Adding Filter Factors- B&W

[timbo10ca]

I've read on a number of occasions that additional exposure should be added on top of a filter factor, originally suggested by Gordon Hutchings- 1 extra stop for an orange #16 and 2 extra stops for a red #25. I've never done this when shooting 35mm, spot metering TTL and filter, and have had no obvious problems. I have also not applied additional exposure when using infrared film (HIE, EIR or SFX).

Since moving to LF I haven't added filters to the equation until recently. When I did however, I found that when I metered as usual (through the red filter) and did not apply additional exposure, the neg was underexposed and the contact print quite dark. I'm thinking I should have added that extra 2 stops of exposure.

Do many people apply that extra exposure? To all formats, or just LF? Why would I notice an effect in LF and not in 35 mm?

Thanks,
Tim

 

[keithwms]

Are you compensating for bellows factor? That is the #1 source of exposure confusion for most people when they begin working with a bellows-focusing camera. Of course, at infinity focus, it shouldn't be an issue. The other sort of "new" exposure issue that comes up in LF which you may not have noticed before is falloff and/or image circle limitations. I am assuming you don't mean that.

I am not sure I understand what you mean by "the neg was quite dark"... do you mean thick? Thin? When you say a neg is dark that sounds like it is too "thick" which means overexposed not under. Perhaps you mean the print is dark?

I do not add additional exposure on top of the filter factor- I just round up for simplicity if the factor is not in full stops. Almost always better to err on the side of slight overexposure with neg film.

 

[brian steinberger]

Tim,

I've had problems in the past with filters for black and white as well. I shoot mostly medium format. Format doesn't matter, although in large format you always have to take into consideration bellows extension and your apertures are going to be smaller than the smaller formats so you take that into consideration too.

What I did was set up a gray card in sunlight (make sure there's no glare) and take a shot without any filter. The put your first filter on, take an image with the filter factor from the manufacturer and then take another adding an extra 1/2 stop, then another with another 1/2. Then move on the the next filter. I ran this test both in the sun and in the shade. Then when you develop the film, you make a contact sheet and compare your other shots to the one shot with no filter and looking at your notes (you need to keep good notes!) figure out how much extra exposure to give when you put on a certain filter. Once I figured them all out I wrote on the filter case for each one the exposure increase needed. For example, I found that I need to give 3 1/2 stops for my #29 red filter, so I wrote on the plastic filter case +3 1/2. I found that many of my filters needed up to a stop more exposure than the manufactures filter factor.

Note also, that at sunrise and near sunset the light is warmer, and therefor the extra exposure needed for the warmer filters (deep yellow, orange, and red) will be decreased. If you don't compensate for this you'll end up with very light tones. I learned this the hard way. Since I shoot medium format though I can bracket when I think the filter factor needs to be cut back.

I hope I didn't confuse you too much. Others will chime in as well.

 

[Chuck_P]

Do many people apply that extra exposure? To all formats, or just LF? Why would I notice an effect in LF and not in 35 mm?

Thanks,
Tim

I have never added an "additional exposure" on top of the stated filter factor either. Something was obviously done wrong between the factor(s) and/or the reciprocity adjustments from LF to 35mm. I agree with Keith that it is better to err on slight overexposure than under with neg film.

An example of how I applied these considerations for picture I took on this last Saturday would be:

1st consideration. Basic Exposure was 8sec @ f/64

2nd. #8 Filter used, factor 1.5x

3rd. no bellows ext factor needed, but if it were, multiply it by the filter factor to arrive at a total factor

4th. the adjusted exposure is: (1.5 x 8sec) = 12sec for: 12sec @ f/64

5th. reciprocity adjustment for TMX in seconds = +15sec

6th. so, the exposure given = 27sec @ f/64

I developed the negative and it looks really good.

 

[timbo10ca]

Are you compensating for bellows factor? That is the #1 source of exposure confusion for most people when they begin working with a bellows-focusing camera. Of course, at infinity focus, it shouldn't be an issue. The other sort of "new" exposure issue that comes up in LF which you may not have noticed before is falloff and/or image circle limitations. I am assuming you don't mean that.

