Not matching light meter values

[Fatih Ayoglu]

Hi all,

So I have 3 light meters;

• Sekonic L308X, incident and reflective light meter
• Sekonic L-28C, incident and reflective light meter
• Pentax Digital Spotmeter, reflective light meter

When I measure reflective light, it all shows the same EV value, which is great.

But when I measure incident light while the Lumispheres are on with both Sekonics, L308X measures 0.5 EV less than L-28C. Certainly, I can put the ISO value down on L-28C and be fine with it but wanted to understand why it would create this difference.

Cheers,
Fatih

 

[xkaes]

I'm not familiar with those meters, but ONE factor might have to do with how fine a read-out they have. My only digital meter shows 1/3f readings, so two meters might be reading the same but easily show a 1/3f difference. With my analog meters, I don't have that problem.

Another thing to consider is dust on the photocell, the positioning of the dome (assuming it is a dome), battery voltage, etc.

 

[MattKing]

The three meters might have different integrating domes, and different angles of acceptance.

 

[Mick Fagan]

Around 30 years ago I worked in an industrial photo lab, attached to the lab were approximately 14 studios where photographers were, on an industrial scale, photographing product for magazine and catalogue reproduction, two shifts a day.

I don't really know how many light meters we had, but it would have been at least 14, plus some spares. Whenever a new batch of light meters were purchased, they were sent off to be calibrated; specifically for incident reading using flash. In general as I understand it, the difference between them when new was around 1/10th of a stop, with the odd one being 2/10ths of a stop different than the rest.

Once calibrated they were all reading as near to identical as possible. Every few months they would be checked, but from memory I don't think any of them ever deviated from their calibrated state. Being dropped or accidently kicked was another matter, but by and large they were good to go, seemingly forever.

We always used 1/10 of a stop light meter readings, as I do now with my newish Sekonic L308-X.

I recall the saying "Man who has two watches, never knows correct time!"

 

[Mr Bill]

When I measure reflective light, it all shows the same EV value, which is great.

But when I measure incident light while the Lumispheres are on with both Sekonics, L308X measures 0.5 EV less than L-28C. Certainly, I can put the ISO value down on L-28C and be fine with it but wanted to understand why it would create this difference.

Well, all I can do is to guess. I DID look up the two Sekonic meters and see that the L-28C model was apparently made between 1970 and 1976 ( ?). And uses a selenium cell vs a silicon photo diode in the later meter.

Regarding meter calibration, per the ANSI standard a formula is used which includes a so-called 'K' constant for reflection meters or the 'C' constant for incident meters (as I recall). Each of these "constants" has a moderate range of permissible values, set by the manufacturer at their discretion, which affects the reading.

So, first possibility is that the 1970s Sekonic used a different constant for incident readings than the "modern" Sekonic. (Per the ANSI standard, as I recall, the meter should be labeled with the constants used, or perhaps only listed in the user manual.) Or maybe some generic repair shop recalibrated one of the meters to their test setup and not to the factory spec.

A second possibility for the discrepancy is related to the spectral sensitivities of the two different sensor types. I'm presuming that you used the same light source when you did your testing, where the reflective readings seem to rule out such an issue. But... perhaps the ~50-year-old lumisphere has taken on a color that is problematic?

A third possibility is that that (at least) one of the meters is not linear with light intensity. Loosely meaning that they may agree/disagree more at different light levels.

Personally I would probably just use the newer Sekonic. But if you wanted to investigate more you might do a couple of tests. I'd suggest to compare meters under several types of light "color," as indicated by the "color temperature." A shaded area on clear sunny day will be lit mostly by the blue sky, with color temp probably over 10,000 Kelvin. Direct sun, same day, probably around 5,500 K (?). Or an old-style 100 Watt tungsten bulb around 2,800 K. Fwiw the metering standard is, as I recall, about 4,700 K.

Another test would be at different light levels. Which ought to be fairly easy to do.

 

[Fatih Ayoglu]

The three meters might have different integrating domes, and different angles of acceptance.

Yes, old Sekonic has a much larger dome so probably it takes more/different angle lights.

