Light Meter Reliability in Sunlight?

[peter_bongard]

Hi!

After having tested light meters of different Brands (Sekonik and Gossen), I

noticed on both models a severe underexposure (about 1 1/2 to 2 stops) when

used in strong sunlight. I'm talking of the incident method here, with the

white dome over the cell. My question is whether this is normal - so that I

always compensate the reading when using the meter in strong sunlight - or not.

Any help would be apprechiated.

Thanks!

Peter

[maury_cohen]

In sunlight your reading should always be close to f/16 at a shutter speed 'equivalent" to your ISO; e.g. 100 ISO would be 1/100 (or 1/125) at f/16.

[Ed_Ingold]

Incident light readings are among the best and most consistent readings you can take. I'm more inclined to trust the Sekonic meters over the Gossen meters which have a small dome that slips over the reflected light port.

 

Make sure you are using the meter correctly. The dome should be at the subject and pointed toward the camera. In sunlight, this will place the dome half in shade, which might account for your error. Any light meter requires skill on the part of the operator. You need to keep score and learn what works and what doesn't.

[helenbach]

An incident meter reading is equivalent to a reflected reading from an 18% grey card, if the two are done correctly. An 18% grey card is only two and a half stops down from a diffuse perfect reflector - ie 100% reflecting. That means that, other than specular reflections, it is impossible for anything in the picture to be more than two and a half stops brighter than the value you are metering. This is often a waste of the 'overexposure latitude' of the film.

 

Similarly, the darkest object in full sun will probably be around 3% reflecting, or about two and a half stops down from 18%. That range would be easily accomodated by film. Objects not in full sun, however, will probably be more than two and a half stops darker than the 18% grey in full sun because of the difference in illumination. They may then be underexposed.

 

There are two easy techniques to get a better match between the scene brightness range and the film latitude:

 

Set the incident meter to half the box speed of the film and meter in full sun, pointing the meter at the camera. This places the brightest possible non-specular object three and a half stops over the metered value, or

 

Set the incident meter to the box speed and meter in the open shade (not deep shade), such as in the shadow of your body, but still pointing towards the camera. This places the likely darkest tone (except for those in deep shade) two and a half stops down from the metered value.

 

Of course you can refine the meter settings based on your experience or on tests.

 

As long as the lighting conditions are the same, you don't have to meter from the subject position, you can meter from the camera position. Just point the meter in the same direction as it would be pointing if it was in front of the camera, pointing to the camera.

 

Best, Helen

[david_henderson]

I'd be interested to know how you know the incident readings are inaccurate and by how much? Are you taking shots at the exposures indicated by the meters in incident mode, and judge that these are variously 1/2-2 stops under ? Or are you comparing the results of meters in incident and reflective mode?

 

Where I'm coming from is that incident and reflective meters measure different things. I would not expect a reflective reading and an incident reading to be the same unless the subject read by the reflective meter were a mid -tone.

 

I'm not trying to promote incident metering here- in fact I never use it myself because I use high contrast slide film and find multiple spot-meter readings far more useful- but that's not the same as saying that the readings are inherently wrong. And if you're getting similar results from two different brands of meter it is making me think that there's an interpretation issue here rather than two bad meters.

 

Let me give you an example. I'm sitting here in a fairly gloomy room in England with no direct sun at this time of day. I have a video box nearby which seems to require rather more than 4 seconds at f11/ISO 400 on an incident reading. A spot meter however tells me that I need 2 seconds at the same aperture/ISO. Is one wrong? Not at all. The reflective reading tells me what I need to make the box look like a mid-tone. But it isn't- it looks to me like its a good stop lighter than a mid tome and so to represent it naturally, as the eye sees it, I need to open up a stop or so to about 4 sec to render it lighter. Which is what the incident reading tells me.

[peter_bongard]

I noticed that a) some of my shots taken in direct sunlight were underexposed, b) the reading differs from the one in my Leica M6, and c) the reading differs from the reading in my Digi-SLR (Canon 1d Mk2). Its histogram indicates that there's an underexposure by 1 to 1 1/2 stops. However - when measuring during overcast sky, the sekonic meters correctly.

[dan_fromm2]

Peter, your meters have aged and gone nonlinear. Quality Light Metric can make them right.

[demianov]

I have THE SAME problem with my Sekonic L-608! I've noticed it about 2 years ago and hardly done anything since. The readings, however, are WAY off, by as much as one and a half stops. My meter has been reliable and consistent for over 4 years for ANY kind of readings EXCEPT this: incident (Lumidome/disc) in DIRECT sunlight. The lower the sun the more underxposure I'd get relying blindly on the readings provided. MAYBE, just MAYBE it has to do something with the photo-diode being sensitive more than needed to UV light. Or maybe IR, who knows? I'm thinking that because normally not much of IR or UV light is present in scenes not lit by direct sunligh while the sun emmits plenty of both kinds.

[helenbach]

As I explained, if you use an incident meter in direct sunlight without making any film speed compensation you are placing your midtone only two and a half stops below the brightest possible white in the scene. In most circumstances that results in underexposure. In Zonespeak you are placing the brightest non-specular highlight on Zone 7.5.

 

Best,

Helen