I am not sure I understand what you mean by "the neg was quite dark"... do you mean thick? Thin? When you say a neg is dark that sounds like it is too "thick" which means overexposed not under. Perhaps you mean the print is dark?

I do not add additional exposure on top of the filter factor- I just round up for simplicity if the factor is not in full stops. Almost always better to err on the side of slight overexposure with neg film.

Sorry- to explain better, this was a landscape photo, so I was at infinity, so no bellows factor was necessary. I should have also said that the neg was underexposed, causing the contact to be quite dark (underexposed).
I will fix my post with an edit- thanks.

Tim

 

[keithwms]

Okay.

There are a few other possible ways that you might see slight differences in exposure between your 35mm and your LF camera.

[1] The "true" lens speed of your LF lens may not be as stated. Usually a negligible error but if it's an older lens, it may have yellowed or whatnot. If that's the case then you might well want some additional exposure compensation on top of the filter factor. If you suspect this then I would set up some rigorous tests, beginning with polaroid or fuji instant film. N.b. these films may not have sensitivity curves that are analogous to that of your sheet film.

[1b] More likely is that the shutter speeds of your LF gear are not what they should be, this almost always gives you underexposure. You can have your speeds checked. You might say: wait, I didn't see any speed problems when I exposed without the filter, so why should I see them with? Well, when the speeds are off they are not all off by the same amount so if the red filter forced you to a 1 sec exposure, that might be off while the 1/4 sec exposure is ~fine. With LF you can leave no stone unturned! That's the fun of it!

[2]Another thing is that you do need to take nto account the film's sensitivity curve when you apply a filter factor. Extreme case: suppose you use an ortho film which has ~no red sensitivity and then you put on a red filter... of course then the filter factor will be way, way more than two stops... in fact it may not even be possible to get an exposure at all! My suggestion is to have a look at your film's sensitivity curve and make sure that red filtering isn't pushing you off the edge of the sensitivity. If it is then all bets are off, as far as stated filter factors are concerned... and you'd need to do your own testing.


Finally, there may be differences in the way you develop....

 

[Mick Fagan]

With red filters, as I understand it, light meters don't read the amount of light correctly.

I mainly run Nikkor glass filters and I know that the correction factor is three stops under daylight. My Gosson Profisix meter nearly always only gives a (close to) two stop reading when the reading is done through the red filter. (Under Tungsten, it is 2 1/3 a stop and that is also correct from my exposures)

With filters I have always metered with my light meter then added the correction factor, I get good to excellent results.

Bellows factors, are another fun bit of LF photography!

 

[JBrunner]

Tim,

How long are your exposures averaging, with the filters and stuff?

 

[DaveOttawa]

I've read on a number of occasions that additional exposure should be added on top of a filter factor, originally suggested by Gordon Hutchings- 1 extra stop for an orange #16 and 2 extra stops for a red #25. ...
Do many people apply that extra exposure? To all formats, or just LF? Why would I notice an effect in LF and not in 35 mm?

Thanks,
Tim

I use B+W orange & red filters for 35mm & MF. I have found, from experience, that I need to add a stop extra exposure more than the stated filter factor for the orange using Delta 400 & FP4 to get simlar neg densities to unfiltered shots. This is with TTL reflective or incident metering.
Like other aspects of exposure I suspect you will need to figure out a recipe that works for you.
The other factor may be reciprocity failure arising as you go to longer exposures (>1 second) due to using smaller f stops with the LF camera (as the poster above me may be suspecting).

 

[23mjm]

The way I use filters doesn't change from one format to the next 35mm to MF to LF. I meter manually figure out my exposure then add 3 stops for a Red #25, 1 2/3 stops for my Orange, 1 stop for a Yellow K2, 2 stops for a green and 2 stops for a polorizer. Normally I just increase the time because i am using the f-stop for depth of field. Once I ge the new exposure time I then adjust for reciprocity and or bellows extension.

