Discussion on the grounding requirements for signal circuits in HG specifications

Thread Content

This post was last edited by kylestone on 2017-4-27 at 14:13. An important concept here is that of isolated signals and non-isolated signals. After doing some research online and asking people around me, I found that many people are not clear about this concept; some think that signals that have passed through isolation transformers are non-isolated signals, but I don’t agree with that. The isolation signal specification states that they are \"mutually insulated, insulated from ground, and have independent power supplies.\" So I understand that, for example, any 2-wire analog signal instrument connected directly to a DCS card via two wires, whether there is an isolation barrier in between or not, constitutes an isolated signal circuit; this is because such a circuit is completely independent, and the electronics within it only go into the DCS card and never anywhere else. In actual engineering projects, such instruments are grounded only through their enclosures and the cable shielding layers; no other grounding is performed. Any instrument that outputs dry contacts, and which then uses a relay to connect to the DCS, requires 24VDC+ busbars and 24VDC- busbars for operation. Obviously, these busbars share the same power source, and there is no insulation between the circuits; electrons in one circuit can move to another circuit through the busbars. This is what is referred to as a non-isolated signal. The 24VDC- terminal must be grounded, and this is indeed how it is done in practical engineering applications; hence it is also called the COM busbar. Since in actual engineering projects the negative terminal of the DC power supply is always connected to ground, it is as if all signal circuits that use this negative terminal have already been grounded. Therefore, even though many people are not entirely clear about this concept, there will be no problems following standard design practices.