i need opto-isolator help please

[dacflyer]

i have a strange opto-isolator circuit i need help with.
its for a special continuity tester i am making.

when the opto isolator is on, i need a circuit that will de-energize a relay.
i am having a brain fart, i cannot remember how to create a transistor off circuit

so when the led goes out, the relay will energize.

my power source is 12v. for the relays. i will be making 4 of these circuits.

the led is powered by 18v. resistored of course.

[russlk]

What you need is a normally closed opto relay such as Omron G3VM353A; it is open when energized. Mouser #653-G3VM-353A.

[MrAl]

Hi,


Well, you can use a transistor to drive the relay, then have the output of the opto 'steal' the base current from the transistor thus turning it off and thus turning off the relay. The transistor is biased 'on' with a resistor. The opto shunts the resistor current to ground when it is 'on'.
Not too difficult, but if the relay needs more current you may have to use two transistors to get more gain so the opto doesnt have to do much.
Dont forget the catch diode for the relay.

[dacflyer]

i do not quite understand, i need a visualization of some sort..

[sghioto]

Here's what MrAl was explaining in schematic form

Steve G.

[dacflyer]

thanks for the clarification for the life of me, i have had the worst brainfarts this week..i blame it on this flu and the meds..
thanks again...

[MrAl]

Hi,

Thanks for drawing that Steve

dac:
If you happen to need a little more current for the relay you can use a PNP output stage driven by the NPN collector plus small resistor and tie the PNP emitter to V+ and take the output from the PNP collector to ground for the relay coil.
What happens is that there is an interplay of gains here, where the opto sometimes can not sink enough current to turn off the NPN. That simply means increase the base resistor of the NPN, but then that leads to less max collector current (due to the gain of the NPN) so another stage may have to be added. This of course depends on how much the opto can sink and how much current the relay requires for proper turn on.
Also, there is an aging factor associated with most opto couplers which means you should design the current to be half of the maximum obtainable, or even less. For example if the opto can handle 10ma output with 5ma input (LED) then shoot for only 5ma output with 5ma input. That's means longer life for the circuit. If this is not done then the aging of the opto means that in the future it may only be able to put out 5ma with 5ma input and that means the circuit no longer works if the design expects to get 10ma output with 5ma input. I think this is clear.

[dacflyer]

the relay is a small micro relay, 600Ohm coil @ 12v.
the transistor drives it just fine
i have a 10K resistor on my base.
and i have a 2K driving the opto @ 18V, everything seems to be working just fine.
i need to make 4 of these circuits

[MrAl]

Hi dac,

Oh ok sounds good. Just one more small reliability note.

To make sure the opto can always pull down the base of the transistor low enough to turn it off, sometimes a diode is inserted between the transistor emitter and ground. This means that now the opto does not have to pull the base down to 0.4v but raises this required level to something like 1 full volt DC. That makes sure that it always works over any temperature range.
Your circuit may not need this, but if you are going to build four of them you want to make sure they always work with no problem so you dont have to go back and modify several circuits when something goes wrong. The diode cost is about 5 cents. It also causes a slight drop in voltage getting to the relay coil which usually doesnt matter.

[dacflyer]

ok, thanks for the tip..

i am building this as a long distance continuity tester ( 100 - 200 ft )
a tool to help make my job easier and faster when i am testing out ped buttons at cross walks.
in the cabinet there is approx 18v across the button terminals, when the switch is closed it places a call to the controller. so the 18v powers the opto-led, when the button is pressed it causes the led to go out, and the relay kicks in to sound a loud alarm ( radio shack 2 tone piezo indoor siren #49-490A ) so it is loud enough that i can clearly hear it across noisy traffic and know that each button is working fine,,
and not having to wait and see if the ped changes
(lost work time waiting to see results of pushing the button)
with my tester i get instant results.