What Jack Ganssle refers to as “The Ghosts in our Machines” have always been a part of the embedded engineer’s life. These are the unanticipated glitches that occur due to software “hiccups,” noise in the signals received and processed, unpredictable electrostatic discharge (ESD), and electromagnetic interference (EMI) that plague every design no matter how much work has been done to minimize their occurrence, severity, and effects on signal integrity.
As you will note in this week's Tech Focus Newsletter, a wide range of techniques, tools, and strategies have been developed to deal with these issues. In addition to Jack's informative blog on “Electromagnetics for firmware people , ” my Editor's Top Picks are:
A standards-based approach to capacitive-sensor EMC problems
By following basic principles based on industry standards, it is possible to implement capacitive sensing designs with high signal-to-noise ratios that will withstand a barrage of noisy abuse.
Software techniques for comprehensive EMC testing of embedded systems
The variety of ways in which a software engineer can provide help with embedded system electromagnetic compatibility (EMC) conformance and testing.
ABCs of signal integrity for embedded developers rules and methods
A series of articles exploring some of the noise, ESD and EMI challenges of designing for signal integrity and possible solutions..
Will the wealth of tools available be up to the challenges ahead? I am not sure about that. I see major EMI and ESD problems ahead not only for next generation wireless embedded devices and smartphones, but also for the many Internet of Things applications that are being talked about.
First let's look at ESD. To get more performance and functionality, smart phone manufacturers are pushing towards more functionality with multicore SoC designs that are pushing the limits of lithography. Most such designs are in the nanometer-sized transistor range, and because of competitive pressures they are asking for even more functionality, higher performance, AND lower power. But as supply voltages are pushed to 1 volt and below, the ability of the digital logic to discriminate between a zero and a one becomes more of a challenge, as does the ability to determine what is a harmful electrostatic discharge and what is not..
In current portable devices such as touch interface-based smartphones, designers have done a magnificent job of dealing with the electrostatic spikes that are only a fingertip away from one of the biggest and most constant producers of ESD – the human body. But it is not clear if the tools developed so far will be adequate in the consumer IoT segment, where developers are talking about more personal and wearable electronics on our bodies, in our clothing, and in constant direct contact with our skin.
The closer such devices are to the skin and the human body's natural electrostatic field, designers will have to factor in their effects in an environment that produces a constant rain of particulates (flakes of skin) and corrosive salt-laden moisture (sweat) – on the reliability of these devices.
As to EMI, the problems there will also increase. In his Ghosts blog, Jack writes about an incident in which embedded electronics in an electric stove turned on every time the home owner’s cell phone rang. That was only a few years ago and it was during a period when mobile phone use was not as prevalent as it is now.
Now, with the number of cell phones in use about equivalent to the population of the earth, I can only imagine the EMI problems will only increase. While my kitchen oven has not turned on when I use my cell phone, the voice quality of phone calls I receive in my electronics-laden home office is often pitiful.
Based on what a few of my neighbors have told me, I am willing to bet that many of the homes with wireless home networking and automation systems are having similar problems. And the numbers of such systems is small compared to the pervasiveness of the Internet of Things that everyone is predicting.
If used imaginatively and with due diligence, many of the techniques and tools that are currently available will lessen the severity and occurrence of such problems. But there are a lot of unknowns ahead that will require that we not only use our current knowledge in new ways, but look for better tools as well.
What do you think? What has been your experience? What tools and methods do you trust enough to deal with the challenges ahead? Which have run out of steam and need to be set aside? What new techniques and tools are you using?
Embedded.com Site Editor Bernard Cole is also editor of the twice-a-week Embedded.com newsletters as well as a partner in the TechRite Associates editorial services consultancy. He welcomes your feedback. Send an email to , or call 928-525-9087.