Radiated Immunity Device Test Orientations - IEC 61000-4-3

[rothlis]

It appears to be standard practice to test four 90° "side" device orientations during radiated immunity (61000-4-3) and then also the "top" and "bottom" if the device can be used as such. This is reflected in several guidances I've encountered, and is presumably the inference behind clause 8.2 of 61000-4-3, where it says that "all sides shall be exposed ... In other cases, as determined for example by the type and size of EUT or the frequencies of test, more than four azimuths may need to be exposed".

From an electrical perspective, it would make sense to test orientations for all six sides of a cube to ensure all possibly orthogonal fields are covered by at least a 45° angle of incidence, on the assumption that any one wire or circuit in the device could be oriented in any direction and could be vulnerable to interference. So why isn't it standard practice to test all six orientations for every device? Is it because large devices aren't easily tested in the "top" and "bottom" orientations in a typical EMC lab setup?

As I see it, to exclude the "top" and "bottom" orientations is to presume that the real-world interference is never going to originate from a vector above or below the device and, unless I'm missing something, that doesn't seem like a justifiable assumption.

 

[phase90]

What I understand is it relates to the feasibility of testing. If the DUT is a smaller hand-held type of device that can operate in any orientation, it can be tested on all six sides. If it is a unit that cannot be tipped onto its side and operate, then four sides is all you can do. I guess until someone configures a chamber to be 10 meters high and have the antenna above the DUT, that is the reality of what can be done.

 

[rothlis]

What I understand is it relates to the feasibility of testing. If the DUT is a smaller hand-held type of device that can operate in any orientation, it can be tested on all six sides. If it is a unit that cannot be tipped onto its side and operate, then four sides is all you can do. I guess until someone configures a chamber to be 10 meters high and have the antenna above the DUT, that is the reality of what can be done.

Thanks for the input. That's what I suspected, but I also perceive tradition has limited testing to four sides whenever top and bottom orientations aren't obviously possible during normal operation, regardless of whether testing is feasible with the device on its side. And it seems to me that 61000-4-3 contributes to this oversight by inferring that four sides is the expected set of orientations, when really that is only appropriate to a minority of devices.

 

[phase90]

I just pulled up edition 3.1 of 61000-4-3. In clause 8.2, it says the "test shall normally be performed with the generating antenna facing each side of the EUT. When the equipment can be used in different orientations (i.e. vertical or horizontal) all sides shall be exposed t the field during the test. When technically justified, some EUT's can be tested by exposing fewer faces to the generating antenna. In other cases, as determined for example by the type and size of EUT or the frequencies of test, more than four azimuths may need to be exposed."

So really, they are saying if it can operate in any orientation, it needs to be tested on all side, otherwise only the four sides as that is what is feasible.

Yes, technically, floor standing equipment may be exposed from above or below out in the field, but the typical exposure will come from a source horizontal to the equipment. A EMC test house engineer kind of told me years ago that was why they only test four sides. Nowadays, I guess one would have to look at it from a risk analysis perspective to determine if other faces are needed.

 

[rothlis]

I read that excerpt as follows:
The "four azimuths" in the last sentence are equivalent to "each side" in the first sentence, so that "normally be performed" = left, right, front, back. The second sentence is then saying that if the device can be used without the top and bottom facing the ceiling and floor then the top and bottom also need to be exposed, and the last sentence is essentially saying the same thing in more general terms that include other conditions. So exposure of the top and bottom is treated as the exception rather than the rule.

It may be true that the majority of interference sources would come from a horizontal plane, but there is no shortage of cases where devices which have a defined upright position during use (and thus not treated as operating horizontally) would easily be exposed from the top or bottom (e.g., any pole or wall mounted device could have sources mounted above or below it, and any cart or similar device could have sources sitting directly above it).

I'm inclined to think that there is a deficiency in the language of the standard that has led to an accepted practice which doesn't align with the intent of the test. But maybe I'm over-analyzing this.