Grounding Cable Shields on 4-20ma Instruments

4-20 ma current loop is pretty resistance to interference. One thing that could cause interference is current flowing in the shield. So if you ground the shield at either end but not both ends (a ground loop), you should be OK.

I was leaning that way too - I thought that either end should work. I just wonder what we do on the next plastic housed instrument we get. (few and far between, but hey do exist)

The possible problem is the introduction of a ground loop. This can increase your power distribution (60 Hz) noise on this or another sensor. Every once in a great while the phase of the induced current is nearly opposite the power line noise from another path thus reducing the measured interference instead of increasing it.

I have experienced the same wiring method and I have the same concerns that you have.

In some cases where there is a remote possibility of an electrical fault such as an arc flash occurring and stray medium or high voltage could enter the instrument wiring, there are actually two separate grounding systems.

This system has noise shielding with a drain wire that is connected at the panel end and also an outer braided ground shield for personnel & equipment protection that functions as part of the equipment grounding system which is grounded on both ends.

It is my understanding that with this method there must be a ground rod or ground grid at the instrument location and the instrument ground must be bonded/attached.

To me this presents the possibility of creating circulating ground current and/or induced voltage in the instrument signal wires.

Grounding either end is applicable as long as equipotential is established between the control panel and the instrument. Possibility of Any ground loop is thereby minimized, eliminated when both sides (instrument and control panel) are sitting on the same ground plane.... In my rusting experience of instrumentation... God luck.

Then don't show the shield on the drawings. After all, there is mucky 110VAC mains travelling on two of the cores within the vendor cable!

4-20mA and simple switch contacts simply don't need shields.

The 110VAC (120 here) is in the power cabinet - never in the 24VDC control cabinet and never run together in wire ways (trays) with the DC stuff. Except for a purchasing error, no instrument is ever powered with 120VAC - only 24VDC. (occasionally there is a slip up and a flow meter gets ordered in 120VAC power) Usually, the 120VAC powers small pumps and UV lights and such.

I agree on the switched 24VDC, no shield necessary, but I think you do need a shield on the 4-20ma and definitely on RTD signals. Our customers all believe we need the shields, so we provide them.

The really mucky thing is the VFD power to the pumps (2 to 3 of them) on the skids and the mixer on the tanks. Ever put a scope on that output? You have to wonder how the motor functions with that noise. I've heard several motors over the last 25 years that made a whistling sound when the VFD was powering them. Noise.

Unfortunately, it is common practice here in this little niche of industry to show a loop around the conductors of any cable coming to terminals in the control panel with a line from the loop to PE. Of course all this is in a nice bright green. Tough to avoid showing that when it is on ALL the drawings we do with signals.