Camera shutter speed tester

[goodoltup]

Greetings forum members,
I would like to design a circuit to test the shutter speed on my old film cameras. There are a multitude of circuits available on the web detailing how to use a phototransistor and a battery to output a voltage spike, which can be read on your computers sound card. I have tried this, and the result is satisfactory. I would like to improve on the circuit a little, and I am asking for some advice.

The circuits that I have found allow the phototransistor to operate in its simplest form, just varying the voltage from low to high in response to light. This, combined with a capacitor sometimes, creates a tiny voltage spike that can read when the transistor first sees light, and again when it stops seeing light. I would like to use a photo sensor to trigger a voltage rise for the entire time the sensor sees light. And, I would like it to respond when the sensor sees ANY light, meaning a very, very low threshold. And, I would like it to have very fast response time. The camera shutters that I am testing are not generally faster than 1/1000th of a second, but I do have some cameras that are faster than that.

What should be used, a photodiode, or a phototransistor?
How can either be designed in a circuit to act as a binary switch?
If another transistor is used to make it act like a switch, will the response time be slowed down?

Thank you for any and all suggestions.
Cheers, Josh

 

[alec_t]

I would like to use a photo sensor to trigger a voltage rise for the entire time the sensor sees light.

If you do that you will have to totally exclude all ambient light. The cap was probably used so that the circuit responded to a light change, making ambient light exclusion less critical.

I would like it to have very fast response time

Phototransistors have a response time much faster than any camera shutter period (think microseconds). An extra transistor will add only a microsecond or so to the response time and will probably be essential to amplify the photo-sensor signal.

 

[goodoltup]

Thank you Alec,
I plan on mounting the photosensor on some PCB, and then sealing the board against the camera body, so it will in total darkness until the shutter opens. That way ANY light that touches the sensor will operate it, meaning that you will only need a flashlight, not an IR source or laser or something.
Could I use a voltage comparator or something to get a square wave signal? As in, when the voltage comparator sees even the slightest rise in voltage it could trigger a transistor to open a gate?

 

[Reloadron]

Yes, you could use a comparator circuit along the lines of what is here. Scroll down to the LDR circuits for some ideas where you would be using a photo transistor and not an LDR. The link should be useful. You will get a pulse out and the width of the pulse is the shutter time.

Ron

 

[goodoltup]

Something like this, maybe?View attachment 61069

 

[Reloadron]

You don't need that transistor in there? Just a pull up beyond the LM339 and pretty much you have it. Later this evening I can do a drawing iof it would help. I am having a busy work day so can't get to it till later.

Ron

 

[goodoltup]

Any help or drawing would be appreciated! I'm in a university course for Mechanical Engineering, and we are just beginning to work with circuits. This is just some extra curricular work for fun. I have a rudimentary understanding of electronic principles, and I have been following published circuit diagrams up until now. This is the first time I have tried to just make something up.

 

[Reloadron]

You want a circuit like the attached. The LM339 is a quad comparator so for just a single you would use a LM311 which is a single comparator 8 pin chip. Initially try 10K Ohms for R1. I used a 10 K Ohm pot to set the reference on the comparator which would be adjusted so you get a nice transition low to high when light strikes the NPN photo transistor. Just about any generic NPN photo transistor should work.

The circuit is just a variation of the other attached circuit taken from the link I posted earlier. When light strikes the photo transistor the voltage at the junction of R1 and the photo transistor collector increases. When it exceeds the reference set with the 10K pot the comparator output will toggle positive. Thus a positive pulse with a duration for the time the shutter is open. The leading positive going transition of the pulse would start a counter and the negative going transition would stop a counter.

Make Sense?

Ron

 

[goodoltup]

Thank you, that is of great help. I was working off the diagram below that one, the Relay Driver Output Schematic, but that one uses a PNP transistor. I think I understand what it happening in your diagram; pin 7 on the LM339 ( or 311 ) is the reference voltage, and pin 6 is the voltage it reads. I may not fully understand the comparator, though. I thought that the output either connects to ground or stays open depending on the states of pins 6 and 7? But what you are saying is that the output is either zero or positive? I have read through the data sheets, but have not found an easily apparent answer.
I would like to use a CR2032, would this diagram work on +3V?
Would it be possible to use a photodiode instead of a phototransistor? (I read that the response on the former is faster than the latter)
Would this circuit output the voltage that is found at the LM311 power pin? I would need to keep the output voltage to 1 or 1.5 VDC to keep within the specifications of the sound card.
Thank you,
Kind regards,
Josh

 

[Reloadron]

Hi Josh, you are using the audio input and software like Audacity?

Ron

 

[goodoltup]

Yes, exactly. I use a Macbook Pro with Audacity. I may also get an o-scope app for my phone. I could use the scopes at school, but just for testing the circuit, as I don't want to bring all my cameras into school.

 

[Reloadron]

OK, this will change a few things. I was looking at a few circuits that use the audio input of a PC and it dawned on me... Duh, the audio inputs to a sound card are capacitive coupled (as audio should be). That will not set well with what I tossed out there. You can't run a DC level into a sound card. The input capacitor will block the DC component level. That is why when doing it, they measure an interval from a blip to a blip. Much unlike measuring a pulse width on a scope or counter. All of a sudden that dawned on me as you are running a signal into an audio input of a PC and not a counter or scope. I was a little slow on the uptake on that one.

Let me think about this and I'll be back.

Ron

 

[goodoltup]

I have an idea,
What if I did something like this, and set one reference limit close to 0, and the other just a little bit higher than that. That way I would get a nice, but quick, square wave form at the beginning and the end of the cycle.
Remember, I am just doing this project for fun; learning and doing is as important as the result! I know I could just put a photodiode and a cap on some strip board and call it a day, but that would be very boring, everyone has already done that. Do you think that the sound card would tolerate two quick DC square waves that are 1V or less?

 

[alec_t]

Do you think that the sound card would tolerate two quick DC square waves that are 1V or less?

Be aware that the sound card / Audacity combo will have a sharp cutoff frequency ~20kHz. That implies pulse rise/fall times faster than 1/20 mS won't be represented accurately, so this may limit the accuracy of your measurements.

 

[Reloadron]

The problem is as I mentioned. What I forgot to consider is a PC audio input uses capacitor coupling of the incoming signal. While this is wonderful for audio, it sucks as to getting a pulse through. Remember, a pulse (in our case) would go from 0 volts to a DC level and then after a period of time return to 0 volts. Later today or tomorrow I'll post exactly what I am getting at with a picture or two. I am trying to think of a good workaround.

Plus what Alec mentions tossed in also.

Ron

 

[goodoltup]

And thank you both!
Hey, why not make this even more complicated and incorporate a counter circuit that could just measure the time of the full DC pulse, and read out on an LCD? Could maybe use an oscillator and a frequency counter?

 

[Reloadron]

That can be done. It is not overly difficult to build a simple counter to measure period. I'll get a few things together for you this weekend. It has been one of those long work weeks.

Ron