How does shutter timing work on an FE?

[Ariston]

I sent my FE off for a CLA several months ago. It looked fine when I got it back, including new light seals. But I just noticed today that my 1-second shutter sounded fast. So I set it to 8 seconds and timed it, and it closed in 5.5 seconds. So, it is about 1/3 stop off at that speed.

Does anyone know... does this mean I should expect it to be 1/3 stop off across the range of shutter speeds, or is it non-linear? I am not sending my camera back, because this is the second bad experience I've had, but I don't have the money to send it to someone else right now, either.

 

[shutterfinger]

The FE uses a vertical travel electronically controlled shutter. Its impossible to say if all speeds will be off or if only some will be off without testing the shutter with a shutter speed tester.
A 1/3 stop exposure error is just detectable in a wet print. AA's film speed test is in 1/3 stop increments. Set up a gray card in flat lighting and make as many equivalent exposures on 100ISO B&W film as you can, process the film, scan it yourself in 16 bit RGB if possible, then use the color sampler of a photo editing program to read the RGB value of the gray card in each frame. This will tell you if the shutter speeds are in tolerance or if some are way off. A difference of 5 or less in any color channel is very good, 10 acceptable, 15 or more is excessive variance. The variance from frame to frame should be a similar amount, not a large jump along the line.

 

[Ariston]

The FE uses a vertical travel electronically controlled shutter. Its impossible to say if all speeds will be off or if only some will be off without testing the shutter with a shutter speed tester.
A 1/3 stop exposure error is just detectable in a wet print. AA's film speed test is in 1/3 stop increments. Set up a gray card in flat lighting and make as many equivalent exposures on 100ISO B&W film as you can, process the film, scan it yourself in 16 bit RGB if possible, then use the color sampler of a photo editing program to read the RGB value of the gray card in each frame. This will tell you if the shutter speeds are in tolerance or if some are way off. A difference of 5 or less in any color channel is very good, 10 acceptable, 15 or more is excessive variance. The variance from frame to frame should be a similar amount, not a large jump along the line.

Thank you for that information. I normally would not be bothered, but I just shot Velvia for the first time (different camera). And bracketing by just 2/3 stops gave me surprisingly different exposures. I knew slide film was supposed to have somewhat less latitude, but the results were more pronounced than I expected.

 

[film_man]

I'm surprised that an electronic shutter is not accurate. Does anyone know why that would be the case?

 

[etn]

I'm surprised that an electronic shutter is not accurate. Does anyone know why that would be the case?

If a crystal is used as time reference, several reasons can be thought of: physical damage to the crystal, aging, wrong electronic adaption of the crystal load (however I wouldn't expect this if the design engineers did their work correctly) etc.
Crystals have an accuracy of about +/- 100ppm or less, including aging.

Of course, any other time base could have been used instead of a crystal. Those would be much less precise. Are the schematics of the Hasselblad available somewhere?
(as a side note, I wouldn't expect Hasselblad engineers to use anything but a crystal on a camera in this price class... but who knows what they might have thought?)

Apart from the time reference, any other element of the chain could cause inaccuracies. This seems improbable as your camera just got CLA'd, though...

 

[Chan Tran]

You do not have the adjustment for each shutter speed. The adjustment if for the whole range. I am not sure but perhaps only 2 range or 1. When you do that all speeds change so you have to pick the best compromise. To reduce this problem one must very carefully adjust the shutter curtain travel time. Both first and second curtain. If this is not perfectly correct and you make the speed adjustment to get correct fast shutter speed you get wrong slow shutter speed.

 

[film_man]

If a crystal is used as time reference, several reasons can be thought of: physical damage to the crystal, aging, wrong electronic adaption of the crystal load (however I wouldn't expect this if the design engineers did their work correctly) etc.
Crystals have an accuracy of about +/- 100ppm or less, including aging.

Aha...didn't think of that. Can't they just use dilithium crystals?

