Why PLC systems often fail EMC tests at interfaces

A PLC rarely fails EMC testing because its processor cannot tolerate RF. More often, the disturbance enters through an interface cable, changes reference potential and reaches sensitive circuitry by a path the design team did not model.

Why interfaces dominate PLC EMC failure

A programmable logic controller is an interface-heavy product. Even a compact unit may have DC power, protective earth, digital inputs, relay outputs, analogue channels, encoder connections, Ethernet, serial communications and an expansion bus. Each port creates a possible noise coupling path.

The enclosure may provide good shielding while the connected cable harness bypasses it. A cable can collect radiated energy, conduct switching noise from external equipment or carry common-mode current generated inside the PLC. Once that current reaches a connector, the result depends on filtering, shielding, PCB reference structure and the impedance of the return path.

This is why a PLC EMC failure often appears to move when cables are repositioned. The product has not changed, but the cable's effective antenna geometry, common-mode impedance and capacitive relationship to the reference plane have changed. Treating that movement as random behaviour usually leads to wasted debugging time.

Common-mode current is usually the first suspect

Differential signals are intended to travel along a defined conductor pair. Common-mode current flows in the same direction on multiple conductors and returns through chassis, earth, stray capacitance or nearby structures. Even a well-balanced industrial bus can carry common-mode current if its transceiver, connector or cable screen termination is asymmetric.

Small common-mode currents can produce significant radiated emissions when they reach a cable with a favourable electrical length. Conversely, during radiated or conducted RF immunity testing, the same cable can collect or receive RF energy and convert part of it into a differential disturbance at the receiver.

Filtering only the line-to-line path may therefore achieve little. A differential capacitor across a 24 V supply does not control RF current flowing from both supply conductors towards chassis. Common-mode chokes, feedthrough structures, cable-screen bonding and controlled chassis capacitance address different parts of the problem and must be selected according to the measured coupling path.

I/O protection can create an unintended RF route

PLC I/O ports normally contain surge, transient and electrostatic discharge protection. Those components are needed, but their placement and return connection determine where disturbance current flows.

A transient suppressor connected to digital ground can divert a fast interface disturbance through the logic reference before it reaches chassis. The protection device may survive while the PLC resets, corrupts an analogue conversion or drops a communications link. At high frequency, even a short PCB trace has inductance, so a protection path that looks satisfactory on the schematic may develop enough impedance to raise the local reference voltage.

Protection components also add capacitance and leakage. On analogue, high-speed or isolated interfaces, an unsuitable suppressor can degrade signal integrity or bridge an intended isolation barrier at RF. Component selection must consider normal signal voltage, transient waveform, parasitic capacitance, clamping behaviour and the physical current loop.

Isolation does not eliminate noise coupling

Galvanic isolation interrupts a low-frequency conductive path. It does not create infinite impedance at RF. Transformer interwinding capacitance, optocoupler capacitance, isolated DC-DC converter capacitance and PCB creepage structures all allow displacement current to cross the barrier.

An isolated I/O module can consequently pass common-mode noise from the field side into logic circuitry. The effect becomes worse when the isolated side floats without a deliberate RF reference. Its voltage then moves until parasitic capacitances establish a return path, sometimes through the least desirable part of the circuit.

Do not add a chassis capacitor without checking electrical safety, insulation coordination and leakage-current requirements. Where such coupling is permitted, component type, voltage rating, placement and connection inductance matter. A long route from capacitor to chassis can prevent it from acting as an effective RF return.

Cabling and grounding mechanisms to investigate first

Cable screens need a low-impedance termination

A screen terminated by a long pigtail may provide acceptable low-frequency bonding but performs poorly as frequency rises. Its inductance develops impedance, allowing screen current to create voltage at the connector and couple into internal conductors. A short, circumferential connection to chassis generally offers a lower-inductance RF path where the product construction permits it.

Whether a screen is bonded at one end or both ends cannot be decided from a universal rule. The answer depends on frequency, equipotential bonding, signal type, installation environment, safety requirements and the applicable manufacturer instructions. One-end bonding may control low-frequency loop current, while both-end bonding can provide better high-frequency screening. Industrial installations sometimes require additional bonding conductors so that cable screens are not forced to carry substantial power-frequency current.

