When You Need a Little EMC Help: A New USB 3.0 EMI Stick from Wurth
This in-line adapter can be a helpful design tool when your USB 3.0 device is having noise problems.
This in-line adapter can be a helpful design tool when your USB 3.0 device is having noise problems.
In this article, I want to look at a product that is quite unusual. And that’s saying a lot, because these days there is a seemingly endless supply of electronic devices made by numerous different manufacturers, and I would say that it’s quite difficult to find a product that is unique with respect to its functionality or performance characteristics. I can’t say with confidence that the EMC stick is truly unique, but I’d be surprised to find an equivalent part from a different manufacturer.
USB 3.0
Before we discuss the USB 3.0 EMC stick from Wurth, let’s say a bit about USB 3.0. It’s a much-higher-speed version of USB 2.0. The original USB 3.0 specification supported transfer rates of 5 Gbit/s, which I would describe as really fast. Apparently, that wasn’t good enough, though, because USB 3.0 was followed by USB 3.1, which offers a maximum transfer rate of 10 Gbit/s—officially in the really really fast category. Alternatively, you can use the name chosen by the USB people: they refer to the 10 Gbit/s mode as “SuperSpeed USB”.
I suppose this sort of bandwidth is beneficial in certain situations, but transfer rates such as these are almost comically high compared to the (often perfectly adequate) 115.2 kbps that I used back in the RS-232 days.
Another issue is that high frequencies are troublemakers—schematic design becomes more complicated, PCB layout becomes more complicated, testing becomes more complicated. And then there’s EMC.
EMC
It stands for electromagnetic compatibility, and it’s a seriously nontrivial aspect of professional electronic design. The general idea here is that an electronic product needs to 1) not generate excessive amounts of electromagnetic radiation (EMR) and 2) function properly despite EMR generated by other devices. The former is referred to generally as the “emissions” side of EMC, and the latter is “susceptibility.”
Digital signals are notorious sources of unwanted EMR, and higher frequencies only make the problem worse. Radiated EMI is proportional to the square of signal frequency (more information here). Also, higher frequencies diminish the designer’s ability to use slew-rate limiting as an EMI-mitigation technique. Rapid transitions contain large amounts of high-frequency energy, and emissions can be reduced by intentionally extending the rise and fall time of a digital signal. However, the maximum practical rise and fall times depend on frequency—very high frequencies will require short transition times.
Spread-spectrum modulation is an effective way to deal with EMC problems, but it’s not always practical. You can also try filtering, and that is the approach used by Wurth’s USB 3.0 EMC device.
The Stick
As you can see in the photo, this is an in-line device.
Photo taken from the datasheet.
One side plugs into the host USB port and then you plug your device into the other side. What exactly is the purpose here? The rather succinct part description indicates that the EMC stick is intended for those who are “having trouble with noise” and want a shortcut to EMI and ESD compliance. Obviously you can’t market a product that meets EMI and ESD requirements only with the help of an EMC stick, so apparently the idea here is to help a designer determine that a device will function adequately with the help of ESD protection and an EMI filter—because that’s what the Stick is, an EMI filter and a TVS diode array.
Diagram taken from the datasheet.
The part description makes the rather bold claim that “in general, this stick will allow the board to pass all EMI and ESD tests.” I certainly can’t verify that claim, but I also have no specific reason to question it. I hope they’re right, and if I ever have to develop a USB 3.0 device (not likely), I hope that I can just incorporate the Stick into the product housing and call it good. My original design would presumably fail miserably when EMC testing came around and, if I tried to duplicate the Stick’s filter, I would probably botch it somehow and fail again.
It’s hard to know exactly what’s going on in the following comparison, but as far as I can tell the effects of this seemingly simple device are quite significant:
Image taken from the datasheet.
Do you have any high-frequency EMC tips? Or any thoughts on designing USB 3.0 devices? Let us know in the comments.