 

[Chan Tran]

If a crystal is used as time reference, several reasons can be thought of: physical damage to the crystal, aging, wrong electronic adaption of the crystal load (however I wouldn't expect this if the design engineers did their work correctly) etc.
Crystals have an accuracy of about +/- 100ppm or less, including aging.

Of course, any other time base could have been used instead of a crystal. Those would be much less precise. Are the schematics of the Hasselblad available somewhere?
(as a side note, I wouldn't expect Hasselblad engineers to use anything but a crystal on a camera in this price class... but who knows what they might have thought?)

Apart from the time reference, any other element of the chain could cause inaccuracies. This seems improbable as your camera just got CLA'd, though...

The error in shutter speed of a typical camera is rarely ever due to the quartz crystal frequency shift. It has the most to do with the shutter curtain travel time which is a function of the spring tension, damper (if any) and lubrication.

 

[Chan Tran]

In the repair manual the shutter speed of 8 sec is considered in spec if it's between 7 and 9 seconds.

 

[etn]

The error in shutter speed of a typical camera is rarely ever due to the quartz crystal frequency shift. It has the most to do with the shutter curtain travel time which is a function of the spring tension, damper (if any) and lubrication.

Makes sense. I was looking at the question with my (biased) eyes of electronics engineer
But to nitpick a bit, aren't those points supposed to be taken care of during the recent CLA ? or is there an inaccuracy inherent to those shutters? Or are those not touched at all during a CLA?

 

[BrianShaw]

Makes sense. I was looking at the question with my (biased) eyes of electronics engineer
But to nitpick a bit, aren't those points supposed to be taken care of during the recent CLA ? or is there an inaccuracy inherent to those shutters? Or are those not touched at all during a CLA?

I don’t think that’s nitpicking. Yes, on a full overhaul it’s righteous to expect that aspect to have been addressed. I’m not always sure if folks advertising CLA mean a full overhaul or not... but that’s what I’d expect. Any time I’ve had service work done I judge accuracy by eyeballing a negative rather than measuring. Perhaps I’m engaging in self-protective behavior as a way to avoid conflict or disappointment.

 

[Ariston]

If a crystal is used as time reference, several reasons can be thought of: physical damage to the crystal, aging, wrong electronic adaption of the crystal load (however I wouldn't expect this if the design engineers did their work correctly) etc.
Crystals have an accuracy of about +/- 100ppm or less, including aging.

Of course, any other time base could have been used instead of a crystal. Those would be much less precise. Are the schematics of the Hasselblad available somewhere?
(as a side note, I wouldn't expect Hasselblad engineers to use anything but a crystal on a camera in this price class... but who knows what they might have thought?)

Apart from the time reference, any other element of the chain could cause inaccuracies. This seems improbable as your camera just got CLA'd, though...

Just to be clear, I am talking about a Nimon FE.

In the repair manual the shutter speed of 8 sec is considered in spec if it's between 7 and 9 seconds.

Well, at about 5.5 seconds, mine is out of spec, then. I think I used a bad tech. It's my fault for prioritizing price, but I have limited means and had heard good things from others online about this guy.

 

[BrianShaw]

Have you contacted your tech, expressed your concern, and allowed an opportunity to correct the situation?

 

[etn]

Just to be clear, I am talking about a Nimon FE.

Ooops, my bad... totally confused with Hasselblad 203FE. I didn't realize that your post was in the 35mm, not medium format, forum.

 

[Ariston]

Have you contacted your tech, expressed your concern, and allowed an opportunity to correct the situation?

No, it has been a few months since the service... and I had already sent back another camera he worked on and couldn't get a resolution. He was courteous enough, but I decided I should go elsewhere from then on. I found a local guy I am trying out now that will hopefully work out better. He's a little more expensive, but I suppose you get what you pay for.

 

[Chan Tran]

Just to be clear, I am talking about a Nimon FE.

Well, at about 5.5 seconds, mine is out of spec, then. I think I used a bad tech. It's my fault for prioritizing price, but I have limited means and had heard good things from others online about this guy.

