Olympus-OM
[Top] [All Lists]

Re: [OM] Shutter Tapering

Subject: Re: [OM] Shutter Tapering
From: HI100@xxxxxxx
Date: Wed, 18 Apr 2001 05:36:20 EDT
abutler@xxxxxxxxx writes:

<< 
Tim,
    I was hoping you would chime in on this subject. 
>>

Mike,
       The reason you get the transient on the two edges is that the audio 
boards are AC coupled and you are just seeing the input capacitor and 
resistor time constant decaying on the square pulse edges.  For example if 
the low frequency cut off frequency of the audio board is 20Hz then the Time 
Constant should be about 1/(20*2*pi) sec or about 8 mS.  If you used this 
system it would be hard to see what was happening at shutter speeds of around 
8mS (1/125 sec) since it would look like a single distorted pulse. At 
1/1000sec it would look much more like a narrow squarish 1 mS pulse, and at 1 
sec you would see two opposite polarity pulses of about 8mS width.

    Since audio boards are pretty cheap you could try modifying the board by 
increasing the input coupling capacitor. This would give you a different TC 
but it is difficult to get a long enough TC to test say a 1second speed and 
still get a square pulse. It might be possible to actually short out the 
coupling capacitor and get a square pulse even for long shutter periods. This 
depends on the chip design and how the AC coupling is implemented. Might be 
worth trying?
  
     A third way would be to gate a clock oscillator signal with the photo 
detector. This would record a burst of pulses with the overall burst width 
being the shutter speed. The clock frequency would need to be strictly less 
than the audio board sampling frequency (say 1/3 to 1/4 of sampling rate, 
which limits resolution a bit).  One comparator (or op-amp),3 resistors and a 
capacitor would make a suitable astable oscillator. You could drive the 
pullup resistor of your photo transistor directly from the op-amp output. 
This would produce a slightly different output: a gap in a continuous train 
of pulses during the time when the shutter was open. There will still also be 
a transient associated with the edges unless you add another R-C coupling 
circuit with a rather short time constant in front of the audio board. 

      Using the parallel printer port or game port as a single bit input is a 
better way of doing this, if you are prepared to write a little software to 
time things. 
   

                  Regards,
                 Tim Hughes
                >>Hi100@xxxxxxx<<

< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >


<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz