Shooting on Bulb

[velvetjones]

One issue that I can't seem to figure out is how to properly expose shots on bulb setting...or longer than 30 seconds. Is there some sort of formula to use...will a light meter calculate that for me?

[Scottes]

Easiest way is to set the camera on manual and the aperture wide open, and see what the meter says. If necessary punch the ISO up too, until you get an exposure under 30 seconds. Then start adjusting back to the aperture and ISO you really want, doubling the time for every full stop.


So let's say you have to set the aperture to f/4 and ISO 400 to get a reading under 30 seconds - for simplicity we'll say that the camera picks exactly 30 seconds. But you really want f/11 for good depth of field and of course you want ISO 100.

f/4 to f/11 is 3 stops - to f/5.6, f/8, f/11. So you have to double 30 seconds 3 times - to 1 minute, 2 minutes, 4 minutes.

ISO 400 to 100 is 2 stops - to 200, 100. So we double 2 more times - to 8 minutes, 16 minutes.

So for for our example you'd set the camera for f/11, ISO 100 and an exposure of 16 minutes.

Since 16 minutes is a long time to wait I'd cut that in half - thus underexposing 1 full stop - and do a picture at 8 minutes and check the histogram. Judging the histogram will give you an idea for the next try. Sure, the 8-minute shot might be an extra, unnecessary step, but why wait 16 minutes to find out it's messed up? :)


Sure, a light meter would do this, too, but this way is quite a bit cheaper.

[scottbergerphoto]

Nicely done! Good explanation!
Scott

[velvetjones]

Thanks for the great explanation...It seems pretty clear. However, I don't get what you mean by histogram...I don't have a digital if that has anything to do with it.

As far as the light meter goes, my friend has one and told me that it cannot meter for exposures longer than about a second...whatever that means. Are you saying that I could just buy a light meter and (taking a cityscape shot at night) simply hold it up to get a reading if I need a 2 or 3 min. shutter speed? I am thinking about getting one anyway, and I just wanted to confirm that it is in fact capable of doing that...

[Scottes]

Thanks for the great explanation...It seems pretty clear. However, I don't get what you mean by histogram...I don't have a digital if that has anything to do with it.

Ow! I feel ashamed for making the assumption. A histogram exists on many/most digital cameras, and it shows a graph of the exposure ranging from light to dark.

With film this does make it more difficult, but it's still possible exactly as explained above, but really requires bracketing. Film has a "problem" known as reciprocity failure, which basically means that long shutter times will cause film to begin to underexpose as the time increases. The above example shows a final exposure time of 16 minutes, but at that length of time film may underexpose by 1 or even 1-1/2 stops. So you have to adjust the shutter time and add another stop (at least) and go to 32 minutes.

Because reciprocity failure varies by film, experimentation is necessary, and thus bracketing is necessary until you get used to that film at long shutter times. Because it's so uncertain the above example would probably require at least 1 extra stop. I would then take at least one more picture at +1-1/2 stops, 48 minutes. Hopefully one of the two would come out correctly. Being paranoid I'd take 4 pictures - 24 , 32, 48, and 64 minutes.

There are links out there which should list reciprocity failure adjustments for the more popular films.

As far as the light meter goes, my friend has one and told me that it cannot meter for exposures longer than about a second...whatever that means. Are you saying that I could just buy a light meter and (taking a cityscape shot at night) simply hold it up to get a reading if I need a 2 or 3 min. shutter speed? I am thinking about getting one anyway, and I just wanted to confirm that it is in fact capable of doing that...

If the light meter maxes at 1 second then calculating a 3-minute exposure would be very difficult, if not impossible. But not all light meters max at 1 second. The Sekonic L-558 (http://www.sekonic.com/Products/L-558.html) lists:

Film speeds:
ISO 3 to ISO 8000 (1/3 steps)

Shutter speeds:
Ambient: 30 minutes to 1/8000 second (full, 1/2 or 1/3 steps) Plus: 1/200, 1/400

Flash:
30 minutes to 1/1000 second (full, 1/2 or 1/3 steps) Plus: 1/75, 1/80, 1/90 1/100, 1/200, and 1/400

(Granted, that's a pretty darn fancy $640 light meter that does 1-degree spot and incident metering.) But that means that this particular meter can do what you want with no issues. And you *do* have a meter in your camera. Because of reciprocity failure it's going to take some experimentation anyway, so experiment a bit and see if it works before spending a few hundred bucks on a meter.

