DesignSpark Electrical Logo

Innovation for Every Engineer

  • Brought to you by
  • RS Components
  • Allied Electronics
TekExpert

August 3, 2016 07:14

Reducing the Cost of EMI Pre-Compliance with Real Time Spectrum Analyzers

Greetings!

My names is Lee Morgan. I’m a business development manager for IoT and Power applications at Tektronix. I’m spending this month as a visiting blogger on DesignSpark, talking about all things wireless. Recently I spoke about the growing problem of interference in the development of IoT products and how Real Time Spectrum Analyzers, such as the RSA306B, were accelerating the design cycle of such products. Traditional RF engineers now have to grasp the time-varying nature of modern digital RF signals, and they’re switching from Swept Spectrum Analyzers to Real Time Spectrum Analyzers (RSAs) to give them insight they need to get the job done. This week I’d like to spend a short time discussing EMI Pre-Compliance testing, its importance and methods to reduce its costs.  

EMI Compliance Background

EMI regulations are in place around the world to ensure the reliability and safety of electronic devices. To ensure compliance with these regulations compliance tests are typical done by accredited, independent labs with specialized test facilities. EMI testing is broken into two main paths, Radiated Emissions which refers to both the intentional and unintentional release of electromagnetic energy from your electronic device, and the Conducted Emissions which refers to the mechanism by which EM energy is created in an electronic device and coupled to its AC power cord. For both, specific measurements, limits and test methodology are defined. Most compliance labs will utilizes expensive equipment including anechoic test chambers, EMI receivers with Quasi-peak (QP) detector and pre-amplifiers, antennas and line impedance stabilization networks (LISNs), to name just a few. The test house typically give a plot of the radiated or conducted spectrums relative to the pass/fail limits, along with calibrations and correction factors for each of the components in the system, as shown in Fig 1a and 1b 

Fig 1a – A typical Compliance Spectral Plot with Pass Limit

title

Fig 1b – A Typical Tabular View of Results Along With Correction Factors

title

With the range of expensive equipment and difficulty in making these measurements, compliance test houses can easily charge between $1000 and $3000 per day for their services. Also, a failure of a product at this stage in its development cycle can cause costly design changes and a company being late in bringing its product to market.  

Pre-Compliance Testing Challenges

Because of this, pre-compliance testing and debugging is a major part of the development cycle in order to catch potential EMI problems early, correct them and increase the probability of passing at the certification stage. It’s important to recognize that the goal of pre-compliance testing is not necessarily recreate the full compliance setup, but to simply to find potential EMI problems. The equipment used can be non-compliant, with less accuracy, greatly reducing the cost of pre-compliance testing. In many situations a spectrum analyzer can be used in place of a dedicated EMI receivers.

EMI compliance specifications define the measurement bandwidth size and filter shapes during testing. In the case of the spectrum analyzer this is driven by the resolution bandwidth (RBW) setting. The bandwidth of this filter is defined as some amount of power down from the peak, and this will ultimately dictate the shape of the filter. So for example, a 100KHz 3db filter is one where the 100KHz width occurs 3db down from the peak and the 6db filter must attenuate twice as fast for the same bandwidth, giving it a different shape. Fig2 below shows a comparison between both. Spectrum analyzers typically specify 3db filters while EMI receivers typically define 6db filters for most measurements. These different filter shapes will ultimately alter the results between the two instruments for the same filter setting. While the peak values will correlate, the overall measured noise will be lower between 3 and 6db.

Another important factor is the detector method used. The detector is used after the filter to calculate a single point on the spectrum, and could be based on the peak negative or positive level in an RBW bin, the RMS value or on a QP value. Compliance labs use QP detectors for certification. QP detectors serve to detect the weighted peak value of the envelope of a signal. This has the effect of weighting a signal depending upon its duration and repetition rate. Signals that occur more often, or last longer, will result in a higher QP measurements than infrequent, short pulses. However during pre-compliance, a peak detector can be used which will allows for more conservative test margins. In Fig 3 we can see the example of a signal with an 8uS pulse width and a 100Hz repetition rate seen with both the peak (blue) and QP (yellow) detectors. The resultant QP measurement is 10db lower than the peak value. A good rule to remember is that a QP will always be less than or equal to a peak detect, never larger, hence it is not critical to use a spectrum analyzer with a QP detector in a pre-compliance setup!  

Fig 2 – 3db And 6db Filter Shape Comparison          title 

Fig 3 Comparison Between Peak And QP Detectors

Pre-Compliance Setups

The setup for pre-compliance radiated emission testing is shown below in Fig 4a. First, a location with low levels of background RF noise is selected for the testing. The device under test is setup anywhere from a few cm to 1m away from the antenna, which in our example is a low cost PC board log periodic antenna. Then there is an optional pre-amp to improve the signal strength, followed by the spectrum analyzer. In the case of the RSA306B  the antenna factor (AF) and cable losses for field strength correction can be entered in the user interface. The test steps involve inputting these correction values, turning on the peak detector and setting the limit lines. Prior to any pre-compliance it’s important to evaluate the test environment before turning on the device under test (DUT) in order to ensure there is enough signal room between the limit line and your ambient noise floor. Once satisfied, turn on the DUT and measure the difference between the background noise floor and the emitted RF energy from your device as shown in Fig 4b. 

Fig 4a – Example Of A Pre-Compliance Test Setup For Radiated Emissions

title

Fig 4b – Example of Pre-Compliance Test Results Using The RSA306B for Radiated Emissions

title

For conductive emissions the setup is similar, however the antenna effectively replaced by a LISN. The job of the LISN is to create a known impedance between the DUT and the spectrum analyzer. It is placed between the ac or dc power supply and the device. It also acts as a low pass filter to isolate unwanted RF signals from the source.

 

The typical setup and results are shown in Fig 5a and 5b. Like with radiated emission, peak detectors and limit lines are used, and the background noise floor levels are characterized before the measurement. The test again are typically a relative measurement between the noise with the device turned on and the background levels on their own.

 

Fig 5a - Example Of A Pre-Compliance Test Setup For Conductive Emissions

title

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig 5b – Example of Pre-Compliance Test Results Using The RSA306B for Conductive Emissions

title

 

Summary

While EMI certification is a critical and challenging porting of the design cycle, pre-compliance testing greatly reduces the risk to project time schedules. Modern low-cost spectrum analyzers are great tools for completing both radiated and conductive emissions tests. For more information on this topic, including a look at the difference between near field and far field measurements, and the use of near field probes for debugging, I recommend checking out following application notes from Tektronix, 37A-60141-1 and 37W-60228-1. Next week I will finish off this series of blogs by taking a look under the hood of a real time spectrum analyzer, how actual works and some of its unique features.   

Thanks for read! Please feel free to comment below

Lee Morgan

About Author:

Lee Morgan is an Account Manager with Tektronix UK Ltd, with over 16 years’ worth of experience in the Test & Measurement sector, covering a multitude of roles in the Mobile Telecommunication, Electrical & Power industry, he has an excellent insight into how modern Test & Measurement Equipment can aid the engineers of today in creating the products of tomorrow.

 

TekExpert

Lee Morgan is an Account Manager with Tektronix UK Ltd, with over 16 years’ worth of experience in the Test & Measurement sector, covering a multitude of roles in the Mobile Telecommunication, Electrical & Power industry, he has an excellent insight into how modern Test & Measurement Equipment can aid the engineers of today in creating the products of tomorrow.