Film Shutter Speed a simple test method

[Robin Guymer]

I have tried using the iphone app method but found it confusing to decipher the readout. So I thought what about just using a digital camera through the film camera shutter and comparing the resulting exposures. I also found that this will show up any open or close curtain drag.
My method is to use a Sony Nex with an adapted small prime lens. In the photo below I have used a 90mm Leitz. This I mounted on a tripod and the lens is pushed against the film camera with the back removed mounted on a second tripod (or you could put this on a table). The film camera has no lens mounted. Before each shot a cloth is draped over the intersection of the two to block out excess light around the front of the lens and through the film camera eye piece.

1st Photo
Set the Digital camera at - No Lens, ISO to suit, Aperture priority to get the speed you want to test (say 125th), 10 second delay. Put film camera on B press shutter to hold shutter open and take a digital photo through the film camera at whatever you have set it up to focus on.

2nd Photo
Set the Digital camera at - Manual and dial it to say 2 seconds, 10 second delay(for stability). Adjust film camera to the same speed that the Digital camera registered for the previous photo (was 125th). Press the shutter on the digital camera and when it fires then fire the film camera. You have now used the film cameras shutter to take the photo onto the digital sensor.

3rd Photo
Do exactly the same as the 2nd Photo but this time adjust the film camera to one stop faster so in this instance 1/250th. Take the 3rd photo.

Now compare photos 2 and 3 with the 1st photo.

I generally find that my good film cameras have good exposure comparison between photo 1 & 3. Photo 2 is generally brighter which would mean that the shutter on the film camera is a bit slow but I think the digital sensor staying open for 2 seconds gets a bit more light - just a theory. If photo 3 is still brighter than photo 1 then take another at 2 stops more which would be 1/500th. If this looks more like photo 1 then it is likely that you film camera shutter could be slow and your shots will over expose on film.

What do you think of the viability of this method?
(Hope this is okay in this forum as I figured it is about testing/repairing analogue cameras - the digital aspect is just another tool to do this - actually you could use a tried and true film camera as the rear testing camera so then the negatives would give a very accurate result with no wondering if the sensor does this or that at 2 seconds or the software still manipulates the image etc.)

 

[Bill Burk]

Seems like a viable test.

 

[Ian Grant]

I do something similar with my roller blind shutters and it works well. The only difference is it's simpler in my case as the shutters fit the front of a lens, I ve done the same with Luc, Gitzo, Le Mignon, etc, front mounting leaf shutters.

It's only a relative speed test and relies on the supposition that the Digital cameras speeds are accurate.

Ian

 

[vickersdc]

This is interesting... I need to find out what shutter speeds I have with a shutter fitted to my c1896 1/2-plate camera. i wondered if there was a way to do it by shining a light through the lens, firing the shutter and recording the resultant light on to a video camera. Not sure how accurate that would be though.

I definitely haven't got a lens with a big enough front element for 1/2-plate though! But I might be able to unmount the shutter from the camera and try something like your idea out

 

[ic-racer]

Couldn't you just use some film? Why use a digital camera to guess how the film will respond.

 

[vickersdc]

Couldn't you just use some film? Why use a digital camera to guess how the film will respond.

Because I need a starting point on which to base my exposure... not sure I trust my guessing the difference between 1/15th and 1/30 second with any degree of accuracy. Especially as it might not be either of those.

 

[awty]

I have tried using the iphone app method but found it confusing to decipher the readout. So I thought what about just using a digital camera through the film camera shutter and comparing the resulting exposures. I also found that this will show up any open or close curtain drag.
My method is to use a Sony Nex with an adapted small prime lens. In the photo below I have used a 90mm Leitz. This I mounted on a tripod and the lens is pushed against the film camera with the back removed mounted on a second tripod (or you could put this on a table). The film camera has no lens mounted. Before each shot a cloth is draped over the intersection of the two to block out excess light around the front of the lens and through the film camera eye piece.

