PDA

View Full Version : Sound Card Shutter Tester



kev curry
23-Nov-2008, 12:20
I fancy trying to put one of these together but can I rely on the accuracy of one of these home made devises? Any experiences?

muchas gracias:)

http://www.davidrichert.com/sound_card_shutter_tester.htm

aduncanson
23-Nov-2008, 14:59
The rise and fall times of the phototransistor are 3 millionth's of 1 second.
The sound card and Audacity samples the signal 44,000 times per second by default.

These figures should allow you to time a 1/500th second shutter speed to reasonable accuracy, say to within about 1/60th of a stop.

Is that good enough for your color transparencies? ;)

Greg Lockrey
23-Nov-2008, 15:06
They're accurate enough for photographic purposes. :)

kev curry
23-Nov-2008, 15:38
Smashing guys thanks.

Now to get the bits a bobs and start soldering. Me no need no stinking overpriced Calumet tester:)

Ron Marshall
23-Nov-2008, 16:14
Instead of building an optical set-up, would it be possible to use a microphone with the same software.

I assume since this is simpler and not being done that there must be a downside?

I only need to test speeds from 1 to 1/30.

aduncanson
23-Nov-2008, 19:21
Instead of building an optical set-up, would it be possible to use a microphone with the same software.

I assume since this is simpler and not being done that there must be a downside?

I only need to test speeds from 1 to 1/30.

I had read somebody suggest this and tried it once with the very loud shutter of my Fuji GW690III. I may not have persevered sufficiently and my microphone was the worst possible, but I was never able to differentiate the opening and closing of the shutter. I am not sure that I was using Audacity at the time either.

I would suggest that you give it a shot. It sure doesn't cost a lot to try.

Ron Marshall
23-Nov-2008, 19:36
I would suggest that you give it a shot. It sure doesn't cost a lot to try.

Unfortunately it entails a probably extensive search for the microphone that came with the computer, which in any case is not good quality. Probably cheaper for me to make the optical tester than to purchase a good mic. Thanks for the info.

C. D. Keth
23-Nov-2008, 19:50
I wouldn't go with the microphone. How do you know what part of the noise of the shutter actually coincides with the opening and which part goes with the closing? If you record a shutter well, you will hear much more than 2 clicks.

Bill_4606
23-Nov-2008, 20:01
I've used the Audacity program with a very cheap microphone and found it more than adequate for shutter measurements longer than 1/250 sec. I also have a photo transistor apparatus similar to the one in the project diagram referred to in the original post. Instead of connecting to audacity, I connected it to a very high speed digital storage scope. I found the slew rate (speed) of the photo transistor to be a problem at speeds faster than 1/500 sec -- but still usable.

Back to the Microphone... for shutter speeds longer than 1/125 I would definitely use the microphone. I even used it with my di$%al SLR camera and could see in the Audacity wave form where the mirror raised, shutter clicked (start and stop), and mirror returned. The lesson here is that you get a little more information with the microphone than the light sensor.
Regards,
Bill Riley

Ron Marshall
23-Nov-2008, 20:13
I've used the Audacity program with a very cheap microphone and found it more than adequate for shutter measurements longer than 1/250 sec. I also have a photo transistor apparatus similar to the one in the project diagram referred to in the original post. Instead of connecting to audacity, I connected it to a very high speed digital storage scope. I found the slew rate (speed) of the photo transistor to be a problem at speeds faster than 1/500 sec -- but still usable.

Back to the Microphone... for shutter speeds longer than 1/125 I would definitely use the microphone. I even used it with my di$%al SLR camera and could see in the Audacity wave form where the mirror raised, shutter clicked (start and stop), and mirror returned. The lesson here is that you get a little more information with the microphone than the light sensor.
Regards,
Bill Riley

Thanks Bill. I will dig out my mic and give it a shot.

Ivan J. Eberle
23-Nov-2008, 20:28
A better phototransistor like one I bought for about $10 will have a response time of about 3 millionths of a second (cheap ones from Radio Shack don't react as swiftly). I was able to wire this inline with a single 1.5V AA battery and a scavenged headphone jack. The hardest part of it was cutting a hole and epoxying the phototransistor flush with the project box I mounted it in (IOW, it was really simple and easy to make).

A microphone would seem not to be worth the bother for shutter testing purposes; the phototransistor will also be useful to time an SLR shutter (moving slit).

Alan Davenport
23-Nov-2008, 23:23
Instead of building an optical set-up, would it be possible to use a microphone with the same software.

I assume since this is simpler and not being done that there must be a downside?

I only need to test speeds from 1 to 1/30.

I think it is unlikely that the moment a shutter makes noise, will coincide with the moment it actually opens or closes. If there's any difference in the timings of those events (sound vs light) then the error will be magnified at faster speeds. I suspect the data you'll get will be ... suspect. (OTOH, it don't cost nuthin to plug a mic into the sound card's mic jack.)

Sound vs light timing is guaranteed to disagree if you're measuring a leaf shutter, since the effective timing (of a leaf shutter) starts when the shutter blades are halfway open, and ends when they are halfway closed, NOT the points where you'll hear the click.

