Hi,
I want to drive a 12V 3A solenoid valve from an 5V Arduino pin. I am mostly interested in the power side of things, so I hope this is the right forum.
I have designed a circuit using a 4N35 opto and a TIP120 Darlington NPN transistor.

From the 4N35 datasheet the pins 4 and 5 (CE) has a forward voltage of 1V at 10mA , while the TIP120 Darlington has a BE voltage drop of 2.5V.
So there is a total of 3.5V dropped by the transistors. So 12V-3.5V = 8.5V across the current limiting resistor.
From the TIP120 datasheet, 10mA is well within the max Base current, so then R=V/I
R = 8.5V/10mA = 850 ohms, so lets use 820 ohms

https://www.vishay.com/docs/81181/4n35.pdf
https://www.onsemi.com/pub/Collateral/TIP120-D.PDF

Does this sound right? I tried using Spice simulator, but I couldn't find a Darlington transistor to use to simulate the circuit.

 

If you don't require electrical isolation, meaning the Arduino and solenoid can share the same ground, you don't need an optocoupler.

 

hi pico,
The LTSpice npndarlington model should be in the '(misc)' folder, so F2 , misc, npndarlington.
On the circuit diagram change TIP100 to TIP120.

You can also post 'asc' files to the thread.
E


EDIT:
Added a darlington test asc file.

 

Or a logic level Mosfet.
Max.

 

The 1N4001 is pretty slow. Better to use a UF4001 or similar.

 

If you don't require electrical isolation, meaning the Arduino and solenoid can share the same ground, you don't need an optocoupler.

The opto isolator has 1 or 2 attractions whether you need isolation or not, someone suggested a LL MOSFET - which could also make life easier regardless.

 

The 1N4001 is pretty slow. Better to use a UF4001 or similar.

You don't need speed in this application.
All diodes turn on rapidly, which is the speed of interest here.
Fast diodes turn off faster but that's not needed for a solenoid damper diode.

 

You don't need speed in this application.
All diodes turn on rapidly, which is the speed of interest here.
Fast diodes turn off faster but that's not needed for a solenoid damper diode.

True, very true.

 

True, very true.

Sometimes I put faster diodes in TV vertical flyback energy recovery circuits because the retrace is supposed to be pretty steep - but there probably wasn't all that much wrong with the opiginal standard recovery types.

The big bad number is reverse recovery time - unless you hit the solenoid again very soon, there's plenty of time for carriers to dissipate from the junction. If anything - it might help damp any ringing.

 

The opto isolator has 1 or 2 attractions whether you need isolation or not, someone suggested a LL MOSFET - which could also make life easier regardless.

What would be the advantages of a LL MOSFET over a Darlington in this application?

 

What would be the advantages of a LL MOSFET over a Darlington in this application?

zero input current as long as you don't pulse it fast.

 

What would be the advantages of a LL MOSFET over a Darlington in this application?

Base current over gate voltage.
Max.

 

With the current design, that is using a Darlington, do I calculate the power dissipated by the transistor as the voltage drop across CE multiplied by the current being passed from C-E ?
That's how I did it, 1.7V across CE and passing 3A, that gives 5.1W being dissipated. Except that the Darlington doesn't get hot, so I am assuming I have missed something here...

 

1.7V across CE and passing 3A, that gives 5.1W being dissipated. Except that the Darlington doesn't get hot, so I am assuming I have missed something here...

That's the correct calculation.
So if it's not getting hot, then either Vce or Ice (or both) are less than you calculated.

 

That's the correct calculation.
So if it's not getting hot, then either Vce or Ice (or both) are less than you calculated.

I previously measured Vce to be 1.7V so I'll get an ammeter in line with the Darlington to measure the current.