Using an opto-isolator to control an AC-coil relay

[Peter S. May]

I've mentioned this project previously, but I wanted to run a particular
schematic by you all to see if there's anything severely wrong with it.

I want to actuate a 120VAC coil relay from HC logic. I'm not using a
solid-state relay because the coil relay is far cheaper for the current
rating, and I'm not using a 5VDC coil because, being connected to mains
as it is, there's no reason the device should be drawing power from the
logic circuit's supply.

This is to be accomplished using an opto-isolator, and a Fairchild
MOC3022 400V triac opto-isolator happens to be handy.

I've never used a triac before, so I'm leery of my kneejerk idea for a
schematic:

MOC3022
.............................
Vcc <---/\/\--.--+ +---.--------------------+
. | | . |
. V emitter ----- . 120V |
. - v ^ . AC ~
. | phototriac ----- . |
. | | . relay coil |
/ON signal <-.--+ +---.-----@@@@@@@@@@-----+
.............................

Does this work, or am I entirely off-base?

Thanks
PSM

 

[Tom Bruhns]

I've mentioned this project previously, but I wanted to run a particular
schematic by you all to see if there's anything severely wrong with it.

I want to actuate a 120VAC coil relay from HC logic. I'm not using a
solid-state relay because the coil relay is far cheaper for the current
rating, and I'm not using a 5VDC coil because, being connected to mains
as it is, there's no reason the device should be drawing power from the
logic circuit's supply.

This is to be accomplished using an opto-isolator, and a Fairchild
MOC3022 400V triac opto-isolator happens to be handy.

I've never used a triac before, so I'm leery of my kneejerk idea for a
schematic:

MOC3022
.............................
Vcc <---/\/\--.--+ +---.--------------------+
. | | . |
. V emitter ----- . 120V |
. - v ^ . AC ~
. | phototriac ----- . |
. | | . relay coil |
/ON signal <-.--+ +---.-----@@@@@@@@@@-----+
.............................

Does this work, or am I entirely off-base?

Thanks
PSM

We've been using a somewhat similar but a bit more complicated circuit
to control power contactors (in a charity project we've been working
on for several years, involving load control for a small hydro power
system, but that's another story). There are problems with your
simple circuit. Note that the data sheet for the MOC3022 tells you,
I'm pretty sure, that you should be using it to trigger bigger triacs;
it's not designed to drive other loads directly. If the relay coil is
low enough current you MIGHT be able to get by without the other
external triac. We are after reliability and never wanted to push it,
so haven't tried it--little TO-92 (or even TO-220) triacs are cheap
enough it just wasn't worth worrying about. The power contactor coils
are pushing the current rating of the MOC3022 in any event. But what
you probably will NOT get by with is omitting a snubber network, to
keep the high dv/dt that results when the triac turns off with an
inductive load from re-triggering the triac. In our circuit, we have
a series R-C in parallel with the triac (the external one), calculated
for the worst-case expected inductance. You should be able to find ap
notes on the web that will tell you how to design snubbers for triacs
with inductive loads.

You can assume the relay is just an inductance; worst-case would be to
assume it's pure inductance of a value to give you the rated coil
current at the applied voltage and frequency. There's one little
problem that could possibly bite you: when the armature actually
releases (some milliseconds after you turn off the coil drive), the
inductance changes rather suddenly since the magnetic circuit changes,
and if there is residual magnetism, that will cause a voltage spike.
The triac should have been off for long enough that it's much less dv/
dt sensitive by that point, but it's something to be aware of.

If you're doing this as a one-off project, you may find that surplus
solid state relays from a place like www.mpja.com are cheap enough and
save enough hassle that they are worth considering. And if you are
switching modest-current 120VAC loads, why not just use the MOC3022 to
drive something like a 3A or 6A TO-220 triac? Of course, either the
SSR or the triac will have quite a bit of power dissipation because of
the voltage drop, and maybe you want to avoid that.

Cheers,
Tom

 

[Ken S. Tucker]

I've mentioned this project previously, but I wanted to run a particular
schematic by you all to see if there's anything severely wrong with it.

I want to actuate a 120VAC coil relay from HC logic. I'm not using a
solid-state relay because the coil relay is far cheaper for the current
rating, and I'm not using a 5VDC coil because, being connected to mains
as it is, there's no reason the device should be drawing power from the
logic circuit's supply.

This is to be accomplished using an opto-isolator, and a Fairchild
MOC3022 400V triac opto-isolator happens to be handy.

I've never used a triac before, so I'm leery of my kneejerk idea for a
schematic:

MOC3022
.............................
Vcc <---/\/\--.--+ +---.--------------------+
. | | . |
. V emitter ----- . 120V |
. - v ^ . AC ~
. | phototriac ----- . |
. | | . relay coil |
/ON signal <-.--+ +---.-----@@@@@@@@@@-----+
.............................

Does this work, or am I entirely off-base?

Thanks
PSM

What's the current rating to keep the relay on?
Ken

 

[Jamie]

I've mentioned this project previously, but I wanted to run a particular
schematic by you all to see if there's anything severely wrong with it.

I want to actuate a 120VAC coil relay from HC logic. I'm not using a
solid-state relay because the coil relay is far cheaper for the current
rating, and I'm not using a 5VDC coil because, being connected to mains
as it is, there's no reason the device should be drawing power from the
logic circuit's supply.

This is to be accomplished using an opto-isolator, and a Fairchild
MOC3022 400V triac opto-isolator happens to be handy.

I've never used a triac before, so I'm leery of my kneejerk idea for a
schematic:

MOC3022
.............................
Vcc <---/\/\--.--+ +---.--------------------+
. | | . |
. V emitter ----- . 120V |
. - v ^ . AC ~
. | phototriac ----- . |
. | | . relay coil |
/ON signal <-.--+ +---.-----@@@@@@@@@@-----+
.............................

Does this work, or am I entirely off-base?

Thanks
PSM

yes, that will work how ever, personnally, I would use a snubber on the
load to help reduce HV flyback issues from the coil if you were for some
reason to lose the 120 in mid stream.
We have heat controllers (eurotherm) with the Triac output option that
actually ties a MOV across the 2 terminals.
My self, I think BiTVS diodes are better for this but that's just my
opinion.

 

[ehsjr]

I've mentioned this project previously, but I wanted to run a particular
schematic by you all to see if there's anything severely wrong with it.

I want to actuate a 120VAC coil relay from HC logic. I'm not using a
solid-state relay because the coil relay is far cheaper for the current
rating, and I'm not using a 5VDC coil because, being connected to mains
as it is, there's no reason the device should be drawing power from the
logic circuit's supply.

This is to be accomplished using an opto-isolator, and a Fairchild
MOC3022 400V triac opto-isolator happens to be handy.

I've never used a triac before, so I'm leery of my kneejerk idea for a
schematic:

MOC3022
.............................
Vcc <---/\/\--.--+ +---.--------------------+
. | | . |
. V emitter ----- . 120V |
. - v ^ . AC ~
. | phototriac ----- . |
. | | . relay coil |
/ON signal <-.--+ +---.-----@@@@@@@@@@-----+
.............................

Does this work, or am I entirely off-base?

Thanks
PSM

That's not the right way (but you might sneak by).
The MOC3022 output should drive the gate of a triac - the
triac drives the relay.

---------
Signal ---------| MOC3022 |---/\/\/\----+--- HOT
| | |
Signal_Ground---| |--->(gate)[Triac]
--------- |
+--[Load]--- Neutral

Above is the conceptual diagram - see the datasheet
for a schematic.

Ed