A millennium ago give or take 954 years I spent my weekdays making measurements to military standards. The original MIL-STDs were open-field radiations – MIL-I-11748 was the Army and Air Force spec, MIL-E-6051C was the Navy spec.

In the middle of a field you set up the EUT, a motor-generator, the measurement equipment and dipole antennas on tripods. If you could find the interference from the test sample in the ambient, you noted it on your data sheet. Each measurement was a problem. You needed to adjust the length of the dipole to resonate at the frequency, then spend enough time at the measurement frequency (plus or minus receiver drift) to estimate the maximum excursion of a microamp meter (S meter).

The most useful tool for idenatifying the test sample emissions was a set of headphones. Generally, if the signal demodulated to Rosemary Clooney ("Come-onna my house, My house a-come on") you suspected that the RF wasn’t coming from the EUT. It rained and snowed in the field. It was hot in the summer and cold in the winter.

To try to solve some of the problems the Air Force came up with MIL-E-6181D. The first thing that had to go was the open field. Indoors it neither rained nor snowed and the temperature was reasonable. This was a big step forward not only for the EMC technician making the measurements but also for the test sample and the measurement equipment which also was out of the rain and in a moderate temperature environment. Then came the piece de resistance. If you’re indoors, why not be inside a screen room where you could control the electronic noise environment? Goodbye, Rosemary.

Could you measure the free space E and M fields in a screen room? The obvious answer was of course not. But the object of the game was to quantify the interference emitted from the test sample. If this interference caused the feedback control system that regulated the fuel to shut off the engine when the EUT was on, the pilot of the plane that was falling toward the jungle could care less about the far-field values of the emitted radiation.

In an exhaustive series of tests the military determined what value of antenna terminal voltage as measured with rod, dipole and horn antennas in a screen room with the EUT on a copper table corresponded to unacceptable levels of interference. The system worked so well that the Army and Navy started to specify MIL-E-6181D in electronics contracts but insisted on redefining limits, measurement antennas, etc. To "standardize" the procedure an Inter-service specification, MIL-STD-461, was developed, and it would have worked fine if the services had left it alone. But anyone who had ever been to an Army-Navy game knew that the Inter-service spec was doomed. Before the ink was dry on the original there was a Note One, a Version C and so on and so forth.

The armed services were not the only ones who had to "make 461 better." The FCC decided that the military didn’t know what they were doing and so they took a giant step backwards to the open field. The Europeans decided that susceptible equipment must be quasi-peak, not peak, detectors. I wonder how they knew that and how they figured that the equipment was installed on 80cm wooden tables – no more Steelcase desks in computer installations, I guess. Obviously specs that work in airplanes don’t work for Detroit automobiles and so we have at least 200 specs out there to quantify the same thing and none of them is better than any other unless you prefer a headache to an upset stomach or vice versa.