Using Spread Spectrum Techniques to Reduce EMI

Digital and power electronic systems can reduce the radiated and conducted emissions profile using spread spectrum techniques. Typically, no more than 10-12dB can be obtained with those techniques but the result can be useful to comply with regulations.

Many EMC radiated and/or conducted emissions failures are related with periodic signals as high speed clocks (small rise and fall times) in digital circuits or very high dv/dt switching waveforms in power systems.

Fundamental and harmonics of that signals (see Figure 1) excite the components, traces, cables, and enclosures of your product creating dangerous situations in both conducted and radiated emissions.

Figure 1: Time and frequency domain for a typical digital or power signal.

Figure 1: Time and frequency domain for a typical digital or power signal.

To minimize emissions at the source level, we can increase the switching time, reducing dv/dt, but usually with limitations because signal integrity (digital) or efficiency (power) specifications.

An alternative or complementary technique is to apply frequency modulation to the switching signal.

The secret for this technique is to modulate the original frequency reducing the peak amplitude (key parameter for EMC problems) at harmonic frequencies. That reduction is because with modulation, the energy of each harmonic is spread into a certain frequency band (Figure 2).

Figure 2: Time and frequency domain for a modulated signal.

Note that because the effects in time and frequency domains, sometimes those modulation strategies are referred as “jitter” or “spread spectrum” techniques.

To modulate the clock or power signal (e.g. driver for the power transistors), we can use frequency modulation using different profiles as random, sinusoidal, triangular, etc. trying to obtain optimum performance. The details to cover that optimal modulation profiles and parameters is out of the scope of this article and many excellent papers can be found to learn that techniques.

To illustrate the spread spectrum techniques, a 900W power resonant converter was analyzed with and without modulation in the driver signals of the transistors (Figure 3).

Figure 3: Basic schematic for the power resonant converter used in this article.

Figure 3: Basic schematic for the power resonant converter used in this article.

A resonant capacitor Cr was connected in parallel with the inductive load in a full bridge topology using four IGBTs (SW in figure). Drive signals DRV1 and DRV2 with safe dead time were generated by the control circuit switching at 60 kHz nominal frequency. The converter was connected to mains (240V-50Hz in Europe) and a POWER BUS circuit is used to obtain a DC bus voltage (VDC) for the bridge.

The conducted emissions (both DM and CM mode) in the 9 kHz to 150 kHz range were measured using an EMCO/ETS-Lindgren 3810/2 Line Impedance Stabilization Network.

Two cases were compared: drive signal WITHOUT MODULATION (Figure 4), and drive signal WITH MODULATION using a sinusoidal profile of 100Hz and 8% modulation index (Figure 5).


Figure 4: Drive signal without modulation.

Figure 4: Drive signal without modulation.

Figure 5: Drive signal with modulation.

Figure 5: Drive signal with modulation.

The emissions from the converter were measured using an E7402A Keysight spectrum analyzer comparing the measured profile with the limits of the regulations (Figure 6).

Figure 6: Conducted emissions with and without spread spectrum applied to the driver signals.

Figure 6: Conducted emissions with and without spread spectrum applied to the driver signals.

Note that peak levels are reduced while energy is spread in a frequency band. Similar results were obtained with other modulation profiles.

A typical 10-12dB reduction in the peak value was obtained. That reduction can be useful to pass the limits of regulations, to simplify the mains filter, or to be used as a safety margin for production.

From the work done and a review of published results in many different applications no more than 10-15dB can be obtained with those techniques. Note that modulation index cannot be increased without limits because signal integrity considerations in digital systems or resonance (efficiency) considerations in power systems.

An additional consideration is that, with spread spectrum techniques more frequencies appear in the circuit so some parasitic resonances could be excited obtaining secondary problems compared with no modulated systems. Be careful!

201604 InCompliance (Spread Spectrum) 300dpi V.1.0

author mediano-arturoArturo Mediano
received his M.Sc. (1990) and his Ph. D. (1997) in Electrical Engineering from University of Zaragoza (Spain), where he has held a teaching professorship in EMI/EMC/RF/SI from 1992. From 1990, he has been involved in R&D projects in EMI/EMC/SI/RF fields for communications, industry and scientific/medical applications with a solid experience in training, consultancy and troubleshooting for companies in Spain, USA, Switzerland, France, UK, Italy, Belgium, Germany and The Netherlands. He is the founder of The HF-Magic Lab®, a specialized laboratory for design, diagnostic, troubleshooting, and training in the EMI/EMC/SI and RF fields!, and from 2011, he is instructor for Besser Associates (CA, USA) offering public and on site courses in EMI/EMC/SI/RF subjects through the USA, especially in Silicon Valley/San Francisco Bay Area. He is Senior Member of the IEEE, active member from 1999 (now Chair) of the MTT-17 (HF/VHF/UHF) Technical Committee of the Microwave Theory and Techniques Society and member of the Electromagnetic Compatibility Society. Arturo can be reached at Web:

Leave a Reply

Your email address will not be published.