Help with infrared focussing please :)

KevinW said:

As others have mentioned, modern 'Infrared' film is not the same as the true infrared film that was available in the past. Don't worry too much about focus. The photo attached was taken with a Bronica S2 on Rollei Infrared and a generic 720nm filter. I hand metered using an ASA/ISO of 6. I don't remember the exposure settings but it was something in the 1/8 or 1/4 second area and shot on a tripod.

Click to expand...

If you shot on a tripod and stop down, no problem. I don't, so focusing is something for me to consider.

xkaes said:

Sure, a tripod is always an option, but if you don't want to use a tripod, you have to open up. I shoot lots of slow film, not just IR, without a tripod -- at a wide aperture. Try it, you'll like it.

Click to expand...

My lenses have seven or eight f/stops on average, some lots more, and I paid for every f/stop the company gave for me to choose from, not just for the creamy bokeh limit!

250swb said:

My lenses have seven or eight f/stops on average, some lots more, and I paid for every f/stop the company gave for me to choose from, not just for the creamy bokeh limit!

Click to expand...

For those of you who always lug around a tripod, you can skip this post.

I use all of my f-stops and shutters as well -- when I can. But when I'm shooting IR, things change. I either have to use such a slow speed that I need a tripod or I need to open up the lens -- a lot. When I don't want the limitations of a tripod, I have no choice but to open up. Opening up is not a problem with normal film, but it is for IR.

Of course, since I always use a tripod for 4x5, this issue is only relevant to me with smaller formats -- which I use to avoid a tripod (most of the time).

Shoving all theory, assumptions, and considerations aside I did what any photographer should do in my world: I tried it out!

With Ilford SFX and a mild 89B filter critical focus wide open at infinity - about 3 km away subject - shifted on a Bronica PG 5.6/250 mm to the first depth-field marking. I used the markings as guideline as they are about 1 mm apart, a good measure for focus-bracketing.
With the PG 4.5/200 mm critical focus was acquired on between the first and second marking. In both cases focus landed about halfway between infinity for white light and the classic IR marking we all know was for HIE and 850 nm.

So if you think, ignoring the issue and stopping down is sufficient, go ahead and more power to you. It will propably work just fine in most cases. I rather use the corrections I found to work even wide open and stop down in addition.

That's great, thanks for reporting the results!

Could you say what distance markings correspond to the classic IR mark and your derived Ilford-SFX+89B mark, for each lens? That is, "IR focus at infinity is X meters on the distance scale", and so on.

Hardly, as I used the depth of field scales as an indication and I wouldn't recommend following the same correction if you aren't using the same make and model of lens. The values depend on the color-correction of the exact model of lens and will probably vary as already outlined in this thread. My recommendation is to sacrifice a film and do your own bracketing with your own lenses. It really pays off!

skahde said:

Hardly, as I used the depth of field scales as an indication and I wouldn't recommend following the same correction if you aren't using the same make and model of lens. The values depend on the color-correction of the exact model of lens and will probably vary as already outlined in this thread. My recommendation is to sacrifice a film and do your own bracketing with your own lenses. It really pays off!

Click to expand...

I understand that the correction is technically dependent on the specific lens model. But there was a discussion upthread about, for example, the table that Mamiya gave for recommended IR focus corrections (for HIE) for RB67 lenses, and how they compared to the old approximate "extend by 1/400 of the focal length" rule-of-thumb. It would be interesting to see if the Bronica lenses behave similarly. One can figure out the distance adjustment by setting the focus scale at infinity and reading the distance off the IR mark.

One could figure out where the Bronica lenses' IR mark is by looking at enough pictures of the lenses (or the physical lens), I just thought I'd ask.

If it is of interest, let me take a look, the lenses are upstairs...
This is the correction for critical infinity-focus on the 4.5/200 mm Bronica PG. Filter 89B, Film Ilford SFX.



And this is the same for the Bronica PG 5.6/250 mm



As you can see, correction lands halfway between visual infinity and the indicated correction for Kodak HIE with R72 filter. But it's hard to tell any actual distance as we are in a small region of the scale where values increase exponential and are hard to estimate.

skahde said:

As you can see, correction lands halfway between visual infinity and the indicated correction for Kodak HIE with R72 filter. But it's hard to tell any actual distance as we are in a small region of the scale where values increase exponential and are hard to estimate.

Click to expand...

Trying to answer the question in numbers for the 200 mm, where the correction lands in the middle between the markings for infinity and 50 m, I estimated quick and dirty on a x^2 graphic-plot: If I assume infinity as 1000 times the focal length i.e. 200.000 mm or 2000 m, plot an f(x)=x^2 function and look for the y=2000 - 50 mark, cut the resulting y in half, go to that value and read the x from there I am at about y=850 m for the 200 mm Bronica lens. One could argue this is irrelevant but its just a bold estimation and may not even follow the assumed f(x) = x^2 on this helicoid and you can see the difference with a loupe on the negative and in the shaprness of the print.

The thin lens equation for focus distance is: 1/f = 1/d_o + 1/d_i , where f=focal length, d_o=object distance, and d_i = image distance.

If we take d_i = f + e, where e = extension past infinity, then we can rearrange it to get an equation for the extension:
e = f^2 / (d_o - f).
For longer distances, d_o >> f, so: e ~= f^2/d_o,
so the extension past infinity is inversely proportional to the object distance.

The amount of turn of the focus ring is linearly proportional to extension past infinity for unit-focusing lenses (but not for lenses with internal focusing, or an internal floating element). This means that the amount of turn goes as the reciprocal of the focus distance, at longer distances. Thus, halfway between the 10-meter mark and infinity is the 20-meter mark; halfway between the 20-meter and infinity is at 40-meters; and so on.

Thank you, at least that part is clear as day now. But still, from this small part of just a distance-scale it is hard to estimate any meaningful numbers. But I already dare to draw the conclusion that for most lenses with an IR-indicator (for HIE as we know) when used with films of extended red-sensitivity to about 720-730 nm and a 89B or R72 filter, a correction about half way into the direction of the IR-index will be sufficiently exact when stopping down a bit. But before pulling out chisel and hammer to carve this in marble I will check my other lenses and would like what others found with lenses from different makes than the Bronica PG-lenses from my inventory.