Figure 1. An Impulse with Radiated EMI Preceding It

Abstract: Measurement of high amplitude impulsive signals, such as ESD, can incur errors due to radiated effects. The characteristics of these effects are discussed and ways to spot such errors in a measurement are given.

Discussion:  Figure 1 shows an example of an impulse accompanied by radiated EMI, electromagnetic interference, from the impulse source itself. Note the high frequency noise in the waveform preceding the main impulse. Although it is possible the waveform represents the true source, more likely, the high frequency noise before the pulse represents error in the measurement.

The noise starts about 10 nanoseconds before the impulse. In this case, the EMI generated by the impulse radiated into the end of the cable at the scope. The impulse itself traveled down the cable at a slower speed (~1.5 ns/foot) while the radiated EMI traveled faster in space (~1 ns/ft) and through a more direct route to the scope input, arriving at the scope about 10 nanoseconds earlier than the impulse itself. Noise like this on a waveform, occurring before an impulse, often is a sign of direct radiated interference into the measurement. Figure 2 shows another example where the noise starts about eight nanoseconds before the impulse.



Mechanisms whereby radiated EMI can affect a measurement in this way include:

• Poor shield connection at the oscilloscope either because of the connector itself (BNC connectors vary widely in quality) or how the shield interfaces the connector (360 degree, secure connection is best).
• Sometimes an impulsive source like ESD can be powerful enough to leak through the shielding in a scope and directly affect internal circuits of the scope.
• The quality of the shielded cable itself can be an issue. I have seen BNC cables labeled as RG-58/U that provide poor shielding effectiveness. In fact, those of you who have attended some of my seminars/classes have seen an experiment that uses just such a cable.
  

Characteristics of radiated EMI on the scope screen include:

• Noise starts before the impulse being measured.
• The noise, as displayed, often contains frequency components that are higher than the rest of the waveform.
• Subsequent events show similar delays between the start of the higher frequency noise and the impulse to be measured.
• The delay between the start of the noise and the impulse is affected by moving the source and/or cable.
  

Summary: Interference from impulsive sources to measurements can be significant. There are several characteristics of such interference, the main one is high frequency noise that starts before the intended pulse to be measured.