Shooting with SFX 200

[Ralphieboy]

Never shot with SFX 200, but "back in the day", I shot a good bit of B&W as well as color infrared film. With the near infrared Ilford should one compensate the focus by using the infrared scale on the lens? Only using a 25a red filter as I don't yet have a 72 filter.

 

[MattKing]

No - with a 25a, all the film is seeing is visible light, so no focus compensation.
Even with a 720nm filter the wavelength of light to which the film responds is so close to the visible spectrum as to make compensating the focus unnecessary.

 

[Mark J]

Give it a try, but as Matt says, an R72 will be better . However if you're going to use an R72, then get Rollei Superpan 200 ( = Rollei IR ) instead , it has more IR sensitivity.
On compensation, I disagree with Matt, but have some nuance to the advice.
I have done some tests last year and reckon that for best sharpness you should use about 2/3rds of the IR offset shown on the older lenses, when using modern IR films. I have a suspicion that the old IR markings were optimised for the Kodak High-Speed Infra-Red film that went out to 850nm. Modern films don't go as far into the infra-red, so the compensation is less.

 

[pentaxuser]

OP, based on the slightly divergent views of the two respondents I'd try both methods by shooting two identical scenes in succession and then look at the corresponding negs and then prints at your normal enlargement to see if you can tell if there is any difference

pentaxuser

 

[Mark J]

If you're only going to use a 25A filter and on the Ilford SFX, then there needn't be any focus adjustment.

 

[Tony-S]

Yup, as noted above, no need for focus compensation. I use a Cokin 007 filter when I shoot it. These are all with a Bronica GS-1.

 

[MattKing]

Give it a try, but as Matt says, an R72 will be better . However if you're going to use an R72, then get Rollei Superpan 200 ( = Rollei IR ) instead , it has more IR sensitivity.
On compensation, I disagree with Matt, but have some nuance to the advice.
I have done some tests last year and reckon that for best sharpness you should use about 2/3rds of the IR offset shown on the older lenses, when using modern IR films. I have a suspicion that the old IR markings were optimised for the Kodak High-Speed Infra-Red film that went out to 850nm. Modern films don't go as far into the infra-red, so the compensation is less.

Yes, the old IR markings on lenses were optimized for HIE.
Somewhere I've seen the results of working through the math, which revealed that the focus offset for 720nm is just a tiny fraction of what is required for 820 - 850nm.
But can I find that calculation? Nope.

 

[Sirius Glass]

I use several filters for infrared film. Each has its own results. From the least to the strongest: R23, R25, R29 and 72 [aka 720]. If one is using small apertures, adjusting the focal length is not necessary unless one can still obtain HIE.

 

[Mark J]

The advice to ignore the focus adjustment and just stop down is commonly given and works OK in a lot of situations.
However it has some problems as follows :

1. The shot may not want a stopped-down look, perhaps you want to shoot at f/5.6 . You will be out of focus at f/5.6
2. IR film is very slow when filtered, stopping to small apertures can put you into several-second exposures, then you have more trouble with things moving in the wind, or some reciprocity effect.
3. The image is still slightly out of focus, and hence your depth of field is now offset and won't correspond with what you set on the scale.
4. Your diffraction limit is worse with longer wavelengths - by about 35% going from a normal visible band to 740nm . So, if you shoot at f/22, you will only get the sharpness of f/32 if using an R72 filter.

That's my take on it. For me it seems easy to spend 2-3 sec adjusting the lens.
It's a bit harder on LF though !

 

[Paul Manuell]

@Ralphieboy
Shot on SFX200 through an IR72 filter. Focussed as normal before adding the filter, with no adjustment made once filter added. F stop of f11 meant focus was pretty good anyway, without having to adjust for the infrared wavelength

 

[reddesert]

Yes, the old IR markings on lenses were optimized for HIE.
Somewhere I've seen the results of working through the math, which revealed that the focus offset for 720nm is just a tiny fraction of what is required for 820 - 850nm.
But can I find that calculation? Nope.

It's not possible to calculate focus offset as a function of wavelength generically, because it depends on the lens design. It might be possible to have a rule of thumb that holds for typical lenses, and that may be good enough for nearly all practical use, but I don't think there is a way to calculate it without going into an actual lens design. Real lenses are achromatized by combining positive and negative lenses of glasses of different dispersion (eg crown and flint) to make a combined lens where much of the dispersion is canceled out from blue to red wavelengths. Since that cancellation isn't perfect, outside that wavelength range the residual variation gets larger in some way that depends on the specific powers and dispersions of the lens elements.

