When a PLC stops working, production stops. Every minute of downtime costs money, and the pressure is on the engineer to diagnose and fix the fault fast. Experienced maintenance engineers use systematic methodologies to isolate faults quickly — methodologies that can be learned, not just developed over years of trial and error.
This guide teaches you the professional troubleshooting workflow used at UK manufacturing plants, water utilities, and process facilities. Apply it consistently and you will diagnose PLC faults faster than colleagues who rely on intuition alone.
The 6-Step Troubleshooting Framework
Step 1: Observe Before Acting
Before touching anything, observe:
- What is the symptom? "Conveyor stopped" is different from "conveyor running but no parts coming out."
- When did it start? After a specific event? At a specific time? After a change?
- Who saw it first? Get the operator's exact description.
- What is the alarm history? Check the last 30 alarms in the HMI/SCADA before doing anything else.
Rushing at this stage leads to hours of wasted diagnosis on the wrong problem.
Step 2: Check PLC Status LEDs
Every PLC has LED indicators for a reason. Read them before doing anything else.
Siemens S7-1500:
- RUN/STOP: Green = running, Yellow = stopped, Red flashing = fault
- ERROR: Red = hardware or diagnostic fault
- MAINT: Yellow = maintenance recommended
- LINK: Green = network connected
Allen-Bradley CompactLogix:
- OK: Green = OK, Flashing Red = major fault, Solid Red = hardware fault
- RUN: Green = running program
- I/O: Green = I/O comms OK, Flashing = I/O fault
Common I/O Module LEDs:
- Green/Yellow on input/output channel: signal present
- Red on module: fault or diagnostic issue
- No LEDs: no power or wiring fault
Step 3: Check the Diagnostic Buffer / Logix Controller Log
Siemens TIA Portal:
- Connect online to the PLC.
- Open Diagnostic Buffer (Online → Online & Diagnostics).
- Read the most recent entries first — they are listed in reverse chronological order.
- Common entries: module fault, communication loss, CPU stop, diagnostic interrupt OB not programmed.
Studio 5000:
- Connect online.
- Controller Properties → Major Faults tab.
- Check the last fault code and description.
- Also check the Program Fault History.
The diagnostic buffer alone solves 40% of faults if you read it carefully.
Step 4: Trace Signal Flow
If the diagnostic buffer doesn't reveal the issue, trace the signal:
- Field device → I/O module → CPU logic → Output module → Field actuator.
- Confirm each stage independently.
At the field device:
- Multimeter check: Is the sensor providing the expected voltage or current?
- Physical check: Is the sensor correctly positioned? Is it powered?
At the I/O module:
- Is the corresponding LED lit when expected?
- With a multimeter, measure voltage at the input terminal.
- Is the module in a fault state?
In the CPU program:
- In TIA Portal or Studio 5000, go online and watch the tag value.
- Does the input tag reflect the field state?
- If input is correct but output isn't triggering, suspect logic.
At the output module:
- Is the output terminal receiving a command?
- Is the LED lit?
- Measure voltage — is the signal going to the field?
At the field actuator:
- Is the actuator receiving the command?
- Is it functioning (e.g., valve opening, motor running)?
Common PLC Fault Categories
1. Power Faults
Symptoms: Dead PLC, random restarts, intermittent operation. Diagnosis: Measure 24V DC bus under load. Check UPS status. Check transformer temperatures. Fix: Replace faulty power supply, clean dust from cooling fan, tighten terminal connections.
2. I/O Module Faults
Symptoms: Single channel or entire module not reading/writing. Diagnosis: Swap with a known-good module. Check module LEDs. Fix: Replace module. Check field wiring for short circuits that may have caused the fault.
3. Communication / Network Faults
Symptoms: HMI shows "communication lost", remote I/O racks drop, PROFINET slaves offline. Diagnosis: Use a network diagnostic tool (Siemens PRONETA, or Wireshark for advanced). Check switch link lights. Test cables with a certifier. Fix: Replace damaged cables, swap faulty switches, check VLAN/IP configuration.
4. Scan Time Issues
Symptoms: Slow response, missed fast signals, intermittent behaviour. Diagnosis: Check scan time in diagnostic buffer. Compare to baseline. Fix: Optimise code, move complex math to periodic tasks, upgrade to higher-performance CPU.
5. Memory / Battery Faults
Symptoms: CPU fault on restart, retentive data lost, program corruption. Diagnosis: Check battery/memory card status LED. Fix: Replace battery (Siemens S7-300/400 legacy), replace memory card.
6. Logic Bugs
Symptoms: Intermittent behaviour, specific scenarios fail, incorrect outputs. Diagnosis: Reproduce the scenario. Watch tags online. Trace logic rung by rung. Fix: Correct the code, test thoroughly, document the change.
Tools Every PLC Maintenance Engineer Needs
Physical Tools
- Digital multimeter (Fluke 117 or similar)
- Clamp meter for AC current (Fluke 376)
- Insulation tester (megger)
- Cable certifier for Ethernet (Fluke DSX-5000) — enterprise-grade, borrow if needed
- Network cable tester (Klein VDV501-851)
- Toner/probe for tracing cables
Software Tools
- TIA Portal and/or Studio 5000 (engineering software)
- PRONETA (Siemens network diagnostic — free)
- Wireshark (network packet analysis — free)
- Rockwell RSLinx (comms configuration)
- Siemens Automation License Manager
- Laptop with all above installed, plus Excel for logging
Diagnostic Reference Materials
- Manufacturer diagnostic manuals (print copies at the plant)
- Hardware catalogues with part numbers and LED meanings
- Your own troubleshooting checklists (build these as you go)
The Professional Engineer's Troubleshooting Checklist
Keep this taped to your toolbox or saved in your phone:
- Safety first. LOTO (Lockout/Tagout) if power work. PPE.
- Read the alarm banner before doing anything.
- Read the diagnostic buffer.
- Check status LEDs on all modules.
- Ask the operator what happened just before.
- Check for recent changes (new code deployed? recent maintenance? new equipment?).
- Identify if it's a field device, I/O, network, or logic issue.
- Verify with measurements (voltage, current, signal presence) at each stage.
- Test in safe mode before enabling production.
- Document the fault, root cause, and fix for future reference.
How to Diagnose Intermittent Faults
Intermittent faults are the hardest. Use these techniques:
- Trend relevant variables for 24 hours. Look for patterns — does it always happen at 06:00 (start of shift)?
- Add diagnostic triggers in the code that log a timestamp and state snapshot when the fault occurs.
- Check environmental conditions. Temperature, humidity, vibration, EMI.
- Check scheduled events — does a maintenance task or backup job coincide with the fault?
- Monitor CPU load. Intermittent scan time extensions suggest memory or comms issues.
Building a Fault Knowledge Base
Every experienced maintenance engineer has a personal fault log. Start yours:
| Date | Plant | Fault | Diagnostic | Root Cause | Fix | Notes | |------|-------|-------|------------|-----------|-----|-------|
Over time, patterns emerge. You will recognise "Oh, that looks like the 3 AM PROFINET glitch we had in April" and solve it in minutes instead of hours.
Training for Troubleshooting
Troubleshooting is a practical skill — you learn it by doing. At EDWartens UK, our Professional Module includes a dedicated troubleshooting day where students fault real PLCs under controlled conditions and diagnose them systematically. This structured exposure compresses years of experience into days.
Ready to learn troubleshooting properly? Book a free consultation with our career team.
