Hb + Unmodded Cam versus Ha + Modded Cam

[Ags]

Based on my absolutely terrible attempts at AP, I would say that my unmodded camera lets through equal amounts of Hb and Ha - obviously there is 3-4 times as much Ha coming in, but the Canon IR filter puts a stop to that.

So let's say I mod my camera and get a 6nm Ha filter (EUR600 in total) - in this situation I will collect 3-4 times as much Ha as Hb - a clear winner! Let's do it!

But wait...

This is forgetting the Bayer matrix, so my boosted Ha light is falling on 1/4 of the pixels...

...and on my unmodded camera the Hb light falls on 3/4 of the pixels.

If that's right, modding the camera will only gather as many Ha photons as an unmodded camera using a Hb filter!

Surely I am wrong? Does the EUR140 Hb filter really match the EUR 600 modding+Ha route?

[davew]

Agnes.... I find your logic most compelling.

But wait....

How do you know the two green pixels will each collect the same amount of light as the blue ?

Does the Hb filter attract the moonlight like an O111 when most Hb filters also have a wider pass band ?

Why not a cheap 7nm filter ?

Self mod for free ?

I await your results

Dave.

[Ags]

I was wondering if the Hb would get transmitted equally on blue and green - as it's on the cusp of both, perhaps you get poor transmission on both types of pixels.

You're right that there are cheaper Ha filters around, and as the Hb filter that caught my eye is 8.5nm, a 7nm Ha is a fairer comparison.

I would not dare mod my camera myself, even though it is a lot of money.

[davew]

Actually I agree on the self mod. I can be really brave and say have a go but never having done it myself, I won't.

What you say is in fact very interesting. I wonder if you'll find a volunteer to try an Hb filter ? Or could you " Borrow " one on a try before you buy ?

Not trying to persuade you either way but there do seem to be a few modded 450D cameras about at the moment. A member here does them. Not sure if he wants to be named though. Around £ 250 / 260.

Dave.

[Ags]

Even if the Hb route is gathering the same number of photons, the signal per pixel is three times less, so perhaps the signal is lost in the noise.

I read sometimes that only certain nebula have Hb but from my understanding, Ha and Hb are always in a mathematical ratio, effects of dust notwithstanding.

[davew]

You may be right about the dust. Hb may get through better than Ha, being shorter. I don't really know.

If a great chunk of Hydrogen gets zapped then I think it'll emit at all the wavelengths. Only those two are properly available to us mere mortals though. The rest go into the UV range.

Dave.

[Ags]

I think the Ha would get through dust better, being longer... after all, IR scopes can 'see through' dust.

[davew]

Just testing

I've never heard of anyone testing out the ratios through dust.

Another thing for you to do. Don't ask much do I ?

Dave.

[johnrt]

Hb has the same dispersion as Ha, but equates to much less signal, I did see a comparison of the wall in the NA nebula done in Ha and Hb and it looked the same, only much dimmer in Hb. I can't for the life of me remember where I saw it though.

If collecting Hb with an unmodded camera was as good as Ha with a ir filter removed camera, everyone woud be doing it. Maybe the fact nobody is tells the story.

[Ags]

THat's the interesting thing - the nebula will be three times dimmer in Hb, but is there a benefit from three times as many pixels contributing a dim signal to the stack?

[Merlin66]

OK, I can help....

I use the spectroscope occasionally with the modded 1000D (and the unmodded 450D)

In theory...the Hb emission is close to 50% the Ha emission - so wherever you see Ha, Hb is there at 50% the intensity.

The camera response curves (due to the Bayer) show a "dip" around 550nm...the response is always higher around the OIII/Hb region than in the Ha - probably (need to check for you) double.

So, in summary, the modded DSLR will record twice as much Hb as Ha - but as there's 50% less Hb they appear as equal......

(forgetting the resolution issue....)

Hope this helps.

[Ags]

In answer to my question about how much Hb gets through the Bayer matrix, i found this:

Obviously it is very rough but it seems that about 40% of the Hb (486nm) would get through on green pixels, and near 100% on blue. So the Ha would win with over 50% more photons getting through (assuming Ha is 3x Hb).

+1 to me for getting the BBCODE for that image right :-)

Merlin66 - our posts crossed. If Hb is actually 50% Ha then things look better. And I see your point about Ha - the bayer matrix really kills Ha (656nm) according to the above graph, so the Hb looks good once more.

Ov course, there is the small matter that no-one is doing the Hb imaging thing, so my sums are probably just fantasy.

[Merlin66]

Agnes,

Caution....Hb will fall on either the blue or the green pixel not both!

So pixel A will record the Hb say in the green, and pixel B next door will record it on the blue. The resulting Hb "image" will then be a collection of green and blue pixel data both at 40% - against the Ha shown at 20%

Hope this helps.

[davew]

50% is not how I've understood the ratio in diffuse nebulae.

See this article - http://articles.adsabs.harvard.edu//full/1994JBAA..104..231B/0000234.000.html

Page 234. Is this not correct ?

