After the repair, the operator screen should return to a stable value or normal status, and the controller should repeat its intended operation without the original symptom. For mixed Fanuc, Siemens, CNC, and NC control cards, the economical route is usually a disciplined bench workflow built around a DMM, oscilloscope, milliohm meter, signal source, and current-limited power supply. A dedicated tester or custom fixture becomes economical when the same board type appears often enough to justify known-good signatures and repeatable test points.
What is the screen actually telling you?
Start with the displayed symptom, but do not treat the display as proof that the PCB is defective. A missing value, frozen value, implausible number, or active alarm can originate at the screen binding, communication driver, controller logic, field wiring, or board hardware.
| Reading or test | Location | Outcome and next check |
|---|---|---|
| Displayed value versus controller value | Operator screen and controller diagnostics | If both are wrong, move toward the controller, I/O, and board. If only the screen is wrong, inspect its tag binding and driver path. |
| Tag identity and live status | Screen configuration or runtime diagnostics | A correct controller value with an incorrect screen value means the tag may be right while the binding, scaling, or data path is wrong. |
| Communication status | Screen driver and controller interface | A failed or intermittent connection can imitate a card fault. Restore stable communications before bench-testing the PCB. |
| Controller diagnostic state | Controller or CNC diagnostics | A channel-specific fault directs the next check toward the associated wiring, connector, or board circuit. |
Record the exact screen state, controller state, machine operating condition, and connector positions before disturbing the hardware. That record becomes the reference for the final verification.
Does the symptom follow the board or stay with the machine?
The next decision separates a board fault from a machine-side fault. With power removed and stored energy discharged according to the equipment service procedure, inspect the board and mating connectors for contamination, heat damage, cracked joints, bent contacts, loose hardware, and previous repair work. A visual finding identifies an area to test; it does not by itself prove the failed component.
If an identical known-good board is available and substitution is permitted by the equipment configuration, compare behavior under the same operating condition. Preserve board-specific settings, memory, and position information before substitution. If the symptom moves with the suspect board, continue with bench diagnosis. If it stays with the machine, return to the power rails, field devices, wiring, backplane, connectors, and controller configuration.
When substitution is not safe or practical, compare measurements at the board connector. Verify supply voltage, reference conductors, input states, and output loading against service data for that exact card. A missing supply at the connector is a system fault until the board is shown to be pulling it down.
Can the fault be reproduced on the bench?
A bench test must recreate the electrical conditions that matter without exposing the board to uncontrolled current. Use a current-limited power supply only after identifying the required rails, polarity, connector pins, and reference points from card documentation or a traced circuit. Do not energize an unidentified connector by analogy with another card.
- Photograph connector orientation, jumpers, switch positions, and removable devices.
- Check resistance between each supply input and its return before applying power. Compare with an identical known-good card when no documented value exists.
- Set the supply current limit low enough to protect the board during initial energization, then raise it only when the measured demand and service information justify doing so.
- Monitor input current and each local rail during startup. Stop if current rises unexpectedly, a rail collapses, or a component heats rapidly.
- Apply only the input signals required to reproduce the symptom, using a signal source with compatible voltage, reference, and loading.
A card that behaves normally on a minimal bench setup may depend on backplane signals, loads, interlocks, or startup sequencing that the setup does not reproduce. That outcome sends the investigation back to connector measurements in the machine rather than clearing the card automatically.
Which measurement isolates the failed stage?
Follow the signal path rather than probing components at random. Begin at the connector, verify the protection and conditioning stage, then trace the signal through conversion, logic, and output stages. Compare each node with the schematic, a known-good board, or a repeatable captured waveform.
| Instrument | Best use | Diagnostic effect |
|---|---|---|
| DMM | Supply rails, diode checks, resistance, continuity, and static signal levels | Finds missing rails, open paths, hard shorts, and incorrect steady-state voltages. |
| Oscilloscope | Clock, reset, ripple, pulses, communication activity, and switching outputs | Separates a valid static level from a signal with wrong shape, noise, or missing transitions. |
| Milliohm meter | Low-resistance comparison on suspected shorted rails or high-current paths | Helps localize a short by comparing resistance along the current path. |
| Signal source | Controlled excitation of an identified input stage | Tests whether the signal propagates through successive stages. |
| Current-limited power supply | Controlled initial energization | Restricts fault energy and exposes abnormal current demand. |
For a collapsed rail, disconnect power and measure resistance to return. If the reading indicates a low-resistance path, use comparative resistance measurements along the rail to narrow the location. For a missing functional signal, probe the input and output of each stage. The first point where a correct signal becomes incorrect defines the repair area.
When does dedicated PCB test equipment make sense?
A universal fault finder faces a basic limitation: different control cards use different connectors, supply arrangements, signal levels, startup conditions, buses, loads, and firmware-dependent behavior. The tester still needs card-specific access information and a definition of correct behavior. Its useful output depends on the quality of that test definition.
Two configurations can work. A general instrument bench is the better starting point for varied Fanuc, Siemens, and similar cards because it adapts to different circuits with modest equipment. A dedicated test rig is better when the same board is repaired repeatedly because it can reproduce supplies, loads, inputs, and test points in a controlled sequence.
Evaluate Boardwizard or any other dedicated system against the actual card mix. Request a demonstrated test on the required boards and determine what adapters, circuit information, reference signatures, and training the system needs. A demonstration that merely identifies an abnormal area is different from one that isolates a replaceable device. Include adapter development and board-data preparation in the cost comparison.
How should the repair and verification be completed?
- Document the failing node and compare it with its input, supply, and reference. Replace or rework parts only after that measurement isolates the stage.
- Inspect the repair under magnification. Check for bridges, lifted pads, damaged vias, incorrect orientation, and disturbed nearby joints.
- Repeat unpowered resistance and continuity checks before applying power.
- Power the card from the current-limited supply and confirm that input current and local rails remain stable.
- Reapply the bench stimulus and capture the repaired node with the same DMM or oscilloscope setup used during diagnosis.
- Return the card to the machine, restore preserved settings, and reproduce the original operating condition.
- Compare the screen value, live tag, driver status, controller diagnostic state, and physical machine response with the initial record.
FAQ
Can I use one PCB fault finder for Fanuc and Siemens control cards?
Only if it has the correct interfaces, power arrangements, signal definitions, and reference data for each card. For a varied card population, a DMM, oscilloscope, milliohm meter, signal source, and current-limited supply provide the more adaptable starting point.
Does a low resistance across a supply rail prove a shorted component?
No. Confirm the reading against circuit information or an identical known-good board, then compare resistance at points along the rail to localize the low-resistance path before removing parts.
Can I verify a repaired CNC card only on the bench?
The bench confirms rails, current demand, and signal propagation, but the machine supplies the real backplane signals, loads, interlocks, and sequence. Finish by repeating the original operating condition and confirming the screen value, tag, driver, controller diagnostics, and machine response.