Different paper speed at different fstops (antique lens or something to do with paper negs)?

Fragomeni · Jan 26, 2022

So I’m seeing something that I don’t think I’ve seen before. I’m also not sure if this is a paper neg question or more to do with the lens. I think it’s the lens as I’ve been working with paper negs for a long time and haven’t seen this before. Twice now, I’ve been able to replicate the same behavior. Currently, I’m looking at Ilford Coldtone VC fiber paper (in this case it’s for direct positive so I’m not using RC). Initially, I tested for speed using the step-exposure method. I used probably f8 or f16 when I did this. I got a paper speed of ISO 6 + 3 stops (I use a Pentax Spot meter that only goes down to 6 so I add the additional stops). This yields a good negative on this paper which would stand on its own as a paper negative and reverses well into a positive image. The problem now is that if I shoot the lens wide open, it is not following the same speed rating and the image area is totally exposed to black (negative) with no image detail. Complete total overexposure. Stop it down and use the same speed rating and the image forms correctly. I’m changing nothing else in the workflow. The only difference is the aperture used for the exposures. Has anyone experienced anything like this?

And as far as the lens, I’m using a antique 10 3/4” Goerz Dogmar f4.5. The only thing I could think of was if the lens (it’s an old lens) somehow has an aperture not conforming to modern standard. It doesn’t have a typical f-stop scale and instead reads: f4.5, f5, f5.5, f6.8, f8, f16, f22

Any insight would be appreciated.

Donald Qualls · Jan 26, 2022

Looks like you've rediscovered reciprocity departure (often called reciprocity failure). Essentially, most photographic materials have a level of light intensity below which exposure doesn't add up the way you expect. For typical B&W negative films, this is pretty well documented by the manufacturer, who then give totally impractical recommendations to open the aperture one or two stops (when the reason you're exposing for six seconds in the first place is that you're already wide open).

What it boils down to, in generally, is that beyond a certain exposure time each stop of extra exposure takes more than twice as much time. There are a number of ways to express this; in the past, I've used formulae like 2.8x for one stop (for a particular film, and this varies from film to film). Ilford recommends an exponent: for exposure longer than one second, after calculating exposure, raise the time in seconds to a power that's characteristic for the film in use (say, 8 seconds would give 8^1.26). Both of these methods require a calculator (unless you can do fractional powers or powers of non-integers in your head).

In the end, though, what you're seeing is just the reciprocity departure of the paper you're using: you've experimentally discovered that paper acts like film, in being slower at low light intensity than with more intense light. Since manufacturers don't document this quality for paper (they expect you'll run enlarging exposures in the 5 to 30 second range and make test strips for each new negative or size), you'll have to do some testing. With a little reading and some math, you should be able to arrive at either a stop multiplier (number larger than 2 to multiply time for each stoop above a few seconds) or a Gainer exponent (power to raise the final exposure, in multiples of the reciprocity limit) for your paper that will be valid until the next time Ilford changes the paper emulsion.

Fragomeni · Jan 26, 2022

Donald Qualls said:
Looks like you've rediscovered reciprocity departure (often called reciprocity failure). Essentially, most photographic materials have a level of light intensity below which exposure doesn't add up the way you expect. For typical B&W negative films, this is pretty well documented by the manufacturer, who then give totally impractical recommendations to open the aperture one or two stops (when the reason you're exposing for six seconds in the first place is that you're already wide open).

What it boils down to, in generally, is that beyond a certain exposure time each stop of extra exposure takes more than twice as much time. There are a number of ways to express this; in the past, I've used formulae like 2.8x for one stop (for a particular film, and this varies from film to film). Ilford recommends an exponent: for exposure longer than one second, after calculating exposure, raise the time in seconds to a power that's characteristic for the film in use (say, 8 seconds would give 8^1.26). Both of these methods require a calculator (unless you can do fractional powers or powers of non-integers in your head).

