Switching a 24VAC solenoid, TRIAC, Relay or SSR?

Thread Starter

ilium007

Joined Aug 24, 2013
139
I am putting together some automation for a garden setting. I need to switch 24VAC solenoids for garden sprinklers. I started looking at standard relays but then saw an open source project using TRIACS. Some more research points toward SSR's. When looking at TRIACS I got incredibly confused by having to work out snubber circuits to combat the dv/dt voltage rise on solenoid turn off. I found the STM ACS120 snubberless, zero cross TRIAC (https://au.rs-online.com/web/p/triacs/6871079/) that I would drive with an opto-isolator. When I looked at the SSR's I found the Panasonic SMD Zero Cross AQH3213A (https://au.rs-online.com/web/p/solid-state-relays/6996512/) SSR. Both seem capable of the sprinkler solenoid current (<1A) but not sure which way to go. The SSR has the opto-isolation built in but not sure if I need a snubber circuit with the SSR. If I do can I just use a 100V bi-directional TVS diode across the load or do I need a proper RC snubber circuit.

I've spent over a week researching this and I'm just not getting anywhere. I haven't done much with AC before and everything I have read about garden solenoids says that AC is the only way to do it due to solenoid core magnetisation if you run DC solenoids.
 

sghioto

Joined Dec 31, 2017
5,380
Either one should work. As long as both are zero crossing the snubber isn't necessary.
The SSR is only rated at 1.2 amps. I prefer a device that will handle at least twice the current draw of the solenoids.
 

Ian0

Joined Aug 7, 2020
9,677
A SSR is a combination of an opto-triac and a triac built into one package.
Hypothetically, you still need a snubber even if the triac is zero-crossing-switched, because the triac will switch OFF when the current reaches zero, but because the solenoid is inductive that will not correspond to voltage zero crossing, and there will be a substantial voltage across the triac at the point it switches off.
However, you'd be hard pressed to get a triac rated less than 400V and so the voltage at switch-off on a 24V solenoid will be in the order of 24V, and that's not going to bother a 400V-rated triac.
A triac or SSR will drop about 1Volt, so will dissipate 1 Watt for each amp of current so you should check heat dissipation.
Switching a DC solenoid with a MOSFET will incur less power loss.
 

Thread Starter

ilium007

Joined Aug 24, 2013
139
A SSR is a combination of an opto-triac and a triac built into one package.
Hypothetically, you still need a snubber even if the triac is zero-crossing-switched, because the triac will switch OFF when the current reaches zero, but because the solenoid is inductive that will not correspond to voltage zero crossing, and there will be a substantial voltage across the triac at the point it switches off.
However, you'd be hard pressed to get a triac rated less than 400V and so the voltage at switch-off on a 24V solenoid will be in the order of 24V, and that's not going to bother a 400V-rated triac.
A triac or SSR will drop about 1Volt, so will dissipate 1 Watt for each amp of current so you should check heat dissipation.
Switching a DC solenoid with a MOSFET will incur less power loss.
I need to find a guide on sizing a snubber circuit. I didn't think about heat factors - thanks.
 

Marley

Joined Apr 4, 2016
502
If its only a small solenoid then it you could simply use relay contacts. Beware of back-emf though. Always good practice to fit a MOV varistor across the solenoid to catch these spikes. Something like this. Select for the supply voltage (24VAC in your case).
 

Thread Starter

ilium007

Joined Aug 24, 2013
139
If its only a small solenoid then it you could simply use relay contacts. Beware of back-emf though. Always good practice to fit a MOV varistor across the solenoid to catch these spikes. Something like this. Select for the supply voltage (24VAC in your case).
I was interested in learning more about TRIACS by implementing one to switch these loads.
 

Ian0

Joined Aug 7, 2020
9,677
Triacs are easy!
For a 24V system you don’t even need isolation.
Use a sensitive-gate triac like BTA08-600TW
Half-wave rectify the 24V, smooth and regulate to -5V. This is the supply your logic runs off.
Connect triac MT1 to your 0V. To switch the triac on, switch the gate to -5V through 470Ω which limits the gate current to 10mA. You can do this with a transistor or even a 74HC logic gate.
Connect the load between MT2 and the other terminal of your 24V supply i.e, not the one you just called 0V.
Done.
 

Ian0

Joined Aug 7, 2020
9,677
The BTA08-600TW is a 3-quadrant triac. i.e. It won’t switch with the gate positive on the negative-going half-cycle. All triacs switch better with a negative gate voltage, even the ones which will switch in the 4th quadrant require more gate current and switch more slowly.
 

BobaMosfet

Joined Jul 1, 2009
2,110
I am putting together some automation for a garden setting. I need to switch 24VAC solenoids for garden sprinklers. I started looking at standard relays but then saw an open source project using TRIACS. Some more research points toward SSR's. When looking at TRIACS I got incredibly confused by having to work out snubber circuits to combat the dv/dt voltage rise on solenoid turn off. I found the STM ACS120 snubberless, zero cross TRIAC (https://au.rs-online.com/web/p/triacs/6871079/) that I would drive with an opto-isolator. When I looked at the SSR's I found the Panasonic SMD Zero Cross AQH3213A (https://au.rs-online.com/web/p/solid-state-relays/6996512/) SSR. Both seem capable of the sprinkler solenoid current (<1A) but not sure which way to go. The SSR has the opto-isolation built in but not sure if I need a snubber circuit with the SSR. If I do can I just use a 100V bi-directional TVS diode across the load or do I need a proper RC snubber circuit.

I've spent over a week researching this and I'm just not getting anywhere. I haven't done much with AC before and everything I have read about garden solenoids says that AC is the only way to do it due to solenoid core magnetisation if you run DC solenoids.
SCRs "latch" on when activated and stay on unless you remove power from all 3 leads. This is OK in an A/C environment because you cross zero repetitively, removing power from all connections on the SCR. There is nothing difficult about a snubber circuit. It is simply a diode in parallel (backwards) with the heavy magnet object you're driving current through. A 'heavy magnet object' can be a motor, a solenoid, an inductor or coil- anything that requires a magnet to be charged and discharged to control something.
 

Ian0

Joined Aug 7, 2020
9,677
There is nothing difficult about a snubber circuit. It is simply a diode in parallel (backwards) with the heavy magnet object you're driving current through. A 'heavy magnet object' can be a motor, a solenoid, an inductor or coil- anything that requires a magnet to be charged and discharged to control something.
We’re dealing with an AC supply here, switched with triacs!
A snubber is a resistor and capacitor in series, connected either across the triac, or across the load: what you described is a flywheel or fly back diode.
However, as virtually every triac you are likely to find is rated at >400V, the chances of a voltage spike big enough to damage it from a 24V AC supply are pretty slim.
 

Thread Starter

ilium007

Joined Aug 24, 2013
139
R19 is a current limit resistor, and R20 (which you only need with sensitive gate triac) stops unintentional triggering, but at 24V you don’t need to isolate it.
Thanks - I got caught up on your reference to -5V, I don't understand why its a negative voltage.
 

Ian0

Joined Aug 7, 2020
9,677
Thanks - I got caught up on your reference to -5V, I don't understand why its a negative voltage.
Because the triac is triggered with a negative current.
Depending on the type of triac, it will only trigger with a positive current if the MT2 terminal is positive, but it will trigger with a negative current whether MT2 is positive OR negative. See previous discussion on 3- or 4-quadrant operation.
 
Top