Symptom
Hum or a shifting offset on an analog signal, noise that appears when large loads switch, a warm or discolored drain wire, or a burned shield after a storm; the shield at the panel measures continuity to ground after it has been lifted.
The short answer
Shield Grounded at Both Ends
For instrument signal cables the shield is grounded at one end only, normally the panel end, and insulated at the field end. A shield grounded at both ends carries whatever current the potential difference between the two grounds drives through it, and that current couples into the signal conductors as hum and offsets, and in a lightning or fault event can burn the drain wire. Find the second connection by lifting the shield at the panel and measuring resistance from shield to ground, which should be open; measure the shield current with a clamp first, because the value tells you how large the ground potential difference is. High-frequency cables such as Ethernet, fieldbus, and drive motor cables are bonded at both ends by design and are the exception.
Key points
- Instrument cable shields: one ground, at the panel, with the field end insulated and taped.
- Both ends grounded turns the shield into a ground conductor, and its current couples into the pair.
- Test by lifting the panel end: shield to ground should read open. Clamp the shield for current before lifting.
- The second connection is usually at the transmitter terminal, in a junction box, or a bare drain touching a gland.
- Ethernet, fieldbus, and drive motor cables are bonded at both ends on purpose; do not apply the instrument rule to them.
Possible causes and what to check
| Possible cause | What to check |
|---|---|
| Drain wire landed on the transmitter ground or shield terminal in the field | Open the transmitter housing and look. Many transmitters have a shield terminal connected to the case; the manual says whether it is isolated. |
| Bare drain wire touching the housing inside a cable gland or conduit fitting | Inspect where the cable enters the housing. A drain wire that is not sleeved finds metal. |
| Shield landed on the ground bar in a junction box | Junction boxes between field and panel are the most common place for a second ground; shields should be carried through on an isolated terminal. |
| Shield connected to ground through a connector shell or a metal-bodied terminal | Connector shells and some terminal styles bond the shield to the mounting rail. Check with the cable disconnected. |
| Shield continued through a splice to a second cable that is grounded | Trace splices; every shield segment should be continuous to the panel and open to ground everywhere else. |
| Panel end grounded twice, at the entry gland and at the terminal | Grounding twice inside one panel is harmless if both points are on the same bus; it is a loop only if they are on different grounds. |
| Large ground potential difference making even a small shield leak matter | Measure AC and DC volts between the panel ground and the field ground after lifting the shield. Several volts is a grounding system problem to fix as well. |
What a shield is for
A shield around a twisted pair is a barrier against electric field coupling. Grounded at one end it holds the field at ground potential without carrying current; capacitive noise that would have reached the conductors is drained to ground instead. It is not there to carry current, and it is not part of the signal circuit. Grounded at both ends it is a conductor in parallel with the plant grounding system, and whatever voltage exists between the two ground points drives current through it. At 60 Hz that current produces a magnetic field around the shield that induces voltage in the pair inside, and the shield resistance times the current appears as a voltage at the signal end if the shield is also the reference. In a lightning event the shield tries to carry the surge current, and a drain wire is not sized for that.
Which end
The convention for instrument signals is the panel end, because the panel ground bus is the reference for the input circuits and is the better ground. The field end is cut back, insulated with heat shrink or tape, and left unconnected. At a junction box the shield passes through on an isolated terminal, still with one ground at the panel. A transmitter with a metal housing sees the shield insulated inside the housing, not landed on the housing ground.
| Cable | Shield practice | Reason |
|---|---|---|
| 4-20 mA, HART, RTD, thermocouple, 0-10 V | One end, at the panel | Low-frequency signals; the shield is a capacitive barrier, not a return |
| RS-485 and RS-232 | One end as a rule; both only where the manufacturer specifies and the grounds are bonded | Common-mode limits are small; ground current on the shield is worse than the noise it blocks |
| Shielded Ethernet | Both ends, through the connector shells | High-frequency shield performance requires low impedance at both ends |
| Fieldbus such as Profibus DP | Both ends, 360 degree clamp at every device | Manufacturer requirement; the network depends on it |
| Drive to motor cable | Both ends, 360 degree bond at the drive and the motor | The shield is the return path for high-frequency drive current, keeping it off the plant ground |
| Coaxial antenna feed | Bonded at the entry to the building | Lightning protection, per the radio and surge protector instructions |
Diagnostic procedure
- 1
Clamp the shield
Before lifting anything, put a sensitive clamp meter around the drain wire at the panel end. A shield with one ground carries no measurable current. Tens of milliamps or more means it is a conductor between two grounds.