Ron Marshall
24-Nov-2008, 07:54
Sound vs light timing is guaranteed to disagree if you're measuring a leaf shutter, since the effective timing (of a leaf shutter) starts when the shutter blades are halfway open, and ends when they are halfway closed, NOT the points where you'll hear the click.


I'm not too worried if my data is out by as much as 1/3 of a stop. I just want to see if any of the slower speeds are really out of whack.

Ivan J. Eberle
24-Nov-2008, 12:25
Chances are good that slower speeds on rim-set leaf-shutters controlled by escapements will come back reasonably close to spec with a thorough cleaning and lube. Once cleaned, high speeds are the more likely ones to be off by more than 1/3 stop due their higher spring tensioning and spring fatigue (if that's the correct terminology). None of which may matter very much except to those using wide taking apertures, e.g. doing portraiture and trying to balance fill-flash with existing light.

Jim Graves
25-Nov-2008, 00:53
I bought the $40+ one that's often for sale on ebay -- built from the on-line directions. I tested it against two professionally timed shutters that I'd just had serviced. I used two different test programs ... Audacity and another whose name escapes me now ... The times up to about 1/100 were very close to the professionally timed readings ... from 1/125 on up they got increasingly divergent ... at 1/500 they weren't even close.

I still use it for testing slow speeds but do not trust it for anything above 1/100.

One thing I did learn in the testing was just how much shutter speed improved after about 10-15 dry fires. Now I always click off 10 or more before I take my first shot out in the field.

kev curry
25-Nov-2008, 05:58
That's very encouraging Jim. I'm only interested in the slower speeds anyhow.

I have another question... I've been around 2 electronic shops today but no ones got a 5k ohm resistor, the closest I can find is either a 4.7 or a 5.6 so I figured that the 4.7 was the one to try thinking that the 5.6k ohm will create to much resistance? I suppose if its crucial to have exactly 5k ohm I could combine 2 or 3 smaller resistors in series to make 5 eg a 3k + 2k = 5k ohm....hmmmm.
If this works its cost me less than 8 quid and thats including a groovy little black box to house it all in:)

aduncanson
25-Nov-2008, 07:53
That's very encouraging Jim. I'm only interested in the slower speeds anyhow.

I have another question... I've been around 2 electronic shops today but no ones got a 5k ohm resistor, the closest I can find is either a 4.7 or a 5.6 so I figured that the 4.7 was the one to try thinking that the 5.6k ohm will create to much resistance? I suppose if its crucial to have exactly 5k ohm I could combine 2 or 3 smaller resistors in series to make 5 eg a 3k + 2k = 5k ohm....hmmmm.
If this works its cost me less than 8 quid and thats including a groovy little black box to house it all in:)

I am not familiar with application details for that phototransistor but I would not worry about 4.7 vs 5.0 Ohms. Electrical designers work in dB like photographers work in stops and 4.7 vs. 5.0 is only .27dB. The resistor is probably just there to limit the current draw from battery and the actual value is not critical for the accuracy of the system.

Ivan J. Eberle
25-Nov-2008, 09:04
The phototransistor that I used wouldn't work with a resistor (of any value) and my MBP 15" sound card--so I built my tester without one.

I'd suggest setting the recording sampling rate of Audacity to the highest setting your sound card hardware supports. I think I've got my set at 96Kbps (96000 divisions per second) for better high-speed shutter accuracy.

But it could well be that fine and fussy accuracy of a sound card shutter timer might depend on the response time of the sound card and speed of the processing cycle at higher speeds . I'm reasonably confident mine is accurate to 1/500s, but to confirm this I'm thinking I might dig out the electronically timed Nikon F5 to run some tests. (F5 shutter self-diagnoses, won't even fire if the timing gets even slightly out of whack).

P.S. --oops... the F5 won't fire the shutter with the back open without a hack to the AF selector button contacts. Gonna have to be one of my Pentax LXs, also electronically timed but will wait til some later date as they're mid roll--

kev curry
25-Nov-2008, 11:04
Eureka............ '' Its alive Its alive!! ''

Well that was straight forward, no problems and to think just yesterday I never knew a resistor from a transistor:) I must say that I followed the schematic drawing that shows one 1.5v battery, if you notice in the photos of the tester in the link above it shows 2 1.5v batteries and 2 wires on the E (Emitter) leg of the photo-transistor....I don't understand what that's about exactly so I followed the schematic as written... the photos seem very confusing. I now need to try and interpret the graph data on Audacity:confused:

kev curry
25-Nov-2008, 14:27
I've sussed the basics of Audacity program it was easy even for me... the ultimate PC dummy! I still cant get over how cool this little tester is...finally I know the shutter speeds of my little Zeiss Ikon folding camera. Now to test the large format shutters;)

S. Preston Jones
25-Nov-2008, 14:49
Would the results obtained using the sound card tester be much different from the results which would be obtained using the Calumet shutter tester? It appears that it would not be too hard to build a sound card tester. Thanks.