Static & Crosstalk
Dr. E. Thomas Chesworth
In electronic equipment and in EMC measurement laboratories ground loops have been a constant pain in the neck or perhaps a few feet south of there. To some extent the problem is going away with the blackberries, boysenberries and cherries. These little circuit boards with batteries included couldn't care less about ground loops, ground planes and tables made of wood or styrofoam. They usually are a no-brainer. You just put them on the table a few centimeters from the edge and push the button. The equipment dutifully records its internal noise floor and you collect your fee, u nless the banana, potato, mango or whatever is a wireless wonder. In this case you should find the operational frequency where the box is chirping and whistling its mating call to boxes of similar feather so that they can flock together. The box may well be susceptible to a 200 volt per meter factory whistle.
In fact, things are getting out of hand with these Wifi and zagbee dudes. I bought a wireless whatnot for my laptop so I could use it with the printers and Internet connections in our basement office from anywhere in the house. This radio shack wonder was supposed to operate at a minimum distance of 400 feet and it did. I live in an older "Leave it to Beaver" neighborhood with trimmed lawns and pin oaks but I found that it also had enough Wifi and zigzag bees to cause my laptop to go ape trying to figure out who's on first and who's the shortstop. I can't use a wireless connection in my own house!
Lest you think it's only me, the computer whips in town ‚ mostly at the university ‚ have the same twisted traffic problems. The newer and better Billy Gates system doesn't help matters. Every computer with this operating system wants to be the conductor of the orchestra. Since all of them are set up to be a chief and none are Indians, they expect the other devices to take orders and everyone gets sick if they don't.
Seems to me that the commercial folks need a whole new set of compatibility tests. Sometimes I wish I had gotten into the EMC testing business. Then I remember why I didn't. The equipment costs an arm and a leg and you have to pay the loan payments whether or not you have test customers. Once you have, say, a quarter of a million dollars worth of equipment that works just fine and allows you to measure to the EMC specks, some government weenie decides that a spectrum analyzer is no good and you have to scrap a perfectly good, functional piece of equipment that gives the right results because this weenie has a brother-in-law in the EMI super-heterodyne, tuned receiver business.
Meanwhile back at the ground plane. Can you measure the plastic box with no input other than buttons and no output other than a display in a chamber without a ground plane? You better believe that you need to have a ground plane that meets CISPR-UR12 requirements. I find the requirements on ground planes particularly amazing. From the point of view of deciding whether the EUT meets the specification, the requirement to measure the site attenuation completely characterizes the ground plane. Why should I have to have a ground plane at all if the linoleum floor is properly reflecting the waves, never mind specified numbers of connections per meter?
I have sat in more than a few meetings where pundits have proposed to disallow spectrum analyzers or require solid ground planes or a new 10-centimeter table. If you dare to suggest that this is an undue fiscal strain on the labs, they hoot you down. They are, after all, not responsible for anyone's business problems.