DSLR full spectrum pros and cons

[traveller]

Hi all, I have been happily using a Baader modified canon 40d.
Now I am tempted to get the canon 1100d and modify that as well.
I am tempted to do a full spectrum mod for the 1100d, either with a replacement clear glass or just leave the sensor exposed.
Obviously if I replace with a clear glass I preserve the refractive index and therefore the auto focus function, the other advantage being keeping the sensor clean.
What are the other advantages are there?
A naked sensor is more vulnerable but also cheaper, but misses out on the auto focus.
Are there anything else I should be aware of?
What brands clear glass replacement filters would members recommend?
Cheers,
Bo

[MrB]

Full spectrum mod doesn't actually leave the sensor exposed as you might expect, the sensor still has a protective clear glass cemented directly to it. In the 1100D this has proved virtually impossible to remove. Earlier bodies can be removed with little or some effort. You will not need to remove this glass unless you wish to attempt a mono conversion too.

[Merlin66]

Bo,
I have and use a FULL full spectrum modded 1000D camera. Both filters removed.
There's no real issues with a "bare" chip..
I use a Clear Astronomik clip-in filter to maintain focus with the standard EF lenses. It also keeps the dust out!
Great for spectroscopy.....

[traveller]

Thanks ken and Simon,
I intend to use the 1100d as a dedicated astro cam, so was only worried about the exposed sensor. Didn't know about the glass protector over the sensor, I take it is a full bypass glass and not filtering out any IR or UV.
Judging by your comments, I just need to remove both the clear IR/UV bypass filters and the blue filter and that should do the trick.
Bo

[MrB]

That is correct.
There is some debate about potential star bloat from refractive optics and UV/IR, so you may want to try with the UV/IR filter left in and removed and then choose which works best for you :thumbsup:

Edit: Just re-read and realised my comment could be misinterpreted....
"from refractive optics and UV/IR"
By this I meant that some optics(including reflectors with corrective optics) may not be well corrected at the UV and IR ends of the spectrum and may cause bloat without the filter. Not that bloat was caused by the UV/IR filter ;)

[rcheshire]

Like Ken, both filters removed, plus an Astronomik clip-in UV/IR filter for use with camera lenses - maintains focus - 1000D.

[rustigsmed]

i have clear glass modified 600d, definitely degraded captures of the planets and introduced star bloat (powermate or coma corrector induced), but mostly corrected with a clip in eos UV/IR filter.

[amitk75]

I have some confusion: I modified my EOS 1100D yesterday. I just removed the filter assembly which was screwed in on top of the sensor (i.e. I didn't use knife etc to remove anything stuck). Even I have doubts on the clear glass stuck on the sensor. If it's not blocking anything, am I safe to assume that I have full-spectrum modification? One doubt is that most of the posts say that auto-focus wont work without a replacement filter/glass, but its working fine in my camera.

[Merlin66]

Amit,
It sounds like you have only removed the front filter element, there is a second colour correction filter mounted on the chip.
For the full spectrum mod, this also needs to be removed.

[glend]

The blue cast filter is the one to remove for full spectrum mod, you do not have to remove the clear anti-alysing filter. Suggest you have a look at Gary Honis' excellent online tutorial on how to mod your camera. Gary's instructions cover many of the popular Canons and all of the filter setups are the same.

[Merlin66]

Glen,
To achieve a FULL full spectrum mod both the AA filter (which blocks the UV/IR) and the colour correction filter need to be removed

[glend]

Merlin, that is not true according to Gary Honis' information. The AA filter (also known as Low Pass 1) is clear and can be left as protection for the sensor and in terms of the spectrum has negligible effect. The AA filter is not a UV/IR block, that function is performed by the IR Cut filter (the blue cast one). In fact Gary recommends that you install a Baader IR Cut which has the cut point down below the Ha frequency (this can be done with a clip-in or with a 2" screw in on the T-adaptor - which is the one I use. Details on this page undre Filter imaging chip disassembly.



http://dslrmodifications.com/rebelmod450d4.html

[Retrograde]

Hope this is not too much of a thread hijack but I was wondering if removing the AA filter is a worthwhile mod in itself?
Would you get any worthwhile increase in resolution on astro targets?

[glend]

IMHO you will lose the sensor cleaning protection that the coating on the low pass1 provides. Normally the auto piezo vibration is tuned off when imaging but you can manually do it from the menu. The surfac coating is suppose to repell dust. If you have nothing there then you risk getting the sensor face dirty at the very least.

[Merlin66]

Glen, et al,
I 've previously posted detailed spectral data on both the filters used in the Canon cameras.
I can assure you that the front AA filter does indeedv act as a UV/IR filter, no matter what Honis says.

[amitk75]

I am confused :confused2:

I have attached picture of the filter assembly I took out.

If the glass I left inside is indeed a UV/IR filter, any idea which wavelengths will it block? Is there any test I can perform to measure what I have achieved and what I am missing?

I was checking the IR sensitivity earlier just to see if I have really unblocked IR, I pointed the camera to a TV remote with and without filter and it appears like a flash without the filter.

