Testing the lightmeter accuracy on a 35mm camera

unwantedfocus · Feb 11, 2022

Hello, as some of you know people DIY shutter speed testers, is there a equivalent to test the lightmeter in a camera? What would be the best way to go about it? I do have a digital camera and a handheld flashmate lightmeter. Also there are spotmeters and matrix like the minolta models would this be differently measured? Any suggestions would be helpful.

Paul Howell · Feb 11, 2022

First question is your handheld meter accurate, a flash mate, I guess a flash meter? I have a Gossen, several Westons and a few older GE meters that I use. I had my Gossen calibrated by my local camera repair and use it as a standard for the remaining meters and my 35mm bodies. I check my later models 35mm in average mode, there is often some differences between average matrix or spot metering, my working assumption is that if an TTL meter is good in average it is good in other modes. My later TTL meters are all within a 1/2 stop of my Gossen, my older machinal bodies, quite a bit of variance up to 2 stops. I also check with sunny 16, noon local time, ISO 125th of a second at F16. This is not the most accurate but works for me.

unwantedfocus · Feb 11, 2022

Thanks, with flashmate, I mean the Sekonic L-308S. I read through these post. There is a light source that I could acquire so it would be more accurate? But I don't know what a gizmo is...language barrier?!?!?!

Cholentpot · Feb 11, 2022

unwantedfocus said:
Thanks, with flashmate, I mean the Sekonic L-308S. I read through these post. There is a light source that I could acquire so it would be more accurate? But I don't know what a gizmo is...language barrier?!?!?!

A gizmo is a gadget. It can also be a doohickey or a dongle. Thingamabob is also correct. But I tend to call it a thingamajig.

Autonerd · Feb 11, 2022

The most accurate way would probably be to meter an 18% gray card and compare what the camera says to a good digital hand-held light meter.

A less accurate way would be the ol' Sunny 16 rule. Compose a scene on a sunny day, something with neutral colors (say grass, trees, sky, all in sunlight), and with ASA dial set to 100, your exposure should be around 1/125 @ f/16. That'll at least let you know if you're in the ballpark.

Aaron

Light Capture · Feb 11, 2022

There are quite few factors when comparing meters.
Different manufacturers actually used slightly different percentages of gray values for calibration. When color response is added into this mix, it makes comparisons between meters even more difficult.

While I did try detailed tests at one point, they didn't give me much additional information due to issues above. User manuals for each meter state how much to compensate for situations that are out of gray values range.
What proved useful is to test camera meter in different light levels. This was much more useful since it showed some of the cameras to be non-linear after so many years.
For example, a meter in a camera would be accurate in daylight outside but metered several stops off indoors. If meter is consistently off, compensation can be dialed in by fixed amount and it's likely to be working well.

Paul Howell · Feb 11, 2022

The best light source is the sun, read up on Sunny 16, if you handheld meter meters correctly in good daylight around noon, then it is likely metering correctly within in range. If your handheld meter is off then you need a manual to open it up and make adjustments, or find someone in Germany who can calibrate it for you.

https://en.wikipedia.org/wiki/Sunny_16_rule

Sirius Glass · Feb 11, 2022

Cholentpot said:
A gizmo is a gadget. It can also be a doohickey or a dongle. Thingamabob is also correct. But I tend to call it a thingamajig.

Do not forget What-Cha-Call-It.

Bill Burk · Feb 11, 2022

One of these. There’s two like it on eBay one is less than $75 free shipping, the other is like mine with the timer jimmied for about $100

runswithsizzers · Feb 11, 2022

short version: take reflected readings from a uniform surface - preferably, a gray card - and see if your camera gives the same reading as your Sekonic L-308s. Be sure to get in close when taking the readings - but not so close to cast shadows on the gray card. You don't want either the camera or the L-308 to see anything outside of the edges of the gray card. If you don't have a gray card, open up something like a cereal box and use the unprinted side of that. If you are just comparing the camera to the Sekonic, it won't matter if the reflective surface is 18% or not. But if you want to use the Sunny 16 rule to confirm the Sekonic reading, then an 18% gray card card is recommended.

too long don't read:
My Sekonic L 308S Flashmate gives the same reflected reading from my Delta brand gray card as it does when used as an incident meter. I am lucky enough to have a second hand-held meter - a Gossen Luna Lux* - and it matches the Sekonic L-308, both reflected and incident. So for me, if my camera meter gives me the same reading from the gray card as either of my two hand-held meters, then I figure the camera meter is accurate. One of my two Pentax MX cameras does match the hand-held meter readings, the other one is about 1/2 stop different. For that camera, I adjust the ISO dial until it gives the same reading from the gray card as the others.

I also have a couple of Fuji digital cameras. I have not done enough testing to really know how the digital cameras compare to the hand-held meters and the two Pentax film cameras - but in some quickie tests, the digital cameras were somewhat anomalous. Until I can do more testing and research, I am not using the digital cameras to test the meters on my film cameras. There may be some differences in how the ISO standard is implemented by different digital camera manufacturers which might explain the differences I think I'm seeing - but I have not done enough research to understand it.

Personally, the Sunny 16 rule does not work that well for me. Under conditions that should give me a Sunny 16 reading, I usually get something closer to f/11. Then again, my cameras do not have a 1/100 sec shutter speed, so maybe the fact that I am using 1/125 sec is a factor?

