Cable Shutter Test Results

[joshuapd]

Hi guys,

After a bit of work and an unfortunate mishap of pressing refresh while making the post... here's my work on using vfmoto's, another user on APUG, cable shutter tester. I have no affiliation with vfmoto, I am merely reviewing it for the sake of both the communal understanding of this tester and continued accuracy of my tests as well, I'm no expert... so please help me out if I'm wrong

I'll be outlining my steps and method, with pictures and screenshots of my actual results. Hopefully this will help me understand what's going on on the mechanical end of my equipment and enable me to photograph with my predictability.

Working Materials:
· Crown Graphic
· Symmar-S 150mm f/5.6
· Cable Tester provided by vfmoto
· Desk lamp light source
· PC w/Audacity program


Method:
1. I started off with making sure I had a proper rig according to the manual of the shutter tester. Here's what my setup looked like:

2. I made sure to have a constant light source at an adequate space away to avoid any possible error, according to the manual
3. With tester lined up behind the rear element, light towards the front element I can now start testing

Testing Procedure:
1. Set shutter to “1/xxx”, xxx being the desired testing shutter
2. Prepare Audacity program to record
3. Cock and release “1/xxx” shutter speed
4. Stop recording
5. Find start of curve and end according to the point of inflection (where curve changes from down to up)
6. Record results & repeat
*The decimal equivalent is found on the lower part of the audacity program middle column*



Results:
At first I was quite scared I had done the test wrong because my 1/250, 1/125, and 1/90 speeds were a bit off, thinking I did it wrong I tested multiple times and with almost identical test results, I came to the conclusion these speeds have gotten slower.

​

[TABLE="class: cms_table, width: 0"]

[TD="align: center"]Testing Shutter Speed[/TD]
[TD="align: center"]Recorded Shutter Speed[/TD]
[TD="align: center"]Actual Shutter Speed[/TD]
[TD="align: center"]Benchmark Shutter Speed Equivalent[/TD]
[TD="align: center"]% Error[/TD]


[TD="align: center"]1/500[/TD]
[TD="align: center"]0.004[/TD]
[TD="align: center"]1/250[/TD]
[TD="align: center"]0.0020[/TD]
[TD="align: center"]100%[/TD]


[TD="align: center"]1/250[/TD]
[TD="align: center"]0.006[/TD]
[TD="align: center"]1/175[/TD]
[TD="align: center"]0.0040[/TD]
[TD="align: center"]50%[/TD]


[TD="align: center"]1/125[/TD]
[TD="align: center"]0.011[/TD]
[TD="align: center"]1/90[/TD]
[TD="align: center"]0.0080[/TD]
[TD="align: center"]37.5%[/TD]


[TD="align: center"]1/60[/TD]
[TD="align: center"]0.018[/TD]
[TD="align: center"]1/60[/TD]
[TD="align: center"]0.0166[/TD]
[TD="align: center"]+8.4%[/TD]


[TD="align: center"]1/30[/TD]
[TD="align: center"]0.035[/TD]
[TD="align: center"]1/30[/TD]
[TD="align: center"]0.0333[/TD]
[TD="align: center"]+5.1%[/TD]


[TD="align: center"]1/15[/TD]
[TD="align: center"]0.062[/TD]
[TD="align: center"]1/15[/TD]
[TD="align: center"]0.0666[/TD]
[TD="align: center"]-6.9%[/TD]


[TD="align: center"]1/8[/TD]
[TD="align: center"]0.125[/TD]
[TD="align: center"]1/8[/TD]
[TD="align: center"]0.1250[/TD]
[TD="align: center"]0%[/TD]


[TD="align: center"]1/4[/TD]
[TD="align: center"]0.241[/TD]
[TD="align: center"]1/4[/TD]
[TD="align: center"]0.2500[/TD]
[TD="align: center"]-3.6%[/TD]


[TD="align: center"]1/2[/TD]
[TD="align: center"]0.514[/TD]
[TD="align: center"]1/2[/TD]
[TD="align: center"]0.5000[/TD]
[TD="align: center"]+2.8%[/TD]


[TD="align: center"]1”[/TD]
[TD="align: center"]1.075[/TD]
[TD="align: center"]1”[/TD]
[TD="align: center"]1.0000[/TD]
[TD="align: center"]+7.5%[/TD]

[/TABLE]



In the table above, the left two columns are my recorded results of the tested shutter speed. The third column values are taken from the values listed in the manual as shutter equivalents in decimal form. You can also get these equivalents using a calculator and making an equation of the shutter speed, i.e. 1 / 500 = 0.0020 (Obvious, but adding in just for clarity). Last, is the margin of error; error can be seen as good or bad depending on how much it deviates from the actual speed it was set.

