Long exposures with a bessa - light meter?

[daveozzz]

Hi,
I'm wondering what other people do about this. For taking very long exposures with a Bessa R obviously the TTL won't help you since anything over 1 sec will not register.
I have a Minolta spotmeter F which doesn't really perform much better in low light (I'm talking urban nighttime light and less here ie. 10sec+ exposures).

What do people do to get an idea of how long to expose these types of shots?
Is there a light meter that's particularly suited to very low light conditions?

Thanks,
Dave.

 

[oftheherd]

I can't help you with the Bessa R itself, since I don't own one. However, many of the newer Sekonic meters should work well, as well as combining incident and reflected as well as flash.

Two of the classics which still work well for low light are the Gossen Luna Pro and the Luna Pro SBC. I have both and can recommend both. The SBC has an edge both in sensitivity and in low light response as it has less memory. Either one should work well for you however. The SBC has an advantage of being able to take accessories such as a 1 degree spot accessory, flash accessory, etc. Frankly, however, if you want a system meter I expect those such as Sekonic, Minolta, etc, which combine all in one meter would not only be less expensive, but smaller. The SBC with its 10/5/1 degree accessory is a little hefty.

 

[RML]

Bracket! Take 1sec, 2 sec, 4 sec or whatever period shots. I do it with my R-D1 as well and can immediately finetune the length of exposure. I usually start with a 1sec, 4sec and 8sec exposure (simply counting 21-22-23-24) using a remote release and the iso200 setting. In light polluted areas (like most of Holland) you'll be hard-pressed to need much longer exposures unless you use iso50/iso25 (or less) film.

 

[hugivza]

I can certainly endorse using the Gossen Luna Pro Profisix SBC as a very useful meter - mine has a scale which reads up to 8 hours exposure. Does anyone know though if with a meter such as the Gossen, it takes into account reciprocity failure: the non linearity of exposure time for long time exposures, or is this a matter of trial and error?

 

[oftheherd]

I can certainly endorse using the Gossen Luna Pro Profisix SBC as a very useful meter - mine has a scale which reads up to 8 hours exposure. Does anyone know though if with a meter such as the Gossen, it takes into account reciprocity failure: the non linearity of exposure time for long time exposures, or is this a matter of trial and error?

I think since all films potentially have different reciprocity responses, you would have to compensate based on the particular response of the film you are using. I haven't done that type of exposure to worry about it in many years. I used to sort of SWAK it. I seem to recall that in fact it is non-linear so you have to compensate based on the time. If it is linear, then on the SBC and I think the Luna Pro itself, you can use the setting as you would to dial in filter factors. I don't have the manual here so I don't remember if they also recommend using that part of the dial for reciprocity settings or not, but I think they did. But that would presumably be after determining the difference based on the recommended time from the meter. It sure doesn't hurt to bracket if you can when doing very long exposures (not eight hour ones of course 😀 ).

 

[sabrewolf]

Provia 400 has one of the most forgiving reciprocity failure time (up to 2 minutes I believe)

 

[Kin Lau]

I also chime in that the Luna Pro is very good for low-light. After all, that's how it got it's name, because it was good enough to meter by moonlight.

As others have already mentioned, all films have different reciprocity failure characteristics but it's seems that + 1 stop is about standard for about > 1 sec to < 1 min.

I believe Fuji Acros (b&w) is suppose to be have no reciprocity failure out to several minutes as well.

If you're thinking of using a digital camera to proof the scene (a good idea btw), keep in mind digital sensors do not have reciprocity failure. The response is linear, so you still have to multiply the times.

 

[daveozzz]

Some fantastic info there... thanks very much for your time.. off to ebay to look for a luna pro!

 

[rbiemer]

I have used the "Blcak Cat Exposure Guide" a paper calculator booklet thingy. Some thing like $20 US and it has 25 or so descriptions of different lighting conditions and you then match these up with you film speed and you get a range of "correct" shutter speed/ f-stop settings.
And Bracket like mad.
Rob

 

[Toby]

If you've got a spotmeter f why not take a suitable highlight reading (the brighter part of a building?) and go from there

 

[pvdhaar]

When you're doing very long exposures under low light, there's much less change to get a wrong exposure than many people think. The trick is not to use a lightmeter reading with reciprocity factored in. That is, if you don't want to have your image look like it was taken in broad daylight.

If you want to keep the atmosphere of the night, then any any exposure less than that corrected for reciprocity will give you a usefull image. This holds especially true in an urban environment where there's artificial lighting.

Streetlights and lit subjects will register at almost any exposure, and the rest will turn black anyway. That's because light under a certain level will never register on film. This non-linear behaviour of film will also make that there's very little difference between brackets. You'll really have to have these more than a single stop apart (i.e. for instance 1,2,4 seconds is not spread enough). Better go for three stops to the least (e.g. 1,8,64 seconds)

 

[daveozzz]

That's intersting info... yeah I sometimes forget just how much elapsed time there is between stops once you get over 8secs.

Toby my spotmeter (apart from being ill at the moment) is surpisingly rubbish at registering a highlight at night. I've had on highlight mode, set to 3200 with some huge shutter speed and pointed it directly at a streetlight in an effort to get some anchor point on exposure but just get the old "E".
Then again maybe it's just a dodgy meter.