Grounding the 0V DC reg

[vjr0512]

There is a thyristor control panel for a Heater in one of the package and is mounted remote in a conditioned room. The panel vendor has derived 24V DC power for the package instrumentation ( Temperature transmitters). There has been a comment from Client that the 0V (-24V DC) has to be grounded.

Hence package vendor is proposing to connecting it to PE.

There has been objection from Instrumentation specialist that the PE being for AC , connecting the DC is not to be done as very high fault current if any in the AC power system may induce spikes in the DC system also. Vendor informs that it is not a common practice to separate 0V DC from AC ground. They are common and connected to the back panel.

My Queries

If the Individual Instruments in the package ( Transmitters) are already earthed separately, is there a necessity to ground the 0V (-24VDC) also?

Who is right in terms of connecting the AC and DC together to the common earth?

Thanks to clarify.

 

[LiteYear]

Technically it depends on how the DC is derived from the AC, but generally speaking your specialist is correct - if the AC is connected to PE, connecting the DC to PE can cause a high fault current and other stability issues. The reason being that generally the rectification method means that the 0V DC is not at the same voltage as the 0V AC. It is indeed quite common, in certain industries, to separate 0V DC from AC ground.

 

[Skogsgurra]

It depends a lot on what is "P.C." for the time being. Opinions have changed over the decades. Insulated was the rule for a long time. But that became more and more difficult to maintain in practice and ABB/ASEA started using the opposite rule (ground everything everywhere, Rudolf Anders and Sten Benda) in the seventies. Siemens stayed insulated until it wasn't possible any more. The change came in the eighties when VFDs were introduced and it was necessary to connect everything to ground potential to avoid interference in the systems. That grounding/bonding with necessity included all DC supplies and they are mostly connected tp GND/PE today. The ideal situation is to have a huge ground plane where everything is connected. That is, for practical reasons, usually not possible. But the next best thing, a mashed ground grid with averything connected to it, is good enough in most cases.

Grounding and bonding is the rule today. There are some exercises trying to use separate PE, TE, GND and N systems and they are sometimes successful. But there is usually a common point where they are all connected together. Except for IT grids, where PE and N are always separate.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Isolated grounds always scare me. In the event of a serious ground fault causing potential gradients in the earth, a potentially lethal voltage may be transferred to exposed conductive surfaces of equipment on the isolated ground.
Potentially lethal touch and step voltages are a known hazard. Isolated grounding electrodes may be subject to gradient induced voltages several times higher than the corresponding touch and step potentials and conduct those lethal voltages to exposed metal.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[controlsdude]

Allen bradley website
Publication 1770-4.1 – February 1998

I think there are too many what if conditions on grounding dc common. Worth a read on if your applies to an example.