 

[CBG]

This topic deserves a book, and it's late and I'm running out of steam, so forgive a ramble.

Metering is a can of worms. Add filtration and it gets worse. Each problematic individually, and the sum more so. Meter through the filter and you have more complexity yet. There's several things I just don't trust involved, and the sum is too shakey to stake my time and film on.

I really don't trust the spectral sensitivity of the meter. Is it full spectrum and matched to my film?
I really don't trust aiming the meter at strongly colored subjects since I have no idea what the meter really sees.
I don't completely trust that I know the spectral response of the film. Which style sensitization am I using?
I don't completely trust the filter factor. It is a numeric compromise averaged for an arbitrary ordinary film response.
I don't completely trust the spectral composition of the incedent light. What color light an I dealing with?

But, what do I trust? I trust that, if the meter is any good at all, it's response to white or neutral light will be roughly proportionate to the film's response. That's about all I trust.

I know some folks meter through the filter and they report it works for them, but... I think they are taking chances. I urge that you do not do so with any of the stronger filters; 15, 21, 23, 25, 29, 47, 58, 61 etc. I won't do it with any filter that has an appreciable influence on color balance.

To hack through the thicket of questions this brings up for me, I meter the least color saturated parts of the scene, wherever that's practical. I only really trust aiming the meter at neutrals, whites, grays and blacks, and take readings aimed at strong colors with a large grain of salt. When I have the time to do it right, I really trust incedent readings, adjusted for the subject at hand by added reflected readings.

I don't trust that the meter will see the same as the film. The only way it is really trustworthy is if you know the meter's sensor has the same spectral sensitivity as the film you are using. Since various films, B&W and color have differing spectral response, there is no way the meter cell matches differing pan emulsions, so that's a poor bet.

Which film is the meter matched to? Plain vanilla panchromatic film? Extended red sensitivity film? An older pan film with less red response? Or a more idiosyncratic response? If I remember correctly, the old Pentax Spot Meter when modified by Zone V! had a filter added to adapt it's response more accurately to their presumptive average film response.

I'm sure this is one more area that could be reduced to numerical exactitude with enough testing and a hand held computer with spreadsheets, but there soon comes a point of diminishing returns for me where testing is involved. So I draw the line here: I meter neutrals in the subject where possible, and the incedent light, with no filter over the meter and add a filter factor later, adjusted as mentioned by previous posters for the light at hand.

C

 

[timbo10ca]

Tim,

How long are your exposures averaging, with the filters and stuff?

This one negative in particular was shot in bright daylight at Red Canyon. I can't remember what the settings were offhand (they may be in my notes at home), but I don't think the time was over 1 second, which is the cutoff for FP4+. I was using tilt to keep as wide an aperture as possible. I figured a red filter in red canyon against a blue sky would really lighten up the rocks while darkening the blue sky to really bring out the contrast. Everything ended up dark on the contact print. I haven't tried printing the neg otherwise, but it's obviously underexposed. I'm pretty sure I would have adjusted for reciprocity, but who knows- I've been known to forget to close my aperture before pulling my darkslide too!

It sounds like nobody follows the advice given by Hutchings here, adding that extra exposure. The table is reproduced in Steve Simmons' book. My understanding is that a red #25 (ff 3 stops) is advised to get 5 stops total extra on top of metered reading, and an orange (ff 2 stops) should get 4 stops extra total (as a general rule, not taking time of day, shadow coverage, or zone system adjustment into account).


Tim

 

[JBrunner]

This one negative in particular was shot in bright daylight at Red Canyon. I can't remember what the settings were offhand (they may be in my notes at home), but I don't think the time was over 1 second, which is the cutoff for FP4+. I was using tilt to keep as wide an aperture as possible. I figured a red filter in red canyon against a blue sky would really lighten up the rocks while darkening the blue sky to really bring out the contrast. Everything ended up dark on the contact print. I haven't tried printing the neg otherwise, but it's obviously underexposed. I'm pretty sure I would have adjusted for reciprocity, but who knows- I've been known to forget to close my aperture before pulling my darkslide too!