Protective earth is not automatically an RF ground

A long protective-earth lead can meet safety requirements yet present considerable RF impedance. Connecting an EMC filter's chassis terminal through that lead may prevent the filter from returning high-frequency current locally. Noise then finds another route through I/O cables, communications ports or mounting hardware.

Chassis connections should be short and broad where practicable. Painted panels, anodised surfaces, loose fixings and narrow mounting rails can introduce uncertain impedance. The consequence is poor repeatability between bench testing and installation, even when the wiring diagram appears identical.

Cable layout changes the test article

Bundling a noisy relay-output cable with a thermocouple input creates capacitive and inductive coupling. Routing Ethernet beside a variable-speed drive output can expose the port to high common-mode voltage slew rates. Excess cable coiled on the reference plane changes inductance, capacitance and antenna efficiency.

Formal and pre-compliance configurations should follow the applicable product or product-family standard, test plan and representative installation instructions. IEC 61131-2 may be relevant to PLC equipment, while other industrial control products can fall within different standards or contractual requirements. The latest active edition, port applicability, operating modes, test levels, limits and performance criteria must be confirmed rather than inferred from a previous project.

Typical scenario

Consider an illustrative modular PLC with 24 V DC power, analogue inputs, relay outputs, Ethernet and a remote I/O connection. It operates normally on the engineering bench but resets during conducted RF immunity testing when a field cable is attached.

If the applicable product standard calls up IEC 61000-4-6, a coupling and decoupling network, or another injection arrangement permitted by the method, is used to apply conducted RF disturbance to the selected cable. A CDN must match the port and conductor arrangement. It is not interchangeable with a bulk current injection probe simply because both introduce RF onto cables. BCI methods are associated with applicable automotive, military, aerospace or product-specific procedures and require the prescribed probe, calibration fixture, placement and monitoring method.

The team should establish which port triggers the reset, whether the failure depends on frequency, and whether the PLC loses power, processor supervision or communications. Monitoring only the front-panel status LED is weak evidence. Analogue values, bus errors, output state, watchdog activity and supply rails may need simultaneous observation.

A current probe can help identify common-mode current on individual cable groups during engineering investigation. If bonding the remote I/O screen correctly removes the susceptibility, that points towards screen-current conversion or inadequate chassis termination. If a temporary local RF bond changes the result, the existing return path deserves inspection. Any temporary modification must be documented and assessed for safety before it informs a production change.

Early pre-compliance work allows the team to compare connector filtering, bonding and cable arrangements before tooling and PCB layouts are fixed. EMC Hire can support this through equipment hire, practical setup advice, on-site testing, access to test facilities and formal compliance testing where appropriate. Hiring also avoids buying a generator, amplifier, coupling network or monitoring equipment for a short investigation, together with the associated storage, servicing and calibration burden.

Building a defensible test and debugging process

Start with a port matrix. Record each physical interface, cable type, screen arrangement, maximum representative length, connected auxiliary equipment and operating mode. Map the applicable emissions and immunity requirements to each port. This prevents an Ethernet configuration being tested while the likely worst-case analogue or expansion interface remains inactive.

For conducted emissions on relevant power ports, use the LISN specified by the applicable method and confirm its impedance, current rating, voltage rating and frequency coverage. Conducted disturbance voltage is commonly investigated from 150 kHz to 30 MHz in many standards, but that range is not universal. A LISN is an emissions measurement network, not a conducted immunity device.

Radiated emissions require a suitable antenna, measurement receiver or analyser, controlled geometry and test-distance discipline. Peak scans can accelerate investigation, but final detector choice, resolution bandwidth and measurement procedure must follow the applicable standard. Quasi-peak and average results cannot be inferred reliably from an arbitrary peak trace.

Automation improves repeatability when it controls the correct instruments and preserves configuration data. EMC Hire provides information on RadiMation test software and Dare test software, alongside broader guidance on commercial EMC test systems. Software does not repair a poor test plan, but it can reduce transcription errors and make frequency, level, dwell and EUT response records easier to review.

Equipment used for relevant measurements should have suitable, current calibration. EMC Hire uses test equipment with calibration traceable through an appropriate ISO/IEC 17025 accredited calibration provider. This supports measurement accuracy, repeatability and comparison between development and formal testing. The accreditation relates to the calibration provider or calibration activity, not to the physical instrument or EMC Hire's test facility.

