help understanding the monster I created :-)

Yes, that's correct. The LM339 is very sensitive: it responds to very small differences between the inputs. As the photocell resistance passes through the critical threshold, the slightest "bounce" is enough to momentarily push it back below the threshold. To overcome this, you should add a little hysteresis by connecting a very large resistor (at least 100k, maybe even 1M) between the output and the non-inverting (+) input. The on-off and off-on transitions will then occur at slightly different points e.g. like a thermostat.

Reply to
Andrew Holme
Loading thread data ...

I've been working on a silly gadget using the cut & paste bits of other circuits technique combined with a large does of ignorance :-).

The gadget details can be found here:

formatting link

With my breadboarded circuit, I have been observing a phenomena that my tiny little brain can't figure out.

When I expose the cadmium sulphide photoresistor to light, my relay triggers once (briefly) as expected and desired.

However, fairly often (not all the time), when I put the lens cap back over the resistor and block off the light, the circuit will trigger the relay again.

The only theory I have (with no access to a scope that can record whats happening at different points and show me) is that the photoresistor resistance actually bounces around some as it gets dark rather than simply smoothly going back up. Could that be it? Or has the ignorance part of my design process merely led me to do something silly along the way?

Any ideas for improvements to my design or explainations of this puzzling behavior gratefully accepted. (I doubt it is gravity waves though :-).

(Perhaps chuck the whole thing and detect incoming calls via the headset interface is a better plan - but I'd have to find out about interfacing to cell phone headsets for that to work...).

Reply to
Tom Horsley

i just looked at the circuit you have there. from what i can see, puting a small cap from the output of the OP-amp to the

- input could increase the response time thus absorbing the quick pulses you may be getting.

Reply to
Jamie

Thanks! That did indeed do the trick. In another Doh! moment, I notice that the web page I point to with info on comparators even describes the reasons I'd want to add hysteresis, but I never made the connection :-).

the comparator page I mention is at:

formatting link

--

email: snipped-for-privacy@worldnet.att.net icbm: Delray Beach, FL | Free Software and Politics

Reply to
Thomas A. Horsley

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.