News
N

News

News

Four Tests Every Control Panel Should Pass Before Packing

Column: Technical Notes    Time: 2026-02-19

Panel leaves the workshop with a test sheet or it does not leave. Four checks cover the faults that would otherwise surface on site, and each one has a document behind it so the result can be verified by someone who was not present.

1. Continuity and polarity

Every conductor buzzed from end to end against the wiring schedule, with the other end identified by its label rather than by assumption. This catches the errors that a visual inspection cannot: a wire in the right terminal but the wrong strip, two cores swapped inside a multicore, a terminal that was tightened onto insulation.

What the sheet records: circuit reference, from-terminal, to-terminal, result. On a 400-wire control panel this takes about two hours with two people working from opposite ends.

Polarity matters most on DC circuits and on instrument loops. Reversed 24 V on an analogue input card is a dead card, and it is invisible until the instrument is connected.

2. Insulation resistance

500 V DC insulation test between live conductors and earth, with drives, surge protection and electronic devices disconnected or isolated — a megger into a drive's DC bus is an expensive way to test a panel.

Acceptance we work to: 1 MΩ minimum, and in practice a clean dry panel reads far higher. Low readings point to a tracking path — cable damage, swarf in a duct, moisture, or a wire trapped at a gland. On a panel that reads 0.5 MΩ, do not re-test hoping for a better number; find the fault.

Test each section separately when the panel is large: incoming, distribution, control, field terminals. Section results localise a fault in minutes where a whole-panel test gives a single unusable number.

3. Function test under power

Panel energised, with a simulation of field signals where the real instruments are not present. This is where the specification is actually proven:

  • Every selector, push button and HMI command exercised in both directions.
  • Contactor and relay sequence confirmed against the drawing, including interlocks and safety circuits.
  • Drive parameters loaded and run command verified, at minimum to a test motor or with the output disabled.
  • Analogue loops driven from a signal generator at 4, 12 and 20 mA, with the reading checked at the PLC and on the HMI.
  • Alarms forced and the annunciation confirmed at the terminal, not just on the screen.
  • Timers and delays timed with a stopwatch where the process depends on them.

The sheet records simulated value and observed response, so a commissioning engineer can compare against what the panel did before it left.

4. Visual and mechanical check

Least technical and hardest to pass first time. Written as a checklist because it is otherwise skipped:

  • Torque check on a sample of terminals per strip, with the screwdriver type recorded — a torque value means nothing if it was set with the wrong tool.
  • DIN rail devices seated, end stops fitted, nothing loose when the panel is tapped.
  • Door swing clear, hinge pins secure, gasket undamaged, lock aligns without force.
  • Labels present, printed, and matching the schedule — read one section against the drawing, not all of them by eye.
  • No swarf, no offcuts, no tools left inside. Photograph the open panel before closing.
  • Rating plate and warning labels fitted, and the enclosure rating stated on the plate.

Why the paperwork matters as much as the test

Test that is not recorded is a test that did not happen, as far as the next person is concerned. Three uses for the sheet after delivery:

  • Commissioning compare: the site engineer knows what normal looked like in the workshop.
  • Warranty discussion: a fault that matches a passed test points to transport or site wiring rather than to the panel.
  • Repeat orders: the sheet format is what makes batch ten identical to batch one.

We issue one sheet per panel, serial-numbered and signed by the person who ran each check. It travels in the door pocket along with the drawings, so it cannot be separated from the panel in transit.

What we do not claim

Factory testing proves the panel, not the system. Motor rotation, instrument calibration, loop tuning and site earthing are all outside what a workshop can verify. Where a customer wants a simulated full system test, we build a rig for it and charge for the time — which is cheaper than discovering the problem during a plant shutdown.

More News