1st Photo
Set the Digital camera at - No Lens, ISO to suit, Aperture priority to get the speed you want to test (say 125th), 10 second delay. Put film camera on B press shutter to hold shutter open and take a digital photo through the film camera at whatever you have set it up to focus on.

2nd Photo
Set the Digital camera at - Manual and dial it to say 2 seconds, 10 second delay(for stability). Adjust film camera to the same speed that the Digital camera registered for the previous photo (was 125th). Press the shutter on the digital camera and when it fires then fire the film camera. You have now used the film cameras shutter to take the photo onto the digital sensor.

3rd Photo
Do exactly the same as the 2nd Photo but this time adjust the film camera to one stop faster so in this instance 1/250th. Take the 3rd photo.

Now compare photos 2 and 3 with the 1st photo.

I generally find that my good film cameras have good exposure comparison between photo 1 & 3. Photo 2 is generally brighter which would mean that the shutter on the film camera is a bit slow but I think the digital sensor staying open for 2 seconds gets a bit more light - just a theory. If photo 3 is still brighter than photo 1 then take another at 2 stops more which would be 1/500th. If this looks more like photo 1 then it is likely that you film camera shutter could be slow and your shots will over expose on film.

What do you think of the viability of this method?
(Hope this is okay in this forum as I figured it is about testing/repairing analogue cameras - the digital aspect is just another tool to do this - actually you could use a tried and true film camera as the rear testing camera so then the negatives would give a very accurate result with no wondering if the sensor does this or that at 2 seconds or the software still manipulates the image etc.)

View attachment 254726

Thanks for the detailed information.
Could you tell me how curtain drag will show up?

 

[jonmon6691]

Just to throw my hat in the ring, I used a photoresistor and a little Arduino nano with its analog input to measure shutter speeds on all my bodies and LF lenses. Here's the code, also includes a wiring diagram:

https://github.com/jonmon6691/arduino_shutterspeed

I'm a little suspicious of the readings at 1/1000th and 1/2000th, I haven't experienced any exposure issues with the F-1 or the T90 at those speeds so I'm thinking the response time of the photocell or the resolution of the timer in the Arduino is starting to hit their limits. The error for each speed is given in thirds of a stop under or over "ideal". This was a nice exercise in grounding me while I've been doing film testing, the margin of error, at least for my bargain bay equipment is in the 2/3rds of a stop range and I shouldn't fuss over anything less.

 

[snusmumriken]

Just to throw my hat in the ring, I used a photoresistor and a little Arduino nano with its analog input to measure shutter speeds on all my bodies and LF lenses. Here's the code, also includes a wiring diagram:

https://github.com/jonmon6691/arduino_shutterspeed

I'm a little suspicious of the readings at 1/1000th and 1/2000th, I haven't experienced any exposure issues with the F-1 or the T90 at those speeds so I'm thinking the response time of the photocell or the resolution of the timer in the Arduino is starting to hit their limits. The error for each speed is given in thirds of a stop under or over "ideal". This was a nice exercise in grounding me while I've been doing film testing, the margin of error, at least for my bargain bay equipment is in the 2/3rds of a stop range and I shouldn't fuss over anything less.

View attachment 317765

Thanks for sharing this. Supposedly the limit to accuracy of micros() is 4us. There is a library to get 8 times this here. But isn't it more likely that the photoresistor is limiting accuracy?

 

[ic-racer]

With horizontal focal plane shutters one really needs to determine the speed of the first curtain and the size of the gap at the beginning and at the end. Those are the parameters that can be adjusted. Without knowing those parameters, adjustments are only a guess.

