Hi Fred, i think you are partially correct.. only partially
If you look at Christians page, and Bobby's tutorial, he removed the Phaser Layer Infrared-absortion glass and replaced it with the Baader BCF Filter. The Phaser Layer Infrared-absortion glass suppose to block the IR range. By replacing this piece of glass with the BCF filter, it blocks out the UV range and a slight amount of IR range. (see attached BCF Filter graph). As you can see in the graph, the IR range with the BCF filter spikes at 1100nm to 80% then drops at 1160nm to 43%. This is the spike which needs to get filtered out. That is how the Baader BCF filter is designed that it doesn't block this IR range hence my image (see attach pic 2) i think it is in the middle of your 350D image and the 40D image. I can tell that your 350D image hasn't got any filters inplace that is why it is soo red compared to mine (not as red) but still red.
The Original Phaser Layer Infrared-absortion glass also blocks the much needed H-Alpha signal.
The lowpass filter is the filter which does the Anti-dust shake so that is why Bobby left it as this filter should not block out the required H-Alpha range. At this stage, my camera hasn't seen first light on astro objects so far.
to fix this for daylight use, I have gone with the Hutech Daylight front filter. For astro use, I still have the Hutech LPS2 filter which also blocks the UV/IR range. That is why in my other post, holding the LPS2 filter infront of the lense, the image looks more normal than without because the LPS2 filter does indeed blocks the UV/IR range.
try this fred, put the UV/IR filter infront of the lense, leave the camera in AWB mode and take a shot of a white piece of paper. Use this image and set it as CWB then set the camera to dial manual mode. Now take a picture. The picture should now come out normal.
try this fred, put the UV/IR filter infront of the lense, leave the camera in AWB mode and take a shot of a white piece of paper. Use this image and set it as CWB then set the camera to dial manual mode. Now take a picture. The picture should now come out normal.
Hi Fred and Eric
You must be in spot metering and cover the spot circle with white.
You must also follow the instructions on page 68 of the manual otherwise the image will not register as a CWB
The phaser layer IR absortion glass blocks out the H-Alpha range. This has to be removed for great HA sensitivity. Thanks Steve yes, the CWB picks up the central spot metering and not the entire image. Best is to ensure the white paper is under sunlight and that it covers the entire image.
Heres my white balance image under sunlight. Others will be different as my IR range is not completely blocked..only slightly as mentioned in my previous posts.
OK Eric, Ill try that tomorrow, but, if you make a CWB with any filter combination and then image using it, it should always look normal. Isnt that the point of the CWB, to compensate for different colour balances, how will that prove which filters are in line ?.
Your pic still doesnt make sense to me, you say the Baader BCF filter has a spike at 1100 (as per the graph) and that is why the pic is redder than my 350D pic with a blocker, but you also said the shaker low pass filter 1 is still there, which has a steep roll off at 700. The spike at 1100 then would be completely eliminated by the low pass filter 1.
I think I have taken out the phaser layer because the red histogram changed enormously after its removal. Also, the leaves are still somewhat green on the 350D pic with just the UV/IR blocker, that to me is fairly conclusive, its a known quantity and very similar to the 40D pic. The shaker layer on its own (no phaser layer) is the only combination that matches the UV/IR blocker filter response so closely.
Youve got me thinking about this now Eric, the different results we get is very interesting, a real puzzle, something is still not right .
Whats really, really anoying is this that this diagram describes the Phasor layer and low pass filter 2 the opposite from the description on the page it is in, we keep switching these terms around, very confusing. I think to "2" should apply to the phaser layer.
OK Eric, Ill try that tomorrow, but, if you make a CWB with any filter combination and then image using it, it should always look normal. Isnt that the point of the CWB, to compensate for different colour balances, how will that prove which filters are in line ?.
.
It depends on what filters are being used on the modded DSLR. If the filter doesn't consist of an IR block and just UV, it will not CWB anything. your images will turn out red only a full IR blocking filter can fix this back to normal (or close to normal) whereas with my BCF filter, it doesn't fully block the IR range. only certain nm's.
Your pic still doesnt make sense to me, you say the Baader BCF filter has a spike at 1100 (as per the graph) and that is why the pic is redder than my 350D pic with a blocker, but you also said the shaker low pass filter 1 is still there, which has a steep roll off at 700. The spike at 1100 then would be completely eliminated by the low pass filter 1.
yes thats correct. the low pass filter 1 rolls off at 700 then after that, we are guessing it stops there? the graph doesn't show anything after that. I might shoot back up at 1000nm? more research is required on the first filter (low pass) as that graph doesnt show IR range.
I think I have taken out the phaser layer because the red histogram changed enormously after its removal. Also, the leaves are still somewhat green on the 350D pic with just the UV/IR blocker, that to me is fairly conclusive, its a known quantity and very similar to the 40D pic. The shaker layer on its own (no phaser layer) is the only combination that matches the UV/IR blocker filter response so closely.
Youve got me thinking about this now Eric, the different results we get is very interesting, a real puzzle, something is still not right .
Ok, lets ignore the UV/IR filter. take a AWB shot with the 40D and a AWB shot with the 350d and see if the leaves are the same colour. I'm betting the 40D leaves will turn out green whereas the 350d leaves will turn on pink
yesss this is VERY VERY interesting, keep at it
Phaser 2 = filter 2 and filter 1 = low pass filter. check figure 7 from christians site. filter 2 (phaser) blocks out the HA thats definitely the filter you want to remove.