//Edit: Thinking about the various dark shadows and bright lights in a night cityscape, you're probably going to want a fairly tight spot meter to ensure that you don't blow out streetlights and lit windows. A 1- or 2-degree spot meter won't be cheap.

[Scottes]

Nicely done! Good explanation!

Hush! Talk like that is going to ruin my reputation around here, and people might start thinking that I actually know something. :wink:

Thanks.

[IndyJeff]

Scottes I am impressed. I would know how to do something like that but I don't know if I could explain it.

Do you know of the F16 rule?

[PhotosGuy]

Hey, Scottes! Great job!

Another thing to consider is what are you taking pics of? For instance, if I had to do a skyline of a city (using slide film) I'd shoot some 2-3 stops underexposed while there was still some light in the sky. Then I'd double expose over them with the actual night shot.

This could even be done with a Nikon by marking the film when I loaded it so that I could reload it later in the same 'registration'. If you try it, be sure that the tripod is firmly attached to something that won't move! It's wouldn't hurt to have an assistant there, too (bathroom breaks).
:)

[nosquare2003]

Easiest way is to set the camera on manual and the aperture wide open, and see what the meter says. If necessary punch the ISO up too, until you get an exposure under 30 seconds. Then start adjusting back to the aperture and ISO you really want, doubling the time for every full stop.


So let's say you have to set the aperture to f/4 and ISO 400 to get a reading under 30 seconds - for simplicity we'll say that the camera picks exactly 30 seconds. But you really want f/11 for good depth of field and of course you want ISO 100.

f/4 to f/11 is 3 stops - to f/5.6, f/8, f/11. So you have to double 30 seconds 3 times - to 1 minute, 2 minutes, 4 minutes.

ISO 400 to 100 is 2 stops - to 200, 100. So we double 2 more times - to 8 minutes, 16 minutes.

So for for our example you'd set the camera for f/11, ISO 100 and an exposure of 16 minutes.

Since 16 minutes is a long time to wait I'd cut that in half - thus underexposing 1 full stop - and do a picture at 8 minutes and check the histogram. Judging the histogram will give you an idea for the next try. Sure, the 8-minute shot might be an extra, unnecessary step, but why wait 16 minutes to find out it's messed up? :)


Sure, a light meter would do this, too, but this way is quite a bit cheaper.

Scottes, you are starting a good reputation of knowing a lot. 8)

But if you want to judge the exposure only, could it be done by f/4, ISO400 for 30 seconds?

[Scottes]

Scottes, you are starting a good reputation of knowing a lot. 8)

Noooooooooooooooo!
:)

Next thing you know people aren't going to accept my bird identifications anymore and my "justabird" IDs are going to have to be replaced with real names.

But if you want to judge the exposure only, could it be done by f/4, ISO400 for 30 seconds?

Yep. It's the same exposure. Ignoring reciprocity failure - which is a huge thing to ignore for film at such long shutter times - f/4 ISO 400 for 30 sec would produce the exact same exposure as f/11, ISO 100 for 16 minutes.

If I were in this situation - having a digital camera with a histogram - I'd do f/4 ISO 1600 for 7.5 seconds - also the same exposure - and then check the histogram and LCD image. If it looked good THEN I'd kick back for 16 minutes. Patience is a virtue. I have no virtues.

[Scottes]

Do you know of the F16 rule?

Yep. Problem is that I am absolutely useless at judging light, so this is useless to me unless I'm out on a sunny cloudless day between 10 and 4.

[RichardtheSane]

Another thing to remember, on film, when doing exposures of more than a minute or so you have to take reciprocity faliure into account.
Basically the longer you expose film to light the less sensitive it becomes to light. So if you calculate your exposure to be 5 minutes, you will probably have to exposeit a bit longet that five minutes to get a balanced exposure.

Some fairly heavy reading bleow :)

http://home.earthlink.net/~kitathome/LunarLight/moonlight_gallery/technique/reciprocity.htm

also google on reciprocity failure
:)

[velvetjones]

Thanks for the great info...I really appreciate it. I am excited to try some of these out....

[Scottes]

Be sure to let us know how it worked out, and share if at all possible.

I'm kinda curious if I was correct about the above procedure since I've never tried it. :)