I need to weigh now if it is worth opening the camera again to take that glass out or no.

[glend]

Perhaps reposting the link to that your spectral data would be useful. Rudy Kokich tested one of Gary Honis' full spectrum modded Canons and compared to stock early this year using the GRISM spectroscope and published the spectral performance here:

http://dslrmodifications.com/DSLRcomparison.html

just scroll down to spectral chart on that page.

[AlexN]

Just thought I'd chime in here. I'm about to mod a 450D for astronomy only duties and intended to remove all the filters (aa and uvir hot mirror.) I do however intend to image primarily with my 300/2.8 (nikon mount with adapter focused via home made ascom stepper focuser.) lens. Will I not be able to reach focus with this setup? I will also be imaging with my 5" newt a bit..

My other intention was to get an astronomik clip in ha filter for imaging in the suburbs.


Anyone got any tips or advice.

[Merlin66]

I can't seem to link to a previous post with the iPad......
You will need a clear clip in filter to maintain focus with standard lenses.
Glen see the post on uv -ir imaging

[AlexN]

To maintain af accuracy or to retain the ability yo reach focus at infinity?

[amitk75]

I found a video for 1100D modification on youtube: https://www.youtube.com/watch?v=gQyT0XoryW4

Unfortunately I dont understand the language. But what I can see that after the filter assembly is removed, there is a clear glass on the sensor frame: see the video at about 11:33. Somewhere I read it is a protective glass. The guy in the video is saying something about the glass, but cant understand it :(

As per the diagram given earlier, The LPF1 should be the first filter facing the lens? If yes, then it will be part of the filter assembly.

I my case, I just removed the filter assembly completely, which makes me believe that there is no filter left now.

[Merlin66]

Alex,
Just for the focus.

[AlexN]

Ahh, well that's awesome. I'm not concerned as my 300/2.8 and my 500/4 are both manual lenses anyway and are being modified with computer controlled focusers. I would have been pretty upset to find myself unable to reach infinity focus after modifying the camera.

[AlexN]

Amit. There are three filters in the assembly. The front most of them is an anti aliasing filter)(essentially robs you of sharpness but prevents a few artefacts that are inconsequential to astrophotography. In my opinion, removal of this is beneficial. The next is two filters sandwiched together that make up the uv/ir block.. This needs to be removed to sucessfully complete the mod.

[rcheshire]

Reflective optics should achieve focus with both filters removed, unlike refracting optics. I used a 1000D for several years with both filters removed and a clip-in UV/IR filter with no loss of auto-focus or infinity using a Canon lens. Some cameras are not so sensitive. The 1000D and 450D are almost identical in structure.

[amitk75]

Alex,

So what you means is that when I remove the filters assembly, i have removed all 3 filters right? And the front means facing the lens?

Is there any test for checking the antialias filters presence or absence?

Sorry if I am sounding dumb and asking same questions again and again... Just need to be sure on what i have done

[rcheshire]

http://www.iceinspace.com.au/forum/showpost.php?p=1170915&postcount=39

Is this helpful visualizing the filter assembly? It's a Canon 1000D sensor.

The glass in the bag is the AA filter (the very front piece of glass with the dust cleaning element) removed from the plastic sensor frame. The other images show a method to remove the IRcut filter and replace the AA filter so that the sensor face is sealed and protected from moisture - you can ignore this part, if you wish.

To remove the filter assembly you need to remove the two screws and pry the plastic frame off the sensor face - is this what you really want to do? Even doing a full spectrum mod I would leave the frame in place.

http://www.iceinspace.com.au/forum/attachment_browse.php?a=181527 IRcut filter following removal stuck to the double sided tape.

[amitk75]

Yes I have removed the filters with frame. I will use the camera with this, and if there are any issues then will put the empty frame back. Thanks!

[glend]

Ok I see what your saying in that other thread that you bumped up. It would appear that the AA filter has the same spectral profile as the Baader UV/IR Cut that Gary Honis recommends to install as an alternative to eliminate the extreme ends of the spectrum, and the AA prevents moire pattern artifacts by slightly smearing the light thus AA removal can result in slight increase in sharpness whiich may or may not be noticable. Obviouslly the AA anti-dust shake function is lost on removal. I'll try it without the AA and see if I can detect improvement. Thanks

[glend]

Well I gave it a go - removing the Low Pass 1 (AA) filter on my 450D (that already had the blue cast filter removed. Prior to removal of the AA filter the camera had worked perfectly for six months with no issues at all - I mention this because of what follows.

Taking Merlin's advice I removed the AA filter, not just the glass but the filter (filter frame which the glass is bonded to, and the little ribbon cable attached to activate the Cleaning shake), I left the black plastic frame in place so that the same aperture to the sensor was kept.

