Exposure in IR Photography....

[FujiSigmaNolta]

Hi all,

I didn't know where to post this, but I was trying out one of my Sigma SD10s which I removed the hot mirror from. I went out today to take some IR photos and this camera is great for that, it can even focus through the filter and all. Now the question is, should I take a meter reading before the filter or with the filter on? I did it without first and I was getting massive over exposure, then I put it about 3-5 stops down and the results looked OK on the camera screen as I could see detail and it looked like they would come out alright, but when I ran it through the software they were very underexposed and had lots of noise. So I had to resort to using 3rd party NR software (which being a purist, I hate doing) and resizing to hide introduced softness (ditto). I also had the WB mistakenly set at "Incandescent". Would this would have contributed to the added noise? Any help/suggestions appreciated.

[beejaybee]

The meter is balanced for visible light with an average colour temperature (daylight) and will almost certainly misread when the visible light is removed with a "black" IR filter. I think you're going to have to bracket, it would make most sense to meter without the filter & work from that, I'd suggest trying -1 -2 -3 stops find out which one looks best & refine from there.

Softness - have you remembered to correct the focus by using the distance scale visible & IR indices? It may help to stop right down but getting the right focus in the first place is surely the key.

[FujiSigmaNolta]

The meter is balanced for visible light with an average colour temperature (daylight) and will almost certainly misread when the visible light is removed with a "black" IR filter. I think you're going to have to bracket, it would make most sense to meter without the filter & work from that, I'd suggest trying -1 -2 -3 stops find out which one looks best & refine from

I have done so, but it seems that I on average underexposed for about 5 stops, just looking at the pics now. The Sigma does have an issue though, when you use the wrong white balance. Even in normal photography it is imperative that you use the right white balance settings.Correction is difficult if you don't and can lead to increased noise at times, which I think it may have been one of the contributing factors.

Softness - have you remembered to correct the focus by using the distance scale visible & IR indices? It may help to stop right down but getting the right focus in the first place is surely the key.

The lenses I used apart from one did not have the IR index scale but the Sigma SD10 without the hotmirror does a great job autofocussing through the IR filter I have to say,the softness was introduced because of the excess noise and the subsequent use of Noise Reduction software. I have a few that survived and here's one:

[attack_donut]

>>...focus by using the distance scale visible & IR...<<

I have this issue, my lenses don't have the markings for IR on them, how do you do this otherwise?

[El_Sid]

Simplest method is to stop down the aperture. How much depends on the focal length but with shorter focal lengths (say 135mm or less) f8 is normally enough to ensure that the subject remains acceptably sharp. This is fine for general landscapes but if you wish to separate the subject from the background by using shallow depth of field then a body converted and re-calibrated for IR may be a better solution...

[Polyglot]

My apologies for reincarnating this thread at a late stage, but someone may still be following it...

As a rough guide to focusing, for infinity shots on IR you set the focus at 10 metres; if the subject is at 5 metres, set the lens at 4m. I just checked this on an old Pentax SLR lens, which has a red dot on the rotating aperture scale which you have to align with the normal focus position - presumably it applies to any other camera also.