Has anyone tried this? Easy and no soldering required.
http://decisivemomentum.blogspot.com...er-tester.html
Has anyone tried this? Easy and no soldering required.
http://decisivemomentum.blogspot.com...er-tester.html
Yes John, I followed that fella's suggestions and used my Canon DSLR to test some focal plane shutters on Graflex cameras. Worked great after a bit of fiddling.
I had done some work on one of the shutters with a buddy's oscilloscope/photocell setup and the numbers agreed. I also used a Schneider lens in Copal shutter as proof of the method. So I am happy to proceed to test the next shutters that cross my path.
Go ahead and give it a try ... certainly easier than messing with Audacity which I hated ... good luck ... J
I tried the Audacity method as well. It seems to work OK for slower shutter speeds like under 1/125. Anything over this it becomes hard to sort out the curve and tell when the shutter opens and closes. My computer has a 5 VDC output on microphone input. I just wired this through a phototransistor and it generally worked but as I said at the lower speeds. I tried messing around with resistors and capacitors like shown on the schematics all over the internet but it didn't make a difference. Unless some one can jump in and let me know I need to do get reliable shutter readings via my sound card I guess I will have to bite the bullet and buy the real thing.
Art
I've used the Audacity method also. It seemed to work fine, and I thought up to about 1/250th on most of my cams (Pentax SV, Nikon FE/FM) and LF lens. However, I noticed that when I tried to use it on my Nikon F, it was very difficult to sort out the start and end. I'm wondering if it can't vary by camera shutter design and implementation also. Anyway, seemed good enough for gov't work.
The Audacity method only works as well as your phototransistor and the selected sampling rate allow. I bought a fast-switching transistor (~10 microseconds and 10X the price of a Radio Shack one but still only $13).
For the money and the minimum of effort, this has to be one of the best hacks ever.
where did you buy the fast switching phototransistor?
Yes, I tried this, but could only get a reading at very slow speeds, Above about 1/20 second the wave form was unreadable, and I never was able to get the square wave shown on the website. Perhaps I had the wrong capacitor, photocell, or had not set up the program the way I should? Oh well...
I just built and used a shutter tester of this sort. I was able to measure the 1/400 speed on my graphex shutter (which apparently a full stop slow, c'est la vie). With a bit of improvement to the design I'm sure it could measure a few stops faster fairly reliably.
My soundcard was supplying 2V on the ring, which I understand is likely through a 2k resistor internally. I used an NTE 3034A phototransistor that I bought at Fry's (NTE is one of the more common semiconductor suppliers for retail stores.) I hooked it up as a common emitter amplifier with a DC blocking capacitor to the connector tip. (power to the collector, collector to the cap to the output, emitter to ground.) I'd show a schematic, but I don't have a good way to draw it up at the moment. Use the smallest ceramic cap you can find (I think mine was a 1.5uF cap). At the faster shutter speeds I had to be careful where I measured from as the charging time of the cap was noticeable (about 1ms).
Using an external power source might be necessary to reduce the source impedance so that I can get a faster charge time on the cap. I could possibly try another configuration without a cap, but I really don't want DC coupling into my sound card. The sampling rate on the sound card is 44.1kHz, so if the circuit is good, the sound card should be able to easily resolve shutter speeds as fast as 1/2000.
I put together my own circuit for this and was working out a way to make a digital read-out, but the digital camera testing method is SOOOO simple that I quit. Just put the camera facing a well lit circuit and you can test all the shutter speeds really quickly AND get an idea of what the amount of variance really means.
I use older lenses and shutters so it is really useful to see the results of a slightly slower top speed on the image instead of just knowing that it is 0.0138 seconds slow.
I bought one of the below testers that is made like the schematics we see:
Pre-made shutter tester from "Lurchrider"
I plugged it into the mic input of an EMU external soundcard and recorded using Audacity (which is free). I turned the record level control all the way up.
When the diode triggers when light appears, the voltage output is a positive spike. When the light goes away, there is a negative spike. The spike on the graph comes (necessarily) after the change in light reaching the diode, so I just "selected" (in Audacity) from the leading edge of the first (positive) spike to the leading edge of the next (negative) spike. I was able to measure all the speeds of all my shutter easily.
My test technique was to remove the rear cell of the lens, hold the tester up against the rear of the shutter, centered as well as possible, set the aperture to wide open, and then hold the lens board/tester combination in my left hand. With my right hand, I started the record mode of Audacity, cocked the shutter, pointed it to an incandescent desk lamp, released the shutter, cocked and fired the shutter three more times, and then stopped the recording. All four firings easily fit in a 10-second window. Then, I ignore the first firing, and zoom into each of the remaining three. Once zoomed in, I select the sound file between the peaks as described above, and read the information line about the selection, which includes the elapsed time. I recorded that in a spreadsheet.
I set up the spreadsheet to average those three readings, and then provide an error in stops by taking the log of the ratio of the actual time and the nominal time and dividing that by the log of 2.
It took me three hours to test 8 lenses, including downloading and installing Audacity, building the spreadsheet, and conducting all the tests. Repeat tests will take only an hour at most.
I was able to test even 1/500 speeds with good reliability. My general result was that my Compurs were pretty good at the fast speeds and draggy at the slow speeds, while the Copals were pretty good at the slow speeds but optimistic at the fast speeds. The best shutters were within a quarter stop up and down the range of useful speeds, but none were perfect. I now keep the resulting table in my bag.
Yes, I could have built the tester for maybe half what this one cost, but I would have had to order the photo-transistor--Radio Shack only sells IR-sensitive photo-transistors in their typical floor plan. Buying it complete was worth it just for the time savings, and would be especially so for those who don't own soldering irons.
Rick "who'll start sending in the worst ones for a CLA" Denney
Posting Permissions
Reply With Quote
Bookmarks