I have a pretty good collection of 35mm cameras that I like to keep in the rotation, and I also have high standards for getting my exposure correct, even when shooting B&W film. As you can guess, the differences between my various camera meters can be an issue.
The obvious answer here would be to spot meter everything with the same Reveni Labs spot I use for my MF and LF work when shooting cameras that don't have a built-in meter. But when have I ever gone with the obvious, safe answer? In truth, I just figure if I'm shooting a smaller format, being able to trust my built-in meter is a huge added convenience.
So far the slightly more modern cameras with modern batteries (AE-1, AT-1, OM-G) seem to just always dead-on match my Reveni Labs spot meter in all conditions. Cool. Of course, the two cameras I have the most interest in shooting (OM-1, FTb) have older meters that want 1.35v batteries. I have modified both meter circuits with a schottky diodes to allow for use with a 1.55v silver oxide battery.
I calibrated the camera meters using their built-in adjustment controls and comparing them against the Reveni Labs at EV 15, 12, and 8. I used a Neewer LED light panel as my metering target—this procedure and even this exact panel was recommended to me by an old school repair technician. At 100% brightness and 5600k white balance, it's advertised as EV 15 at the diffuser, and my spot meter measured it dead on at that exact brightness. EV 12 therefore should be 12% power, and EV 8 unfortunately ends up being a fraction of 1%, so I achieved that EV by instead evenly illuminating a white wall with the panel and metering off the wall.
It was easy enough to adjust the cam on the OM-1's galvanometer to match the spot meter reading at EV 12 (when filling the frame with the light panel through a 50mm lens). The diode modification seems to have succeeded, since I found linear readings with OM-1 (after adjusting to match the spot meter at EV 12, I found that EV 15 and EV 8 also matched within less than 1/3 of a stop with no further adjustment). Good news, too, since the only way I could find to fix nonlinearity with this meter was desoldering and replacing some resistors in the meter circuit. Not exactly convenient, especially for such a "guess and check" type process.
The FTb has 3 potentiometers to adjust the galvonometer response, giving the ability to correct for nonlinearity. I tweaked those until I was also matching the spot meter at EV 15, 12, and 8 using the light panel.
Fast forward nearly a year. I've put several rolls through both cameras (mostly HP5+), with largely good results. But lately I have noticed extra density and contrast coming from my FTb in particular, so I decided to do a more exhaustive round of testing. My hypothesis is that the overcast outdoor conditions I frequently work in are producing a different response in my camera meters than my spot meter due to a different makeup of wavelengths making up natural daylight vs the "daylight balanced" LED panel.
Sure enough, I found that my FTb is matching my spot meter only at about EV 13.5 in open shade conditions. This was confirmed by metering an 18% gray card with both the camera and the spot meter outdoors in open shade. I checked again at several different EVs at different times of the last few days, as well as checking an EV15 reading in full bright direct sunlight on the gray card. Above EV 13.5, the FTb wants to underexpose by as much as 2/3 of a stop. Below EV 13.5, it wants to overexpose by as much as a whopping 1.5 stops in the low end of the EV scale. Most of my shooting with this camera is done in the neighborhood of EV 9 to EV 12 or thereabouts. No wonder I'm getting significant and noticeable extra density in my negs. As a precaution, I also confirmed that my battery is actually outputting the correct voltage, since it's several months old at this point. Yep, 1.58vdc on my multimeter.
Some nonlinearity and inaccuracy due to a less-than-optimal method of calibrating the meter isn't anything strange. Here's the weird part though. Out of curiosity, I pulled out my LED light panel and ran several comparisons at both 5600k and 3200k. The camera meter saw only a very slight (~1/4 stop) difference between any two EV values at different color temperatures. But for all readings, the camera wanted to underexpose by somewhere between 1/2 and 1 full stop.
So in open shade, the FTb meter will overexpose by 1/2 stop at EV 12. When pointing at an LED light panel at 5600k, the FTb meter will underexpose by 1/2 stop at EV 12.
I guess the lesson is... don't trust an inexpensive LED panel as a reliable metering target when adjusting your camera meter. Looks like I'll be taking the FTb and OM-1 out on the back porch next time I get a nice overcast day, to spend some quality time with tiny screwdrivers, a spot meter, and a gray card.
