The short answer
Noise Problems
Noise reaches a control panel by conduction along wires that share a path, by capacitive coupling from fast voltage changes, by magnetic coupling from changing currents, and by radiation into wiring that acts as an antenna. A panel keeps it out with segregated wireways by signal class, a single ground bus with shields landed at one point, bonded doors and mounting plates, suppression on every relay and contactor coil, drives in their own section with shielded motor cable, and DC commons bonded once. When a panel is noisy, the search runs from the symptom to the coupling path: correlate the noise with a source, find where the source wiring and the victim wiring share a route or a ground, and fix that path rather than filtering the victim.
Key points
- Four paths: conducted, capacitive, magnetic, radiated. Each has its own signature and its own fix.
- Segregation by class inside the panel matters as much as outside; the wireway is where most coupling happens.
- Unsuppressed coils are the most common noise source inside a panel and the cheapest to fix.
- One ground bus, shields at one point, DC common bonded once, doors and plates bonded: the panel grounding checklist.
- Drives belong in their own section, with shielded motor cable and their control wiring routed away from the motor leads.
- Find the coupling path and fix it; a filter on the victim hides the problem and leaves it for the next signal.
How noise gets in
| Path | Mechanism | Signature | Defense |
|---|---|---|---|
| Conducted | Noise current shares a conductor with the signal: a common, a ground, a power supply | Affects every circuit on the shared conductor; correlates with a load on that conductor | Separate commons and supplies, single-point grounding, filtered power |
| Capacitive | Fast voltage change on a nearby conductor drives current through the capacitance between them | Affects high-impedance and voltage signals; worse with parallel length; drives and contactor arcs | Shields grounded at one end, separation, distance from drives |
| Magnetic | Changing current in a nearby conductor induces voltage in a loop | Spikes on motor starts, hum near transformers and heavy conductors; shields do little | Twisted pair, separation, steel conduit, avoiding loops in wiring |
| Radiated | Wiring acts as an antenna for radio frequency energy | Correlates with radios, modems, and antennas; steps rather than noise | Bonded enclosure, closed doors, shield bonded short at the entry, ferrites |
What a quiet panel has
- Wireways segregated by class: low-level signal and communications in one, discrete control in another, power and motor wiring in a third, with drive output wiring in its own path. Where classes must cross, they cross at right angles.
- Input and output cards placed so that signal wiring reaches them without passing through the power section.
- One ground bus bonded to the enclosure and to the equipment ground, with a shield termination bar tied to it at one point, and every shield landed there and insulated at the field end.
- The DC supply common bonded to the ground bus once, at the supply, and nowhere else.
- Suppression on every relay and contactor coil: a diode across DC coils, an RC snubber or varistor across AC coils, fitted at the coil.
- Drives in a separate section or enclosure, with shielded motor cable bonded at both ends, a short ground strap, and the control wiring to the drive in shielded pair routed away from the motor leads.
- Doors, back plates, and sub-panels bonded to the enclosure with straps, not through the hinges or the paint.
- Antennas and radios outside the panel, or in their own section with the antenna cable routed out directly and a bonded surge protector at the entry.
- Control power from a transformer or filtered supply rather than shared with motor loads, and surge protection at the panel feed.
The sources inside the panel
A panel can be wired perfectly and still be noisy because of what is in it. A relay coil that is de-energized without suppression produces a voltage spike of several hundred volts and a burst of radio frequency energy as the contact opens, and every time it happens the controller inputs on the same wireway see it. Contactors are worse. A drive without its motor cable shield bonded puts its common-mode current into the panel ground bus. A switching power supply with a failed filter capacitor puts its switching frequency on the DC bus that feeds every input. A radio inside the panel transmits into everything. Any of these produces symptoms that look like a field problem, and the fix is inside the panel.
Finding the path
- 1
Characterize the symptom
Which signals, what the noise looks like, when it happens. A single signal points at its own wiring or a coupling path; several signals on one card point at the card common or the wireway; every signal in the panel points at the ground, the DC supply, or a strong source.
- 2
Correlate with a source
Trend the worst signal at a fast rate and operate suspected sources one at a time: a drive, a contactor, a radio, a heater. The one that moves the trend is the source.
