Checking Light meter accuracy- Did I do something wrong?

Discussion in 'Beginner Questions' started by johnmikka, Sep 7, 2020.

  1. So I just got this light meter Sekonic 358, and I was hoping to use it to check the accuracy of my old Nikon FM with a built in spot meter. But obviously it's either I did something wrong or the Sekonic is faulty. Can someone help me?

    I know for a fact that my Nikon's meter is more or less accurate, but no where as far as to 3 stops. The Sekonic is a new thing to me.
    Both at ISO 400, T=1/30,
    My Nikon reading is F-11
    My Sekonic is F-5.0 with sphere down
    F-6.3 with sphere up.
    Both readings are against the same spot on the window curtain.

    MEDIA REMOVED - Video is "private"
    Last edited by a moderator: Sep 8, 2020
  2. Also, I understand that the SLR meter is averaging the whole area where the lens(50mm) sees, but even so I use the Sekonic to meter around the area, the highest reading is f-6.3, lowest is 5.0 so I don't understand how the SLR got the F-11 reading..
  3. Welcome to I think your meter and camera are working correctly. You need to understand the difference between reflected light metering, and incident light metering.

    In reflected light metering, the meter measures the light being reflected back from the subject. Incident light metering measures the intensity of the light falling on the subject. The camera is measuring the reflected light from the curtain. With the dome in position, the Sekonic is measuring the incident light coming from the curtain. You need to remove the dome from the Sekonic and point it towards the curtain, when you should find much better agreement - although different meters rarely agree exactly.
    William Michael and NetR like this.
  4. thank you for the response John, I still don't quite understand. I know the camera is measuring the reflected light, and the meter is measuring the light falling into the dome. But I'm pressing both meter against the curtain, so why is it not in unison?
    How does removing the dome help the reading, since I've already kept the dome tucked in to prevent surrounding light getting in from the sides.

    Thanks for the help.
  5. A followup video, makes me even more confused to think what if I didn't have the meter in the camera, and only rely on an incident meter, how will I expose the curtain correctly?

    MEDIA REMOVED - Video is "private"
    Last edited by a moderator: Sep 8, 2020
  6. William Michael

    William Michael Moderator Staff Member

    Reflected light meter readings are affected by the colour and saturation of the object from which the light is reflected:Incident light meter readings are not.

    Because the curtain is white, or close to white: see the above explanation. If the curtain were black the SLR's meter would indicate about F/2.8.

  7. Ok, I looked at your video and then a quick look at that Sekonic meter (on Amazon). Best I can tell, you are using the Sekonic solely as an incident meter. As described (on Amazon) when the dome is retracted, it is an incident reading for a flat surface or light from a specific direction. With the dome out, it takes an incident reading for a 3D object (like a persons face). Pointing the dome at your curtin (a light source), is measuring the light falling on an object lite by that window, not the proper light reading for a photo of the curtin. The sekonic reading would be fine if you wanted a photo of something right in front of the window, taken from right outside the window.

    To take a reflective meter reading with the Sekonic, I believe you need to buy a Lumigrid attachment .
  8. In other words you are saying, if my camera doesn't have a spot meter built in there is no way I can expose the curtain properly with a incidence meter?

    But I don't understand being a white curtain will have 3-4 stops in difference?!?!?
  9. Also, I noticed that in this scenario, the llight mainly comes from the window which is behind the curtain, there aren't much light shinning upon from the other side, so I'm not sure if the reflected light from the curtain being white will cause a F-11 vs F2.8 or a difference? help me to understand... i'm so confused...
    I tested the meter and camera spot meter again, under the sunlight over a surface, and both reading seems to match, so I think there is something I'm not understanding correctly here.
    My main concern is if this curtain situation were a real life scenario, and my camera didn't have the internal meter, how would I meter it with only an incidence one, am I supposed to think if it's a white curtain I have to +3/4 stops???
  10. William Michael

    William Michael Moderator Staff Member

    Correct: you have answered the question.

    The camera's light meter is reading the reflected light off the curtain and also the light shining through the curtain from the outside of the window.

    The incident light meter when held against the curtain and looking into the room, is measuring the light falling on the curtain, sourced from inside the room.

    That would account for a large difference in the reading.