In the case of 500, 250, and 125, the margin of error deviates far enough to decrease the speed from 1 stop and less. That’s where the initial happy/sad moment came. With multiple tests, the near-consistent decimal results led me to believe these speeds have lost their luster.
The weird parts of the graph where the graph reads a whole bunch of changes could be a result of the light source or some unaccounted aberration, but because it was consistent I decided to make it negligent to my test results. With more than 30% of change in speed from speed 125 and up, I’ll be sure to make some changes to my shooting technique.

Below I’ve also included screenshots of each of my highlighted work areas of the specific shutters. The only thing I’m concerned with what I’ve done is that my working test results don’t match the shutter “good” shutter spikes included within the manual. I’m not sure why, but from what I’ve understood these can be considered equal, I have no way to completely verify the results other than the help from the community to improve my tests.

Total Tests:

---

Test area for 1/500: 0.004

---

​Test area for 1/250: 0.006

---

​Test area for 1/125: 0.011

---

​Test area for 1/60: 0.018

---

​Test area for 1/30: 0.035

---

​Test area for 1/15: 0.062

---

​Test area for 1/8: 0.125

---

​Test area for 1/4: 0.241

---

​Test area for 1/2: 0.514

---

​Test area for 1": 1.075

Conclusion:
From what I’ve gathered, the test results have proven to be accurate and consistent with multiple instances and would say the tester works great. With shutter speeds in the faster end degrading more than the slower speeds, it does make sense that they would be slower, but seeing the degree of change makes me just a bit more happier that I can confirm it's slower haha. Ideally the control would be my digital camera since I’ve just everything serviced for them so it should be in good order with a very small margin of error, but I have to test myself to make sure. I really enjoy doing the tests because it helps me better prepare for a proper exposure that I would have envisioned at the time of the photo.

I hope this helps others out in some small way, but please if there is anything wrong with my testing kindly point it out

 

[mmerig]

Thanks for the review, and your procedure looks fine. The faster speeds on old shutters tend to be slow, and your results are not surprising.

Without knowing how the shutter tester actually works (both the light source and the sensor) it is difficult to know how accurate it would be. Apparently, the signal frequency is about 1/120 of a second (count the cycles of the sine wave in the slower speed tests). The signal also appears clipped in the higher speed tests, so the boundary for the signal amplitude is uncertain to me.

Ideally, in a light-source and photo-sensor system, you would want either a consistent correlation between signal amplitude (i.e., voltage) to the shutter's speed (time), or a contrast in signal amplitude between open and closed conditions. With the former system, a consistency in light intensity is important; with the latter, light intensity only has to be bright enough to exceed the threshold of the sensor and amplifier, so you get a signal when the shutter is open and no (or negligible) signal when closed.

I have built and used both types of testers. The voltage versus time tester was not very accurate, so I built one that uses a 1/10000 pulsed light source that is sensed by a photo-transistor, and the transistors signal is read into a sound card (I use Audacity). The pulse can be much faster, but 1/10000 was enough for the speeds I was testing (up to 1/2000 second).

The advantage of the pulsed-light system is that if the signal frequency is fast enough, the shutter's open-close boundaries are very easy to see on Audacity, and the precision is defined by the frequency rather than light intensity (voltage amplitude) alone. Yet there is some difference in signal amplitude as the shutter opens and closes, so the signal trace on Audacity indicates a shutter's efficiency (the relative time for the blades to open fully and close fully) at different shutter speeds.

Perhaps vfmoto can let us know more about the tester. I am curious about the signal frequency.

I can send screen shots of my tests results if you are interested.