The obvious answer here would be to spot meter everything with the same Reveni Labs spot I use for my MF and LF work when shooting cameras that don't have a built-in meter. But when have I ever gone with the obvious, safe answer? In truth, I just figure if I'm shooting a smaller format, being able to trust my built-in meter is a huge added convenience.
So far the slightly more modern cameras with modern batteries (AE-1, AT-1, OM-G) seem to just always dead-on match my Reveni Labs spot meter in all conditions. Cool. Of course, the two cameras I have the most interest in shooting (OM-1, FTb) have older meters that want 1.35v batteries. I have modified both meter circuits with a schottky diodes to allow for use with a 1.55v silver oxide battery.
I calibrated the camera meters using their built-in adjustment controls and comparing them against the Reveni Labs at EV 15, 12, and 8. I used a Neewer LED light panel as my metering target—this procedure and even this exact panel was recommended to me by an old school repair technician. At 100% brightness and 5600k white balance, it's advertised as EV 15 at the diffuser, and my spot meter measured it dead on at that exact brightness. EV 12 therefore should be 12% power, and EV 8 unfortunately ends up being a fraction of 1%, so I achieved that EV by instead evenly illuminating a white wall with the panel and metering off the wall.
It was easy enough to adjust the cam on the OM-1's galvanometer to match the spot meter reading at EV 12 (when filling the frame with the light panel through a 50mm lens). The diode modification seems to have succeeded, since I found linear readings with OM-1 (after adjusting to match the spot meter at EV 12, I found that EV 15 and EV 8 also matched within less than 1/3 of a stop with no further adjustment). Good news, too, since the only way I could find to fix nonlinearity with this meter was desoldering and replacing some resistors in the meter circuit. Not exactly convenient, especially for such a "guess and check" type process.
The FTb has 3 potentiometers to adjust the galvonometer response, giving the ability to correct for nonlinearity. I tweaked those until I was also matching the spot meter at EV 15, 12, and 8 using the light panel.
Fast forward nearly a year. I've put several rolls through both cameras (mostly HP5+), with largely good results. But lately I have noticed extra density and contrast coming from my FTb in particular, so I decided to do a more exhaustive round of testing. My hypothesis is that the overcast outdoor conditions I frequently work in are producing a different response in my camera meters than my spot meter due to a different makeup of wavelengths making up natural daylight vs the "daylight balanced" LED panel.
Sure enough, I found that my FTb is matching my spot meter only at about EV 13.5 in open shade conditions. This was confirmed by metering an 18% gray card with both the camera and the spot meter outdoors in open shade. I checked again at several different EVs at different times of the last few days, as well as checking an EV15 reading in full bright direct sunlight on the gray card. Above EV 13.5, the FTb wants to underexpose by as much as 2/3 of a stop. Below EV 13.5, it wants to overexpose by as much as a whopping 1.5 stops in the low end of the EV scale. Most of my shooting with this camera is done in the neighborhood of EV 9 to EV 12 or thereabouts. No wonder I'm getting significant and noticeable extra density in my negs. As a precaution, I also confirmed that my battery is actually outputting the correct voltage, since it's several months old at this point. Yep, 1.58vdc on my multimeter.
Some nonlinearity and inaccuracy due to a less-than-optimal method of calibrating the meter isn't anything strange. Here's the weird part though. Out of curiosity, I pulled out my LED light panel and ran several comparisons at both 5600k and 3200k. The camera meter saw only a very slight (~1/4 stop) difference between any two EV values at different color temperatures. But for all readings, the camera wanted to underexpose by somewhere between 1/2 and 1 full stop.
So in open shade, the FTb meter will overexpose by 1/2 stop at EV 12. When pointing at an LED light panel at 5600k, the FTb meter will underexpose by 1/2 stop at EV 12.
I guess the lesson is... don't trust an inexpensive LED panel as a reliable metering target when adjusting your camera meter. Looks like I'll be taking the FTb and OM-1 out on the back porch next time I get a nice overcast day, to spend some quality time with tiny screwdrivers, a spot meter, and a gray card.
