Control system troubleshooting and PLC fault finding for process and power plants in the UAE, Saudi Arabia and across the GCC. We trace faults with a structured method, find the root cause and give you a clear report with corrective actions, so the same trip does not happen twice.
Control system troubleshooting is the disciplined search for why a plant does not do what its design says it should. A trip, a frozen reading, a valve that hunts or a network that drops out are symptoms. Our job is to find the cause quickly, restore production safely and stop the fault coming back.
Locating the failed part of a loop, panel, network or program. We work from the symptom back to the cause with measurements, not guesswork, so parts are not swapped blind and spares are not wasted.
Asking why the part failed. A blown fuse is a fault. The wet cable gland that let water reach the terminals is the root cause. We document the chain so your team can fix the system, not just the symptom.
Remote first-line diagnosis and engineers on site across the UAE and the wider GCC. We support Siemens PCS 7 and S7-400H systems and work across Rockwell, Schneider, ABB, Emerson, Honeywell and Yokogawa platforms.
PLC fault finding and DCS diagnostics share one principle. Every signal has a path, and every point on that path can be measured. A 4 to 20 mA input travels from the transmitter through a junction box, a marshalling cabinet and a barrier or relay to an I/O channel, then through scaling and logic to an HMI tag. If you can read the value at each point, you can find where it goes wrong. The same logic applies to a digital output, a PROFINET device or a Modbus register.
Experienced engineers look fast because they follow a fixed sequence. The sequence keeps people safe, avoids making the fault worse and produces evidence for the root cause analysis afterwards.
Most call-outs fall into ten fault classes. Knowing the typical failure for each class tells the engineer where to put the first test point.
| Fault class | Typical causes | How we diagnose it |
|---|---|---|
| Power and earthing | Low or rippled 24 V DC, failed redundancy module, tripped electronic breaker, shared 0 V, earth loops between instrument and power earth. | Measure under load at the far end. Check the redundancy module alarm contacts. Measure AC ripple and 0 V to earth. SITOP RED1200, E-T-A REX12D |
| I/O and wiring | Open circuit, short to screen, reversed polarity, loose or partly inserted push-in terminals, wrong channel configuration. | Read mA and V at the terminals. Compare with channel diagnostics for wire break and short circuit. NAMUR NE43 limits |
| Communications | PROFIBUS reflections, missing termination, PROFINET device name or IP conflicts, fibre losses, ring or PRP redundancy faults. | Bus analyser, oscilloscope, switch port statistics, fibre light meter, topology view. IEC 61158, IEC 62439-3 |
| Firmware and configuration | Mismatched firmware after a spare is fitted, wrong GSD or device description, hardware configuration not downloaded. | Compare online and offline projects. Read module identity data and revision levels. GSD, GSDML, EDD, FDI |
| Logic and sequences | Interlocks that block a start, wrong scaling, timers too short, latches never reset, unauthorised changes. | Online monitoring, cross-references, version compare against the last approved backup. IEC 61131-3 |
| Instruments | Drift, plugged impulse lines, wet DP legs, coated level probes, sensor failure reported over HART. | Loop calibrator, HART communicator, comparison with a second reading. NE107 status |
| Control valves | Stiction, deadband, positioner calibration lost, air supply problems, actuator spring or diaphragm failure. | Step tests, valve signature from a smart positioner, compare demand with travel feedback. Bump and step tests |
| Nuisance trips | Trips with no process cause: noisy transmitters, missing delay timers, setpoints too close to normal operation. | Sequence of events records, first-out logic, historian at high resolution. ISA-18.2, IEC 62682 |
| Intermittent faults | Loose connections, heat related failures, vibration, faults that clear on reset. | Data logging over days, thermal imaging, tap testing with the loop monitored. Trend at 100 ms or faster |
| EMC and noise | Spikes on analogue inputs when drives or radios operate, comms errors near VSDs, poor screen termination. | Oscilloscope, screen and earth checks, cable route survey against segregation rules. IEC 61000-6-2, NAMUR NE21 |
A fault that clears on reset is the hardest to find and the most expensive to ignore. We set up high-resolution trending, log the controller diagnostic buffer and, where needed, leave a data logger on the suspect loop for several days. Correlating the fault time with ambient temperature, a pump start, a radio transmission or a shift change usually points at the cause. Heat is a frequent factor in the Gulf, so we often check panel temperatures at the hottest part of the afternoon.
