Ground is presumed to be a “safe haven” for sensitive components. All grounded surfaces and connections are supposed to be “equipotential” i.e. with no voltage difference between them. This assumption may work only for DC. At high frequencies a wire is an inductor presenting impedance, and isolated wires and metal parts are plates of capacitors – good conductors for high frequency signals. At high frequencies voltage between different grounded parts, especially in automated equipment, may reach several volts, which can cause electrical overstress (EOS) damaging the devices.
For test, ground is supposed to be the reference for all data, metrology, communication, interface – in essence, all operations. If ground reference isn’t zero, then all other signals are referenced to something else – an invitation to a problem. Here is a screenshot of voltage between two “grounds” in a wire bonder where the resistance between these grounded parts is sub-ohm and no mains or DC voltage is found. This voltage is unacceptable according to SEMI E.176 and IRDS guidelines.
Inside equipment measure voltage and, if it is feasible, current between grounded parts, especially where components are being handled. Note that the measurements are relevant only when the equipment is operational (see further in the text).
On a facility level measure voltage between equipment’s chassis’ ground and facility ground and between chassis’ grounds of adjacent equipment. http://www.onfilter.com
First, this Newsletter focuses only on conducted EMI (i.e. on wires and cables), mostly because conducted emission carries much more energy than the radiated one, and it doesn’t diminish with distance as much. Radiated EMI deserves a separate discussion since its implication, its measurements, and its mitigation vary significantly from that of conducted emission.
The key primary parameters to measure during an EMI Audit are EMI voltage and EMI current. For component damage it is mostly EMI current that is prevailing cause. For interference with equipment operation and with test, it is mostly EMI voltage. These two parameters are intertwined. Voltage is much easier to measure, especially in a working tool, and it offers a good representation of a resulting current. In most situation EMI voltage is the parameter we will focus on.