How to make an eight hour exposure...

[Karl K]

Help is needed before I go out for this shoot. I am planning to photograph the interior of Grand Central Terminal's Main Floor. My camera will be situated next to one of the main staircases on either end, if I can get the necessary permit to set up a tripod for eight hours. The camera is a Leica M3. The lens is a 12mm Voigtlander. The film is Ilford SFX. Thanks to an APUG member, Simon R. Galley, I now have an Ilford SFX filter mounted in a Cokin P holder, which I will mount in front of the 12mm lens. I estimate that the average exposure during the eight hours is four seconds @f/22, without compensating for reciprocity failure and the 16X filter factor. I can only make one exposure in one day, so I want to get as close to a perfect exposure as I can. Does anybody have experience with this sort of thing? Can anyone calculate the correct exposure factor? Thanks in advance.

 

[Donald Miller]

Do you plan on staying with the camera? Are your plans to have the image devoid of people? A long exposure will do that.

Are you planning on the shutter remaining open for 8 hours or do you have some means of remotely or automatically tripping the shutter?

Those are questions that immediately come to mind. There may be others.

I have checked the Ilford site and find no reciprocity information on the film. I would recommend contacting them for guidance on this.

 

[Dave Parker]

I am a bit confused???? Are you going to be making one 8 hour exposure, on one frame of film, or several shorter exposures over an 8 hour period on one frame of film, or several exposures over 8 hours on different frames of film?

 

[Karl K]

Do you plan on staying with the camera? Are your plans to have the image devoid of people? A long exposure will do that.

Are you planning on the shutter remaining open for 8 hours or do you have some means of remotely or automatically tripping the shutter?

Those are questions that immediately come to mind. There may be others.

I have checked the Ilford site and find no reciprocity information on the film. I would recommend contacting them for guidance on this.

Yes, of course I plan to stay with my Leica...this is NYC, after all!

 

[Karl K]

I am a bit confused???? Are you going to be making one 8 hour exposure, on one frame of film, or several shorter exposures over an 8 hour period on one frame of film, or several exposures over 8 hours on different frames of film?

This is a one exposure assignment, so that Grand Central Station will be completely devoid of people, except for that homeless guy who is asleep in the corner and will remain there for the entire eight-hour exposure.

 

[argus]

I don't really get your initial point. You speak about an 8 hour exposure, but you also mention a 4 second exposure at f22.

I bet that with an exposure of 8 hours, the reciprocity failure correction must be within the scale of days ;-) What f-stop do you plan to use? f1024?
Is it the filter that requires that much compensation?

I've got the data here for FP4/HP5 (same factors): 30 seconds exposure will become 157 seconds with reciprocity correction.
I have NO idea what SFX with a filter and its correction will require... Ilford.com does not mention anything about it: http://www.ilford.com/html/us_english/pdf/sfx200.pdf

G

 

[Poco]

Unless the assignment explicitly calls for an 8 hour exposure, I strongly recommend shooting wider open and getting the time down to the one hour (at most) area. That should still get rid of the people for you while giving you time to bracket. I once shot a scene that required an unfiltered 5 hour exposure at f11 and it was a nightmare ...took me three attempts/days to get it right. Seriously, try to get your time down.

 

[reellis67]

... I estimate that the average exposure during the eight hours is four seconds @f/22, without compensating for reciprocity failure and the 16X filter factor...

This seems to be what is confusing people, at least it is confusing to me. Is the base expsoure, meaning no filter factor and no reciprocity, 4 seconds at f/22? The way it is phrased in your question is a bit confusing but I think that that's what you are saying.

If so, I think that you will need to add some neutral density as well, because even with a 4 stop filter, you are looking at 64 seconds if I did the math correctly.

I have notes on Ilford traditional black and white films having a 1.6x reciprocity factor for long exposures, so let assume it's the same for SFX (your guess is as good as mine, perhaps better!) so this will give you 102.4 seconds @ f/22. So now we have something like 1.7 hours. If you added ND to give an additional 1 stop change, you would be around 3.4 hours, and ND for 2 stops would land you around 6.8 hours- pretty close...