Here they are, and both with no external filters. As you predict, the 350D pic is pink. But thats exactly what I expect you see, it makes sense to me, the shaker filter on the 40D cuts off at 700nm, so the leaves are still somewhat green (the IR extended response to 700 would not make that huge a difference to green leaves, but a spike at 1100 sure would). The mod must be right, because the 40D pic here is very close to the 350D with the external UV/IR blocker. ie the UV/IR blocker has the same curve as the shaker filter, hence the same pic, it couldnt be any other way.
And....... your 40D MUST produce the same result, because the shaker filter in your cam cuts off at 700nm, the response of your internal filter past 700nm is irrelevent, the shaker filter has already blocked the IR past 700nm.
Again, I just cant see how your shaker filter is still present due to the above. Not that it matters mind you, the external filter will make the shaker filter irelavent anyway, and you can do IR photography.
I maybe wrong BTW, that wouldnt be unusual , fun trying anyway
The graph represents Christians graph with the BCF filter.
Grey is filter 1, maron is filter 2.
Filter 2 is designed to eliminate all the IR's. Lets predict that the maron line goes down at 750 then goes flat all the way pass 1200nm but looking from 300 - 400nm range, it doesn't block any UV's just mainly IRs.
Filter 1 blocks the UV range (300 - 400nm) then goes up to cater for visual light including H-Alpha, then it goes back down again at 700nm. why would canon put 2 filters which does the same thing? To me, filter 1 doesn't block IR just the UV whereas filter 2 doesn't block UV but blocks the IR.
Filter 2 also blocks the H-Alpha range therefore changing filter 2 is more ideal than changing filter 1.
Therefore with my cam, taking filter 2 out will unleash the IR power but adding the BCF filter will slightly cuts some IR signal.
OH, OK, you have a point there, the shaker glass graph doesnt show response past 700nm, so maybe it has a spike, BUT it doesnt matter I think, I get similar pics from the 350D and 40D with the external filter (flat after 700nm), which makes any response after 700nm on the shaker filter irelavent.
Anyway, based on your analysis, I must have removed filter 2, because I got a large red boost on its removal (and the histogram was really obvious), this would not have happened if filter 1 was removed (assuming as you say, it doesnt block IR). Christians picture and position/description of filter 2 is the same as the filter I took out, I really now dont doubt thats the one I took out. But still leaves us with two 40Ds with different IR responses
Here they are, and both with no external filters. As you predict, the 350D pic is pink. But thats exactly what I expect you see, it makes sense to me, the shaker filter on the 40D cuts off at 700nm, so the leaves are still somewhat green (the IR extended response to 700 would not make that huge a difference to green leaves, but a spike at 1100 sure would). The mod must be right, because the 40D pic here is very close to the 350D with the external UV/IR blocker. ie the UV/IR blocker has the same curve as the shaker filter, hence the same pic, it couldnt be any other way.
I don't believe the shaker filter (filter 1, low pass) has the same curve as a UV/IR block filter maybe just UV, my guess is it spikes back up at 1100nm or even maybe before that. christians graph stops at 700nm. I don't think both filters (1 and 2) cuts all the IR range only one of them and that is filter 2.
our 40D has different responses is because mine's got the BCF filter which doesnt block 1100nm leaving my images with a red tint and my leaves are not green whereas your one is.
hmm something is not adding up though ?
need to think... need to think..
ps. we need a 3rd person opinion i see if i can draw a graph after 700nm . easier with graph.
Night Fred, we talk about it tomorrow (although i get flat out busy at work) ill try to reply between then.
attached is another graph with my 40d camera prediction after 700nm.
Taking filter 2 out will inleash h-alpha at around 652nm with filter 1 inplace. then, the BCF filter gets put in which cuts down IR range from 1000nm to 1100nm (blue line) but it doesnt cut from 1100 - 1200nm causing my image to have red tint.
Regarding your camera, looking at visual spectrum, green doesnt' occur anywhere above 700nm so how did you get your leaves green by taking filter 2 out? *SHRUGS....MEGA SHRUGS* it is green because filter 1 has been taken out with filter 2 still in place? i dont know...
Eric. as here for the 350D (the 40D would be similar) the green and blue channels are unaffected by filters, they both pass through (the removal of filters is just to extend red response) red and green colours are presnt in both moded and unmoded versions (so images are still RGB). What makes the colour balance so strange with no filter is the over all extra red sensitivity and big green and smaller blue response spikes at 7000 (and another blue spike at 8200), which causes red, green and blue all to accept IR light as extra data at 7000 and beyond, causeing a washed out effect way out side normal balance. The huge red curve gives the overall red hue with something like 3 times ( more than GB) the light collection over a large frequency range.
The UV/IR blocker cuts out the 7000 and above spikes and returns colour balance with just a slightly extended red response to just before 7000. Green and blue response is then back to normal as in an unmodded cam. This is what you want for astro RGB, normal GB and a bit higher red respose to capture Ha.
That is why in my other post, holding the LPS2 filter infront of the lense, the image looks more normal than without because the LPS2 filter does indeed blocks the UV/IR range.
.
