Measuring shutter speed

Jesito · Expire: 2015-01-07

Hi,
After my unsuccessful trial of a simple shutter speed meter, I've to get back to the basics, so I've been looking for info and reading.

Related to the accuracy of a shutter speed meter (My main concern).

Accuracy Issues

The performance of an electronic shutter tester is dependent upon a number of factors. Let's consider some of these factors, using an real world example at a shutter speed of 1/1000.

1. Velocity of Shutter. This can be calculated by dividing curtain travel time by distance. The horizontal travel curtain in Minolta SRT 101 should be adjusted to cross the 36mm focal plane in 13.5 to 14 milliseconds. Lets assume 14 milliseconds and divide by the distance of 36mm. The result is 14milliseconds/36mm = .38889 milliseconds per millimeter.

2. The time between the opening of and the closing of the shutter. This should be equivalent to the shutter speed, in our example is 1/1000th of a second or 1 millisecond.

3. Combining the velocity with the exposure will result in the distance between the opening and closing shutter curtains which in our case is 1millisecond/.38889milliseconds/mm = 2.57mm.

4. Now consider the size of the photo transistors or other electronic sensors that is needed to switch the circuit on and off. Assume the detector is 1mm across. For a 1/1000 or 1 millisecond exposure, the detectors relationship to the opening between the opening and closing shutter is 1mm/2.57mm or 37 percent. It should be clear that a smaller sensor is better than a larger sensor.

5. Phototransistors have a lag time between being switch on and or off and visa versa . Each particular phototransistor design is different in this regard, but it is easy to find off the shelf photo-transistors that have a lag time of 10 micro seconds or less. This response time is similar in both turn on and turn off functions, so basically cancels itself out. This fast response time and the fact that it is pretty much symmetrical for turn off and on should result in only small error.

6. Phototransistors and other photoelectric devices in a real world circuit don't instantaneously turn on or off the moment a single ray of light touches the sensor. Phototransistors have a curve that allows more current to flow as more light reaches the sensor, at least until it is saturated. As the opening curtain travels across the focal plane, it allows more and more of the light to reach the sensor, which in turn, allows more current to flow, until enough light reaches the sensor to saturate it. The circuit in which the phototransistor is placed will switch on after a predetermined amount of current flows. Bottom line is that the circuit turns on sometime after light begins hitting the phototransistor, not immediately. The circuit turn on point depends on both light intensity and the circuit in which the photo transistor is placed.

7. There is a similar lag time for turn off of the circuit. A properly designed circuit will turn off in a inverse function compared to the turn on. If this is the case, the lag for turn on and turn off will mostly cancel each other out.

An analysis of these factors reveals that to minimize error, one item is the most critical, assuming that the sensor designed is as small as practical.

1. Symmetrical electronic turn-off/turn-on performance. In other words, it is most important that the function that turns on the sensor output is as close as possible to the function that turns it off.

This is really handy and clarifying for the curtain shutters, but what about the blade based ones?. Any special considerations?.

Jes.

bob955i · Joined: 15 Apr 2007 Posts: 2495

Jes:

Some DIY speed testers use two phototrannies instead of just one for greater accuracy.

That said, even with the best will in the world, you'll never get all the speeds spot on - this is where a consistent speed is more important than a one time accurate one that may drift with time and temperature later on.

For critical exposure, a shutter speed that is consistently say half a stop fast, or half a stop slow, is much easier to deal with than a speed that drifts all over the place - if you know that it's half a stop fast or slow, you can allow for that, but you can't allow for a speed that drifts as you'll never be sure by how much it's out.

Jesito · Expire: 2015-01-07

bob955i · Joined: 15 Apr 2007 Posts: 2495

Farside · Expire: 2013-12-27

Jes; I tried Rick Oleson's method and it works quite well. 1/1000s and 1/500s were very close to his pics, but there seemed to be an irregular gap from 1/500s to 1/250s, so I'll wait to see what a test film tells me.

If you start building shutter testers, I might be interested in one.

Jesito · Expire: 2015-01-07

Farside · Expire: 2013-12-27

Farside · Expire: 2013-12-27

I just found this...
http://www.kyphoto.com/classics/shutterspeedtester.html
Seems dead simple, but it does depend on having a known accurate shutter to set it up to start with.

Jesito · Expire: 2015-01-07