Two non-destructive current sensing methods for measuring current

We almost always need to measure some type of current. In the last article, I introduced the two main reasons for measuring current, and several methods of measuring using loss current sensing technology. This article will focus on the lossless current sensing technology.

    Use existing circuit components! We will introduce two methods for current sensing using existing circuit components. The two methods are inductor DCR sensing and FET sensing.

    Inductor DCR sensing is not special, but it is sufficient. The DCR of an inductor is usually approximately +/- 10%. The temperature coefficient changes due to copper, and you may get some very inaccurate measurements. The good thing is that after the DCR network, you can finally get a very clean signal without switching noise. Figure 1 is the network required to extract current information from the inductor's DCR.



    Select the components of the network using the following formula:



    There are several factors to consider when setting up a DCR network:

    1. The amount of semaphore that the controller or sensing circuit can handle — may need to be split.

    2. Temperature compensation — Components with a negative temperature coefficient can be used to help keep the DCR constant when the temperature changes.

    DCR sensing is usually used in multi-phase configurations to achieve circuit mode control. Using this technology, it is easy to implement current sharing between multiple phases.

    Use FET sensing, but you need to pay close attention to switching noise! When the FET is switched, a lot of noise is generated. This noise needs to be filtered. Figure 2 shows the FET sensing scheme and how noise interferes with the measurement. There are several ways to reduce noise, but they all have drawbacks.



    1. Use the filter of the RC network. It can be used, but it will round the current signal to make the edges unclear. The rounded current sense signal will cause jitter and other noise problems.

    2. Leading edge blanking. This is a technique that ignores part of the current sensing signal. The main problem with this method is: it will lead to on-time, if there is a problem, the duty cycle can only be reduced by so much.

    Figure 3 shows the shortcomings of the above two methods.



    Lossless current sensing technology does not use high-precision resistors, but can avoid efficiency and power loss issues. Generally, for applications where the efficiency gain ratio is more important, the lossless method is more suitable. The current sensing method is by no means limited to the methods introduced in the last two blogs. Please leave a comment and let us know how you did it!