It sounds like nobody follows the advice given by Hutchings here, adding that extra exposure. The table is reproduced in Steve Simmons' book. My understanding is that a red #25 (ff 3 stops) is advised to get 5 stops total extra on top of metered reading, and an orange (ff 2 stops) should get 4 stops extra total (as a general rule, not taking time of day, shadow coverage, or zone system adjustment into account).


Tim

I was thinking reciprocity might be the culprit, but it sounds like you have a handle on that. What speed are you rating the FP4 at, and what developer are you using and what is your processing and times?

 

[keithwms]

It sounds like nobody follows the advice given by Hutchings here, adding that extra exposure. The table is reproduced in Steve Simmons' book. My understanding is that a red #25 (ff 3 stops) is advised to get 5 stops total extra on top of metered reading, and an orange (ff 2 stops) should get 4 stops extra total (as a general rule, not taking time of day, shadow coverage, or zone system adjustment into account).


Tim

Tim, I would go with the basic filter factor + reciprocity + some fudge factor for response of the film. "Fudge factor" depends on the kind of light you're getting (e.g. morning, noon... spring or summer or winter... clear or fogged...) and the spectral sensitivity of your film and your filter cutoff. The fudge factor can be as little as zero or as large as several stops. Extreme cases:

*** You're shooting in the canyon in late summer and there are high clouds or mist or Santa Ana / LA crap blowing over the canyon. You know there is significant scatter of higher wavelengths and IR because of the clouds... the scene looks "bluer" than normal. In this case I would pick a fudge factor of at least one extra stop or two. Yeah, I know without precise metering it's not so scientific, but.... this is how I would rationalize my confidence in my exposure and hence my need to bracket.

***Other extreme case: really clear, spring weather over the canyon, no LA crap, no high clouds, the scene looks its normal pure, reddish colour. Fudge factor = zero. No real need to bracket the exposure.

I adopted this idea of a spectral sensitivity fudge factor when I started doing IR. I found that in the canyon, in the early summer, it was surprisingly hard to get a decent IR exposure because of the LA crap and high clouds. I got home, developed my negs, and found them to be lacking at least 2 or 3 stops. I think that would affect your reds in an analogous way because the spectral sensitivity of the film + the cutoff of the filter is fairly constant across deep red / near IR. And the LA crap does scatter well into the visible reds.

 

[Bruce Osgood]

I have recently been exchanging email's with Schneider Optics regarding B+W filter factors and Wratten filter numbers.

Schneiders' position is that a Factor is a Factor is a Factor! It makes no difference what emulsion you are using. T-grain emulsions are factored exactly as conventional emulsions. If your results are disappointing it is due to operator error, not the filter factor.

 

[timbo10ca]

I have recently been exchanging email's with Schneider Optics regarding B+W filter factors and Wratten filter numbers.

Schneiders' position is that a Factor is a Factor is a Factor! It makes no difference what emulsion you are using. T-grain emulsions are factored exactly as conventional emulsions. If your results are disappointing it is due to operator error, not the filter factor.

The filter factor here is irrelevant, as I'm metering through the filter. My question is whether people tend to apply the "Hutching's Factor" as well when filtering.

 

[timbo10ca]

I was thinking reciprocity might be the culprit, but it sounds like you have a handle on that. What speed are you rating the FP4 at, and what developer are you using and what is your processing and times?

I used HC-110 H, developing unfiltered negs from Arches and Zion (using the same method of spot metering and zone placement) at the same time. Only the filtered neg was screwy.

Tim

 

[timbo10ca]

Tim, I would go with the basic filter factor + reciprocity + some fudge factor for response of the film. "Fudge factor" depends on the kind of light you're getting (e.g. morning, noon... spring or summer or winter... clear or fogged...) and the spectral sensitivity of your film and your filter cutoff. The fudge factor can be as little as zero or as large as several stops. Extreme cases:

*** You're shooting in the canyon in late summer and there are high clouds or mist or Santa Ana / LA crap blowing over the canyon. You know there is significant scatter of higher wavelengths and IR because of the clouds... the scene looks "bluer" than normal. In this case I would pick a fudge factor of at least one extra stop or two. Yeah, I know without precise metering it's not so scientific, but.... this is how I would rationalize my confidence in my exposure and hence my need to bracket.