In the end, though, what you're seeing is just the reciprocity departure of the paper you're using: you've experimentally discovered that paper acts like film, in being slower at low light intensity than with more intense light. Since manufacturers don't document this quality for paper (they expect you'll run enlarging exposures in the 5 to 30 second range and make test strips for each new negative or size), you'll have to do some testing. With a little reading and some math, you should be able to arrive at either a stop multiplier (number larger than 2 to multiply time for each stoop above a few seconds) or a Gainer exponent (power to raise the final exposure, in multiples of the reciprocity limit) for your paper that will be valid until the next time Ilford changes the paper emulsion.

Hugely useful. Thanks for this. Do you have any links or info from your experiences on how I might go about determining the factor I need to determine actual exposure per f-stop? I’m assuming the way to go with this would be to start with a test strip wide open and then figure out the factor from there to be applied which each stop on the scale. I felt like I was going to end up having to test for ideal iso at each f-stop which is fairly annoying but if there’s a way to shorten that process by defining a factor, that would be great. Strangely though, I haven’t seen this before with paper negs although now I’m unsure if I was previously just testing for speed more broadly (I don’t think i was). I’ve usually been able to just use the determined iso across the scale. I’ve definitely never seen testing at f16 and then wide open is completely overexposed to black.

Thanks again for this info. Really appreciate it. This was (still is) driving me nuts.

Don_ih · Jan 27, 2022

Given the odd aperture markings for your lens, maybe you should measure the diameter of the aperture and see if those markings are correct.

Fragomeni · Jan 27, 2022

Don Heisz said:
Given the odd aperture markings for your lens, maybe you should measure the diameter of the aperture and see if those markings are correct.

Those aperture markings were commonly used with lenses of this vintage and the lens is in original condition with the aperture working flawlessly. I don’t think it’s a matter of the aperture diameters but it is something I will probably check just to be certain.

Donald Qualls · Jan 28, 2022

I think the intention there was to point out that there were multiple different aperture scales in use as late as the 1920s -- until ASA/ANSI and DIN, in the US and Germany respectively, established standards that eventually everyone got on board with.

For instance, there was USA aperture, where each halving of area (which we'd be used to seeing as the change from, say, f/4 to f/5.6) doubled the number -- so you'd go from USA 4 to USA 8 as a single increment. There was at least one other that was also based on the focal ratio (focal length divided by entrance pupil) by that used a different starting value -- so instead of 1, 1.4, 2, 2.8 and so forth with powers of the square root of 2, they ran (IIRC) from 4.5 to 6.3 to 9 to 13 etc. Those at least work the same as what we're used to, they're just 1/3 stop off all the way along.

But unless your aperture scale has each number at 1.4 times the previous, closing down may be reducing the light more than you think it is.

Two23 · Jan 29, 2022

I regularly use old lenses. The scale you mention was used on pre-war German cameras/shutters. I've found it to be accurate.

Kent in SD

Donald Qualls · Jan 29, 2022

Two23 said:
The scale you mention was used on pre-war German cameras/shutters.

That sounds right, and one of the reasons f/6.3 was so common on mid- to low-cost models (and f/9 was the largest aperture on box cameras with more than one choice).

rknewcomb · Mar 30, 2022

Wasn't that older scale called US Stops? The way to match those to current stops was I believe that f16 was the same value in both systems.

Also, i believe that Don is probably correct in his idea of reciprocity failure coming into play. May be a combination of two factors.
FWIW
Robert

Fragomeni · Mar 30, 2022

Meant to reply before. Donald was right in that I was running up against reciprocity. I charted it out for the paper and all has worked well since doing that. Thanks everyone.

Donald Qualls · Mar 30, 2022

Glad to hear it's working for you.

Ayne · Mar 30, 2022

Fragomeni said:
f-stop scale

https://en.wikipedia.org/wiki/F-number
There were many different numbering sequences for fstops the charts on this page are good.

Different paper speed at different fstops (antique lens or something to do with paper negs)?

Privacy & Transparency

Privacy & Transparency