Testing Auto Thyristor strobes

[mattk]

Am I doing this right? 1) I set up the Sekonic flash meter some distance away. 2) I set the flash (metz 45, sunpack, and the new nikons I have) to auto 3) Pop the flash and read the meter. Probelm is the actual settings are not matching the meter measurement--usually between a stop or 2 slow. I assume this is a function of the thyristor reading the sum of the reflected light and much will depend on the overall reflectivity of the scene. I just want to get a feel for how accurate this mode will be without burning through a few rolls of film on the RB. Is my test functional and realistic?

Matt

 

[archphoto]

Direct the flash at a 18% grey card at 1 meter and measure on the greycard.

If you do it diferently, e.i. a white wall, the auto setting will assume that it is 18% grey a expose accordingly.

Hope this helped a bit.

Peter

 

[Ian Cooper]

Point the flash at an even toned wall. Rather than using the sekonic flash meter in incidence mode, use it in reflectance mode held alongside the flash gun pointing at the same wall. This means both the auto-thyristor and the flash meter will be assuming an 18% reflectance exposure value from the same scene.

By using the meter as an incident meter you will be measuring the exact light output of the flash gun, however, if the scene infront of the auto-thyristor isn't 18%, then the flash will give either more or less light output until it does sense 18% reflected back, hence why the flash and the meter will give different results.

 

[dpurdy]

It seems logical that your flash meter should give the same reading except there is the variable of what is reflecting the light back to the thyristor. If you have your flash meter propped up in a dark area then perhaps the thyristor is seeing a need for more light. If you have the meter set in a bright reflective background then the thyristor will see the need for less light. If you have your meter set in an area where the surface that bounces the light back to the thyristor is further away then the thyristor will see the need to pump more light in. Maybe you should try the test in a variety of settings.

 

[Pinholemaster]

Better to switch the flash to manual to measure the flash with your meter. When set on 'Auto' the flash meter will not read the light the same way was the 'Auto' sensor.

 

[Chan Tran]

To test the flash with an incident meter you would need to point it at a gray card. But there is a catch if you point it at the gray card where can you put the meter? So may be a large gray wall and with a small meter in the center it won't affect the reading much? The readings are likely to be lower because I read Metz publication somewhere that their flash expect a subject of 50% reflectance and not 18%. In reflected mode, you can point it at a blank wall and expect the reading to be close. Again I found most flashes give lower readings.
To test the flash in manual, all you check is their power and not how well their auto mode operates.

 

[benjiboy]

Better to switch the flash to manual to measure the flash with your meter. When set on 'Auto' the flash meter will not read the light the same way was the 'Auto' sensor.

Thats right, Pinholemaster , you don't need the auto function the flash from several flashes will effect all the thyristors present , set all the flashes on full power (manual) and use the flash meter to calculate the exposure.

 

[John Koehrer]

You think the meter is reading the sum of the light? This suggests you're using more than one in auto. Is this correct?
I don't think you can get a predictable result with multiple units all set to Auto unless it's something like Nikon or Canon's ttl systems that were meant to work in multiples.

 

[mattk]

I am only using one flash. I thought he incident meter (with the dome on the 558) was giving you a setting to correctly expose a 18% gray card. I will try metering a gray card and see how the results differ. My thought is in auto mode the flash cuts out the light when it determines enough light has reached the subject to properly expose an 18% gray card. Doesnt the meter do the same thing? Maybe I just burn a roll through the 645 to see real results.

 

[MattKing]

This is in essence just restating what others have posted.

The flash meter used in incident mode measures the light output from the flash at the subject's position.

If you have the flash set to auto, the auto function will vary the light output according to how much of the light output bounces back to the sensor on the flash. This depends on distance and, among other things, what the reflectance is of the subject and its background. Depending on how you set things up, the presence of the flash meter itself in the scene may affect how much light bounces back to the sensor.

If your subject and the surrounding scene happens to average out to a reflectance of 18%, your meter will indicate the same setting as the auto aperture set on the flash (assuming everything is properly calibrated). In the real, non 18% world, the meter will indicate something different.

If you are going to use a flash meter, you need to use the flash on manual.

Matt

 

[Q.G.]

To test the flash with an incident meter you would need to point it at a gray card.

You just need to slide the dome aside and point the meter at whatever the sensor in the flash unit is seeing. No matter what that is: pitch black, gray, purple with yellow stripes, or white.

 

[eddym]

What is the purpose of your test?
Are you trying to determine the strength of your flash? If so, then you need to set it on manual, not auto.
Or are you trying to determine the accuracy of the auto flash sensor? If so, then you need to take the reflected reading as others have mentioned.
The strength of your flash is good to know; the accuracy of the flash's auto sensor depends on the subject it is reading. Both are important in the real world, but for different purposes.

On the other hand, if you are not really testing, but trying to take a picture, then take an incident reading with your flash meter and go with it. Then you don't care what the background is, whether you are in auto or not; your meter tells you what the exposure will be.

 

[BrianShaw]

Or are you trying to determine the accuracy of the auto flash sensor? If so, then you need to take the reflected reading as others have mentioned.

The best way I've found for "testing" the accuracy of autoflash sensor is to shoot a roll of slide film in a controlled manner. Shoot pics of scenes at the extremes of the auto sensor range (if the range is 10 ft to 45 feet, shoot a scene 10 feet away and another 45 feet away and maybe one in between) and see if they are adequately illuminated.