One of my colleagues has a desk drawer full of I/O boards that have been burned quite severely. Why did they burn? The I/O boards are in energy-limited SELV circuits. There should be no possibility of fire.
When we closely examine the boards, we find that the ground trace from the I/O connector is the trace that was overheated. It can only overheat when it conducts lots of amps. But, we all know that the ground is not a current- carrying conductor. At least, it is not a current-carrying conductor under normal conditions.
How can the trace overheat when there is no current?
There must have been a fault condition. Can we determine what it was?
When we check out the circuit with an ohmmeter and a voltmeter, we find everything is okay: zero ohms, and zero volts.
My colleague performed the traditional grounding continuity test using a 30-amp source. It passed.
Well, my colleague went a step further. He attempted to duplicate the failure with a new board. He kept cranking up the current until the board burned just as his drawer full of boards. It took 100 amps!
There is no way to get 100 amps through a 120 volt cord-connected product on a 20 amp branch circuit.
The board did burn. It did take 100 amps to bum the board. Those 100 amps had to come from somewhere.
Before we explore this, let’s turn to a different phenomenon.
Have you ever measured the potential difference between the neutral and the ground? You probably measured a couple of volts.
To get a few volts potential difference, there must be a few amps of current passing through a resistance, somewhere.
Since the ground has no current, the current must be in the neutral. When we make this measurement, the ground wire acts as a remote contact to the end of the neutral wire. So, we are measuring the voltage drop across some portion of the neutral conductor.
Now, let’s return to the original issue: How can current get into the ground under normal conditions? Normal conditions are the only conditions in which we can get continuous current in the ground conductor.
The answer is found in the National Electrical Code Handbook. One neutral may be grounded at more than one point! See Figures 250- 7 and 250-8 (pages 193 and 194 of the 1987 Handbook) and Figures 1 and 2.
Figure 1: 3-wire 120/240-V AC single-phase secondary distribution system (From 1987 NEC, Fig. 250-7.)
On a 2-wire or 3-wire single-phase acsecondary distribution system, groundingconnections are made on the secondaryside of the transformer and on the side ofthe service disconnecting means.
Figure 2: 4-wire, 3-phase 208Y/120-V secondary distribution system (From 1987 NEC, Fig. 250-8.)
The neutral is grounded at each service and also on the secondary side of the transformer on this 4-wire, 3-phase, 208Y/120-V secondary distribution system. When 3-wire, 3-phase service equipment is installed for power purposes on this type of ac system, the grounded (neutral)conductor is required to be run to the service equipment.
What does this mean?
If the neutral is connected to ground at more than one point, then the neutral and ground are connected in parallel between those two points. In accordance with Kirchoff’s laws, such connection makes the ground a current-carrying conductor under normal conditions!
What does this mean for the I/O board?
The I/O includes a signal ground. When the I/O is connected to another piece of equipment which is grounded at another location, then the signal ground wire, because it is grounded at two points, parallels the ground and neutral wire! Thus, the neutral current gets divided into three paths: the neutral wire, the ground wire, AND the signal ground! VIOLA! Lots of amps in the signal ground wire! The I*I*R causes the traces on the boards to burn. Depending on the distribution transformer size, the distance between I/O ports, and wire sizes and lengths, it is indeed within the realm of possibility to have 100 amps in the signal ground wire!
And, we have a fire hazard.
Two or more neutral ground points necessarily connect the ground in parallel with the neutral. Signal grounds are always in parallel with the ground. Whenever the neutral is grounded at two or more points, the signal ground between two points, especially remotely located units, may be in parallel with the neutral. When this occurs, some portion of the neutral current will be in the signal ground. If the neutral current is high enough, it can cause overheating on the I/O board.
What can be done to prevent this situation?
In order for a grounding conductor to not be a current-carrying conductor in parallel with the neutral, the neutral must be limited to one grounding connection. Fortunately, single point grounding is permitted by both the NEC and the CEC. 
 |
Richard Nute is a product safety consultant engaged in safety design, safety manufacturing, safety certification, safety standards, and forensic investigations. Mr. Nute holds a B.S. in Physical Science from California State Polytechnic University in San Luis Obispo, California. He studied in the MBA curriculum at University of Oregon. He is a former Certified Fire and Explosions Investigator.Mr. Nute is a Life Senior Member of the IEEE, a charter member of the Product Safety Engineering Society (PSES), and a Director of the IEEE PSES Board of Directors. He was technical program chairman of the first 5 PSES annual Symposia and has been a technical presenter at every Symposium. Mr. Nute’s goal as an IEEE PSES Director is to change the product safety environment from being standards-driven to being engineering-driven; to enable the engineering community to design and manufacture a safe product without having to use a product safety standard; to establish safety engineering as a required course within the electrical engineering curricula. |
