Abstract: Common mode current measurements are
widely used to predict emissions from equipment cables. Mat
Aschenberg and Charles Grasso have written an analysis relating common mode current
measurements to radiation comparing three different methods. Their treatment is unique in that it addresses
the case where the cable is longer than one half wavelength. The
results are presented as a pdf file.
Discussion: Radiated emissions from a cable are directly
related to common mode current flowing on the cable. Figure 1 shows the
use of current probes to measure common mode current on cables. About
15 microamps of current flowing at the center of a one half wavelength dipole will result in radiated
emissions that come close to international emission limits for class A
equipment (industrial as opposed to consumer, or class B). By
extension, one can say that a cable that is a significant fraction of a
wavelength long may have the same emissions problem if the same current
is flowing on the cable. As a general rule of thumb, if a cable
is carrying more than about
15 microamps at a frequency, one can expect that class A (industrial as
opposed to consumer) radiated emissions requirements may be exceeded.
This is somewhat of an oversimplification, but will serve as an example
for this article.
Electromagnetic compatibility engineers often work in dB, so 15
microamps would be about 24 dB above a microamp or 24 dBuA. The current
probes used to measure common mode currents are rated in transfer
impedance, the ratio of the output voltage to the current through the
probe. A popular probe used below 300 MHz is the
Fischer F-33-1
probe. It has a transfer impedance of about 5 Ohms or about 14 dB above
one Ohm or about 14 dBOhms. To relate the voltage out of the current
probe for a 15 microamp current, one can multiply the current by the
probe transfer impedance or add dBuA of the current to dBOhms of the
transfer impedance to get the final level.
So for the F-33-1 we have:
Probe voltage output (dBuV) = 24 dBuA + 14 dBOhms = 38 dBuV
When adjusting a spectrum analyzer, I just put a display line at about 40 dBuV as the level I worry about.
Mat Aschenberg and Charles Grasso, both of Echostar, have performed an
accurate analysis
with supporting data comparing three different methods of relating
common mode current to radiated emissions from a cable. They analyze
the
case where the cable is longer than a half wavelength, which is often
true for many system cables. They have written a short paper with the
results of their analysis in pdf format titled "Radiation from Common
Mode Currents -
Beyond 1 GHz."
Click here to view or download the paper. In the paper, Mat and Charles address the relative accuracy of the three methods.
Summary:
Common mode current measurements can be used to predict radiated
emissions from cables. Such a measurement can avoid unnecessary
failures during EMC testing. The linked paper by Mat Aschenberg and Charles Grasso compares three
methods for relating common mode current to emissions and presents data
to suggest a reasonable method to use.
If you like the information in this article and others on this website, much more information is available in my courses.
to see a listing of upcoming courses on design, measurement, and
troubleshooting of chips, circuits, and systems.