Optocoupler question based on Fairchild H11G1

[financecatalyst]

Do I need to put a diode across the terminals of this opto based on the attached configuration (I have to switch from the high side!) ? I selected this opto thinking that it comes with a diode already between emitter and collector but have been loosely advised that it is just for EMC purposes, so would like to make sure of this. Please I need some advice and thanks in advance.

8211

[Dave]

No. It is better to place the diode in the same polarity but across the relay to shunt the back emf when the relay is disabled by the opto. If you notice you are just paralleling the internal diode on the opto.

[financecatalyst]

No. It is better to place the diode in the same polarity but across the relay to shunt the back emf when the relay is disabled by the opto. If you notice you are just paralleling the internal diode on the opto.

Hi Dave, thanks for your answer! Yes, I already have a diode across the relay, I forgot to draw it on the diagram.

I am just not sure if the interval diode in the opto can resist a punch from the fly-back voltage/current from the relay or it will give way and render the opto useless after a few switching cycles. I am not experienced but hold basic knowledge about this and could not really find anything in the data sheet to give me some confidence towards removing the parallel diode from the opto.

[Charlie]

The diode you have drawn is not required. The diode across the coil you forgot to draw is the critical one. I'd also recommend you put the opto between the relay and GND rather than between the relay and +24, although it should be O.K. where it is. By the way, if you are only dealing with +24V, why use an opto to drive the relay? A far cheaper transistor or FET would easily do the job.

[AvionicsMaster1]

For what it's worth, and I'm probably wrong but, I don't think your worry about the opto is properly directed. I think the concern should be placed on the PIC. If you don't have the flyback diode on the relay the back EMF will be rectified through the opto into the PIC. The reverse diode in the opto will not dispose of the relay field collapsing only rectify it to the PIC. You stated the diode on the relay is there so IMWO the diode drawn across the opto is unnecessary.

I would appreciate it if when the real engineers tell me I'm wrong that you do it gently.;-)

[richard]

it would be a very poor quality opto that could not provide the isolation necessary to protect a pic from a relay coil spike . as others have said the "flyback protection" diode goes across the coil .

ps the opto's o/p transistor may not survive the kick but isolation should be maintained

[Demon]

Take this for what it's worth:

I'd use the PIC to drive the opto, to drive a MOSFET, then you can drive the relay from there if you like, or just use the MOSFET.

The opto protects the PIC, the MOSFET can take the current (or is it the voltage - ailing memory fragments says it's current that's the key).

(of course you could just drive the mosfet directly from the pic if you wanted, don't those things act like transistors and draw practically nothing from the gate?)

Google proper mosfet circuitry, I'd hazard a total guess a resistor is at the gate .

Robert

EDIT: I'd start reading this stuff:
http://www.microchip.com/forums/m445886.aspx

And this stuff here:
http://bfy.tw/59xe

I really like this one:
http://electronics.stackexchange.com/questions/125488/control-relay-from-bluetooth-on-attiny85