 

[Fatih Ayoglu]

Well, all I can do is to guess. I DID look up the two Sekonic meters and see that the L-28C model was apparently made between 1970 and 1976 ( ?). And uses a selenium cell vs a silicon photo diode in the later meter.

Regarding meter calibration, per the ANSI standard a formula is used which includes a so-called 'K' constant for reflection meters or the 'C' constant for incident meters (as I recall). Each of these "constants" has a moderate range of permissible values, set by the manufacturer at their discretion, which affects the reading.

So, first possibility is that the 1970s Sekonic used a different constant for incident readings than the "modern" Sekonic. (Per the ANSI standard, as I recall, the meter should be labeled with the constants used, or perhaps only listed in the user manual.) Or maybe some generic repair shop recalibrated one of the meters to their test setup and not to the factory spec.

A second possibility for the discrepancy is related to the spectral sensitivities of the two different sensor types. I'm presuming that you used the same light source when you did your testing, where the reflective readings seem to rule out such an issue. But... perhaps the ~50-year-old lumisphere has taken on a color that is problematic?

A third possibility is that that (at least) one of the meters is not linear with light intensity. Loosely meaning that they may agree/disagree more at different light levels.

Personally I would probably just use the newer Sekonic. But if you wanted to investigate more you might do a couple of tests. I'd suggest to compare meters under several types of light "color," as indicated by the "color temperature." A shaded area on clear sunny day will be lit mostly by the blue sky, with color temp probably over 10,000 Kelvin. Direct sun, same day, probably around 5,500 K (?). Or an old-style 100 Watt tungsten bulb around 2,800 K. Fwiw the metering standard is, as I recall, about 4,700 K.

Another test would be at different light levels. Which ought to be fairly easy to do.

As all read the reflective light, which is the same surface at the very same time, (I think) the photo sensitive diodes are working fine. The disparity happens only with the dome attaches for some reason but then the old Sekonic has much larger dome.

I have ordered a Kodak Gray Card. Once it arrives, I will photograph that with a digital camera at the setting showed by both meters. That should give me some idea. OTOH, both meters has an accuracy +- 1/3EV so in between both of them, 0.5EV difference is within the manufacturers spec sheet.

 

[Fatih Ayoglu]

Around 30 years ago I worked in an industrial photo lab, attached to the lab were approximately 14 studios where photographers were, on an industrial scale, photographing product for magazine and catalogue reproduction, two shifts a day.

I don't really know how many light meters we had, but it would have been at least 14, plus some spares. Whenever a new batch of light meters were purchased, they were sent off to be calibrated; specifically for incident reading using flash. In general as I understand it, the difference between them when new was around 1/10th of a stop, with the odd one being 2/10ths of a stop different than the rest.

Once calibrated they were all reading as near to identical as possible. Every few months they would be checked, but from memory I don't think any of them ever deviated from their calibrated state. Being dropped or accidently kicked was another matter, but by and large they were good to go, seemingly forever.

We always used 1/10 of a stop light meter readings, as I do now with my newish Sekonic L308-X.

I recall the saying "Man who has two watches, never knows correct time!"

Sometimes I forgot my L308X in another bag, so I wanted to keep an analogue one in my backbag just in case But great saying

 

[BrianShaw]

Between the

• Sekonic L308X, incident and reflective light meter
• Sekonic L-28C, incident and reflective light meter

which one results in better negatives when used in the incident metering mode?

 

[chriscrawfordphoto]

Hi all,

So I have 3 light meters;

• Sekonic L308X, incident and reflective light meter
• Sekonic L-28C, incident and reflective light meter
• Pentax Digital Spotmeter, reflective light meter

When I measure reflective light, it all shows the same EV value, which is great.

But when I measure incident light while the Lumispheres are on with both Sekonics, L308X measures 0.5 EV less than L-28C. Certainly, I can put the ISO value down on L-28C and be fine with it but wanted to understand why it would create this difference.

