Symptom
The analog value on the HMI moves continuously, jumps erratically, or wanders slowly without a corresponding change in the process.
The short answer
4-20 mA Signal Is Unstable
An unstable 4-20 mA reading is usually one of four things: a genuine process fluctuation, induced electrical noise from nearby drives or motors, a ground loop from the circuit being earthed at more than one point, or insufficient loop voltage for the total loop resistance. Determine which by observing whether the instability correlates with equipment operation and whether it appears at the transmitter or only at the panel.
Key points
- First establish whether the process itself is actually moving.
- Noise that correlates with a drive running points to induced interference or a ground loop.
- Measure at the transmitter and at the panel to locate which segment is adding the instability.
- Instability only near full scale suggests insufficient loop voltage.
- Filtering in the controller hides the symptom and should be the last step, not the first.
Possible causes and what to check
| Possible cause | What to check |
|---|---|
| The process is genuinely fluctuating | Compare against a local gauge or a second measurement. A pump cycling or turbulence at a probe produces real movement. |
| Induced noise from a VFD or motor cable | Observe whether the instability starts when a drive runs. Check signal cable routing and separation from drive output conductors. |
| Ground loop | Measure AC voltage between the panel ground and the field device ground. Confirm the cable shield is grounded at one end only. |
| Insufficient loop voltage | Calculate total loop resistance against supply voltage and transmitter compliance. Instability or clipping near 20 mA is the signature. |
| Loose or corroded terminal | Inspect and re-torque every termination in the loop. Flex the cable at each termination while watching the reading. |
| Failing transmitter or power supply | Substitute a loop calibrator for the transmitter. If the reading stabilizes, the transmitter is the source. |
| Turbulence or installation position | For flow and level, check whether the sensing point sees a disturbed flow profile or a falling stream. |
| Analog card channel fault | Move the loop to a spare channel and observe. This isolates the input hardware in two minutes. |
Start by characterizing the instability
Before touching a wire, describe what the signal is doing. The shape of the instability narrows the cause more than any single measurement will.
| What it looks like | Points toward |
|---|---|
| Fast, small, constant fuzz | Induced electrical noise |
| Large jumps that coincide with a motor start | Ground loop or severe induced noise |
| Slow wandering over minutes | Genuine process movement, or a thermal effect |
| Steady until the reading approaches full scale | Insufficient loop voltage |
| Intermittent dropouts to zero | Loose termination or an intermittent open |
| Stepped movement rather than smooth | Poor resolution, or heavy filtering already applied |
| Only unstable in wet weather | Moisture in a junction box or conduit |
Diagnostic procedure
- 1
Confirm the process is not actually moving
Check a local gauge, a sight glass, or a second instrument. It sounds obvious and it is regularly the answer, particularly on discharge pressure with a pump cycling or on a level probe near an inflow stream.
- 2
Correlate with equipment operation
Note whether the instability starts and stops with a specific drive, pump, or blower. This one observation separates electrical interference from everything else.
- 3
Measure at the panel non-intrusively
Read voltage across the input sense resistor. If it is unstable here, the instability is real on the wire and you continue outward. If it is stable here while the HMI is not, the problem is in scaling, filtering, or the SCADA tag.
- 4
Measure at the transmitter
Use test jacks or a series measurement at the transmitter. Stable at the transmitter and unstable at the panel means the cable run is picking up the disturbance. Unstable at both means the transmitter or the process is the source.
- 5
Substitute a calibrator
Disconnect the transmitter and inject a fixed 12 mA. If the panel reading is now rock steady, the transmitter or the process is responsible. If it is still unstable, the wiring, the card, or a ground problem owns it.
- 6
Check grounding
Verify the shield is grounded at exactly one end. Measure AC voltage between the two ground references. More than a few hundred millivolts is worth pursuing.
- 7
Check loop voltage headroom
Calculate maximum allowable loop resistance from supply voltage and transmitter compliance voltage, and compare against the actual total. Marginal headroom produces instability that gets worse as the signal rises.
- 8
Move to a spare channel
If everything above is clean, move the loop to a different analog input. A failing channel is uncommon but not rare, and this test is quick and conclusive.
Common mistakes
- Adding filtering in the PLC first. It makes the display look better and hides a genuine fault that will get worse.
- Replacing the transmitter before proving the transmitter is the source. It is the most expensive step and frequently the wrong one.
- Grounding the shield at both ends "to make sure it is grounded". This creates the ground loop you are trying to eliminate.
- Testing with the drive stopped and declaring the problem solved. Reproduce the condition before believing a fix.
- Ignoring a small persistent instability. A signal that has become slightly noisy over months is often a termination that is slowly failing.
Fixes, in order of preference
- 01Correct the actual cause: re-route the cable, re-terminate the connection, fix the shield grounding, or raise the loop supply voltage.
- 02Add a signal isolator where a second ground cannot be removed.
- 03Improve the installation: separate the signal cable from drive output conductors, use proper twisted shielded pair, and add distance where possible.
- 04Move the sensing point if turbulence or installation position is the cause.
- 05Only then, apply modest filtering in the controller, and document that you did and why.
Frequently asked questions
- Why does my analog signal only get noisy when a pump runs?
- Either the drive output cable is coupling into the signal cable, or the motor current is exciting a ground loop. Check cable routing and separation first, then shield grounding, then add an isolator if the second ground cannot be removed.
- How much filtering is acceptable?
- Enough to remove measurement noise without hiding a real process change the control loop needs to see. On a wet well level a few seconds is usually harmless. On a pressure loop controlling a pump, heavy filtering will destabilize the control.
- Can a bad power supply cause an unstable analog signal?
- Yes. A loop supply with excessive ripple or one operating near its current limit puts that variation directly into the loop. Measure the supply output with a meter on AC volts to check for ripple, and confirm it is not overloaded.
- The reading is stable at the panel but jumps on the HMI. What now?
- The field is fine. Look at the controller scaling, any filtering, the update rate between controller and SCADA, and the SCADA tag configuration. A slow poll rate showing a fast-changing value can look like instability that is not there.
Related topics
- 4-20 mA Current LoopsWhy the industry standardized on current instead of voltage, how a two-wire loop is powered, what the live zero buys you, and how to read a loop with a meter.
- 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 Test a 4-20 mA LoopProve a current loop end to end with a meter and a calibrator, without interrupting a process you were not authorized to interrupt.
- VFD Noise on Analog SignalsAnalog readings that jump or wander when a drive runs, and speed references that make a drive hunt: how the drive output couples into signal wiring, why the input turns kilohertz noise into a slow wander, the tests that pin it on the drive, and the fixes.
- Transmitter Reads Wrong ValueThe reading is steady, plausible, and wrong. How to prove it against a reference, then separate a scaling mismatch, a range change, a zero shift, an installation effect, and a genuine transmitter fault, with the order of checks that finds it fastest.
- Signal Pegged High or LowA 4-20 mA signal stuck at a rail or a fault current: a transmitter in fault, a process truly out of range, an open or shorted loop, a wrong range, a wiring error, an input channel fault, or a scaling clamp. How the current itself tells you which.
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.