Please feel free to correct me on any of this - I rarely do exposures longer than a handful of seconds...

- Randy

 

[Dave Parker]

This seems to be what is confusing people, at least it is confusing to me. Is the base expsoure, meaning no filter factor and no reciprocity, 4 seconds at f/22? The way it is phrased in your question is a bit confusing but I think that that's what you are saying.

If so, I think that you will need to add some neutral density as well, because even with a 4 stop filter, you are looking at 64 seconds if I did the math correctly.

I have notes on Ilford traditional black and white films having a 1.6x reciprocity factor for long exposures, so let assume it's the same for SFX (your guess is as good as mine, perhaps better!) so this will give you 102.4 seconds @ f/22. So now we have something like 1.7 hours. If you added ND to give an additional 1 stop change, you would be around 3.4 hours, and ND for 2 stops would land you around 6.8 hours- pretty close...

Please feel free to correct me on any of this - I rarely do exposures longer than a handful of seconds...

- Randy

Randy,

You hit the nail on the head! I was sitting here playing with the calculator, and comming up with the same numbers as you are, but was still confused a bit with how he phrased his question, thank you, you have restored faith in my ability to think!!!

LOL

Dave

 

[ras351]

If so, I think that you will need to add some neutral density as well, because even with a 4 stop filter, you are looking at 64 seconds if I did the math correctly.

Your response makes more sense than the original posters question.

If the above is what is being asked the remainder of my comments are directed at Karl. Firstly if you use an SFX (or other IR) filter you'll need to check to ensure that the filter factor for any ND filters you also use applies to IR frequencies as well. Some ND filters only block visible light and let pretty much all the IR frequencies through. Also getting an exact exposure is going to be questionable at the best of times with an IR sensitive emulsion and filter unless you have a meter which properly meters those frequencies. For an indoors estimate it'll be even more of a problem where I couldn't even begin to make a guess as to the amount of IR radiation and hence exposure compensation needed. I don't wish this to sound negative but what I'm suggesting is that if you only have one shot at this it would be worth testing the film, filters and camera/meter in a similar environment beforehand to resolve as much of this as possible.

Roger.

 

[Helen B]

I also think that there is no alternative but to do tests. Presumably each film has a point at which the intensity of light falling on it is too low to form a stable latent image - so there is a danger that in designing for an extremely long exposure time you could end up with no image. Extreme reciprocity failure resulting in no viable image centres except in the brightest parts of the scene. This can only be determined by testing, surely.

Best,
Helen

 

[Bob F.]

I would also suggest that you trial it. Meter shadow & highlights in the station during the day and set up in your bedroom/garage/etc with the same lighting levels and burn a few rolls of film. That will let you hone in your development times too as the kind of reciprocity error you are hitting is going to do interesting things to the film curve...

Good luck getting a permit if you are going to be where people need to walk past - I'd concentrate on getting that first.

Cheers, Bob.

 

[Helen B]

I don't know why I didn't think of this before, but the obvious answer is to use a development technique that dramatically reduces the sensitivity of the film, rather than decreasing the intensity of the light falling on the film. This gets round the extreme reciprocity failure problem, but it could lead to other problems.

So folks, what suggestions? Open the aperture and see if you get an image without development (ie raise the intensity of light on the film)? Probably not enough. Post fixation physical development?

Best,
Helen

 

[Karl K]

Eight Hour Exposure continued

Sorry, guys, I thought I made it clear. The basic (no filter) light meter reading is four seconds @f/22, with 200 ASA. I am using the Ilford SFX filter, which has a filter factor of 16. I want to make the exposure 8 hours. Why? Because that will eliminate all traces of movement. I can do the math for a filter factor of 16. I can't figure out the reciprocity and the filter factor of 16 together. I also have available the proper neutral density filter, should that become necessary to use. That would make the calculation even more complicated. Basically, I'm worried about receiprocity for such a long exposure. Ilford says that they don't recommend that SFX be exposed for eight hours, but I want to give it a try anyway.