    This scenario is not a good example of the question that you're asking "What if I wanted to photograph...?

    it is unlikely that you would want to photograph a curtain with light shining through it. Perhaps a close example would be a stained glass window, from inside a building with the sun illuminating the glass from the outside.

    In this situation a reasonable method would be frame the window only in the shot (i.e. no walls) and to use the 'Matrix' (Nikon) metering in the camera for the exposure reading for the stained glass.

    If the shot was critical, then consider bracketing three shots - I'd go one stop over what was indicated and one stop under what was indicated, this is assuming you're using negative film, which typically would have a reasonable exposure latitude.


    OK, what is more likely reality is taking the curtain scenario and putting a person in front of or in the place of the curtain, then if you wanted to use the INCIDENT meter, typically you would meter near the face of the person and point the incident meter to the camera; this would provide the exposure reading for the portrait of the person.

    However, depending upon the intensity of the light falling through the window, the background might be blown out, which is typical of a "back lit" scene - see example below:


    Last edited: Sep 7, 2020
  11. "In other words you are saying, if my camera doesn't have a spot meter built in there is no way I can expose the curtain properly with a incidence meter?" I would not use an incident meter to take a light reading of a light source. It is designed to measure light falling on an object from light sources. A spot meter is a reflective meter with a small area of measurement. Using your Nikon's center-weighted meter close up to that curtain is not much different than if it also had a spot meter and you used it at a distance to take a meter reading of the curtain.
  12. But if you see the video, I have pointed the incidence meter facing towards window, and got F-5.0 still 3 stops in difference.
  13. I meant the real life situation where I want to control how much the curtain will be blown out (or not), how would I do that.. Not everyone shoot/meter around the human subject.
  14. I'm sorry, but I'm so confused right now.. and it's not your fault, But

    I was told when you take an incidence reading (esp, in studio setting) you should point it towards the light source for a more regional reading.... but let's say there is a person standing near this curtain, wouldn't I point the incidence meter towards this same way as I did in the video I posted? facing to the window to get the amount of light projecting through the curtain?
  15. I think I figured out what went wrong here.
    If you see my 2nd video with the dome down pointing towards the window reads F2.8
    My camera reads F5.6, I guess if I dial the dome to up position will get me half a stop over? And the rest can be explained by the white curtain factor affecting my matrix meter inside the camera?
    I will experiment more tomorrow when the sun comes up again :D
  16. William Michael

    William Michael Moderator Staff Member

    That technique is used in a studio setting to get incident readings of the different light sources.

    For example if you have a Portrait Shot,which is front lit, and you have a Key light and a Fill Light - you can take two incident readings one with the meter pointing towards the Key light and the other with the meter pointing towards the Fill light. This method was taught in structured classes and is seen text books. From those readings you can a calculate the "Lighting Ratio".

    Typically, using the method above, you would use the Key Light reading as the exposure for the portrait, thus the Fill Light side of the subject would be rendered darker, in shadow.


    Correct. Again that's because the incident meter is reading the light from the curtain.

    I think that you're not understanding a fundamental concept - in simple and basic terms:

    You use an INCIDENT meter by placing it in front of the Subject you want to photograph and point the dome towards the camera.

    This will give you an exposure reading for that Subject, irrespective of the colour of the Subject, because the meter reading is based upon the light illuminating that Subject.

  17. Yes. I just did some experiment myself. I guess I didn't think it would actually be THAT much of an exposure difference only due to the color itself
    . But apparently it does... valuable lesson learnt, again for free through the internet :D

    MEDIA REMOVED - Video is "private"
    Last edited by a moderator: Sep 8, 2020
  18. But does this also mean if I'm filming a person wearing a white shirt and black pants, it will be almost impossible to get both piece of clothing in the "correct" exposure, (or similarly a really black person and a really white person)?
  19. I guess to answer my own question, is if "correct" means neutral grey... then yes, it's impossible.
  20. AJG


    If you're using an incident meter correctly then the shirt should be white and the pants should be black in the photograph that you take. The situation you're describing is precisely where a reflected meter, especially a spot meter, can give you the wrong exposure. An incident meter isn't fooled by a scene that is very dark or very light--it gives you an exposure that should render those scenes accurately.
    William Michael likes this.

Share This Page