 

[Jim Jones]

It appears to me from the waveforms that severe clipping is occurring. This would tend to indicate longer shutter speeds than the actual effective speed. This is further aggravated by measuring just the center of the light cone transmitted through the rear of the lens instead of the entire area of the lens. Testing of between-the-lens shutter speeds with a photo diode is best done with a set-up that accurately shows the intensity of the light through the shutter as well as the duration of the exposure. Then we see that, at the highest speeds, the opening and closing of the shutter can take a significant fraction of the total exposure time. With the lens wide open, exposure on film begins just after the shutter blades begin to open, and continues until they are closed. When the lens is well stopped down, no light reaches the film until the blades have moved a significant amount, and the light is cut off while the blades are still moving. A shutter tester that merely indicates when the blades start to move and when they are closed will indicate a slower shutter speed than the film sees. especially at the highest shutter speeds.

Using a pulsed light source (or a gated photo diode) as mmerig describes is one way of getting numerical values for shutter speeds. However, using an oscilloscope to plot the light intensity through the full diameter of the lens against time is more useful. It does require interpretation, while a simple sound measurement can give a less accurate digital result. An oscilloscope can also measure flash delay and shutter bounce.

 

[AgX]

I'm puzzled by the term "cable tester" in this context. Can someone explain?

 

[joshuapd]

Thanks for the review, and your procedure looks fine. The faster speeds on old shutters tend to be slow, and your results are not surprising.

Without knowing how the shutter tester actually works (both the light source and the sensor) it is difficult to know how accurate it would be. Apparently, the signal frequency is about 1/120 of a second (count the cycles of the sine wave in the slower speed tests). The signal also appears clipped in the higher speed tests, so the boundary for the signal amplitude is uncertain to me.

Ideally, in a light-source and photo-sensor system, you would want either a consistent correlation between signal amplitude (i.e., voltage) to the shutter's speed (time), or a contrast in signal amplitude between open and closed conditions. With the former system, a consistency in light intensity is important; with the latter, light intensity only has to be bright enough to exceed the threshold of the sensor and amplifier, so you get a signal when the shutter is open and no (or negligible) signal when closed.

I have built and used both types of testers. The voltage versus time tester was not very accurate, so I built one that uses a 1/10000 pulsed light source that is sensed by a photo-transistor, and the transistors signal is read into a sound card (I use Audacity). The pulse can be much faster, but 1/10000 was enough for the speeds I was testing (up to 1/2000 second).

The advantage of the pulsed-light system is that if the signal frequency is fast enough, the shutter's open-close boundaries are very easy to see on Audacity, and the precision is defined by the frequency rather than light intensity (voltage amplitude) alone. Yet there is some difference in signal amplitude as the shutter opens and closes, so the signal trace on Audacity indicates a shutter's efficiency (the relative time for the blades to open fully and close fully) at different shutter speeds.

Perhaps vfmoto can let us know more about the tester. I am curious about the signal frequency.

I can send screen shots of my tests results if you are interested.

It appears to me from the waveforms that severe clipping is occurring. This would tend to indicate longer shutter speeds than the actual effective speed. This is further aggravated by measuring just the center of the light cone transmitted through the rear of the lens instead of the entire area of the lens. Testing of between-the-lens shutter speeds with a photo diode is best done with a set-up that accurately shows the intensity of the light through the shutter as well as the duration of the exposure. Then we see that, at the highest speeds, the opening and closing of the shutter can take a significant fraction of the total exposure time. With the lens wide open, exposure on film begins just after the shutter blades begin to open, and continues until they are closed. When the lens is well stopped down, no light reaches the film until the blades have moved a significant amount, and the light is cut off while the blades are still moving. A shutter tester that merely indicates when the blades start to move and when they are closed will indicate a slower shutter speed than the film sees. especially at the highest shutter speeds.

Using a pulsed light source (or a gated photo diode) as mmerig describes is one way of getting numerical values for shutter speeds. However, using an oscilloscope to plot the light intensity through the full diameter of the lens against time is more useful. It does require interpretation, while a simple sound measurement can give a less accurate digital result. An oscilloscope can also measure flash delay and shutter bounce.

Hm, that's something quite new to me, the pulsed light source reasoning brought out a lot of new things I didn't really think about. My understanding is that vfmoto designed this to give us those that don't have as much understanding of testing (like me), and so that tester was designed to give just enough accuracy, I'm assuming.