***Other extreme case: really clear, spring weather over the canyon, no LA crap, no high clouds, the scene looks its normal pure, reddish colour. Fudge factor = zero. No real need to bracket the exposure.

I adopted this idea of a spectral sensitivity fudge factor when I started doing IR. I found that in the canyon, in the early summer, it was surprisingly hard to get a decent IR exposure because of the LA crap and high clouds. I got home, developed my negs, and found them to be lacking at least 2 or 3 stops. I think that would affect your reds in an analogous way because the spectral sensitivity of the film + the cutoff of the filter is fairly constant across deep red / near IR. And the LA crap does scatter well into the visible reds.

This would be similar to the fudge factor that AA mentions (which I found ironic) I don't think atmospheric conditions were at play, but I can't be sure- we're talking about the middle of nowhere here. I shot some EIR at the same place the day before (same conditions) and found it to be dramatically underexposed (I actually posted here on the subject a couple months ago). Mabey it's related, but I somehow doubt it. I took many other photos (unfiltered) that were fine over the course of the week. I think I just wasn't careful enough with the EIR (using Sunny 16 and the occasional incident reading).

Tim

 

[keithwms]

Tim, I think if you're working at either extreme end of the visible spectrum (blue or red) or indeed beyond that (UV or IR) then this "fudge factor" I mention can become very important. A really robust way to deal with this would be to use a digital sensor and to plug in a sensitivity curve that mimicks your film sensitivity + the filter's transmission curve. But most of the time you don't need to go to that level to get a good exposure.

Atmospheric conditions matter a lot when it comes to determining how much light in a given wavelength range makes it to your film, especially through a narrow filter. When you red filter you of course select out a particular narrow band of reddish wavelengths. If the reds are underrepresented in your illumination due to atmospheric conditions, then the appropriate exposure comp value can be something quite large. The filter factor concept is a rather dangerous one: it is merely a recommendation, and one that can be waaay off if the illumination isn't "ideal" and the film doesn't have broad and fairly uniform spectral sensitivity.

Haze appears less in IR exposure because the relevant atmospheric scattering goes as lambda^4 (lambda=wavelength), so IR light scatters much less in the atmosphere than bluer light. Hence the perceived better sharpness and contrast in IR. But notice, visible red light scatters quite a bit more than IR... the dependence is to the fourth power, after all. So I allege that I can wave my hands around several stops difference when shooting a scene like the canyon.

Metering through a filter has the obvious problem that the standard meter doesn't know the sensitivity curve of your film. Sometimes that is an issue. I think the very safest way to proceed is to learn how to estimate how much "spread" (uncertainty) there is in your exposure estimate and bracket accordingly. That doesn't mean that you have to bracket every damn shot, which would be expensive in LF. You could just bracket a representative scene and develop a bracket for that one shot, and determine how you will alter the rest of your development, if necessary. This is the practice that I have most recently been trying to adopt, in lieu of a fancy meter into which I can program a sensitivity curve.

 

[JBrunner]

Tim, I think if you're working at either extreme end of the visible spectrum (blue or red) or indeed beyond that (UV or IR) then this "fudge factor" I mention can become very important. A really robust way to deal with this would be to use a digital sensor and to plug in a sensitivity curve that mimicks your film sensitivity + the filter's transmission curve. But most of the time you don't need to go to that level to get a good exposure.

Atmospheric conditions matter a lot when it comes to determining how much light in a given wavelength range makes it to your film, especially through a narrow filter. When you red filter you of course select out a particular narrow band of reddish wavelengths. If the reds are underrepresented in your illumination due to atmospheric conditions, then the appropriate exposure comp value can be something quite large. The filter factor concept is a rather dangerous one: it is merely a recommendation, and one that can be waaay off if the illumination isn't "ideal" and the film doesn't have broad and fairly uniform spectral sensitivity.