First test subs through the RC08 revealed strong colour blooms on the lower left of the frame in every shot, which people suggested was AMP Glow or light leaks. I researched Amp Glow on the 450D and found it doesn't have any and then looked at light leaks. I put the camera in a black bag thinking that maybe I had created a leak when I last disassembled but the colour bloom was still there. I then looked inside the RC08 and found there was a potential for light to enter via the focuser collimation ring, taped that up and reshot. Colour bloom still there. At this point I figured it was coming from inside the camera, and did some test shots with my other 450D and they were perfect.
Tore down the camera again looking for mistakes in the last assembly, there were none. I replaced the AA filter, and reassembled. Testing last night showed the camera is back to perfection - no colour bloom in any shot and subs of over seven minutes show no evidence of light leaks or bloom. I can only conclude that the AA filter, its frame, circuit etc, had something to do with the colour bloom - as strange as that may seem to some. So I am leaving the AA in place. As Merlin's spec study showed, the cutoff points for the AA are almost exactly the same as the Baader UV/IR Cut filter that is used on many full spectrum mods (and which I run on the front of my T-Adaptor anyway) so I am not losing anything other than the fringe frequencies which I don't want anyway.

For me it was a learning experience, and I cannot recommend anyone remove the AA filter from a 450D (which is exactly Gary Honis' conclusion). Sure the theory is that it gets you marginally sharper pixels due to the light smear effect of the AA construction but even close zoom examination of my test images (ignoring the bloom regions) failed to see much, if any, difference in pixel resolution.

[AlexN]

I removed the whole lot from my 450D and I get none of that Gary.

[AlexN]

I actually did temple the black plastic frame too. I pulled the whole lot out. Aa filter, uv ir block, the clip frame, the black plastic frame. Everything. Obviously, the camera will be more sensitive to light that you can't see. Possibly light entering the viewfinder assembly.

I've removed the mirror and completely blacked out the viewfinder prism so there is no chance of leaks there.

[glend]

Well it's a mystery to me, I can only report what I found. I did test the camera in a black cloth bag, and the viewfinder window is normally taped up as well. I suppose it is possible that the sensor is getting some sort of signal from nearby components and representing that as light - is that possible. Is it possible that the cooling components of my cold finger mod are impacting the image. The sensor was running at -10C, with the copper finger pressed against it's back. On the camera there is the TEC running via a remote PWM controller and the fan motor is onboard as well. If for example, the sensor was unevenly cooled it seems feasible that it may represent light colour in different ways. The area to the lower left of the frame is the side the cold finger enters the area behind the sensor, but it is well coated in thermal paste. I suppose that the presence of the AA glass in front of the sensor could be helping ( like double glazing) to hold the cold into the sensor surface area and by removing it I created an uneven temperature gradient on the sensor surface that may yield different colours, if that's possible. I am open to suggestions.

[AlexN]

I will do more testing of mine Gary.
I've only tested it in daylight at this stage so I have no idea how it will perform on longer exposures. Also, I had to drill a hole I the casing next to the viewfinder to access a screw. That's been blocked by bluetac for the time being. I will see how I go if the sky ever clears.

[Merlin66]

Hmmm interesting issue, Glen.
I regularly communicate with many Canon FULL full modded camera users who get the benefit of larger wavelength coverage for spectroscopy. (I don't know of any who use a DIY cooling system)
I've never heard of a similar problem. It may be worth posting on the DSLR forums (or SGL) - see if they have any good ideas.

[AlexN]

I'm planning on dropping my 450D into a cooler box because I don't have the testicular fortitude to do a cold finger myself with my skills.

[rcheshire]

I have noticed that auto focus works without the glass as well, despite the contrary. It may depend on the camera. I did notice with both filters removed and no other replacement that focus toward the IR range was very fuzzy with camera lenses/refractor. Lenses are not designed to focus these wavelengths as a rule, unless using reflective optics, which seem not to be affected.

EDIT: When I think of it, fuzzy toward both ends of the spectrum is more accurate, and moreso UV.

[AlexN]

Photos I've tested today with no aa or uv/ir filter are sharply focused if using live view but af is a little short sighted. Everything does have a very glowing appearance I'm assuming is the strong uv presence. Ir is showing through also with foliage glowing strongly and very bright in the red channel.

[ZeroID]

Glen and I did a lot of Cooler Box tests a while ago. Minimal success. Two problems are:
Getting enough cold to to the sensor, air to hardware inside a camera body is suboptimal to say the least
A lack of control and consistency, the sensor soon warms up to wthin 5* of ambient and the first problem above won't allow it to cool fast enough.

I have a pending mod to another camera I can afford to butcher, my old KM 7D which involves mounting a copper slab ( ~ 2mm thick) inside within the internal stainless back plate which sits about half a millimetre away from the rear of the sensor. If I can find a silicon thermal pad to act as an insulated medium between the two and get a cold finger onto the copper slab I might have something that works.
I've already milled out the stainless back and milled a shoulder on the edge of the copper so it sits inside the opening and close to the sensor.

Just contemplating how to get a cold finger access without totally destroying the rear. The 7D already has the IR mod done and I haven't really had a chance to test that what with weather, a current rebuild project on my setup, some back problems and a busy rally photographic schedule.
Keeps me outa trouble I guess :P