I am wondering about 61000-3-2 current harmonics pre-compliance test. Say a company designs a single-phase [up to 13A] power converter for low-line 115V and high-line 230V and pass 61000-3-2 pre-compliance test, should the same test be carried out for extreme cases such as 85V for low-line too?

Thank you.


1 Answer 1


Testing current harmonics outside of the prescribed standard limits can be, in my opinion, opening a can-of-worms. It all depends on how you interpret the results.

First, the harmonic content test methodology is defined at >220 VAC input (since this was originally a European standard defined for 230VAC distribution networks), so 230V is widely used. Secondly, the load is required to be set to "normal" operating conditions producing the highest possible harmonics content. This can be, but isn't always, maximum load, but maximum load is also commonly used. Finally, the harmonics limits themselves are defined by the type of equipment you're testing, divided into four classes (three specific classes and one "catch-all" for items not covered).

So what do "failures" mean if you test outside the standard? It doesn't mean you aren't compliant with the standard, since you're not testing to the standard. What limits are you using if you're not testing to the standard? Are you using the same harmonic limits you would use at 230V for testing at 85V? What about frequency?

In my experience, I have observed testing at alternate voltages and load levels, using the standard-prescribed limits. These test results are strictly used to infer the overall operation of the product and generally are not shared with external customers or published in official product qualification documentation (i.e. it's "for internal use" information)

Examples of this: testing very low and very high input voltages can give you a good indication that your input power factor correction (PFC) converter stage has robust and well-designed current feedback compensation, and is not malfunctioning or entering a strange operating mode at these conditions.

At very high input voltage, the PFC will stop boosting (if the rectified input voltage exceeds the normal boost setpoint) and as it approaches this threshold, the current waveform is going to degrade and appear more and more like a non-PFC corrected product. Degradation of the correction is inevitable but it's good to see that the controller isn't going totally bananas in this circumstance.

At very low input voltage the input current will be at its highest, and if you haven't designed your sensing correctly there may be clipping and distortion which will manifest itself as current distortion. Also a good area to check for overall robustness and component stress purposes.

At very light loads, the PFC and DC/DC converters may enter a 'skip' or hystereric mode of operation. This will cause discontinuous input current and may manifest itself as odd-looking harmonic patterns when tested. Again, not a failure, but good to know about (you may end up defining a minimum load for the product in your spec - if it is operated below this threshold, the product will work but cannot be guaranteed to be fully compliant).

As long as you're looking at the test data generated from non-standard conditions objectively, you're good to go. Just don't say "We don't pass harmonics!" if there's an issue at these non-standard points.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.