Cheers,
Fatih

Is the dome dirty or discolored? I've seen a lot of old incident meters with yellowed domes and that cuts some of the light, causing readings to be too low. Also I have read that at some point, Sekonic changed the calibration standard on the Studio Deluxe meters, and the L28C is one of the older ones.

 

[Fatih Ayoglu]

Between the

• Sekonic L308X, incident and reflective light meter
• Sekonic L-28C, incident and reflective light meter

which one results in better negatives when used in the incident metering mode?

Didnt have the chance to use L-28C so I thought the easiest way to test is with a Kodak Gray Card

 

[Fatih Ayoglu]

Is the dome dirty or discolored? I've seen a lot of old incident meters with yellowed domes and that cuts some of the light, causing readings to be too low. Also I have read that at some point, Sekonic changed the calibration standard on the Studio Deluxe meters, and the L28C is one of the older ones.

Yes L-28C is the older one. I have made it sure that it is calibrated, ie when there is no light, like in the film changing bag, it is absolutely zero.

The issue is the old Sekonic measures 0.5 EV more than the new Sekonic. However both domes are clean and bright. I am inclined to think as the old dome is much much larger, it gathers somehow more light from different angles so measure more.

Because again, when I use as reflective meter, all 3 measures exactly the same.

 

[MattKing]

Yes L-28C is the older one. I have made it sure that it is calibrated, ie when there is no light, like in the film changing bag, it is absolutely zero.

That doesn't mean that it is calibrated - i.e. the meter has been professionally adjusted so that the readings match the expected standards.
It just means that the zero point has been correctly set.
Most people don't have professionally calibrated meters. That service is/was used mostly by people working in high volume commercial environments, or the motion picture industry.

 

[Fatih Ayoglu]

That doesn't mean that it is calibrated - i.e. the meter has been professionally adjusted so that the readings match the expected standards.
It just means that the zero point has been correctly set.
Most people don't have professionally calibrated meters. That service is/was used mostly by people working in high volume commercial environments, or the motion picture industry.

I am certainly the one who is not

 

[Chan Tran]

Possible one of the dome gets old and not transmitting the same amount of light. I had to replace the dome on my Minolta meter to get the readings correct.

 

[Bill Burk]

You may get better agreement if you will get a 12.5% gray card and meter off it in reflected light mode, and compare it to the incident light mode.

If you only have an 18% gray, the difference between 18 and 12.5 is expected and means that your meter agrees (if the difference can be accounted for by that much) and you should leave it alone. Treat it as OK if they differ by that much.

 

[Bill Burk]

The delta EV from 18% to 12.7% is 0.5

These patches are sixth stop apart.

So your meter reading of an 18% gray card “should” read 0.5 EV greater light value than incident reading.

 

[Philippe-Georges]

For incident measuring, I have two Gossen Digipro F's, the first gen., and a much cherished good old Spectra Combi-500. The Spectra is about 1/10 off with the Digipro's who are spot on equal. All measure correct in natural and ambient light; checked on Hp5+/120, pre wetted in ADOX XT-3 1+1 for 14min. @ 20°C.
On this modern emulsion - developer combination, 1/10 is hard to see...

 

[BrianShaw]

Well, all I can do is to guess. I DID look up the two Sekonic meters and see that the L-28C model was apparently made between 1970 and 1976 ( ?). And uses a selenium cell vs a silicon photo diode in the later meter.

Regarding meter calibration, per the ANSI standard a formula is used which includes a so-called 'K' constant for reflection meters or the 'C' constant for incident meters (as I recall). Each of these "constants" has a moderate range of permissible values, set by the manufacturer at their discretion, which affects the reading.

So, first possibility is that the 1970s Sekonic used a different constant for incident readings than the "modern" Sekonic. (Per the ANSI standard, as I recall, the meter should be labeled with the constants used, or perhaps only listed in the user manual.) Or maybe some generic repair shop recalibrated one of the meters to their test setup and not to the factory spec.

...

... a moderate range of recommended values...
That is an informative clause in the standard, not a normative clause. A technicality that does not change the intent of your very correct comment.