Haze appears less in IR exposure because the relevant atmospheric scattering goes as lambda^4 (lambda=wavelength), so IR light scatters much less in the atmosphere than bluer light. Hence the perceived better sharpness and contrast in IR. But notice, visible red light scatters quite a bit more than IR... the dependence is to the fourth power, after all. So I allege that I can wave my hands around several stops difference when shooting a scene like the canyon.

Metering through a filter has the obvious problem that the standard meter doesn't know the sensitivity curve of your film. Sometimes that is an issue. I think the very safest way to proceed is to learn how to estimate how much "spread" (uncertainty) there is in your exposure estimate and bracket accordingly. That doesn't mean that you have to bracket every damn shot, which would be expensive in LF. You could just bracket a representative scene and develop a bracket for that one shot, and determine how you will alter the rest of your development, if necessary. This is the practice that I have most recently been trying to adopt, in lieu of a fancy meter into which I can program a sensitivity curve.

Tim,

The first places I look when someone new to LF is getting under exposure is reciprocity failure, bellows factor, and shooting for box speed with a developing regimen that isn't delivering that speed. Since it appears that you are compensating properly, and have good density on unfiltered negatives, (which also seems to rule out shutter problems) I would suspect an old fashioned exposure error, improper calculation of the filter factor(s) or the specteral sensitivity and atmospheric issues, along the lines of Keith's post. I would suggest testing before you go to the races again. Cheap compared to losing a shot that isn't easily revisited.

A second suggestion, that has saved my bacon, is to shoot one holder per "shot" and develop the film from one side of all of the holders, while holding the #2 shots in reserve. Then you can alter development for the second sheet to help with an exposure problem, plus you have a back up in case of some other problem.

Yes, you shoot more film, and no, sometimes "the moment" is too fleeting to make two exposures, but it has saved me enough that I feel the expense and small effort is worth it.

 

[timbo10ca]

Keith and Jason-

Thanks to you both- definitely food for thought. I'm cofident that I compensated properly for bellows, reciprocity, film speed and development (as the other negs were fine). I hadn't thought that spectral effect could make such a dramatic difference, especially when it's nice and clear out, with no pollution around. I can't imagine how you could test for this and expose properly when you go anywhere then- how are you to know what kind of spectral effect there's going to be at any time in any place??? I would never have thought that just the act of puting a filter in front of the lens could introduce so many variables on exposure, other than filter factor.

I usually do shoot a back-up of all my shots, as I like the security- not sure why I didn't this time....

The weird thing is that when I was hiking last summer (close to home this time), I simultaneously made exposures with my 35mm camera (filtered) and LF camera (unfiltered) and all exposures came out correct. I wish I had tried some filtered of the LF as well now, to compare. Another project for me!

 

[Chuck_P]

T-grain emulsions are factored exactly as conventional emulsions.

Not by the data sheets I have from Kodak. Daylight factors for T-max films are all lower than that rated for Plus-X and Tri-X, except the #25 red and the polarizer that are shown to be the same.

 

[Andrey]

I have recently been exchanging email's with Schneider Optics regarding B+W filter factors and Wratten filter numbers.

Schneiders' position is that a Factor is a Factor is a Factor!

They don't know what they're talking about.

 

[phenix]

I recently asked a question concerning filter compensation factors applied manually vs. TTL automatic compensation. Some answers were so much clever and so much simple! Here is the link to this thread (it is short, only 10 posts):
(there was a url link here which no longer exists)

 

[tomonator60]

They don't know what they're talking about.

You are correct! Filters transmit their own color. The same 25 red would have a greater effect on exposure at altitude with a summer sky than it would at sea level at sunset or when using tungsten lights in the studio. So even if the filter factor doesn't change your exposure better. That is why every chart says the the exposure compensation is approximately 3-5 stops because they don't know what your light source is. J Brunner is right on, when you are not sure, shoot two process one.