- 3
Find where they meet
Trace the source wiring and the victim wiring through the panel. A shared wireway, a shared common, a shared ground point, a shared power supply: that is the coupling path.
- 4
Check the grounding
Ground bus bonded; shields on the bar and open at the far end; DC common bonded once; doors and plates strapped. A clamp meter on the ground conductors and the shields shows current where there should be none.
- 5
Check the sources for suppression
Every coil with a suppressor, the drive with its shield and strap, the power supply clean on a scope.
- 6
Fix the path
Reroute, suppress, bond, separate. Then trend again with the source operating. Filtering the victim is the last resort and gets written on the drawing as a compromise.
Two classic cases
- The controller that resets when the big contactor drops out
- The contactor coil is on the same 120 VAC control circuit as the controller power supply, without suppression. The spike on coil de-energization dips the control voltage and the controller browns out. An RC snubber on the coil and a separate breaker for the controller supply fix it; a UPS for the controller is the second layer.
- The level that steps when the pump starts
- The level transmitter shield is landed on the field junction box and the panel, and the pump starting current changes the ground potential between them. The shield carries the difference, and the input sees it. Lifting the field end of the shield fixes the signal; correcting the ground path between the junction box and the panel prevents the next one.
Preventing it at design
The features that keep a panel quiet cost almost nothing when the panel is designed and a great deal afterward. The layout drawing shows wireways by class and places the cards accordingly; the bill of material includes coil suppressors, shield bars, bonding straps, and shielded drive cable; the specification tells the builder where shields terminate and where the DC common is bonded. A panel checkout before shipment verifies each item. A plant that gets those things in every panel spends its time on the process instead of on noise.
Frequently asked questions
- The panel was quiet when it was built. What changed?
- Usually a modification: a cable pulled through the nearest wireway, a relay added without a suppressor, a shield landed at a field box during a transmitter replacement, a radio installed inside the door. Panels accumulate noise with every change unless the rules are written down and followed.
- Is a separate instrument ground a good idea?
- A separate bar for shields and instrument commons, tied to the main panel ground at one point, is good practice; it keeps shield currents off the bus that carries everything else. A separate ground electrode for instruments, not bonded to the equipment ground, is a hazard and a noise source, and the code does not allow it.
- How do I tell a noise problem from a bad input card?
- Swap in a known-good card or move the loop to a spare channel. A card problem follows the card; a noise problem follows the wiring. Also look at whether the problem correlates with a source; card faults do not care what else is running.
- Where do I put a drive if the panel has to hold one?
- In its own section with a barrier, as far from the I/O as the enclosure allows, with the motor cable leaving directly through its own entry, a short ground strap to the bus, and its control wiring in shielded pair routed away from the motor terminals. Better still, in its own enclosure with a short shielded control cable between them.
Related topics
- Component LayoutArranging a panel so it is safe to work in, cool enough to run, quiet enough for its signals, and possible to modify: power and control separation, heat, wireways, terminals, and the working space rules.
- Control RelaysThe interposing and logic relays in a control panel: why a controller output drives a relay instead of a starter coil, contact ratings and the difference between resistive and inductive loads, coil suppression, ice cube relays versus terminal relays versus solid-state, and the failure modes that make a relay the first suspect in a dead circuit.
- Cable Routing ProblemsNoise that comes from where a cable runs: signal cables sharing a conduit, tray, or wireway with power or drive cables, long parallel runs, and shallow crossings. How coupling works, the separations that prevent it, and how to find the shared path.
- Drive Causes Instrument NoiseWhen many instruments became noisy after a drive was installed, the drive installation is the place to look: the checklist for motor cable, shielding, grounding, filters, and carrier frequency that decides how much noise a drive puts into the plant.
- Shield Grounded at Both EndsA shield connected to ground at the panel and again in the field becomes a conductor between two grounds, and its current becomes noise on the pair inside. How to find the second connection, what to measure before lifting it, and when both ends are correct.
- Radio Frequency InterferenceReadings that jump when a handheld radio is keyed, inputs that flicker near a telemetry antenna, transmitters that drift beside a modem: how radio energy gets into control circuits, how to prove it with the radio itself, and the fixes.
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.