I am trying to finalize the AC voltage and current measurement circuit using a differential amplifier. I have designed two circuits for each. In the high side current measurement circuit 2 and voltage measurement circuit 2, I have connected multiple resistors in series to the inputs of the op-amp. However, in the high side current measurement circuit 1 and voltage measurement circuit 1, I have connected a single resistor to the inputs of the op-amp. I am planning to use 0603 package resistors. As I know that op-amps have very high input impedance, I used a single input resistor in circuit 1. I am confused about which circuit to follow for voltage and current measurement, respectively.

 

AC Voltage will have to be Rectified.

It would be much safer to step-down the Voltage with a small Transformer.

High AC-Voltage on the same PCB with Low-Voltage Circuits is a very dangerous plan.
.
.
.

 

Is this mains voltage AC? What is your DC power supply? You probably need isolation, in which case you need a current transformer or a hall-effect device.
What do you mean by "High-side" in this context? It only makes sense for the high positive voltage of a DC supply.

 

Hi there, It's AC 220V rms mains voltage. The power supply for the opamp is 3.3Vdc, I don't want to use a transformer, current transformer, or hall effect sensor as these are bit costly solutions. High-side sensing means the shunt resistor is connected to the mains and the load comes after the shunt resistor. I tested circuit 2 of voltage and current measurement and it worked. I doubt why I should consider using multiple resistors in series to the inputs of opamp as shown in circuit 2 rather than using a single equivalent resistor as shown in circuit 1?

 

Where does the 3.3V come from? What else does it connect to?

 

Where does the 3.3V come from? What else does it connect to?

3.3V is coming from the SMPS and powering microcontroller, opam

 

3.3V is coming from the SMPS and powering microcontroller, opam

IS the power supply for the microcontroller have any parts that can be touched by the operator?
Does the microcontroller have any other input/output connections?
If the answer to either of those is "yes" then you must isolate the AC inputs - you have no choice.

 

I don't want to use a transformer, current transformer, or hall effect sensor as these are bit costly solutions.

But non-isolation could cost you (or someone else) their life!

 

And if it costs someone else their life, it costs you your freedom (life imprisonment for Gross Negligence Manslaughter)

 

IS the power supply for the microcontroller have any parts that can be touched by the operator?
Does the microcontroller have any other input/output connections?
If the answer to either of those is "yes" then you must isolate the AC inputs - you have no choice.

No part is touched by anyone, apart from the opamp output to the ADC of microcontroller doesn't have any other input/ output connections. I tried low-side current sensing practically and it did work without any mishap. I am still learning, as we know opamp has very high input impedance (in the range of mega ohms) so I believe that there won't be any issue of isolation. Please correct me if I am wrong.

 

No part is touched by anyone, apart from the opamp output to the ADC of microcontroller doesn't have any other input/ output connections. I tried low-side current sensing practically and it did work without any mishap. I am still learning, as we know opamp has very high input impedance (in the range of mega ohms) so I believe that there won't be any issue of isolation. Please correct me if I am wrong.

If it has no external connections, what does it do?

 

No part is touched by anyone, apart from the opamp output to the ADC of microcontroller doesn't have any other input/ output connections. I tried low-side current sensing practically and it did work without any mishap. I am still learning, as we know opamp has very high input impedance (in the range of mega ohms) so I believe that there won't be any issue of isolation. Please correct me if I am wrong.

What happens when the very high input impedance breaks down?
I would design with proper galvanic isolation.

 

No part is touched by anyone, apart from the opamp output to the ADC of microcontroller

So Is not the common for that output connected to the main's neutral?
Is the SMPS output common isolated from the input?

 

If it has no external connections, what does it do?

Microcontroller sends readings wirelessly.

 

What happens when the very high input impedance breaks down?
I would design with proper galvanic isolation.

Hi, Could you please explain the scenarios which could cause the input impedance to break down? And also I have seen a few energy-monitoring smart switches, and energy meters using a similar circuit (They have also used shunt resistors, and differential amplifiers for measurement).

 

So Is not the common for that output connected to the main's neutral?
Is the SMPS output common isolated from the input?

So Is not the common for that output connected to the main's neutral?
Is the SMPS output common isolated from the input?

Yes, SMPS ground is isolated from the main's neutral

 

Microcontroller sends readings wirelessly.

In that case, connect the power supply 0V to mains neutral, and sense current on the neutral side of the mains.
NOTHING may connect to earth.
No connections to the circuit may be touchable.
The circuit must withstand a 4kV test between itself and earth, with no breakdown of any insulation.

Use your original #2 voltage measurement circuit, and split R4 into two resistors of twice the value, one connected to PSU 0V and one to PSU 3.3V.
Use your original #1 current measurement circuit and do the same with its R4.
That will give you a sinewave centred on half-supply, which you should be able to sample and calculate its rms value.

 

In that case, connect the power supply 0V to mains neutral, and sense current on the neutral side of the mains.
NOTHING may connect to earth.
No connections to the circuit may be touchable.
The circuit must withstand a 4kV test between itself and earth, with no breakdown of any insulation.

Use your original #2 voltage measurement circuit, and split R4 into two resistors of twice the value, one connected to PSU 0V and one to PSU 3.3V.
Use your original #1 current measurement circuit and do the same with its R4.
That will give you a sinewave centred on half-supply, which you should be able to sample and calculate its rms value.

Hi, What if I don't connect the power supply 0V to mains neutral and keep the sense current resistor on the live side of the mains (As I am intending to measure current at the high-side)?

 

Hi, What if I don't connect the power supply 0V to mains neutral and keep the sense current resistor on the live side of the mains (As I am intending to measure current at the high-side)?

It might work - it probably won't - it depends on what happens to the common mode voltage.
If you want to sense on the live side, connect PSU 0V to mains live.

 

It might work - it probably won't - it depends on what happens to the common mode voltage.
If you want to sense on the live side, connect PSU 0V to mains live.

When I tried it on the low side sense, It did work. For the high side I will try it without connecting PSU 0V to mains live, If it fails then I will connect it. Apart from the shunt resistor, all the resistors I am planning to use are 0603 package smt. And particularly in the original #1 current measurement circuit As I am going to deal with the mains 220Vac. I am talking about safety, having multiple resistors in series to the input of the opamp or a single resistor to the input of the opamp safe? Also, having input resistors value larger (as gain of opamp can be maintained by varying the feedback resistor value according to Rf/Ri)?