This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

EVM430-F6736 recommended current transformer or shunt

Other Parts Discussed in Thread: EVM430-F6736

Hi,

Could anyone please explain why CT (or shunt) is preferred to use on EVM430-F6736?  

What properties of CT and shunt do I need to care when I select a suitable one for the EVM? I would appreciate if you could tell me the part number that you use on the EVM just  to have a reference for me.

Thank you,

Serdar

  • A current transformer provides galvanic separation. So you can measure three phases without the risk of blowing your device if you’re connected to the wrong side of phase and neutral. (I’ve seen this going wrong even with 1-phase meters)
    However, a CT still needs a shunt to convert its output current into a voltage that can be measured by the MSP.
    Most CTs are X:1 or X:5 converters. That means you’ll have 1A or even 5A running through the shunt at nominal input current. That means low-resistance shunts and still quite some heat.
    We use CTs with an output of 33 or 66mA at a nominal current of 100/400A. This allows us to use cheap 7.5 or 15Ohm resistors as shunts with below 40mW power dissipation.

    Basically, you need to pick a combination of CT and shunt that produces a suitable output voltage at nominal current. The rest is done by a calibration factor inside the software.

  • Jens-Michael, thank you for the reply.

    I have a few more questions, I hope you can help me with those:

    - The EVM that I refer is an one phase e-meter http://www.ti.com/tool/EVM430-F6736 It has two current channels. Here below is the current interface:

    It is enough to use only one current channel right? Why are there two current channels? Is the purpose to test one channel with a CT and the other one with a shunt resistor?

    If I use a CT, then the protection circuit (left side) before the burden resistor (R9 and R10) can be safely removed, can't it?

    In case you have/used the EVM430-F6736, the design guide says CUR2+ and CUR2- are connected to a CT. Do you know what CT is placed on the EVM?

    I have been reading some application notes about CTs. One says that too high turn ratio (Ns/Np) increases the distributed capacitance and leakage inductance, too low turn ratio may distort or drop the output signal which may causes inaccuracies. Could you tell if the range 1:1000 - 1:2000 has an OK ratio?

    Is the EVM polarity sensitive in the voltage input interface? If it is the case, do you know if it is possible to make it polarity insensitive that the end user can just plug/screw the AC cables to the circuit without caring which line is what?

  • Sometimes both, line and neutral, are measured, to detect any leakage to ground. But this is more often used on 3-phase meters (4 current channels). I don’t know why this design has two.

    I don’t know which CT is used in the EVM.

    Whether you can remove the protection circuit depends on the CT you use. On an overcurrent, a voltage peak might burn through your MSP, before the fuse reacts. A current surge of 270mA will exceed 3.6V. And with more than 430mA, the 1k series resistor won’t help you anymore. So it depends on the CT and its iron core and maximum output power etc.

  • Thank you again for the answers. 

    Lastly, is it possible to make the input connection polarity free?

  • What do you mean with ‘polarity free’?

    It is important to know whether you draw power or inject power. Well, if you just want to know the current (not the power), then you can simply not draw the sign in your output :)

  • Here is a screen shot from the design guide. Line and Neutral are connected to V+ and V- respectively. 

    If the end user wants to connect it from the wall outlet, Line and Neutral might be connected to V- and V+ respectively. As I understood from the user guide, this might damage the hardware, is it correct? If it is the case, is it possible to make the voltage input polarity free that the user can connect Line and Neutral to any terminal of voltage input and the device would perform correctly without getting any harm?

  • When you use shunts to measure the current, then the shunts will be tied to neutral (or line) on one side, and are a few mV above (below) on their other side. This requires that circuit GND is tied to the same baseline. If you switch V+ and V- then, you’ll have 220V between the two and this will fry your circuit.
    If you use current transformers, this is no problem, as the output side of the CTs will have no base level.

    For the supply, it is unimportant which side is V+ and V-. The supply will use a bridge rectifier or something like this.

    However, for the metering result, polarity is important. If you twist V+ and V-, you’ll have a 180° phase shift and your results have the wrong sign. Well, if the software detects the phase shift, it may swap polarity in code.
    Else you’ll also have to swap the I+ and I- inputs or turn the CT by 180°.
    A phase shift of 90° means that you have plain reactive power. A phase shift of 180^means that you’re producing energy, not consuming it.

**Attention** This is a public forum