I'd like to break the current of a relay coil in response to a comparator going high.

I have an idea on how to do this, but it's not very elegant and I wanted to see if there were any better ideas.

My idea is to use the comparator output to trigger a bistable 555 which then disables a FET in series with the coil.

I guess what i'm looking for is perhaps some semiconductor like an SCR that changes state and stays stuck that way. (Once the coil is tripped the comparator will go low again). The SCR however gets stuck on. I'd like something that gets stuck off.

 

Something latches the relay initially (manual button?) Now the comparator puts out a momentary pulse that unlatches the relay until the next time it is manually latched again?

 

You just need a relay with multiple contacts.
Use one set of normally open (NO) contacts in series with the relay coil.
Then use a push button to manually energize the relay.

 

What sets the relay on initially?
Max.

 

Hi,

If possible you should use a reverse connected diode for the coil too in order to help reduce contact arc over when the relay opens.

 

Maybe something like this?

 

#12 - I did think of using an SCR to pull the coil current like you've drawn. The disadvantage though is that you're shunting the full coil current to ground, and that can be significant for a larger coil. It would be preferable to have a zero current latching shutoff.

 

Max - good question. Imagine the user applies a power supply to the relay coil to close the relay. I have access to the coil terminals and can interrupt/break that current (with the right circuit!)

 

Meh. Just an idea. That's why there are several of us here. Lots of ideas.

You could have an SCR latch a driver transistor off to use less current, but SCR's have a minimum hold current. If you want zero, it can be done, but not with an SCR.

 

How about using a latching relay?

 

How about using a latching relay?

That's not an option, the relay is already spec'd and it's a non-latcher.

 

I have used a similar method in the past with LM311 , if the output of the the comparator is set to arm a 2n7000 fet that has a N.O. momentary P.B. in series with the drain and the relay coil, a N.O. contact on the relay is wired in parallel with the PB.
If the comparator output is set (armed) the PB will set the relay on and the relay contact will keep it retained until the comparator changes state, the circuit will remain until the cycle is repeated by the PB.
Max.

 

the relay is already spec'd

Does it have spare contacts that could be used for latching?

 

That's not an option, the relay is already spec'd and it's a non-latcher.

Not THAT relay, I mean use a latching relay instead of the triac or whatever. It will keep its setting after power loss and will not require any current throughout the whole time, except for the brief moment when you wish to disable the main relay.

 

I'd like to break the current of a relay coil in response to a comparator going high.

I have an idea on how to do this, but it's not very elegant and I wanted to see if there were any better ideas.

My idea is to use the comparator output to trigger a bistable 555 which then disables a FET in series with the coil.

I guess what i'm looking for is perhaps some semiconductor like an SCR that changes state and stays stuck that way. (Once the coil is tripped the comparator will go low again). The SCR however gets stuck on. I'd like something that gets stuck off.

The easiest way to commutate an SCR when running on DC; is another SCR - the second SCR has a pull-up resistor from anode to Vcc, the 2 anodes are connected together via a capacitor. When your control SCR is latched, its end of the capacitor is pulled down to 0V, the other SCR is off so the pull-up resistor charges its end of the capacitor to Vcc. When the added SCR latches, it pulls the positive end of the capacitor down to 0V - the negative end of the cap then tries to go below the 0V rail. This takes the SCR current down to zero (and then some) so it unlatches.

 

Not THAT relay, I mean use a latching relay instead of the triac or whatever. It will keep its setting after power loss and will not require any current throughout the whole time, except for the brief moment when you wish to disable the main relay.

Oh I see. That's a good idea. I'd like to avoid another relay however, and if possible, go all solid state on the control.

 

The easiest way to commutate an SCR when running on DC; is another SCR - the second SCR has a pull-up resistor from anode to Vcc, the 2 anodes are connected together via a capacitor. When your control SCR is latched, its end of the capacitor is pulled down to 0V, the other SCR is off so the pull-up resistor charges its end of the capacitor to Vcc. When the added SCR latches, it pulls the positive end of the capacitor down to 0V - the negative end of the cap then tries to go below the 0V rail. This takes the SCR current down to zero (and then some) so it unlatches.

I'm having a little trouble visualizing this - what drives each gate?

 

are you describing this? http://www.allaboutcircuits.com/vol_3/chpt_7/5.html#03215.png

 

I'm having a little trouble visualizing this - what drives each gate?

Given the circumstances - perhaps the comparator could drive a C/R differentiator to produce a narrow spike to fire the gate.

Or possibly a long period astable multivibrator to periodically reset the circuit - you could use the differentiator or have a very uneven mark space ratio on the multivibrator to give narrow reset pulses.

 

The circuit in #12 is only four components including the comparator and relay.
Max.