For switching AC with relays, would it be advantageous to switch at 0v?

I figure the life of the relay would be prolonged if I use the phase information to trigger the relay at the optimal time. I’m not really sure if it’s worth the trouble, but it would be a fun challenge.

My strategy to do this is to fully rectify the AC, drop it to logic level with a voltage divider, then input that to the arduino. Then the arduino can know when the AC will be at 0v and it can send the switching signal around 5ms before then to account for the switching time.

Is this commonly practiced? My google searches didn’t turn up anything but idk the exact terminology here.

You would need to switch OFF at the zero crossing of the CURRENT which could vary 60 degrees or more from the voltage crossing in an inductive load. Then, you would have to switch ON at the voltage crossing because there is no current flowing.

Yes. It's pretty common and it generates less electrical noise. You can buy zero-crossing solid state relays and zero-crossing opto-isolator TRIAC drivers.

A TRIAC actually latches-on until current drops to zero and most AC solid state relays are built-around TRIACs so turning-off at the zero crossing is "natural".

Regular AC Light dimmers turn-on at a particular time after the zero-crossing, then off at the next zero crossing.

I don't think you could get a relay to do this. You need timing accurate to microseconds. Relays are mechanical and slow, and take milliseconds to operate.

PaulRB:
I don't think you could get a relay to do this. You need timing accurate to microseconds. Relays are mechanical and slow, and take milliseconds to operate.

That's a good point. I think another issue would be contact bounce. A delay can be compensated for as long as it's consistent, but I'm not sure what I'd do with 5ms of bouncing.

GustavoMcSavy:
My strategy to do this is to fully rectify the AC, drop it to logic level with a voltage divider, then input that to the arduino. Then the arduino can know when the AC will be at 0v and it can send the switching signal around 5ms before then to account for the switching time.

If by AC you mean mains then you need to isolate the mains from the Arduino, you can't just rectify it and connect it to your Arduino through a resistor, well, not unless you want to kill someone or something.

HI,
What you are talking about is called "zero crossing switching".
As has been mentioned, a relay will not be quick or reliable enough to do this and there are SSR that can zero cross switch.

What load are you switching?
What voltage and current?

A Solid State AC relay with zero crossing is what you need.

google

SSR AC zero crossing

Tom... :slight_smile:

TomGeorge:
A Solid State AC relay with zero crossing is what you need.

Or maybe not.

Zero-cross switching a highly inductive load, like a transformer, could be bad.
Leo..

Interesting Leo, thanks.

You may want to do a little study of Relay Specifications- Release Time. The instant current is removed from the coil the magnetic field begins to collapse. When the magnetic force can no longer hold the relay the spring takes over and then there is a small transfer time for the contacts. I have tested some larger relays and got anywhere between 10mSec to 50mSec. Now you figure with 60 or 50 Hz power on the contacts where that is.

If any of this is of a concern then do as suggested, see if a zero crossing SSR will meet the needs but also watch the pitfalls already mentioned.

Ron

Ron_Blain:
You may want to do a little study of Relay Specifications- Release Time. The instant current is removed from the coil the magnetic field begins to collapse. When the magnetic force can no longer hold the relay the spring takes over and then there is a small transfer time for the contacts. I have tested some larger relays and got anywhere between 10mSec to 50mSec. Now you figure with 60 or 50 Hz power on the contacts where that is.
Ron

Not only that but if you have a flyback diode across the coil the release time will be considerably longer than with no diode.

PerryBebbington:
Not only that but if you have a flyback diode across the coil the release time will be considerably longer than with no diode.

Oh yeah, that's right. Did some testing many, many years ago and we measured current decay in the coils. Some interesting stuff when you really get into it.

Ron

Ron_Blain:
Oh yeah, that's right. Did some testing many, many years ago and we measured current decay in the coils. Some interesting stuff when you really get into it.

Ron

Do you know about slugged relays? Relays with a chunk of copper (I think) wrapped around one or other end of the coil? Done to make them slow to operate or release or both.

PerryBebbington:
Do you know about slugged relays? Relays with a chunk of copper (I think) wrapped around one or other end of the coil? Done to make them slow to operate or release or both.

I had heard of them and maybe over the years saw a few, a chunk of copper similar to how it is wrapped on a shaded pole motor. When I was involved with this sort of thing the idea was measuring current decay in the coil. We needed to know where it would release and the time it took. Those numbers had to meet a specification and the time number was included in a software routine as a sort of “fudge factor”. Some were actually breakers of a strange sort like the AQB 250 mostly naval shipboard type stuff. The testing was interesting.

Ron

GustavoMcSavy:
I figure the life of the relay would be prolonged if I use the phase information to trigger the relay at the optimal time. I'm not really sure if it's worth the trouble, but it would be a fun challenge.

You forgot to mention your intended load - as it depends big time what you switch, both the type of load and the current drawn.

Last night I was looking into relays to allow me to switch on/off a 1.5 kW vacuum, i.e. a motor. The devices you need for this are called "contactor", and are way bigger and 10-20 times the price of a relay with the same current rating. Solid state relays are out as I don't want to have to deal with the 10-20W of heat they put out...

wvmarle:
You forgot to mention your intended load - as it depends big time what you switch, both the type of load and the current drawn.

Last night I was looking into relays to allow me to switch on/off a 1.5 kW vacuum, i.e. a motor. The devices you need for this are called "contactor", and are way bigger and 10-20 times the price of a relay with the same current rating. Solid state relays are out as I don't want to have to deal with the 10-20W of heat they put out...

Turned on and off plenty of motors including all sorts of pumps and three phase 480 V flavors. Always used contactors on motors. SSRs had their place and I used them on heaters quite a bit but motors, even large motors, always a contactor which I always viewed as a relay on steroids.

Ron

The main problem is that contactors typically require 12V DC, 24V DC or 220V AC on the coil... so for me that's going to be a small TRIAC to switch the 220V AC coil... no 12V or 24V available.