Impedance added by current measurement

Have you ever wondered how current is measured in PSIM? There are multiple methods for measuring current, with the most obvious ones being the Current Probe, Current Sense blocks, and the Current Flag for some components. However, there are also more unconventional ways of measuring, such as using current-controlled voltage sources.

 

 

The main difference between all these current measuring options is whether they are "invasive" or "non-invasive." Some of them introduce a very small impedance to measure the current, while others do not. Can you guess which ones add the small impedance and which ones do not?

... Yes, you guessed it right! The methods that break the circuit to measure the current, such as the Current Probe, Sensor, and Current Controlled Source (flowing through), are invasive. On the other hand, the Current Flag and Current Controlled Source do not interact with the measured circuit in an invasive way. 

We can confirm this by examining the results of the simulation with the 100kV sources. :

 

 

The error of the measured current is 0.0001% so the next question is why should we care about this impedance?

The answer to this question is quite simple: most of the time, we don't need to be concerned. However, there is one circumstance where this impedance can potentially cause problems. Imagine we have a 18-switch inverter and we use thermal models for the switches. Thermal models offer a pin to measure the switch temperature as well as the power losses:

 

 

Now, let's consider a scenario where we want to verify the thermal losses of the thermal module by measuring the current flowing out of those pins. Ideally, this current should be equal to each switch's losses. Instead of placing down 18 current sensors we assume symmetry and only use one. Then we can multiply the result by 18 to get the overall losses of the inverter:

 

 

By placing this current sensor here, we are essentially interrupting the thermal circuit. Even though the change is minimal (0.0001%), the temperature of the first switch (Q1) may differ from the other upper switches (Q2 - Q3). In reality, the physics of a real inverter would naturally compensate for this difference, ensuring that the upper switches maintain the same temperature.

PSIM also accounts for this kind of physical feedback. As a result we may observe mismatched temperatures and power losses for the first switch (Q1) and even the rest of full-bridge switches (in less degree though):

Temperature of the switches without the current sensor:

 

 

Power losses of the switches without the current sensor:

 

 

Temperature of the switches WITH the current sensor:

 

 

Power losses of the switches WITH the current sensor:

 

 

Notice the significant changes in temperature and power losses for the first switch (Q1) following the addition of the current sensor.

In such situations, it is advised to place a current sensor at all thermal circuits to maintain inverter balance and prevent power loss mismatches. Alternatively, if your thermal circuit includes an external thermal resistance, the current flag can be used.