The right tool turns a guess into a reading. Our engineers carry calibrated field equipment and use the engineering software and built-in diagnostics already in your system.
A true RMS multimeter checks supply, ripple and continuity. A loop calibrator measures mA in series, sources a signal to drive an input or simulates a two-wire transmitter. Intrinsically safe models are used in hazardous areas.
HART rides a 1,200 baud FSK signal on the 4 to 20 mA loop and needs at least 250 ohms of loop resistance. A communicator reads the primary variable, range, damping, sensor diagnostics and device status without breaking the loop.
For PROFIBUS DP we check termination, cable length and signal quality with a bus analyser and oscilloscope. For PROFINET we use topology tools such as Siemens PRONETA, switch port counters and packet capture from a mirror port.
An S7 diagnostic buffer records module faults, redundancy events and stop causes with timestamps. Equivalent logs exist on every major platform. We read them before anyone resets the controller, then compare online code with the approved backup.
Smart devices report one of four NE107 signals: failure, function check, out of specification or maintenance required. Routing these to an asset management view tells maintenance which device needs attention before it trips the plant.
Managed switches such as SCALANCE log link changes, CRC errors and redundancy events. On PRP networks we check that both LAN A and LAN B carry traffic, because a silent failure of one path removes your redundancy without any process upset.
On TenneT's 980 MW HVDC offshore wind converter platform, built by McDermott at Jebel Ali in Dubai, we commissioned two redundant Siemens S7-400H systems with CPU 410-5H processors. One ran integrated monitoring and control, the other the power management system. ET 200SP HA distributed I/O ran over PROFINET on multimode fibre, with SCALANCE XF204-2BA DNA switches for PRP networks and redundant SITOP 24 V DC supplies. That depth of work on H-systems, fibre networks and redundant power is what we bring to every fault-finding call.
Restoring the plant is half the job. Root cause analysis explains why the failure happened, and loop tuning fixes control problems that look like equipment faults but are not.
Quick and suited to simple, single-chain failures. The pump tripped because the motor overloaded. It overloaded because the valve was shut. Keep asking until you reach a cause you can act on, usually a process, design or maintenance gap.
Groups possible causes under headings such as equipment, method, people, materials, environment and measurement. Useful in a workshop with operations and maintenance when several factors may have combined.
Fault tree analysis to IEC 61025 works down from the top event through AND and OR gates. The Apollo method maps cause and effect with evidence for each link. We use these for repeat trips and high consequence events.
A loop that oscillates, overshoots or responds slowly often gets reported as a faulty transmitter or valve. Before touching the PID settings we check the loop is sound: the transmitter is scaled correctly, damping is sensible and the valve moves when asked. A valve with stiction produces a sawtooth controller output and a square-wave process variable. No amount of tuning cures that. The valve needs repair.
Once the hardware is right, we run open-loop step tests to measure process gain, dead time and time constant, then calculate settings using lambda tuning or the SIMC method. Lambda tuning lets you choose a closed-loop response time, which suits level and pressure loops that interact with other units. We trend the result, document old and new settings and change them only under your management of change procedure.
Fault finding on a live plant is governed by the same standards as the original design. We work to them so that a repair never weakens safety or certification.
Safety instrumented systems in the process sector. Governs bypasses, proof testing and modification of safety functions during fault finding.
Root cause analysis guidance, covering methods such as Five Whys, Ishikawa and fault tree approaches.
Fault tree analysis, the formal top-down method we use for complex or repeat failures.
Self-monitoring and diagnosis of field devices with four standard status signals.
Failure signalling on 4 to 20 mA loops, with failure below 3.6 mA or above 21 mA.
Industrial communication networks, including PROFIBUS and PROFINET protocol definitions.
Parallel Redundancy Protocol (PRP) and HSR for zero recovery time networks.
Alarm management lifecycle, used to rationalise nuisance alarms and trips found during call-outs.
Inspection and maintenance of electrical installations in explosive atmospheres, including intrinsically safe loops.
A call-out runs to a fixed pattern, whether it starts with a phone call at night or a planned visit for a recurring fault. Response targets are agreed with each client and written into the support agreement.
