Frequency Steps in the IEC 61000-4-3 standard

EMI/EMC testing per IEC 61000-4-3 is discussed at this URL by:

Engineering Limited, 1 Brassey Road, Old Potts Way, Shrewsbury, United Kingdom, SY3 7FA.

From there, which we may download this pdf file of “A Practical Guide for EN 61000-4-3: Immunity to radiated radio frequencies” and perhaps do some practical revisions of this guide to speed things up during product development time.

On page 14 of that file, we find the following excerpt:

Figure 1 Excerpt taken from “A Practical Guide for EN 61000-4-3: Immunity to radiated radio frequencies”.

The frequencies of incoming RF illumination that get applied to the unit under test (UUT) are stepped from one frequency to the next where each frequency following the first one is the frequency below multiplied by 1.01. The multiplication factor is (1 + p/100) where “p” equaling one represents 1%.

The total number of steps going from a low frequency of Flow to a high frequency of Fhigh may be derived as follows:

 Fhigh / Flow = (1 + p/100)Steps.

With algebraic rearrangement, we get:

Steps = log (Fhigh / Flow) / log (1 + p/100).

Since this calculation will almost always yield a non-integer result, we can choose the required number of steps as:

Steps = 1 + INT ( log (Fhigh / Flow) / log (1 + p/100) ).

Using this last equation, we confirm the theoretical value of 12.7 minutes of test time for stepping from 80 MHz to 1 GHz at three seconds of dwell time, but we can also look at alternative possibilities for developmental use as follows:

Figure 2 Total test time derived from Fhigh, Flow and dwell time.

If our upper frequency were to go from 1 GHz to 6 GHz, then the theoretical stepping time would increase from 12.7 seconds to 21.7 seconds. However, if for product development purposes, it we desired to have ballpark testing take less time, the use of coarser stepping intervals could be of some time saving benefit.

For example, 80 MHz to 1 GHz with three percent per step would cut the 12.7 minutes down to only 4.3 minutes, almost a three-to-one testing time speed-up. You get less time for your coffee and donut break each time you test, but your product development time will probably be faster.

John Dunn is an electronics consultant, and a graduate of The Polytechnic Institute of Brooklyn (BSEE) and of New York University (MSEE).

Related Content


Scroll to Top