* outside of the US, the Gossen Luna Lux may be known as the Lunalight

Paul Howell · Feb 11, 2022

1/100 vs 1/124th is like an 8th of a stop. Here in the Phoenix metro area, most times of the year, clear day I get ISO 100 1/125h f 16 at noon. Mid winter might be somewhere between F11 and 16.

reddesert · Feb 11, 2022

The Sekonic L-308 has an incident dome. Take an incident reading outside on a sunny day and make sure it corresponds to Sunny 16. Late model meters like that rarely go out of calibration, IMO, so it's probably pretty close.

Then take whatever camera you are interested in, point it at a relatively uniform bright outdoor scene, and point the Sekonic in reflected mode at the same scene to compare. Then do the same thing but for a relatively uniform more dimly lit indoor scene. This compares the meters over a range of normal operating brightnesses.

Just point the meters at a relatively uniformly lit wall. Don't mess around with special purpose reference light sources. Unless you know how to do calibrated photometry, diffusion, uniform illumination etc, those will just get you in a heap of trouble.

Moose22 · Feb 11, 2022

Cholentpot said:
dongle.

A dongle is actually not a generic thing, it's a specific thing. A hardware device that connects to a computer port, originally for copy protection. Now sometimes for other things, like datalogging.

And on topic, I compare camera meters by comparing cameras and my sekonic. Just point 'em at a grey card, and sometimes the sky. Also, incident here is a dead nuts match to my agave plants out front, so I sometimes cheat there mid day when I know what to expect.

One of the wiser members here warned me about green stuff, though, as some meters can be fooled by the IR reflectivity. Doesn't seem an issue with modern camera meters, but if I cared about perfect accuracy I'd always use grey cards.

Strange now that I think about it, I've done this with most of my meters and never had one really too far off. Even my Leicameter, which is 50s vintage, is close enough or black and white. I think the only one marginal was the porofinder meter for my C330, though I don't actually use that for the meter.

Alan Edward Klein · Feb 11, 2022

Bill Burk said:
View attachment 298154

One of these. There’s two like it on eBay one is less than $75 free shipping, the other is like mine with the timer jimmied for about $100

How do you know the light is still calibrated? Don;t they get dimmer too.

Alan Edward Klein · Feb 11, 2022

Paul Howell said:
1/100 vs 1/124th is like an 8th of a stop. Here in the Phoenix metro area, most times of the year, clear day I get ISO 100 1/125h f 16 at noon. Mid winter might be somewhere between F11 and 16.

I believe it's more like 1/3 of a stop between 1/100 and 1/125th.
https://photographyuncapped.com/art...so-shutter-speed-chart-subscribe-for-updates/

Bill Burk · Feb 11, 2022

Alan Edward Klein said:
How do you know the light is still calibrated? Don;t they get dimmer too.

The number of hours is reasonably low. The bulb is “under driven” and power on is slow. I haven’t relied on it to calibrate meters. Instead I use it to verify meters and it’s been working for that.

Alan Edward Klein · Feb 11, 2022

Bill Burk said:
The number of hours is reasonably low. The bulb is “under driven” and power on is slow. I haven’t relied on it to calibrate meters. Instead I use it to verify meters and it’s been working for that.

Could a puck used for calibrating a monitor be used to check the light output?

Alan Edward Klein · Feb 11, 2022

I was just thinking that my NEC puck has an ambient sensor to check the light in the room and gives values. Maybe that part could check your light. It's used to set up an ambient area for looking at prints in a particular lighting intensity.

Bill Burk · Feb 11, 2022

Interesting idea. Of course the puck could read light, just not sure about what the software would do.

Alan Edward Klein · Feb 11, 2022

Bill, I just checked the NEC Spectra view II software manual. The ambient portion of the puck measures color in Kelvin and ambient light intensity in Lux. Since there's a relationship between EV shown on a meter and Lux, maybe you can convert the reading from Lux to Ev to see if you can measure the light output.

John Koehrer · Feb 11, 2022

Used to work at a shop that had Kyoritsu(sp?) that used a source similar to Bill's single lamp and ND filters to change brightness.
I'm assuming ND's because the knob to change brightness made clunking sounds when it was rotated.

Alan Edward Klein · Feb 11, 2022

John Koehrer said:
Used to work at a shop that had Kyoritsu(sp?) that used a source similar to Bill's single lamp and ND filters to change brightness.
I'm assuming ND's because the knob to change brightness made clunking sounds when it was rotated.

The ND's would seem to be stable. The question is, is the bulb output consistent, especially as it ages? I suppose you can work it backward. If you know you have a calibrated meter then just measure the bulb output to see what its' output is. If it is not accurate, then adjust the readings on the meter to allow for the luminance divination output of the light.

beemermark · Feb 11, 2022

Here's my easy way to check a camera meter - shoot some film, process film, and see what you get.

beemermark · Feb 11, 2022

beemermark said:
Here's my easy way to check a camera meter - shoot some film, process film, and see what you get. Seems everybody over thinks these things.

Alan Edward Klein · Feb 11, 2022

beemermark said:
Here's my easy way to check a camera meter - shoot some film, process film, and see what you get.

So let's say the whole roll is a little dark. How much do you adjust the meter to get it to read correctly for the next roll of film? 1/3 of a stop, 1/2, a full stop? What if you made a mistake in developing in addition to the meter being off? You really need to calibrate the meter to a known and accurate input.

Testing the lightmeter accuracy on a 35mm camera

Privacy & Transparency

Privacy & Transparency