Welcome to the fourth installment in our five-blog series on EMC basics!
In the previous posts, we walked you through an overview of EMC, chip bead ferrites, and losses.
Today, we will talk about the difference between common mode and differential noise in EMC and how to combat them.
Clamp-on ferrites are typically made with two different types of materials: manganese zinc and nickel zinc.
Nickel zinc can be used in situations with either conducted or radiated noise. Manganese zinc is used mostly for conductive noise.
This image provides a guideline for which material to use depending on your situation. Of course, there are exceptions, but this is what we find to be typical.
Are you dealing with common mode noise or differential noise? This is a common question that you need to be able to answer.
Here's a quick trick to getting you headed in the right direction. It's not 100% accurate, but it helps you get the process started.
Imagine a doctor prescribing you antibiotics without knowing whether you have a bacterial infection or a virus. He does so knowing that if you have a bacterial infection, the medication will work, and the problem will be solved. If the medication does not work, at least he knows you're dealing with a virus, and he will then treat you accordingly.
So in our case, you can simply apply a clamp-on ferrite onto a cable, remembering both lines (Vcc and ground) will be in that cable. If the noise is reduced, or the immunity is increased, then you're experiencing a common mode problem. If there's no effect, you have a differential mode problem.
Therefore, at the board level, you know that if you have a common mode problem, you can use a common mode choke. If you have a differential mode problem, you can use a chip bead ferrite.
Here, we have a visual representation of how a common mode choke works.
The red arrows represent a differential signal going in. This is a useful signal. It creates a magnetic field inside the core going in one direction, according to the right-hand rule.
The differential signal then has to go back to the source, which creates another magnetic field, according to the right-hand rule. These two signals will cancel each other out. So you're seeing an animated crash here, giving a visual representation of the cancellation.
Common mode noise will create a magnetic flux inside the core, but this time it's in the same direction, as shown here by the blue arrows. The core will react to the frequency and convert it to heat. You can see a mushroom cloud in the in the middle of the core as an extreme example, but we're converting the frequencies to heat.
This fact makes it important to choose the right core for the specific frequency you're working with. Because of the winding structure inside the part, we're not influencing the signal; therefore, if we choose the correct core, we'll get a high attenuation of the noise.
What you're seeing here is a visual representation of what a differential mode magnetic flux line actually looks like. You can see there's a constant struggle in opposite directions, which in the end will cancel each other out.
Now, we’ll give you a look at the common mode magnetic flux line. You can see in this image that they're flowing in the same direction.
Something to keep in mind when using a common mode filter is that there will be a differential impedance that could attenuate your useful signal. As shown in this graph, the blue line represents the common mode impedance, and the red dotted line shows the differential impedance.
That means if your signal is at 100MHz, and you're using a common mode solution, you could possibly be unintentionally attenuating this signal that you're trying to use.
So this gives you a graphical representation for a common mode solution. There is actually some differential mode of impedance as well.
Here is a specific example. The blue line represents your common mode impedance, and the red dotted line shows your differential mode impedance. Your useful signal is a thick black line right at 4MHz.
As you can see, you're making good use of the common mode chokes, as it has a high common mode impedance right at that 4MHz and a low differential mode impedance.
Therefore, the impact on the noise is high, and the impact on your useful signal is kept to a minimum. When reading our datasheets, we'll show you both the common mode impedance and the differential mode impedance for that part.
There are two sets of windings in a common mode choke: sectional and bifilar.
Sectional-wound components are most useful in power supply applications. They can also be used as inductors to attenuate differential mode noise. Because sectional components have a higher leakage inductance, they attenuate differential mode noise as well as common mode noise.
Bifilar-wound parts are typically used for signal lines because they don't want to attenuate the differential mode signal (your useful signal). Thus, you have a lower leakage inductance, and that is utilized for filtering noise on a signal.
It's very important to know where your differential mode impedance and your differential mode signal are. Common mode noise can be solved with either a sectional- or a bifilar-wound part.
To summarize: the way it attenuates noise is very similar. In contrast, there's a big difference if you're using a sectional or a bifilar part to attenuate differential mode noise.
You can see these differences shown with the red and blue lines above. Notice that the sectional parts (the red lines) are acting in very similar ways, whereas the bifilar parts (the blue lines) are acting in very different ways when it comes to common or differential mode noise.
And now you know the difference between common mode and differential noise, as well as how to use common mode chokes to reduce unwanted noise.
In the next and last post in our blog series on EMC basics, we’ll cover clamp-on ferrites and why it's so important to design with EMC in mind. Don't miss it!