 

[David]

You don't need 8 hours to eliminate the movement. Having done extensive shooting in cathedrals with ULF cameras I speak with some experience. Needing shadow detail in the ceilings necessitated exposures that averaged 4 hours. One five hour shot had the janitor buffing the center aisle for ~2 hours and he never showed at all.

Also, there is a difference between a 3 and a 5 hour exposure for shadow detail. After 5 hours the reciprosity never seems to catch up with itself and additional time was of limited value.

I was using Bergger BPF 200 at the time and have no experience with SFX. Good luck and don't be afraid to test out the responses before an important shoot. Oh, yeah, and get a really sturdy tripod.

 

[eumenius]

In my eyes, not all the cameras and films are intended to be used with such exposure times - I am thinking about light scattering inside the camera. In eight hours it can be a real foggy trouble using a 35mm camera. With LF and ULF the things are way much simpler thanks just to the bigger volume of bellows, and effective absorption of scattered light by bellows' ridges. But Leica R, 135 film with its less-than-perfect antihalation layer, and extra-wideangle lens... to me, it just doesn't seem to be a proper tool set for the job. Maybe I am wrong, of course.

Zhenya

 

[ras351]

Sorry, guys, I thought I made it clear. The basic (no filter) light meter reading is four seconds @f/22, with 200 ASA. I am using the Ilford SFX filter, which has a filter factor of 16. I want to make the exposure 8 hours. Why? Because that will eliminate all traces of movement. I can do the math for a filter factor of 16. I can't figure out the reciprocity and the filter factor of 16 together. I also have available the proper neutral density filter, should that become necessary to use. That would make the calculation even more complicated. Basically, I'm worried about receiprocity for such a long exposure. Ilford says that they don't recommend that SFX be exposed for eight hours, but I want to give it a try anyway.

Reciprocity failure is usually applied after the metered exposure is known. You're going backwards so if you want an eight hour exposure you adjust for reciprocity from this eight hour value to your metered exposure ignoring all filter factors. You then divide this number by your 4sec base exposure to give you the total filter factor needed. You can then divide this by 16 to adjust for your SFX filter which will give you the ND filter factor required.

The problem is that I doubt that many (if any) have any notion of what reciprocity factor to apply in this case. Ilford don't publish it on their SFX datasheet and I personally haven't had any exposures with SFX longer than about 10 seconds. You will need to be aware that the reciprocity failure adjustment (usually log(-1/x) order) may be infinite for the times you're talking about in which case it cannot be done. Another possibility at these time scales is that an infinitely small change in x (metered exposure) correlates to a significantly large change in y (adjusted exposure) which means a very small filter factor adjustment can be the difference between a completely blank negative and a completely overexposed negative. Because you're dealing with severe reciprocity you may find that somewhere in the middle of these you end up with a lith negative (ie areas which have had light above a certain threshold are completely exposed and anything below are completely unexposed).

Put simply you're going to have to do tests unless you're lucky enough to find someone who's done exactly the same thing with exactly the same film.

Roger.

 

[Karate Dad]

Oh, yeah, and get a really sturdy tripod.

I'm pretty new to this so I was wondering...I've spent a lot of time in Grand Central. With all the people and the trains the floor does vibrate quite a bit. Wouldn't that affect the sharpness of the image or does the long exposure and/or tripod negate that?

..john

 

[Helen B]

...

Also, there is a difference between a 3 and a 5 hour exposure for shadow detail. After 5 hours the reciprosity never seems to catch up with itself and additional time was of limited value.
...

That's exactly the point I was trying to make. For every film there is a light intensity that will never create a latent image no matter how long you expose the film for. What is happening, highly simplified, is that silver atoms are being created in such low numbers at such a low rate that they do not stabilise - they die a lonely death before they find a mate to help them survive.

