ELECTRICAL NOISE

Since the advent of solid state instrumentation, many customers have experienced difficulty in making these devices operate in their applications due to electrical noise. Microprocessor-based controls often are the most susceptible to this problem, which is inherent in any electrical system where computers, variable speed drives, solid state relay switching and switching power supplies operate, or where inductive loads such as contactors, solenoids and relays operate. In addition to the voltage spikes (reactive resonance creation, usually involving capacitors) that these devices generate, harmonics problems can also ensue. The effect can be random memory loss, non-planned resets, loss of setpoint etc in the instrument. Electro-mechanical instruments are generally much more tolerant of these electrical deviants. Also, some brands of solid state instruments seem to be much better than others due to superior input filtering networks.

Voltage Spikes

There are some simple, inexpensive steps that can be taken to minimize noise problems due to voltage spikes, which usually occur in less than a 1/2 cycle. Several can occur at the same time, and the impulse can be therefore additive to the power voltage waveform and can cause capacitors, inductors and transformers to fail. The simple solution is to add noise suppressors to the control circuit of the affected instrument. Several types of noise suppressors are available, including Metal Oxide Varistors (MOV's) and resistor/capacitor networks. The latter are often packaged as "Quencharcs" or R/C Snubber noise and Arc Suppressors. In general, we recommend the use of R/C networks over MOV'S. One type of Quencharc is the Danaher #PBN-2002 (Paktron 104MO6QC100), and it consists of a 2 watt, 100 ohm resistor in series with a 0.1 microFared capacitor (600V).

Ideally, the noise should be suppressed at its source. This usually means that suppressors should be placed across the coils of all inductive loads in the control system. To help suppress any arcing noise, it would also be desirable to have suppression across the contacts of any relays, contactors, starters, or switches.

In order to cut costs involved, it should be determined which devices are switching when the instruments malfunction. It might be possible to suppress the noise problem by taking care of only these components. The suppressor would then be connected across the switching contacts. If the noise is random, the noise suppressors can be put on the input of the affected instrument. The first place to try a suppressor is across the power connections to the processor. If this does not eliminate the problem, a second step is to place suppression on any other secondary inputs (remote reset, start etc terminals). Please refer to the attached installation considerations, Microprocessor Based Installation Practices, and Inductive Load Suppressor, R/C Snubber Noise and Arc Suppressor information.

Electro-Magnetic Interference

The number one cause of erratic instrument operation is EMI. An inexpensive solution to this problem is to install a Line Filter, installed close to the instrument and earthed to the power line round. Please refer to the LFIL attached. Another solution might be the Ferite Suppression Core (see attached). This device attenuates the EMI in the 25 MHz to 200 MHz range.

Any non-linear load, such as a variable speed drive, creates harmonic voltages by the flow of their harmonic currents through the wiring system's impedance's. Any current flowing through any wiring impedance develops a voltage drop, occurring at each harmonic frequency and in proportion to the harmonic current's frequency. This is sent down the pike to all downstream points on the wiring system.