- 2
Lift the panel end
Disconnect the shield drain from the panel ground and measure resistance from the drain to ground. Open, above a megohm, means the field end is insulated. Continuity means there is a second connection somewhere.
- 3
Measure the potential
With the shield lifted, measure AC and DC volts from the drain to the panel ground. That is the voltage between the field ground and the panel ground. A few hundred millivolts is normal; several volts means the grounding system has a fault worth finding.
- 4
Find the second connection
Work from the panel toward the field: junction boxes, splices, glands, the transmitter housing. Disconnect the shield at each point and measure again toward the field until the continuity disappears.
- 5
Insulate it
Cut the drain back, sleeve it, and tape or heat shrink so it cannot touch metal. Reconnect the panel end.
- 6
Verify
Clamp the shield again; the current should be gone. Watch the signal for the noise or offset that started the investigation, with the loads that made it worse running.
Fixing the cause, not just the loop
A shield grounded at both ends only causes trouble when there is a potential difference between the grounds. A large difference means the grounding system is carrying current it should not: neutral current from a wrong bond, drive current returning through the earth, or two buildings with no bonding between them. Insulating the shield end removes the symptom on one signal; correcting the grounding system removes it for every signal in the plant.
Frequently asked questions
- The manufacturer says to ground the shield at the transmitter. Who is right?
- Follow the manufacturer for that device, and then do not ground the same shield at the panel. The rule is one ground per shield, and some transmitters, particularly ones with radio frequency protection built in, are designed for the shield to terminate at the device. What is never right is both.
- Is it acceptable to leave a shield unconnected at both ends?
- It does nothing then, and the pair is unshielded. Connect it at one end. A shield with no ground can also float up to a voltage through capacitance and give a surprising tingle from a long cable.
- Why are drive motor cables bonded at both ends when instrument cables are not?
- Because the motor cable shield has a different job: it is the return path for the high-frequency current the drive pushes through the motor cable capacitance. Bonded at both ends with a 360 degree connection, that current stays in the cable instead of returning through the plant ground, where it would raise the potential differences that cause instrument problems.
- Can I ground shields on a common bar with everything else in the panel?
- Yes, if that bar is bonded to the panel ground bus, which is the reference for the analog inputs. Some designs use a separate instrument ground bar tied to the main bus at one point; that is fine too. The point is one connection to one ground per shield.
Related topics
- Ground Loop SymptomsHow a ground loop shows itself: hum on analog signals, offsets that change when large loads switch, serial links that error, and readings that differ by location. The mechanism, the measurements that prove it, and fixes that never lift a safety ground.
- Ground Loops in InstrumentationTwo grounds at different potentials, one signal circuit between them, and a reading that will not sit still. How to find it and how to prevent it.
- How to Diagnose a Ground LoopProve that a noisy or offset analog signal is a ground loop, find the second ground, and fix it the right way: remove the extra ground, isolate the signal, or correct the shield termination. With the measurements that make the diagnosis certain.
- 4-20 mA Signal Is UnstableAn analog reading that will not settle. How to separate a genuine process swing from induced noise, a ground loop, or a loop running out of voltage.
- Surge Protection for Signal CircuitsProtecting 4-20 mA loops, discrete inputs, network cables, and radio feeds from lightning and switching surges: where surges enter, the device types and how they clamp, the grounding that makes them work, the resistance and capacitance they add, and where to put them at a remote site.
- Drive Trips on Ground FaultA ground fault trip on a variable frequency drive: how the drive detects it, the motor and cable insulation faults behind most trips, the submersible cable problem, the capacitive leakage on long cables and filters that produces nuisance trips, wiring errors, and the insulation test that decides it.
Direct contact
Have a controls question?
Reach Eric Sullivan directly about anything on this site, a controls or automation topic, or one of his personal projects.