L-23C specification:
C=25
K=1.25

L-308X specification:
C (lumishere)=340
C (lumidisc)=250
K=12.5

With a hemispherical receptor, ISO 2720:1974 recommends a range for C of 320 to 540 with illuminance in lux; in practice, values typically are between 320 (Minolta) and 340 (Sekonic). The relative responses of flat and hemispherical receptors depend upon the number and type of light sources; when each receptor is pointed at a small light source, a hemispherical receptor with C = 330 will indicate an exposure approximately 0.40 step greater than that indicated by a flat receptor with C = 250. With a slightly revised definition of illuminance, measurements with a hemispherical receptor indicate “effective scene illuminance.”

Although the constant is unitless, the difference could be from the unit of measurement used in the standard at the time of the meter design/manufacture: cd/m2 in 1974 and lux more recently.

Not sure if this related discussion adds value or not (I think it does, but it's a lot of reading/thinking):

Light meter calibration

So I've been thinking a lot lately about how to correctly calibrate my various light meters. The Internet is chock full of people sharing their rules-of-thumb and get-it-in-the-ballpark methods, pretty much none of which even vaguely resemble what a professional might do. The closest you ever...

www.photrio.com

 

[Bill Burk]

Another entertaining thread on the subject…

Thread 'Different measurements with spot vs incident metering on same target?'

Different measurements with spot vs incident metering on same target?

Hi all, Apologies if this doesn't quite fit in this forum; I figured, though, it's a question that might be helpful to some future medium format shooter... My question: why am I getting two different meter readings when I take an incident reading of my subject, and then a spot reading off a...

www.photrio.com

 

[Fatih Ayoglu]

Another entertaining thread on the subject…

Thread 'Different measurements with spot vs incident metering on same target?'

Different measurements with spot vs incident metering on same target?

Hi all, Apologies if this doesn't quite fit in this forum; I figured, though, it's a question that might be helpful to some future medium format shooter... My question: why am I getting two different meter readings when I take an incident reading of my subject, and then a spot reading off a...

www.photrio.com

I guess I’m more confused after all that. It I think still my test would give some idea about the correct measurement.

However 1 question I have now, with a correct measurement with incident metering, shooting a 18% gray card would produce a histogram with the peak right in the middle or not?

 

[BrianShaw]

I guess I’m more confused after all that. It I think still my test would give some idea about the correct measurement.

However 1 question I have now, with a correct measurement with incident metering, shooting a 18% gray card would produce a histogram with the peak right in the middle or not?

Maybe. If all is equal, it should. It’s not clear that all is equal.

You should test your hypothesis and let us know how close the readings are.

 

[Bill Burk]

If you aim the spotmeter at an 18% gray card and take the picture at exactly the meter recommended f/stop and shutter speed (assuming a nice accurate shutter, and the ability to choose the recommended f/stop). For instance I keep 0.10 and 0.20 ND filters for just this purpose because some of my cameras only have whole f/stop clicks.

In this case I allowed the camera to shoot auto shutter so it should have transferred the exact necessary speed to the shutter.

The resulting negative density of the gray card point that you metered regardless of its tone or angle of reflection to the light, graphed on the characteristic curve, should result in the placement of that point that you metered, on the curve corresponding to the point on the curve where the film received “ten times” the exposure of the 0.1 density speed point.

In this example, the result graphed 18% gray (because that was the gray value of the point where I clicked “Spot”) at 1/6 stop less exposure than the theory I explained says it should.

 

[Bill Burk]

My story doesn’t explain why one of your incident meters is different than the other.

Maybe once you get that gray card things will start to make sense.

 

[Chan Tran]

My story doesn’t explain why one of your incident meters is different than the other.

Maybe once you get that gray card things will start to make sense.

I still think that is because the dome gets old. You eyes don't see it but it transmit less light. I found that out on my Minolta meters. The 12.5 vs 14 is related only to the reflected meter. Both meter are Sekonic and thus should be calibrated to K12.5 while the Pentax should be calibrated to K14 and they all read the same. For the incident with the sphere the Sekonic calibrated with a C340 while some other brand like Minota is calibrated to C330.