In another page the manual listed the lower limit for 8 sec is 6.277 and upper limit is 10.190. A more slack tolerance but still yours is out of specs.

 

[shutterfinger]

There is a factory repair manual for the Nikon FE10 available at http://arcticwolfs.net/
There is a complete schematic diagram for the camera's electronics at the end of the manual. Its all IC's, transistors, diodes, resistors and capacitors on a flex circuit board. No part numbers for the electronic components.
This is an oscillator type timing not crystal. I would pay close attention to battery type (the camera accepts several) as some batteries drop to low voltages under load but appear to be good and contact condition, as incorrect supply voltage and current will cause the oscillator to run incorrectly.

 

[Ariston]

There is a factory repair manual for the Nikon FE10 available at http://arcticwolfs.net/
There is a complete schematic diagram for the camera's electronics at the end of the manual. Its all IC's, transistors, diodes, resistors and capacitors on a flex circuit board. No part numbers for the electronic components.
This is an oscillator type timing not crystal. I would pay close attention to battery type (the camera accepts several) as some batteries drop to low voltages under load but appear to be good and contact condition, as incorrect supply voltage and current will cause the oscillator to run incorrectly.

Wouldn't a low battery make the shutter slower, not faster? It doesn't hurt to check. I will replace them and see.

 

[Chan Tran]

Low voltage shouldn't be a problem.

 

[Chan Tran]

By the way to the OP. You don't have 1/3 stop off across the range. I think the problem is that the tech adjusted to get more accurate higher speed and sacrified the slow speed.

 

[RalphLambrecht]

I sent my FE off for a CLA several months ago. It looked fine when I got it back, including new light seals. But I just noticed today that my 1-second shutter sounded fast. So I set it to 8 seconds and timed it, and it closed in 5.5 seconds. So, it is about 1/3 stop off at that speed.

Does anyone know... does this mean I should expect it to be 1/3 stop off across the range of shutter speeds, or is it non-linear? I am not sending my camera back, because this is the second bad experience I've had, but I don't have the money to send it to someone else right now, either.

1/3 stop off is actually not terribly bad but, I think 5.5 vs8 s is closer to a 1/2 stop. shutter-speed deviations are not linear; they are different for each speed.

 

[Ian C]

In agreement with post 21, 5.5 seconds is 0.54 stops less than 8 seconds.

 

[Ariston]

1/3 stop off is actually not terribly bad but, I think 5.5 vs8 s is closer to a 1/2 stop. shutter-speed deviations are not linear; they are different for each speed.

In agreement with post 21, 5.5 seconds is 0.54 stops less than 8 seconds.

Can you teach me how to calculate this? The time of 5.5 is about 2/3 (0.6875) of 8, so I thought it was 1/3 stop slow.

 

[MattKing]

It is all to do with the fact that the light transmitted is related linearly to the time of exposure, but also linearly to the square of the aperture.
I find this to be a useful thing to remember:
When it comes to apertures, each of these numbers are 1 stop apart: 2.8, 4, 5.6, 8, 11, 16, 22, 32, 45, 64, ...
When it comes to exposure times, each of these numbers are 1/2 stop apart: 2.8, 4, 5.6, 8, 11, 16, 22, 32, 45, 64, ...

 

[Ian C]

To find the difference in stops ∆ from an initial time t1 to a second time t2, calculate

∆ = 1.4427*ln(t2/t1)

Example 1: Moving from t1 = 4 seconds to t2 = 16 seconds, we have

∆ = 1.4427*ln(16 seconds/4 seconds) = 2.00, a 2-stop increase

Example 2: For t1 = 16 seconds to t2 = 2 seconds

∆ = 1.4427*ln(2 seconds/16 seconds) = - 3.00, a DECREASE of 3 stops (indicated by the negative sign).

For your question of posts #1 and #23, t1 = 8 seconds and t2 = 5.5 seconds, so

∆ = 1.4427*ln(5.5 seconds/8 seconds) = - 0.54, a decrease of 0.54 stops.