Which optocoupler to use for 350V DC 100mA load?

I wish to control a pulse controleld DC valve from my arduino digital output. Max pulse voltage on valve is 350V and 100 mA(it requires 10 pulses to swtich .

What optocoupler should i use?

I tried MOC3021(triac) but after it discovered that it needs AC load on emitter side to switch back off, after it is turned on :P

Opto couplers may not be the best choice as they are not normally rated for this voltage or current.

Describe exactly what you are doing and we may be able to make better suggestions.


Look for a DC solid state relay. [u]Here's one I found[/u], but the specs are a little "too close for comfort" (it's rated at exactly 350V).

Most AC solid state relays use TRIACs, so you'll have the same issue as you had with the TRIAC opto-coupler.

...it requires 10 pulses to switch.

If there are any speed requirements, check the speed of the solid state relay.

If you don't need electrical isolation, you should be able to find a MOSFET that can handle more than 350V. Or, you can use an opto-isolator and a MOSFET to essentially build your own solid state relay.

DVDdoug: Or, you can use an opto-isolator and a MOSFET to essentially build your own solid state relay.

Which begs the question - how do you find an optocoupler that is rated to that voltage, in order to drive the MOSFET?

Hmmm? :roll_eyes:

i recalculate the exact voltage and then also measure it on the actualy load, that wil be on the load side and its (since its halfwave signal)around 172V(max, 130V effectiv) and 30mA. i desided to use this SFH619A optocoupler from VISHAY, taht is rated for max 300V DC voltage and 120mA, so also have lots of roo mfor any spikes if thy occur.

Can you provide us with a link to the valve you want to control?

Your requirements have suddenly changed from 350V/100mA to 172V(max, 130V effective) and 30mA.

Why does it need 10 pulses to switch? - Do you mean a minimum of 10 cycles of a half wave rectified AC mains waveform?

The valve itself(sanhua kmv432) is designed for 230V AC/50Hz. But to switch it from state to another you need minimum of 4 half wave pulses(10 pulses recomended) in the right direction. So the voltage on the valve itself is cut approximatly in half, becouse i cut off half of ac signal via 1N4007 diode.

I plugged the valve into 230V 50Hz source, via the diode, and measure the voltage drop on the valve and then changed the direction of diode and measure again. In both circumcates the voltage was 168-172V and current was 21-23mA. This also goes hand in hand well whit what i calculated and expected to see before measurement. I also confirmed that valve actualy switches state when i changed the direction of current flow.

So now that i know that this system works, i just need to wait for my optocoupler darlington(SFH619A) to arrive and use 2 of them to choose what direction the current will flow.

There is circuit diagram of it in attachement. Im turning on/off optocouplers via atmega328P, and the V1 and V2 are terminal to plug in valve in serial connection whit 230V 50Hz power supply.

Brez naslova.png