Camera Labs

Went and made my own version of a spectrascope... much cardboard was involved as well as a random prism I picked up a while ago. This happened:

Caution: I kicked the tripod multiple times trying to get these. I tried to line up the images as best I could but not that well now I see them side by side...


Canon 600D - unmodified


Canon 600D with Astronomik CLS filter.


Canon 450D "full spectrum" modified - all filters removed.


Same 450D with Astromik CLS filter. Note the significant IR response. I need the CCD version which also includes IR blocking.


Same 450D with Hoya R72 filter (IR pass from 720nm)


Same 450D with cheap 850nm IR pass filter


Same 450D with cheap 950nm IR pass filter

Light source was whatever the smaller bulb on a Celestron powertank is. Both cameras set to ISO100 1/1600s "sunny" white balance with lens at f/2.8.

Edit:



I've scaled, aligned and combined the 600D unmodified output (top) and 450D modified output (bottom), along with the RGB channels of each. I've got channel clipping but still I think it is indicative.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi popo,

Thanks for sharing - quite thought provoking.

At first glance the Astronomik CLS filter response pretty aggressive. For comparison here is the plot of an IDAS LPS filter:

Click the image to go to the Hutech website

But it is interesting to compare the Astronomik response with the pass-bands of my Astrodon Tru-Balance E-Series filters:

Click the image to go to the Astrodon website

There is a similarly aggressive notch as seen on the Astronomik filter between 570 and 615 nm (approximately) which is why I can successfully run 1,000 second Green exposures at f/7. Obviously my local street lighting doesn't emit at the IDAS notches at 435 and 545 nm. I have cut my old IDAS filter down to fit my filter wheel but the edges are awfully rough and I've yet to see if it fits well enough to use for luminance work.

So far as astrophotography with the Astronomik filter and a Canon DSLR go it's interesting to compare the filter notch with the spectral responses on Christian Buil's site. The major casualty appears to be the already weak red response.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

The Astronomik CLS filter response is here as pdf. Basically there is a big notch from around 540nm to 640nm to take out Sodium and Mercury emissions. It perks up again in time to pass Ha and friends though. The IDAS LPS manages to squeeze in two narrow passbands in that region but it also means the transmission isn't quite zero where the LP emissions are. Curious tradeoff.

I find Christian Buil's response a little confusing now I think about it some more. His charts show the response dropping to zero at 700nm even for the IR block filter removed case. That is clearly incorrect, and I assume either it is a limitation of the measurement equipment used, or there remains some UV/IR block filter as commonly used in astro modifications. For a chart that appears to match what I see with my modified 450D see here.

My reason for doing this testing was to get an indication of what the filter and sensor combinations do. I know the CLS does not block IR, but how bad is it? Now I know. I should have paid a bit more for the CCD version initially which is supposed to have IR blocking. However my experience shows that even with a CLS filter in place, broadband visible light imaging is not viable from my location for anything other than the brightest objects.

The other part was the performance of IR pass filters. Research so far has shown there is a spike from Sodium lighting in the IR region around 818nm, which if I read things correctly elsewhere, is of approximately 12% the intensity compared to that of the visible emissions. I was wondering if broad band imaging was possible in the NIR region to get around light pollution, and I would have to avoid that emission. That's where the 850nm IR pass filter comes in, and I found a dirt cheap one on ebay. But does it do what it is supposed to? Seems to so far. Next step, I'll have to try getting a spectrum of the sodium lights outside when it gets dark, and see 1: if the 818nm spike there and 2: does the 850nm filter block it? In case you're wondering, I got the 950nm while I was at it in case it comes in handy later.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

And this is the response of the streetlamp on the road outside my house. The scale can't be directly compared against the earlier ones since the distance between camera and prism isn't the same. I guess I can stretch it, but how much?

It's definitely stretching into the IR though by the pink, with red-ish bands between that and the visible yellowish ones. I was rather stupid and forgot to try the 850nm IR filter, which I'll leave for another night now.

Also I found a flaw with my earlier measurements. I forgot that these lenses are not corrected through to IR, so the focus is off if set for visible. To get both more or less in focus I had to stop down to f/8 for the above, but diffraction effects were already kicking in.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.