Every visit leaves a record your maintenance team can use. Good reports turn a single repair into a lasting improvement in plant reliability.
| Document | Content | Reference |
|---|---|---|
| Service report | Symptom, timeline, readings, cause found, repair carried out and loop retest result. | Per call-out |
| Root cause analysis report | Method used, evidence, cause and effect chain, root causes and contributing factors. | IEC 62740 approach |
| Corrective action register | Corrective and preventive actions with owner, priority and target date. | CAPA list |
| Marked-up drawings | Loop diagrams, wiring and network drawings corrected to match what was found on site. | ISA-5.4, IEC 81346 |
| Program change record | Online changes, version compare and saved backup with reference number. | IEC 61131-3 project |
| Loop tuning record | Step test data, old and new PID settings, before and after trends. | Tuning sheet |
| Instrument test sheets | As-found and as-left readings from calibrators and HART communicators. | Calibration record |
| Bypass and force log | Every force or bypass applied, authorised by whom, and when it was removed. | IEC 61511 control |
Plants in the UAE, Saudi Arabia, Qatar, Oman, Kuwait and Bahrain fail in ways that plants in Europe rarely do. We look for the regional causes first.
Summer ambient temperatures above 45 °C push panel interiors well beyond component ratings when cooling fails. Power supplies, I/O modules and switches age faster. We check air conditioners, filters, fan alarms and panel temperature as part of every intermittent fault investigation.
Fine dust blocks filters and builds up on boards. Coastal sites in Dubai, Abu Dhabi, Doha and Jubail see high humidity and condensation overnight, which corrodes terminals and creates leakage paths. Salt-laden air attacks fibre connectors and earth bonds.
Many oil and gas sites are classified zones under IEC 60079, so we use certified test equipment and follow IEC 60079-17. We work to client requirements such as ADNOC specifications, Saudi Aramco SAES-J instrumentation standards and the site rules of QatarEnergy and PDO.
Outdoor work is also shaped by the climate. Summer midday work restrictions apply across much of the region, so we plan field tests for early morning or evening and use remote diagnostics to keep progress moving through the hottest hours.
Fault finding often leads to wider work on your systems.
See all our control systems, instrumentation and power services or contact our engineers.
Short answers to what clients ask us most. For anything specific to your plant, talk to an engineer.
Half-splitting means testing at a point near the middle of the signal path and deciding which half contains the fault. You then test the middle of the faulty half and repeat. It finds the fault in the fewest measurements, so a loop with eight stages needs at most three tests.
First check whether the input is below 3.6 mA, which NAMUR NE43 treats as a failure, or truly 0 mA, which suggests an open circuit or lost supply. Measure loop supply voltage, then put a loop calibrator in series at the marshalling cabinet to split the loop into field and system halves. Continue splitting until the open circuit, failed barrier or faulty transmitter is found.
NE107 defines four standard status signals for field devices: failure, function check, out of specification and maintenance required. They tell operators and maintenance whether a reading can be trusted and how urgently the device needs attention. Most HART, PROFIBUS PA and Foundation Fieldbus devices support them.
Common causes are valve stiction, a controller tuned too aggressively, interaction with another loop or a noisy measurement. A sawtooth controller output with a square-wave process variable points to stiction, which needs valve repair rather than retuning. We step test the process, fix any hardware problem and then retune using a method such as lambda tuning.
We capture evidence over time with high-resolution trends, controller diagnostic buffers and data loggers left on the suspect circuit. Matching the fault times to temperature, vibration, equipment starts or radio use usually reveals the cause. Thermal imaging and careful tap testing on terminals help confirm it.
Yes. We provide call-out support across the UAE and the wider GCC, starting with remote diagnosis where your site allows and mobilising engineers to site when needed. Response targets are agreed with each client and written into the support agreement.
Fault finding locates the failed part and restores the system. Root cause analysis asks why that part failed, using methods such as Five Whys, fishbone diagrams, fault tree analysis or Apollo RCFA. Without it, the same failure tends to return.
We have deep experience of Siemens PCS 7, S7-400H redundant systems, ET 200SP HA I/O and SCALANCE networks. We also work across Rockwell, Schneider, ABB, Emerson, Honeywell and Yokogawa platforms.
Tell us what the system is doing, what it should be doing and when it started. An engineer will review the details, advise on remote diagnosis or a site visit and send a clear scope and price. For urgent faults call +971 54 774 6012 or message us on WhatsApp.