When to Hire EMC Equipment

Hiring is particularly effective when a PLC EMC failure has created a defined diagnostic window. A team may need a current probe and analyser for two weeks, an ESD simulator for design verification, or a complete conducted RF immunity chain for a scheduled test campaign. Purchasing equipment for irregular use ties up capital and leaves the owner responsible for maintenance, storage, firmware management and calibration.

Rental also reduces the risk of selecting equipment around one programme and finding it unsuitable for the next. Amplifier power, generator range, coupling-device ratings, receiver detectors and accessory configurations must match the intended method. A low purchase price does not help if the system cannot achieve the required level into the calibrated setup.

During project peaks, hired equipment can expand an internal laboratory without committing to permanent capacity. Where the setup is unfamiliar or the evidence will feed into formal compliance work, facility booking or on-site engineering support may be more efficient than equipment-only hire. The EMC test guides provide further background, although the current standard and product-specific test plan remain controlling documents.

Common EMC Testing Mistakes to Avoid

Testing an unrepresentative operating mode

An idle PLC may produce fewer emissions and show different immunity behaviour than one scanning high-speed inputs, switching outputs and exchanging maximum network traffic. Testing only the convenient mode creates false confidence and an evidence trail that does not represent foreseeable operation.

Changing cable geometry without recording it

Moving a harness can shift resonances and common-mode current distribution. If cable lengths, positions, terminations and connected loads are not photographed and recorded, a later retest may not reproduce the original failure.

Using the wrong coupling device

A CDN selected for the wrong number of conductors can alter signal operation or inject an unrepresentative disturbance. Substituting a BCI probe where the prescribed method requires a CDN changes the coupling mechanism and invalidates comparison with the specified procedure.

Routing protection current through logic ground

This often preserves the interface component while upsetting the processor. Inspect the complete current path from connector pin to chassis or return conductor, including trace inductance, vias and mounting structure.

Ignoring auxiliary equipment

Laptops, programming leads, unmanaged switches and laboratory power supplies can introduce emissions or provide unintended RF return paths. They may mask a weakness or create a false failure. Auxiliary equipment, isolation arrangements and cable configurations belong in the test record.

Keeping incomplete evidence

A screenshot without detector, bandwidth, transducer correction, cable arrangement and EUT mode is difficult to defend. Record instrument settings, calibration status, environmental conditions where relevant, software versions, observed performance and every modification made during debugging.

Frequently Asked Questions (FAQs)

Why does a PLC pass without I/O cables but fail when installed?

The attached cables create coupling paths and can act as antennas. They also connect the PLC to external reference structures, field devices and noise sources. Investigate common-mode current, cable-screen termination, chassis bonding and whether the installed cable lengths reflect the tested configuration.

Should every PLC cable screen be bonded at both ends?

Not automatically. Both-end bonding can improve high-frequency screening, but the installation's equipotential bonding, low-frequency loop currents, safety requirements and equipment instructions must be considered. The engineering decision should be documented rather than reduced to a blanket rule.

Can a clamp-on ferrite fix a PLC EMC failure?

It can help if its impedance is effective at the disturbance frequency and the cable passes through it in a way that targets the relevant common-mode current. A ferrite fitted without measurement may shift the problem rather than remove it. Saturation, temperature, cable current and production installation also need consideration.

Does passing pre-compliance testing prove conformity?

No. Pre-compliance testing provides engineering evidence and exposes weaknesses before a formal programme, but it does not automatically demonstrate compliance. The manufacturer remains responsible for identifying applicable legislation, standards, conformity assessment routes and technical file requirements.

What should be monitored during PLC immunity testing?

Monitor functions tied to the specified performance criteria, such as I/O accuracy, output state, scan operation, communications errors, watchdog events and recovery behaviour. The monitoring system must not create an additional coupling path or suppress the disturbance being investigated.

Plan the next investigation

A useful PLC investigation starts with the interfaces, representative cabling and return-current paths, not with indiscriminate component substitution. Calibrated pre-compliance measurements can narrow the mechanism, reduce late-stage redesign risk and build stronger data for a technical file, Declaration of Conformity or other self-certification evidence where that route is legally and technically appropriate.

For equipment hire, on-site testing, pre-compliance support, formal compliance testing or space at the EMC Hire test facility, speak with the engineering team on +44 (0)1462 817111 or email sales@emchire.co.uk. The discussion can cover the applicable interface, test method, frequency range, coupling device, monitoring requirement and equipment needed for the planned test window.