Customer call: a critical legacy board fails

A Gulf of Mexico refinery reported a failed legacy circuit board and a potential shutdown. The customer described scarce spares and asked whether the repair team could complete the circuit board repair within 24 hours while a customer representative waited.

The partner record was published on January 24, 2019; the actual repair date is not specified.

Delivery: a customer representative brings the board

After the repair team accepted the rush request, the customer sent an employee by air with the board in his possession. He stayed beside the repair technician while the work was completed. The source describes delivery to the repair provider, not an engineer’s site visit.

Technical guidance: receiving a scarce board should include photographs, full part number, revision, connectors, jumpers and available configuration data. Record the alarm sequence and what happened before the failure. The source does not publish a complete receiving report.

Technical diagnosis: from supply branches to board function

The partner record documents the rush repair and return. For industrial electronics repair teams, this circuit board repair workflow explains where to inspect, how to interpret evidence and what to verify. Actual failed parts and measurements belong in the board’s diagnostic record.

1. Identify the board and preserve its configuration

Record the manufacturer, complete part number, hardware revision, connectors and jumper positions. Photograph both sides and damaged areas. Establish whether failure occurred at power-up, start, operation or shutdown, and whether power loss, communication faults or an external load were involved.

Before replacing memory or logic devices, confirm backups and recovery for programs, calibration and machine parameters. Similar-looking revisions can use different firmware, interfaces or calibration data. Check compatibility before substitution.

2. Inspect contamination, joints and connections

After isolation, stored-energy checks and suitable ESD precautions, inspect terminals, connectors, power-device areas, heavy-component joints and vias under magnification. Record contamination, corrosion, discoloration, cracks and pad damage before cleaning. Use a process compatible with the board and coating; protect labels, seals and finishes.

A burn mark does not identify the first failed component. Inspect suspect joints for continuity and mechanical condition. After cleaning, look for carbonized insulation and lifted copper. A clean surface alone does not establish a repaired circuit.

3. Use unpowered checks to narrow the supply branch

Using the schematic or a confirmed circuit structure, inspect input protection, fuses, rectifiers, regulator branches and output protection. Compare resistance and forward/reverse diode behavior. Isolate a branch or remove a suspect component when needed to eliminate misleading parallel paths.

Low resistance to ground may reflect circuit loading or capacitor charging. A continuity beep alone does not prove a short. If a fuse is open, assess downstream devices, protection and loads before replacement. Repeated power-up with a new fuse can enlarge unresolved damage.

4. Assess capacitors and semiconductors electrically

For electrolytic capacitors, consider capacitance and equivalent series resistance, or ESR, alongside leakage or bulging. Interpret readings against specifications, temperature and test frequency. Parallel components can affect in-circuit tests, requiring an isolated measurement. Appearance and one ESR reading do not establish the condition of the complete supply branch.

For fitted diodes, MOSFETs or other power devices, use the pinout and internal structure to assess abnormal conduction, opens or shorts. A failed device also calls for checks of related drive, suppression, feedback and protection circuits. Replacing the visibly damaged part alone can leave a related fault.

5. Check DC supplies, ripple and startup behavior

After unpowered checks and confirmation of the input and fixture, use manufacturer-appropriate protected power conditions. Observe startup, then assess supply levels, ripple, heating and reset behavior. Acceptable values come from the board documentation, not limits borrowed from a different machine.

Use a meter for DC levels and suitable oscilloscope measurements for ripple and startup transients. Record the node, coupling, probe attenuation and bandwidth. Floating or high-side nodes require an appropriate differential or isolated method; do not attach an earth-referenced ground clip to an unknown potential. Resolve unstable power before condemning logic devices.

6. Separate logic, communications and machine faults

With stable supplies, inspect relevant reset, clock, enable and status signals. A lit indicator proves only a limited response. Comparisons with a known-good board need matching revision, firmware, configuration, load and test conditions.

Where RS-485, CAN or another interface is actually fitted, assess the relevant transceiver, isolation, protection, wiring and protocol settings. Use a compatible fixture or verified load for input/output checks. Record the applied condition and response. If the board passes on the bench but fails in the machine, investigate external sensors, actuators, supplies and interlocks.

7. Rework the confirmed fault and check related circuits

Verify replacement ratings, package, pinout and compatibility. Capacitor selection also involves temperature, ripple-current and service-life requirements. Logic and memory devices can require compatible firmware and preserved parameters. Similar appearance or a partial part-number match is insufficient.

Use a board- and component-appropriate rework process. Inspect pads, tracks, vias and connectors for damage. Check new joints and solder bridges under magnification, address residues and restore required protection using a compatible process. Confirm suspect joints individually; whole-board heating is not a substitute for diagnosis.

Turn findings into the next diagnostic action
FindingCheck nextRepair and verification focus
Open fuseDownstream power devices, protection and loadResolve related faults before replacing to specification
Low supply or abnormal rippleSupply branch, capacitors and load currentSeparate supply and load faults; recheck the waveform
Repeated startup or resetSupply sequence, reset and watchdog signalsPreserve timing evidence before judging the processor
Unstable communicationInstalled interface, wiring, isolation and settingsVerify signals and recovery under consistent conditions
Bench pass, machine failureExternal load, sensors, configuration and interlocksValidate under the equipment owner’s actual conditions

Reported repair: complete the rush work and return the board

The published account reports completion of the work within the customer’s 24-hour request. The board was immediately placed back in its protective case and flown back to the refinery. The provider reports that the rush circuit board repair averted a potential shutdown.

The account does not publish component readings, bench waveforms or an installation acceptance report. Its turnaround and avoided-downtime statements are attributed provider claims, not a guarantee for another board or an independently audited result.

Additional acceptance workflow: bench and machine checks

The following is verification guidance. A bench result and acceptance in the original machine cover different conditions.

  • Before power-up: verify replacement identity, polarity, joints, connectors and abnormal shorts; confirm jumpers, revision and configuration and retain rework photographs.
  • Controlled power-up: record input conditions, supplies, startup and reset behavior; check heating and retain waveform settings with results.
  • Functional verification: list tested inputs, outputs and communications with the fixture, load, firmware and simulated conditions; indicators do not replace functional results.
  • Stability: within permitted conditions, record load, duration, temperature behavior, repeated starts and symptom recurrence; state untested conditions.
  • Site acceptance: the equipment owner verifies parameters, interlocks, communications and process actions under the site procedure and retains installation results.

Handover: document the repair boundary

A useful handover lists actual replaced parts, test conditions, results and limitations. Urgency does not replace evidence. For a legacy system, preserve the configuration and compatibility information before another scarce board fails.

Technical editing: Suntech Robotics Technology Limited. Based on repair records supplied with partner permission.

View circuit board repair services

Related reading: inverter capacitors and bearing currents

For a general method beyond this event record, see printed circuit board repair: industrial fault diagnosis and acceptance tests.