Light Detection circuit Improvements

Thread Starter

JDR04

Joined May 5, 2011
367
Could the members of this forum please make any improvement suggestions to my circuit. They are always appreciated.

I am really interested in finding out how one would determine what specifications the Phototransistor should be.


Thanks a lot-JDR04
 

Attachments

Last edited:

bertus

Joined Apr 5, 2008
22,270
Hello,

I am missing the ground connection of the LM393.
Pin 4 should be tied to ground.
Also the photo transistor will burn when the potentiometer is a 0 Ohms.
Put a resistor of 100 Ohms in series with the potentiometer to protect the transistor.

Bertus
 

mcasale

Joined Jul 18, 2011
210
To prevent oscillations at the switching point, add some positive hysteresis around the comparator. Just a few millivolts is all you need.

The observations that "bertus" makes are good ones.
 

Thread Starter

JDR04

Joined May 5, 2011
367
Thanks mcasale, could you explain how I would go about adding positive hysteresis please. I'd appreciate that -JDR04
 

mcasale

Joined Jul 18, 2011
210
Here are a couple of basic references. Essentially, you add positive feedback from the output to the "+" input so that it shifts a bit after the comparator switches. It looks like an op amp amplifier that is connected wrong, but it's not. It really works!

One note: if the comparator is an open-collector output, like the LM311, you need to include a pullup resistor to Vcc.

http://www.analogzone.com/acqt0327.pdf
http://home.cogeco.ca/~rpaisley4/Comparators.html
 

Thread Starter

JDR04

Joined May 5, 2011
367
Thanks mcsale, I'll have something to read in the bed tonight and better still to tinker with the day after. Really appreciate all the info. JDR04
 

Thread Starter

JDR04

Joined May 5, 2011
367
Thanks wayneh, have always appreciated your input. I'll fiddle around with the circuit and let you guys know how it goes. Thanks again -JDR04
 

WBahn

Joined Mar 31, 2012
29,979
If you do use a multi opamp/comparator chip and don't use all of them, don't just tie the inputs together and leave it go at that. What should the output be? It wouldn't take much noise to cause that unused amp to oscillate (and, once it got going, there might be enough parasitic feedback to keep it going). Drive the output into a known state. This can be down by tying one input to each rail (but make sure that the data sheet shows that the rails are within the allowed input voltage range). Another way is to configure it as a voltage follower and then ground the input. Yet another way is to put three resistors in series between the rails and have the amplifier see the voltage across the middle resistor. The two end ones can usually be large (100kΩ or more) and the middle one can be about a tenth that size (10kΩ).

For amps packaged like the 393, you can use the amp on the top (high-numbered pins) as your active amplifier and then just tie Pin 1 to Pin 2 (making the other amp a unity-gain voltage follower) and tie Pin 3 to Pin 4, making the voltage that is being followed the negative rail, which is within the input voltage range.
 
Top