However, I'm curious to see those screenshots mmerig. My next question after learning about these things is where can you draw the line of finding "practical" shutter speeds, because between the pulsed-light source vs. the cable tester results are different. I'm curious of how significant the difference is between the two, is it enough to impact exposure results if I were to use those shutter speeds?

I'm going to give the tester another go with a variety of light sources and make another conclusion there about how the tester performs for the photographer.

I'm puzzled by the term "cable tester" in this context. Can someone explain?

Dead Link Removed. Basically, a tool for us to use to measure light that passed through the lens once the shutter goes off. There's a bit more to it than that but, it's the general gist. At least, I hope it is if I'm not mistaken haha.

 

[Kilgallb]

I checked my shutter using a photo diode and a storage oscilloscope. I bet AN EE student at the local tech school could do the same.

Also, at higher speeds the opening time of the leaf blades is significant so you can see the rise and fall time to get the true speed. This means the Center of the image is actually more exposed than the edges. Not a big deal at 1 second but a big deal at 1/250th.

 

[shutterfinger]

This tester works like the infrared photo LED/photo transistor tester that has been on the web for years. The only advantage to the cable tester is that you can use a wider variety of light sources.

Also, at higher speeds the opening time of the leaf blades is significant so you can see the rise and fall time to get the true speed. This means the Center of the image is actually more exposed than the edges. Not a big deal at 1 second but a big deal at 1/250th.

A leaf shutter is just a variable aperture that operates at controlled times. I doubt you will see any difference in exposure between maximum aperture exposure and a minimum aperture exposure of the same scene provided the light source is adjusted to match the shutter speed/aperture in use whether the shutter speed be fast or slow.

The amplitude of the waveform in Audacity is determined by the intensity of the light source. It is normal for it to be smaller at faster speeds than at slower speeds.
http://photo.net/large-format-photography-forum/0044cW?start=10
I can no longer guarantee the Radio Shack links to be valid. A deep red filter over the IR LED photodiode allows for use of incandescent and fluorescent light sources that emmit a white light.

 

[mjork]

Results:
At first I was quite scared I had done the test wrong because my 1/250, 1/125, and 1/90 speeds were a bit off, thinking I did it wrong I tested multiple times and with almost identical test results, I came to the conclusion these speeds have gotten slower.

​

[TABLE="class: cms_table, width: 0"]

[TD="align: center"]Testing Shutter Speed[/TD]
[TD="align: center"]Recorded Shutter Speed[/TD]
[TD="align: center"]Actual Shutter Speed[/TD]
[TD="align: center"]Benchmark Shutter Speed Equivalent[/TD]
[TD="align: center"]% Error[/TD]


[TD="align: center"]1/500[/TD]
[TD="align: center"]0.004[/TD]
[TD="align: center"]1/250[/TD]
[TD="align: center"]0.0020[/TD]
[TD="align: center"]100%[/TD]


[TD="align: center"]1/250[/TD]
[TD="align: center"]0.006[/TD]
[TD="align: center"]1/175[/TD]
[TD="align: center"]0.0040[/TD]
[TD="align: center"]50%[/TD]


[TD="align: center"]1/125[/TD]
[TD="align: center"]0.011[/TD]
[TD="align: center"]1/90[/TD]
[TD="align: center"]0.0080[/TD]
[TD="align: center"]37.5%[/TD]


[TD="align: center"]1/60[/TD]
[TD="align: center"]0.018[/TD]
[TD="align: center"]1/60[/TD]
[TD="align: center"]0.0166[/TD]
[TD="align: center"]+8.4%[/TD]


[TD="align: center"]1/30[/TD]
[TD="align: center"]0.035[/TD]
[TD="align: center"]1/30[/TD]
[TD="align: center"]0.0333[/TD]
[TD="align: center"]+5.1%[/TD]


[TD="align: center"]1/15[/TD]
[TD="align: center"]0.062[/TD]
[TD="align: center"]1/15[/TD]
[TD="align: center"]0.0666[/TD]
[TD="align: center"]-6.9%[/TD]


[TD="align: center"]1/8[/TD]
[TD="align: center"]0.125[/TD]
[TD="align: center"]1/8[/TD]
[TD="align: center"]0.1250[/TD]
[TD="align: center"]0%[/TD]


[TD="align: center"]1/4[/TD]
[TD="align: center"]0.241[/TD]
[TD="align: center"]1/4[/TD]
[TD="align: center"]0.2500[/TD]
[TD="align: center"]-3.6%[/TD]


[TD="align: center"]1/2[/TD]
[TD="align: center"]0.514[/TD]
[TD="align: center"]1/2[/TD]
[TD="align: center"]0.5000[/TD]
[TD="align: center"]+2.8%[/TD]


[TD="align: center"]1”[/TD]
[TD="align: center"]1.075[/TD]
[TD="align: center"]1”[/TD]
[TD="align: center"]1.0000[/TD]
[TD="align: center"]+7.5%[/TD]

[/TABLE]

I don't think your shutter has gotten slow. These results are approximately what I'm getting for my leaf shutters, too. I assume you are measuring the time from when the shutter starts to open to when it is fully closed? That's different from the effective shutter speed. At those faster shutter settings, the time it takes to open and close the shutter leaves becomes a major factor. When you set the shutter to 1/500 and measure 4ms, then for most (or pretty much all) of those 4ms, the shutter isn't actually fully open but on its way to open or close. So effectively the light that makes it through corresponds to a 2ms exposure.

What you want to measure is the time from when the shutter is half-way open to when it is half-way closed (with aperture set to max opening).

There is a photography text book that has a nice chapter on exactly this topic:
"Basic Photographic Materials and Processes" by Leslie D. Stroebel et al.
I picked up an older edition of this book used for under $1 on Amazon.

 

[SalveSlog]

(.. but at the smallest aperture it will be half speed. Nowadays fast shutter speeds may be used along with small apertures when using very fast film. They probably never did so in the 50es, though.)

 

[mjork]

(.. but at the smallest aperture it will be half speed. Nowadays fast shutter speeds may be used along with small apertures when using very fast film. They probably never did so in the 50es, though.)

Exactly. If you use maximum aperture, the exposure is going to be what you expect based on the shutter setting. But if you use smaller aperture openings, the measured times are a good indication of the result: exposure will be off by one stop at 1/500 shutter setting.

That's working as designed. The effective shutter speed is a function of the aperture setting. The shutter speed is calibrated based on max aperture opening.

 

[shutterfinger]

I have / had some shutters read their marked fast speeds with the tester linked in my last post. Test with real time exposures of the same scene at various speeds provided the scene brightness is within the speed and aperture range of the shutter. Include an 18% gray card or other even color mid tone object in the scene. The gray card can be read with a densitometer or the scan checked for the same tone in PS.

 

[joshuapd]

Well, after reading all of the responses, which I'm really thankful for all the conversation, I think I'll go ahead and do some more research on the practical applications of this information and refine my results. All this information makes learning this a lot easier than if I were to just read and confuse myself haha.

When I have time this week, maybe I'll check in between differences in times based on different apertures, didn't think about that either. With semester slowly starting, hopefully all this reading will put me in a better position for my advanced lighting class

 

[shutterfinger]

I have tested shutters at both wide open, f4.5 to f6.3, and at f32 and never seen any difference in measured times.
I did some crude testing with a microphone against the shutter housing and listening carefully on headphones to find the start of shutter opening to when it reached full opening. The result from 3 shutters was open time to be .0001 to .0003 second. The delay mechanism holds the shutter blades open for a predetermined time then the main cocking spring forces them closed. I think it will be very difficult for the average person to test the close time once the delay timer has released the blade controller and main cocking spring mechanism. On speeds above 1/100 (1/125) second a booster spring adds additional closing force to the main cocking spring which is applied near instantaneously when the blades reach full open. Slow fast speeds are due to worn blade controller, shutter blades and their pivots, and weakened springs. Springs are weakened from high usage and the shutter being used regularly after it slows to 1/3 to 1/2 slower than the marked speed. Shutters that are run until they quit rarely come back to good above 1/50 speed.

The exposure difference between wide open and f32-f45 is so small that it has negligible impact on exposure unless you are an extreme pursuitist.