detecting on off state of any appliance using a CT sensor

[dr pepper]

Ok so you know the circuit works, step 1.
Looks like noise is an issue, most likely conducted into the circuit via the supply, if the noise were magnetic it'd be there with a battery supply.
If your measuring current draw from a smps then you might have some rf noise issues, rf can produce large voltages between the pri and sec of a transformer, a lot of smps's have a cap that goes from the dc bus on the primary to the output for this reason.

 

[ronsimpson]

Are you certain you have the right wires?
Looks like three wires from the CT.
The right two will have 2 to 10 ohms resistance. (don't really know the resistance but it is mostly the wire resistance in the transformer)
The third wire might be a shield that should be connected to the computer's ground.

 

[MikeMl]

chrischris

I think the confusion in this set of posts is does the intrinsic CT have a built-in rectifier so that it produces a DC voltage (more or less) proportional to the AC current, or if it lacks the rectifier, and just makes an AC signal of a couple of Volts across the burden resistor?

If the former, then connecting to an Arduino ADC pin is trivial, provided that the DC signal stays between 0V and 5V.

If the latter, then there needs to be an interface (consisting of bias supply, a diode, possibly an opamp, resistors, capacitors) to convert the AC voltage across the burden resistor to a DC signal so that a single ADC reading gives a value proportional to the magnitude of the AC current.

Alternatively, the AC signal has to be biased to 2.5V (to put it at the center of the ADC's range), and then take lots of ADC samples for each period of the AC waveform and use Arduino software to convert the series of readings into a single value proportional to the magnitude of the AC current in the primary.

What are we trying to do here?

 

[ronsimpson]

confusion in this set of posts

From two different posts:

Here the "0 current point" should be at mid point of the ADC.
The voltage should be + and - from center. (AC)
There is a claim that the resistor is in the CT.

up to 16V AC

This can not happen. Something is wrong.
The burden resistor might be missing.
I have used these CTs. I do not remember much. lol But they work!

I have CTs like this:

Three wires. One wire is "shield". If the shield wire is not connected to ground of the Micro, then some noise will get in.
BUT
If the shield wire is used for signal, by mistake, then; the readings will be very strange. If the power supply, powering the micro, is "switching" then there will be great noise. If the input is high impedance, volt meter or ADC, the voltage will be large. Noise from the power supply and noise from the power line will be on the ADC input. (noise should be switched to ground)

From my bad memory, I think full current should be about 0.3 volt AC. Not 5V.

 

[MikeMl]

...
I have CTs like this:
View attachment 106023
Three wires. One wire is "shield". If the shield wire is not connected to ground of the Micro, then some noise will get in.
..

Hence my original post #4 regarding tip, ring, and sleeve which chrischris never bothered to respond to.

 

[chrischris]

Honeywell data show the series is available up to 1275 AMPS!!

What are we trying to do here?

Thanks for reply, the output is AC, 300mV hence the circuit

This can not happen. Something is wrong.

16V AC is not from the CT sensor, from CT sensor its 30mV to 300mV AC (for ON and OFF conditions respectively). 16V AC is what noise is on my circuit. So in above circuit, when I connect the pin labeled "to Arduino GND" to my circuit GND , connect pin labeled "to Arduino 5V" to my +5V rail and measure the pin labelled "to Arduino analog" with voltmeter(AC), the voltage keeps on jumping around 14 to 16 V AC.
If I measure the CT sensor output itsself, on its own without connecting to any circuit, it goes from 30mV to 300mV as stated above.

Hence my original post #4 regarding tip, ring, and sleeve which chrischris never bothered to respond to.

My apologies, I was overwhelmed by options and trying to make sense of all the responses. Honestly I don't know what a tip ring is and couldn't understand what do you mean by it, circuit looks simpler I will google a bit about it and experiment.

Thanks all

 

[ronsimpson]

We are trying to find out what you did with: TIP, RING, and SLEEVE.
These three wires are connected to what? I think you are only using 2 wires of the 3.

 

[ronsimpson]

 

[MikeMl]

We are trying to find out what you did with: TIP, RING, and SLEEVE....

Ron beat me to it. I found, and was going to post the exact same figure...

 

[ronsimpson]

Ron beat me to it. I found, and was going to post the exact same figure...

Ha Ha My google is faster than yours.

 

[ronsimpson]

Big question:
What is the resistance of these three points.
My guess that two will measure about 20meg ohms and one will measure about 10 ohms.

 

[Daniel Wood]

Chrischris does mention in #1 that the SCT013-30 is being used. After having a look at the datasheet, it looks like there is no shielding whatsoever.

 

[ronsimpson]

Thank you Daniel. I looked long and hard for that picture. Could not find it.

I do not know why the volt meter measures 14 to 16 volts.
The CT works when in circuit so use it.

 

[chrischris]

Sorry for late reply, I moved houses so was very busy.

We are trying to find out what you did with: TIP, RING, and SLEEVE.
These three wires are connected to what? I think you are only using 2 wires of the 3.

Ok now I understand what you mean (feeling stupid!). There is a 39 ohm resistance between TIP and Sleeve, Ring is open circuit with respect to TIP and Sleeve.

I tried grounding Sleeve and now, the 14V AC noise is gone , but I don't see any reading on TIP or RING. Actually ring stays at about 4V AC regardless current detection and TIP stays at 12mV AC.

Surprising thing is that datasheet suggests 62 Ohm built in load resistance but I don't get 62 Ohm anywhere ?

Any ideas please.

Thanks everyone !!

 

[Les Jones]

My guess is that you are just measuring the resistance across the secondary of the current transformer. This could be JUST the winding resistance OR it could be the winding resistance in parallel with the 62 ohm burden resistor. You need to test it properly by passing a known current trough the primary and measuring the AC voltage across the secondary with a 62 ohm connected across the secondary. If the voltage reading is lower than that expected then measure it again without the added 62 ohm resistor.

Les.

 

[chrischris]

I know it has been a while, but I think I should post the outcome here so it may help others. Atfer trying many things as mentioned in previous thread. This is the conclusion :
The CT sensor was useless as there is no shielding on the wire and the current I wanted to detect is only 95mA AC. There was no way this sensor or any other no invasive sensor (unless I spend $$$).
So I changed the design and decided to cut the wire I wanted to measure current through and put a 67OHM resistor in line with a diode connected parallel to the resistor .
This gives me 2.5V DC when the appliance is on. Fed that to an opto coupler and job done.

Thanks everyone for the help.

 

[dr pepper]

Maybe more a case of a suitable current trans isnt available, less than 100ma is very low for such a device.
I have seen this done with series'd diodes, put in line they drop enough to light the led section of an opto coupler, still invasive though.

 

[MrAl]

Hi,

Just to note, Hall Effect current sensors are invasive physically but they are isolated electrically.