It's useful to look at the simplest achromat, a doublet. A couple of discussions of achromatic doublet design:
https://www.opticsforhire.com/blog/achromatic-doublet-with-flint-crown-glass/
https://phys.libretexts.org/Bookshelves/Optics/Geometric_Optics_(Tatum)/02:_Lens_and_Mirror_Calculations/2.10:_Designing_an_Achromatic_Doublet

The Tatum book section at libretexts is useful because he reduces the design to something one can calculate with a little algebra. He achromatizes it by constraining the focal length at 0.48 and 0.66 microns (red and blue) to be equal. I went through the exercise of recomputing Tatum's achromat with two real glasses for which I can look up the indexes of refraction at a range of wavelengths, N-BK7 crown and N-SF5 flint - these are very common glasses. The goal was to get a doublet of 160mm focal length in yellow light (because that's what Tatum used), and what I got was ~161mm with this wavelength dependence:

microns
wavelength fl_mm diff_mm diff_fractional
0.486 161.46 0.085 0.0005
0.588 161.38 0.0 0.0
0.656 161.46 0.085 0.0005
0.707 161.54 0.161 0.0010
0.852 161.86 0.479 0.0030
1.014 162.23 0.911 0.0056

The third column tells the focus shift in mm and the fourth column is the focus shift as a fraction of the base focal length at 0.588 microns (yellow light). An old rule of thumb for IR photography was to approximate the focus shift as 1/400 of focal length (for HIE wavelengths) and that is 0.0025 which is actually fairly close to the 0.003 fractional shift at 0.85 microns (850 nm) for this achromat.

Obviously I'm not going to put any real optical designers (like Mark) out of work, but real lenses are to some degree made by a similar process of combining crowns and flints to cancel out chromatic aberration. If this example shows anything, it suggests that the focus shift for SFX type film with a true IR filter (so wavelength maybe 700-760 nm, much redder than a 25A) is about 1/2 to 1/3rd of the HIE shift mark (about 850 nm or so). Not totally negligible.

For wide to normal focal length lenses on 35mm, I find the IR mark is usually somewhere around the f/4 or f/5.6 DOF marker. For longer lenses on 35mm, it is at much higher f-stop on the DOF scale - eg about f/20 on a 200mm lens I just looked at. This makes sense with that 1/400 of focal length rule of thumb, and implies that for long focal lengths on a given format, you do need to stop down and take some care with the focus shift.

 

[neilt3]

I like the Rollei Infrared , I use it in 135 , 120 and 5x4 with either a B&W 092 filter on ( 685nm ) or 720nm filters , either screw in ones or Cokin circular ones that fit on the P and Z-pro holders .
When focussing I adjust focus point to just over halfway between the visible light distances and the infrared mark .

Focus seems spot on this way .

I haven't used the Ilford film for IR as when I first researched what to use , it is much less sensitive to IR light than the Rollei .

 

[pentaxuser]

The advice to ignore the focus adjustment and just stop down is commonly given and works OK in a lot of situations.
However it has some problems as follows :

1. The shot may not want a stopped-down look, perhaps you want to shoot at f/5.6 . You will be out of focus at f/5.6
2. IR film is very slow when filtered, stopping to small apertures can put you into several-second exposures, then you have more trouble with things moving in the wind, or some reciprocity effect.
3. The image is still slightly out of focus, and hence your depth of field is now offset and won't correspond with what you set on the scale.
4. Your diffraction limit is worse with longer wavelengths - by about 35% going from a normal visible band to 740nm . So, if you shoot at f/22, you will only get the sharpness of f/32 if using an R72 filter.

That's my take on it. For me it seems easy to spend 2-3 sec adjusting the lens.
It's a bit harder on LF though !

Do I take it that based on my understanding of an earlier clarification from yourself that none of the above applies to using SFX?

The problem as I see it is that we have diverged from answering the OP question and have moved on to focusing as it applied to Kodak HIE but that is not always clear, at least to me it isn't

Thanks


pentaxuser

 

[Tony-S]

Do I take it that based on my understanding of an earlier clarification from yourself that none of the above applies to using SFX?

The problem as I see it is that we have diverged from answering the OP question and have moved on to focusing as it applied to Kodak HIE but that is not always clear, at least to me it isn't

Correct, no focus adjustments for SFX 200. Shoot it like you would any other conventional B&W film.

 

[Mark J]

Yes, Pete, sorry we have strayed a little.
For SFX with a red filter I would do no correction.
For SFX with an R72 filter, I myself would apply a small correction, because you are then working with 720-730nm light. Probably less than half-way to the marker IR point. However I am v.picky and it might not make a lot of difference. The ISO with this filter must be VERY low, though .. '3' ?

Reddesert, your points above are spot-on.
It was good that you mentioned the use of the d.o.f. markers though, I was going to comment on that.
I just went round and checked a few more of my manual-focus lenses ( the non-apo or semi-apo ones ) and they show the 'IR focus' mark at either the right-hand f/5.6 or f/8 depth of field limit.
So for the current crop of Rollei/Adox films then it makes sense to use the right-hand f/4 d.o.f. line as the reference, if your lens does not have a red line.
Good to see that Neil agrees on this .

I tried to find a critical case to check this last year, since I like to pin-down these technical questions. I used 35mm Adox HR50 film in a Canon EOS body and a Zeiss 35 f/2 ZE at f/5.6. On a subject around 1.5m away, it was fairly obvious under the loupe, which adjustment was correct.

