Industrial networks using RS-485 transceivers can be subject to electrical fast transient (EFT) bombardment on a daily base. These transients, typically occurring in the form of bursts, originate from switching transients due to the interruption (switching) of inductive loads, relay contact bounce, etc. These transients can corrupt the data between transmitting bus nodes and even damage transceiver devices, resulting in network downtime. Device internal transient suppression circuits are therefore necessary to ensure a fast recovery from EFT events in order to maintain network operation.
When de-energized, inductive loads such as relays, switch contactors, or heavy-duty motors produce bursts of narrow high-frequency transients on the power distribution system. These fast transients can also be produced when the utility provider switches in or out the power factor correction equipment. A common cause of power line transients is sparking that occurs whenever an AC power cord is plugged in, equipment is switched off, or when circuit breakers are opened or closed (see Figure 1).
EFT Immunity Test Standard
To assess a system’s immunity to EFT bursts, the IEC commission developed standard IEC 61000-4-4 which defines the test voltage waveform, range of test levels, test equipment, test set-up, and test procedure.
Figure 2 depicts the waveform of a single EFT pulse and Figure 3 illustrates the sequence of test pulses during a typical EFT test. Here, the EFT pulse generator creates a sequence of low-energy pulses known as a burst. Each pulse has a rise time of 5ns and a pulse duration (time-to-half value) of 50ns. The burst period is 300ms and includes 75 transients followed by a pause interval. The test allows for two repetition rates of 5kHz and 100kHz. For a pulse repetition rate of 5kHz, the 75 pulses take 15ms, while for a repetition rate of 100kHz, they only take 0.75ms.
Figure 3 depicts the minimum required test duration is two minutes and includes three 10-second windows of positive pulses, each followed by a 10-second pause interval, and three 10-second windows of negative pulses, with a 10-second pause interval. This results in a total of 15000 positive and 15000 negative pulses during a two-minute test sequence.
While the individual pulse represents a low-energy transient, the energy of an entire pulse train does not. For a given test voltage the energy of an EFT pulse train is about 250-times higher than the energy of a single pulse.
To test the EFT immunity of transceiver data ports, the EFT test pulses are coupled into a point-to-point data link via a capacitive clamp (Figure 4). This clamp surrounds both data lines, thus representing a common-mode test.
Figure 5 shows the five test levels for data ports. The first four levels double in voltage with each level increment. The fifth level, x, is a special level that can assume any test voltage. However, the voltage level must be specified in the manufacturer’s data sheet.
EFT Immunity of RAA78815x Transceivers
The RAA78815x transceivers were tested with up to 5kV EFT transients. Before and after an EFT test, the bus and receiver output signals of the transmitting and the receiving nodes were observed for visual comparison.
In addition to the visual confirmation of normal transceiver operation, all transceivers were tested on an Automatic Test System (ATE) for parametric performance. The pass criterion required that a device did not show any parametric shift. The results showed, the RAA78815x family of 5V RS-485/RS-422 transceivers passed all EFT tests with 5kV test voltage, the highest possible test voltage of the AXOS-5 test system, which places this transceiver family into the highest special test level category (Figure 6).