It could be that no one else images in Hb with a non modded camera because they haven't thought of it ?

Dave.

[Ags]

Agnes,

Caution....Hb will fall on either the blue or the green pixel not both!

So pixel A will record the Hb say in the green, and pixel B next door will record it on the blue. The resulting Hb "image" will then be a collection of green and blue pixel data both at 40% - against the Ha shown at 20%

Hope this helps.

Well, both blue and green will have a signal and can be used as data sources in a stack - both the green channel and blue channel would be valid Hb data and could be combined to reduce noise, methinks.

[Merlin66]

Dave,

Well true and not true....

The Hydrogen emissions follow a strict sequence - The Balmer Series. This series defines the wavelengths and relative intensities of all Hydrogen emissions (or absorptions).

http://hyperphysics.phy-astr.gsu.edu/Hbase/hyde.html#c5

You'll see that Hb is approx 50% of Ha.

In the various emission and reflection nebula/ planetary neb etc the ratio probably varies due to the surrounding conditions - dust etc.

So, yes in some the Hb may well be stronger/ less strong than in others....

[Ags]

This is so strange - I am starting to think I really should just get a Hb filter and save my pennies, rather than modding!

[Merlin66]

Agnes,

Certainly worth trialing...map one image to Blue (Hb) and one at twice the intensity to Red (Ha)........

[davew]

Dave,

Well true and not true....

The Hydrogen emissions follow a strict sequence - The Balmer Series. This series defines the wavelengths and relative intensities of all Hydrogen emissions (or absorptions).

http://hyperphysics....se/hyde.html#c5

You'll see that Hb is approx 50% of Ha.

In the various emission and reflection nebula/ planetary neb etc the ratio probably varies due to the surrounding conditions - dust etc.

So, yes in some the Hb may well be stronger/ less strong than in others....

Oddly enough I saw that diagram over a year ago ( Having first gone to wiki ). I had presumed, may be incorrectly, that the experimental laboratory measurements were as they were because the intensity was measured a matter of inches from the tube. I also presumed the intensity of Hb from diffuse nebulae had been measured at approximately 25% to 30%, in most cases, from the telescope. If you've done work on this yourself then I'll go with what you say.

Give it a try Agnes but don't go spending loads of cash !

Dave.

[Ags]

Here is a link to some more graphs of the transmittance of bayer matrices.

http://www.astrosurf.com/buil/50d/test.htm

These ones change the story again - it looks like the first graph I gave showed a combined transmittance for bayer matrix plus the IR cut filter, so it was a bit misleading. This link has graphs showing the transmittance between modded and unmodded cameras, si it suggests that a mod gives a Ha transmittance equal to a Hb transmittance. And a humbling comparison of DSLR to CCD.

I'm not sure I believe it because Canon only gives a Ha transmittance of 30% for the 60Da, which is presumably all a home mod could achieve too.

[Merlin66]

Buil's graphs are based on actual measured performance.

In my fully modded 1000D the Ha response improved by almost 100% - 20% to 40%....

[Ags]

Ah, I see... For the 60Da Canon claims triple Ha sensitivity, with a 20% increase. Which would imply a jump from 10 to 30%.

Thanks for all of your input, it has given me a lot to think about!

It is now far too late and I seem to be getting more and more confused. I think I will cling blindly to the idea a Hb filter might just work a little better than my UHC-S filter.

[Gina]

Very interesting thread I didn't know the H beta signal was as much as half the H alpha until I read this thread. Provokes a few thoughts. Not only regarding DSLRs but for CCD mono cameras and NB. I might change my colour mapping from bi-colour NB with my 314L+ maybe add some Ha to the blue channel. ATM I'm using Ha to red, adding Ha & OIII to form a synthetic green and OIII to blue. More "realistic" might be Ha to red, OIII + 0.25 Ha to green and OIII + O.5 Ha to blue. Or something like that.

[davew]

Further to last night's information I decided to look into it further.

It would appear that the figures from the lab say the ratio is around 44%. Sounds a bit high to me but there it is.

The figures from the Brazell article, citing NASA as the source, are around 30%

My own investigation of The Astrophysical Journal would indicate quite a range. One particular paper I read had a number of observations in a Cygnus region. These ranged from a mere 22% upto almost 40%. Checking all the figures it seems that most observations were in the mid to high 30s.

Now if asked I would say on average it's likely to be around 35% +.

Gina. The fashion used to be Ha to red, O111 to green and O111 mixed with about 15% to 30% Ha for blue ( to taste ). I think synthetic green comes from RGB imaging and not narrowband.

Now I need to lie down and recover.

So Agnes, from my point of view it's a good thread but doesn't answer. " Is Hb affected by the moon like O111 ? "

Dave.

[Gina]

Gina. The fashion used to be Ha to red, O111 to green and O111 mixed with about 15% to 30% Ha for blue ( to taste ). I think synthetic green comes from RGB imaging and not narrowband.

My last go on M42 and surrounding area was much the same but with 33% Ha in the blue - I may try it with less Ha.