Many motorized valves use a capacitor charge to open it at switch off, seems a little crude, but most seem to do it this way.
Max.

 

The ACS712 based sensor circuit seem to be all DC output.
It detects AC & DC current.
Max.

 

The ACS712 based sensor circuit seem to be all DC output.
It detects AC & DC current.
Max.

If the ASC712 is powered by 5V, its output will be centered around 2.5V. i.e. If the current drawn by the load I=Imax.cos(wt), the output of the sensor will be V=2.5 + 0.185Imax.cos(wt)

 

Another observation, the ACS712 appears to use a hall effect isolation from the low level circuit.
But due to the fact you are detecting 240v ac current, even though if you wire it in the neutral side of the AC, I would connect the DC low level circuit comm and valve power common to earth ground for safety.
( you are already using the earth ground symbol anyway on the low voltage!)
Max.

 

If the ASC712 is powered by 5V, its output will be centered around 2.5V. i.e. If the current drawn by the load I=Imax.cos(wt), the output of the sensor will be V=2.5 + 0.185Imax.cos(wt)

A more suitable detector for this circuit would have been a current level (on/off) detector rather than a linear one.
Max.

 

A more suitable detector for this circuit would have been a current level (on/off) detector rather than a linear one.
Max.

Is there anyway (through circuitry) to make a linear detector work as a current level detector? Where I am, it's not that evident to find some kinds of sensors.

 

Not sure if they are available to you but I use the Honeywell CSDA series, they are also non-invasive, they are current transformer style with logic output.
You could use a simple comparater such as LM311 etc to convert to hi-lo.
Max.

 

You could use a simple comparater such as LM311 etc to convert to hi-lo.

I was about to suggest the same thing. Adjusting the comparator's input using a pot as a voltage divider should be fairly simple.

DJ, have you built electronic circuits before?

 

I was about to suggest the same thing. Adjusting the comparator's input using a pot as a voltage divider should be fairly simple.

DJ, have you built electronic circuits before?

Yes, I know some basics

 

Ok, then. For this little project, you're going to have to:

• Draw and fabricate (etch) your own PCB
• Buy the following components:

1. 1. An SSR similar to the one shown in post #7
   2. A trimpot of about 10k
   3. A comparator, preferably the LM311 that Max mentioned, but it can be some other chip too, if you run into trouble finding it.
   4. 7805 linear regulator
   5. PCB terminals
   6. Probably a few capacitors and resistors.
   7. I'm assuming that you already have a 24VDC power supply. How many amps is it capable of delivering?

Where you are, can you find the components that I've just mentioned?

 

I am wondering why the SSR isolation? If the LM311 was used to drive a logic Mosfet it could switch the valve direct, no?
The Valve supply polarity would need to be opposite connection to the dwg, i.e. Mosfet sink to common with the valve +ve fed.
Max.

 

I am wondering why the SSR isolation? If the LM311 was used to drive a logic Mosfet it could switch the valve direct, no?
Max.

Sure it could... it's just that I trust an SSR more for switching inductive loads than a mosfet, even if said mosfet is properly protected. Then again the mosfet option would be cheaper... I'm expecting the SSR to cost about $20.00 dlls. If it's more than that, then the mosfet could be a better option.

 

In this case it may not be a problem, there is no repetitive switching (PWM etc), the load is shown as 1amp and if it has the automatic reverse by capacitor as many of them show, it may absorb any BEMF?
The OP does not show his country of origin so parts may be hard to come by.
Max.

 

The OP does not show his country of origin so parts may be hard to come by.

Good point ... if it were me, I'd definitely go for the mosfet option. I was just trying to make things easier for him.
Well, now he knows there are at least two ways to go around this. Let's see what he decides and take it from there.

DJ, can you acquire a mosfet with a rating of at least 30V and 10 Amps where you are? 40V and 15 Amps would be best, IMHO.

 

Good point ... if it were me, I'd definitely go for the mosfet option. I was just trying to make things easier for him.
Well, now he knows there are at least two ways to go around this. Let's see what he decides and take it from there.

DJ, can you acquire a mosfet with a rating of at least 30V and 10 Amps where you are? 40V and 15 Amps would be best, IMHO.

Don't forget the sensor will put out an AC signal centered around 2.5 volts so you will need to rectify and filter it. I think I saw a current of only 100 ma. If it is really that low I think I would amplify the sensor output a little then AC couple it into the comparator to remove any offset.

 

Don't forget the sensor will put out an AC signal centered around 2.5 volts so you will need to rectify and filter it. I think I saw a current of only 100 ma. If it is really that low I think I would amplify the sensor output a little then AC couple it into the comparator to remove any offset.

I still have to read the sensor's datasheet more thoroughly.

 

AFAIK it is an analogue signal according to several applications of the IC.

The module can measure the positive and negative 20 amps, corresponding to the analog output 100mV / A;

AC or DC current can be measured.
Max.

 

I still have to read the sensor's datasheet more thoroughly.

Data sheet see output slope.
Max.

 

Data sheet see output slope.
Max.

Being AC current driving the motor, I'm guessing then that the sensor's output will be a sinusoid with its inflection centered at 2.5V, isn't that right?

So what I would do is this:

• The OP stated that he was expecting an 18mV output from the sensor, with the motor running unloaded. So the sensor's output should be amplified by a factor of 55, in order to obtain a a sinusoid of 2V, peak to peak, more or less.
  
• Subtract the 2.5V offset so that the sinusoid's inflection lies at 0V. We're now left with a ±1V sinusoid.
  
• To accomplish that, a negative voltage source must be used. See this thread for that purpose.
• Run that signal through a full wave bridge rectifier.
• Make that rectified signal pass through an RC filter, so as to obtain a smooth output voltage.
• Connect that output voltage to one input of a comparator, and the other input to a voltage divider trimpot.
• The comparator's output will either control the SSR or the mosfet driver.

Is that more or less right, or am I too far off the mark? Or perhaps I've just complicated things, and there's a far simpler way to accomplish what he wants?

 

Being AC current driving the motor, I'm guessing then that the sensor's output will be a sinusoid with its inflection centered at 2.5V, isn't that right?

So what I would do is this:

• The OP stated that he was expecting an 18mV output from the sensor, with the motor running unloaded. So the sensor's output should be amplified by a factor of 55, in order to obtain a a sinusoid of 2V, peak to peak, more or less.
  
• Subtract the 2.5V offset so that the sinusoid's inflection lies at 0V. We're now left with a ±1V sinusoid.
  
• To accomplish that, a negative voltage source must be used. See this thread for that purpose.
• Run that signal through a full wave bridge rectifier.
• Make that rectified signal pass through an RC filter, so as to obtain a smooth output voltage.
• Connect that output voltage to one input of a comparator, and the other input to a voltage divider trimpot.
• The comparator's output will either control the SSR or the mosfet driver.

Is that more or less right, or am I too far off the mark? Or perhaps I've just complicated things, and there's a far simpler way to accomplish what he wants?

I kind of like everything except needing 2 supplies. My bad.
So here is what I'm thinking: We could leave it biased at 2.5 volts then 1/2 wave rectify it to only use the positive peaks. We could set the crossing point high enough to allow for some offset.
The other possibility is to only amplify it a little, ac couple it so the negative peak doesn't go below minus 0.3 volts so the comparator is okay.
I'm a little partial to the first one. I think cost and complexity is about the same.
Whatca think?