Best,
Helen

 

[Dracotype]

A quick consultation of the Ilford's reciprocity tables, 64 seconds doesn't even show up. In fact, the curve may approach 64 seconds asymptotically close, but might never actually get there. This has already been stated very emphatically by others here, but it is an important fact. The curve flattens out so much at the extremes that any increase in the exposure time will not affect density at all. Change your aperture to something smaller.

Drew

 

[David]

Because you're dealing with severe reciprocity you may find that somewhere in the middle of these you end up with a lith negative (ie areas which have had light above a certain threshold are completely exposed and anything below are completely unexposed).
Roger.

When I was doing the really long Cathedral exposures I couldn't find anything in the way of good information. The times were just too far out of the norm. You just learn by doing. What I did find through experimentation was that with the really long exposures it was almost impossible to blow out the highlights - something that was both unexpected and counter-intuitive. I develop by inspection but the decelopment times were pretty close to normal (whatever that is).

 

[c6h6o3]

I want to make the exposure 8 hours. Why? Because that will eliminate all traces of movement.

You don't need anywhere near 8 hours to do that. Way, way less time is needed. Take a look at Storrow Drive by David Fokos. (Scroll down. It's at the very end.)

8 hours is a little like going after a squirrel with a .458 Winchester Magnum.

 

[David A. Goldfarb]

You've discussed this with the Station Master at GCT? It's not hard to get a tripod permit for GCT, but they will limit where you can set up (no stairways, as I recall), and not during rush hour in general. Maybe if you can shoot on a weekend, you can find an 8-hour window in which to shoot, if you decide that you really need 8 hours.

I'd definitely do some testing. If you've got more than one camera, you could also set up a range of exposures. With two cameras, for instance, you could try 8, 5, and 3 hours.

Have you considered doing this as a pinhole shot? That and an ND filter will extend your exposure time pretty far. Finney probably makes a pinhole cap for your Leica. Calumet sells them.

 

[Jim Jones]

. . . Have you considered doing this as a pinhole shot? That and an ND filter will extend your exposure time pretty far. Finney probably makes a pinhole cap for your Leica. Calumet sells them.

A 35mm pinhole shot with a 12mm focal length will produce sharp billfold size prints. The goal for such an investment in time could be larger images than that.

The formula I used for determining pinhole exposures of an hour or more with several common films of a few decades ago was:

actual exposure = (metered exposure / .6) raised to the 1.4 power.

A light meter and a cheap cientific calculator saved a lot of tests. However, even back then, some films departed considerably from this reciprocity failure calculation. Astrophotographers may have experience with long exposures on Ilford SFX. A google search for astrophotography may show much better information. Perhaps your ND filter reduces the light to about the level of the night sky. If so, your range of test shots will be narrowed. Remember, the relative IR content of the light may vary from day to night in Grand Central Station.

 

[Claire Senft]

I think David id givinhg good advice in sugessting a test. If GST is close to you got there and take a light reading. Duplicate that light level at home and test and bracket some exposures. If you wish a view with no people take your photo at 2-3 am Monday monning. Granted I have never been to GST but I am guessing that the traffic at time is light enough that you could take several exposures of 2 minutes each without any people in them. With your camera on a tripod 5 feet from the floor I would guess that you can set your focus to have everything thing sharp @ f8 At f22 on a 12mm lens you have a diameter of about 1/2mm. This is a very small aperture and will mosy likely cause a good deal of diffraction. Leitz has. in the past stated that any physical aperture smaller than 4mm will lead to increasing losses thru diffraction. F22 would allow. at a medum green wavelength apprix 64lpm.
F8 would allow 200lpm. Obviously, you will not get either level of resolution because of film characteristics etc. If an exposure that is not corrected for reciprocity is 64 seconds on one hand at f22 it will be 8 seconds at f8. I have not used the film in question but I doubt that more than 45 seconds would be required at f8 when correctiing for recprocity.

Good luck with your photo.