In the last couple of posts I’ve been discussing ground loops and bonding and I made a comment that at the higher frequencies, i.e. VHF/UHF and higher multiple point grounding is used. It occurs at these higher frequencies because of capacitive coupling whether we planned it that way or not. The so-called ground loop exists whenever there are multiple ground connections between equipment. I created a firsthand example last week end when I decided to extend my sound system from the corner of the living room to the far end of the house.
It seemed simple enough and would have probably worked OK if I hadn’t put in the extra amplifier to enhance the Bass. It enhanced it all-right but it was also putting out a wonderfully distorted 60Hz tone. If you could call that growl a tone. The growl was highlighted by a cute little whine whenever the refrigerator turned on. In the good old days turning the AC plug 180 degrees often solved this problem. Now days, the plugs are polarized, so with the help of an old non polarized extension cord I reversed the amplifier plug and made it a lot worse! However I noted that running a long extension cord from the source outlet to the distant amplifier almost solved the problem.
Ground loops aren’t a problem unless there is unwanted current flowing in the loop. The current is developed by differences in grounding potential or by magnetic fields being coupled into the loop. In this case I suspected that there might be a wiring error in the house which splits the return current loop increasing its coupling area or possibly a neutral to green wire short. This is illustrated in the figure.
When doing magnetic field surveys a lot of the problems are caused by ground loops created by incorrectly wired feeds and branch circuits, but sometimes the problem is an equipment problem. For example an RF filter. It’s just like any other EMC problem. It takes a Level III Wizard (or higher) to find the culprit!
I spent most of the weekend looking for the problem and finally found it – in the laundry. When our new water softener was installed the installer who was obviously not an electrician pushed a knock-out into a metal conduit box and fed his wire through the hole. He did not twist off the knock-out plug and when the duplex outlet was reinstalled the neutral wire made contact with the knock-out plug. (Yes, it was a he.) This shorted the neutral to the safety wire. The refrigerator in the kitchen is on that same branch circuit. Murphy strikes again!
Anytime there is unbalanced circuit current we will have EMC emission problems. That why EMC engineers insist that through any wire pair through any connector there must be a net zero current flow and the outgoing and return leads should be kept as close together as possible to minimize loop areas. When/if you have a similar problem to the one I just described associated with power wiring look for:
• Interconnected neutrals from separate power circuits. This causes major return current sharing. The further the interconnect occurs from the source the more devices are involved and the larger the magnetic loop area
• Neutral-safety ground shorts. Can occur anywhere but do check subpanel neutral bus wiring for errors.
• Check three-way switch wiring. Neutral and Line may be fed to different points in the circuit increasing loop area. The further the switches are from the associated load the more likely this is to be true.
Since all of these problems create magnetic fields, it’s amazing how much troubleshooting you can do with an oscilloscope and a clamp on current probe . . . even a DIY one. See The HF Current Probe: Theory and Application by Ken Wyatt in the 2012 Interference Technology Directory and Design Guide for more info on how to do it yourself.
-Ron Brewer
