothing until less than 2 weeks ago.

Hence the sale of the high end gear which I normally would never part with.

Should be interesting.

I intend to equip it with a FLI Atlas focuser, a MMOAG, FLI 5/7 filter wheel (the 4/5 is a pain when you want to image narrowband swapping over filters) and use the Proline or the Trius 694 (for galaxies). I may get a Trius 814 if the 694 works out. QSI camera look interesting with their all in one approach as well.

I think the same as you in that 12 inches F3.8 is close to an ideal format for non remote imaging where you have to setup and remove. Also time at the observatory being limited, then clouds etc. So when you image it really needs to count. 12 inches aperture and F3.8 is still manhandleable and is fast as you'd want to go (no collimation required on a Honders).

First light is not far away. I just have to sort out an Atlas focuser adapter.
I think it needs to be about 40mm long and I may already have something that may be suitable.

Greg.

[gregbradley]

Did Peter sell his AP RHA? I don't think so but could be wrong.

Greg.

[gregbradley]

I think I'll stick with 5nm for now.

Mike you have some 3.5nm Ha etc coming right? I guess that will be an indication.

What concerned me was Berts (Avandonk) 3nm Ha with the Officina Stellare RH200 often had blotchy backgrounds and I wasn't ever sure was it the processing or the result of the filters not matching the very fast F3 scope. Its the look you get when you black clip a Ha image. it goes splotch in the background and stars look speckled.

Greg.

[Slawomir]

That is going to be super-awesome astro gear Greg!
Fingers crossed all will go smoothly and soon you will be treating us with all-inspiring astro-images :)

[strongmanmike]

Awesome Greg :thumbsup:...a rather nice (all be it very expensive) surprise that must have been :) What will be the final landed OTA price anyway?

Re filters, I have a brand new set of the new series 6nm SIIHaOIII Astronomik's here ready to install...I'll whack'em in after I finish my current imaging project :) They are all rated to better than F3.8 with very high transmission.

I'm sure you will be fine with the 5nm Astrodons :thumbsup:

Mike

[RickS]

Greg,

As Clive mentioned, Don Goldman says his filters are good at fast ratios. I think the latest FAQ says they are OK at f/3 but lose 20% transmittance. At f/3.8 the loss would be small but if you're at a dark site a 5nm (or 6nm, Mikey!) would be nice for planetaries with NII.

Slawomir: I'm keeping the guide scope :lol:

Cheers,
Rick.

[SamD]

In theory ...

The combination of the loss of most of the NII, and say a reduced transmission of 90% at 3nm with f/3.8, gives little or no improvement in SNR by my quick calc below. Mind you, the "signal" with 3nm is actually going to be almost all Ha !

Near the centre of M42 the some relevant line intensities are:
654.8nm 10.7 [NII] - Lose half of this in 3nm filter vs 5nm ?
656.3nm 171.5 [Ha]
658.4nm 32.5 [NII] - Lose all this in 3nm filter vs 5nm,
so lose 0.18 of signal detected in 5nm

---
5nm
---
signal 25
skyglow 25 (so square root gives noise = 5)
SNR 5

---
3nm
----
signal 25 * 0.82 (since lose most of NII) * 0.9 (estimate loss due to f/3.8 on 3nm filter ?) = 18.45
skyglow 25 * 3/5 (skyglow proportional to FWHM) * 0.9 (transmission of skyglow also decreased by f/3.8)
= 13.5 (noise 3.67)
SNR 5.02

SNR assumes that skyglow noise is dominant over read noise and thermal noise in both cases, ie sub lengths would need to be approx double on 3nm filter.

[gregbradley]

Thanks Sam. That's a good workout.

It seems 3nm is more suited to light polluted locations and at a dark site unless you want to have pure Ha only there may not be much advantage.


Greg.

[clive milne]

There is a caveat that probably needs to be stated..

Bandpass shift (in interference filters) is a function of the angle of incidence of the light passing through them, which is not always the same accross the gamut of optical systems of the same focal ratio.

Newtonians with a mild (0.95x) focal reduction in the corrector are probably a best case example where the chief ray is parallel to the optical axis.
The following diagram might help illustrate an example where the angle of incidence is actually double what the effective focal ratio might suggest:
http://boson.physics.sc.edu/~rjones/phys153/raytracelens.GIF

fwiw) Here is the fsq106 optical layout for comparison:
http://www.takahashi-europe.com/images/products/fsq106ed/700/FSQ-106ED_lightpath_700.png

[clive milne]

One further point that probably deserves mention wrt extragalactic emission line sources; There will be approximately 1nm of bandwidth shift per 21Mly for Ha & 27Mly for Oiii

You can see that 3nm filters are pretty much out of band for objects as close as Virgo cluster members.

2c

[strongmanmike]

Well...there you go, never thought of that :) so collecting a Ha data set to highlight the HII regions in galaxies outside our local group, using a 3nm Ha filter, will show nothing..?

[clive milne]

Well, to be strictly correct, the bandpass shift is proportional to the recessional velocity which is more or less corellated with distance.

I would expect 3nm filters to be fine out to 30M light years, give or take, without much loss, but at 42Mly the Halpha line is actually shifted to 658.3nm (bang in the middle of a typical Nii filter transission curve, centred on the strong Nii line)
Around this point however, the weaker 654.8nm Nii line will be shifted to 656.8, so you will record it through the Halpha filter... somewhat ironic, hey?

If you are using a faster focal ratio telescope, the higher angle of incidence in the optical train will compound the issue so you will experience a higher reduction in the emission line signal.

[graham.hobart]

So Greg, will you be selling your sx filter wheel and OAG???!!!!!!:D