 

[reddesert]

Just to throw my hat in the ring, I used a photoresistor and a little Arduino nano with its analog input to measure shutter speeds on all my bodies and LF lenses. Here's the code, also includes a wiring diagram:

https://github.com/jonmon6691/arduino_shutterspeed

I'm a little suspicious of the readings at 1/1000th and 1/2000th, I haven't experienced any exposure issues with the F-1 or the T90 at those speeds so I'm thinking the response time of the photocell or the resolution of the timer in the Arduino is starting to hit their limits. The error for each speed is given in thirds of a stop under or over "ideal". This was a nice exercise in grounding me while I've been doing film testing, the margin of error, at least for my bargain bay equipment is in the 2/3rds of a stop range and I shouldn't fuss over anything less.

View attachment 317765

Tolerance for shutter speeds is usually spec'ed to be about 1/3 stop even for a serviced shutter.

The response time of a photoresistor can be of order 10 millisec (or slower), so depending on the response curve and the activation threshold of your circuit it may read slow at fast speeds.

A phototransistor is much faster (few microsec), so if you revised your circuit to use a phototransistor for the pulldown it might improve the time resolution a lot. You could probably put a phototransistor in series with a small resistor (say 1-2Kohm) in place of the photoresistor and get it to work.

At the highest speeds for leaf shutters, "shutter efficiency" means that the effective exposure time depends on whether the aperture is open or stopped down (because it takes a shorter time for the blades to clear a smaller aperture). This sometimes accounts for deviations at high shutter speeds, plus leaf shutters rarely achieve their highest speed in practice.

 

[jonmon6691]

At the highest speeds for leaf shutters, "shutter efficiency" means that the effective exposure time depends on whether the aperture is open or stopped down (because it takes a shorter time for the blades to clear a smaller aperture). This sometimes accounts for deviations at high shutter speeds, plus leaf shutters rarely achieve their highest speed in practice.

That's cool I never thought about that before, I guess its possible to get vignetting from a worn out leaf shutter at high speeds and wide apertures

 

[ic-racer]

That's cool I never thought about that before, I guess its possible to get vignetting from a worn out leaf shutter at high speeds and wide apertures

No, because a leaf shutter should be at the center of the optical axis. It is pretty cool concept, maybe one of the lens designers can define it better, but neither the shutter or aperture causes any change in the angle of view or vignetting (when in the correct position.)

 

[reddesert]

No, because a leaf shutter should be at the center of the optical axis. It is pretty cool concept, maybe one of the lens designers can define it better, but neither the shutter or aperture causes any change in the angle of view or vignetting (when in the correct position.)

Correct, the leaf shutter in a between-the-lens position is at the pupil (the location of the aperture), which is effectively the maximally out of focus place in the optical train. So the shape of the pupil doesn't affect the field illumination (eg on simple cameras like P+Ses there is sometimes only a 2 or 3 blade iris/shutter), although it does affect the appearance of out of focus areas. For fast or wide angle lenses, stopping down the aperture sometimes decreases vignetting caused by the construction of the lens.

A way to think about shutter efficiency is that a leaf shutter correctly operating at 1/500 would uncover the full aperture for 1/500 sec, but it would uncover the small aperture at say f/22 for maybe 1/250 sec.

There is a description of shutter efficiency in the Calumet shutter tester manual, page 6; the manual is a good read for people building their own testers. https://www.kennethleegallery.com/images/shuttertester/CalumetShutterTesterManual.pdf

 

[gone]

FWIW, I used electronic shutter testers for many years and dutifully wrote down the speeds when I checked cameras. Oddly, I rarely got what I would call on the money exposures.

So I stopped using the testers and just gauged things by the way the shutters sounded (you didn't have to worry on electronic shutters, but on leaf shutters the speeds generally ran 1 stop off), and that's how I would set my EI. Worked fine. Why this happened, I don't know.

 

[jonmon6691]

No, because a leaf shutter should be at the center of the optical axis. It is pretty cool concept, maybe one of the lens designers can define it better, but neither the shutter or aperture causes any change in the angle of view or vignetting (when in the correct position.)

Oh duh of course, just the same as a narrow aperture doesn't leave a shadow, only evenly reduces the intensity at the film plane