 

[Andrew O'Neill]

The only film I corrected focus for was for HIE and Efke... but that was only if I used an opaque IR filter, like the 87C. Rollei IR, I've never bothered with a focus adjustment with the strongest filter capable of producing decent IR (Wood) effect, a 720.
Definitely no focus compensation needed with SFX.
I wish HIE was still with us. Such a wonderful film to work with and so much easier than current stock, as far as IR effects go.

 

[pbromaghin]

Ye
I just went round and checked a few more of my manual-focus lenses ( the non-apo or semi-apo ones ) and they show the 'IR focus' mark at either the right-hand f/5.6 or f/8 depth of field limit.

Not all lenses turn the same direction for focusing. Do you mean the farther, or nearer DOF mark? I really want to explore SFX more and my chosen camera doesn't have any IR mark. It's a Voigtlander Bessa I With a Color-Skopar in 6x9, a meterless "guess" focus camera, so it's all up to me to make the right decisions.

I have no idea how the old Tessar-type lenses compare in this aspect to the very complicated modern multipleelementfullymulticoated lenses of today.

Thanks,
Peter

 

[reddesert]

The IR focusing mark is virtually always in the sense of the lens has a longer focal length at IR wavelengths. That means the lens has to move a little further away from the film, which means you turn the lens in the direction that puts the center mark at a closer distance. In other words, if the subject is at infinity, you're going to put the center mark at 40 or 80 meters or something like that.

(You have probably a 105mm lens on 6x9, and if we follow the old 1/400 shift rule of thumb for HIE-wavelengths, you would need to move the lens out by ~0.26mm, which means focusing at ~40 meters. But with SFX or Rollei IR, you probably only need to shift by half that or less, so the shift is like from infinity to focus distance 80 meters.)

Nearly all conventional lens designs shift in this direction. In a previous thread, photrio member AgX mentioned that the only lens he knew of with focus shift in the other direction was the Schneider Angulon (which is sort of a weird design). That's a large format lens usually without a focusing barrel, so no marks. I have never tested its focus shift.

 

[Mark J]

Ok yes sorry, forgot about Nikon, Pentax etc which focus the other way. Advice as above, it would be the left f/4 d.o.f. mark in that case.

Reddesert, I think your numbers above are an under-estimate. Most photo lenses have more secondary spectrum than a doublet. On a Zeiss planar 50/1.4 I just checked, the HIE marker means a shift from infinity to 10m. This is 1/200 of the efl.

 

[TheFlyingCamera]

What's the consensus on best film speed to rate SFX at with a 720 filter? 6? 12? 3?

 

[reddesert]

Ok yes sorry, forgot about Nikon, Pentax etc which focus the other way. Advice as above, it would be the left f/4 d.o.f. mark in that case.

Reddesert, I think your numbers above are an under-estimate. Most photo lenses have more secondary spectrum than a doublet. On a Zeiss planar 50/1.4 I just checked, the HIE marker means a shift from infinity to 10m. This is 1/200 of the efl.

This is fair, the doublet calculation was really to get a sense of the dependence of longitudinal color (secondary spectrum) on wavelength for a real lens (if only a doublet), as I could not find a simple discussion of that anywhere.

The "add extension by 1/400 of the focal length" is an old rule of thumb for HIE IR photography that one can find in books. I can't find a book on IR photography at the moment, but I found this rule quoted in John Schaeffer's "Ansel Adams Guide to Basic Techniques of Photography," and in the old Kodak Reference Handbook (circa 1940s-60s) section on IR and UV photography. While in the section on Kodak lenses, some of the lenses have a recommended focus shift, which can be smaller or larger than 1/400.

I looked at several SLR lenses and found that in a sample of a few, wide-angles at 24, 28, 35mm had a HIE IR mark focus shift around 1/200 of efl, while the two 50mm I checked (a Nikon and Vivitar) had a shift around 1/400 efl.

 

[neilt3]

What's the consensus on best film speed to rate SFX at with a 720 filter? 6? 12? 3?

I've never bothered with SFX as it's nowhere as sensitive to IR as the Rollei Infrared film .

The Rollei stuff I use with a 720nm filter I take a meter reading as ISO 6 to get the woods effect . And I like the results .
I'd suggest buying that .

If you insist on using SFX , try it as ISO 3 or less .

 

[retina_restoration]

What's the consensus on best film speed to rate SFX at with a 720 filter? 6? 12? 3?

My experience tells me to expose at 10-12 ASA during the summer half of the year, and more like 5-8 ASA though the winter half.

 

[Paul Manuell]

What's the consensus on best film speed to rate SFX at with a 720 filter? 6? 12? 3?

I just set the ISO to box speed, ie., 200, then overexposed by 4 stops from the unfiltered reading. So for example, at my chosen aperture of f11, a reading of 1/250th without the filter was manually adjusted to 1/15th once the filter was added

 

[Mark J]

( Response to retina_restoration )
Those are the numbers I'd use with Rollei IR 400 / Superpan 200 , so I'm surprised.