Yaskawa Vfd Brand Logo

Custom Wire Harness Checklist: 10 Quality Gates Before You Approve Production

VFD engineering technical article

Who this custom wire harness checklist is for

If you are sourcing a custom wire harness, a wiring harness for car program, an ODM robot cable set, a medical connector harness, or supporting UAV wire harness production, the first-article approval process is where money gets saved or lost. This is the 10-step checklist I use as a quality/brand compliance manager at an industrial automation distributor. We source cable assemblies that connect Yaskawa VFDs, PLCs, contactors, relays, and timers, and we review every wiring harness cable assembly before it goes to a customer—roughly 200 unique builds a year. I have rejected about 18% of first deliveries in 2025, mostly for documentation gaps and crimp inconsistencies.

It took me three years and about 150 first-article inspections to understand that the drawing is not the product. The process is the product.

Use this checklist when you are about to approve a sample, release a pilot order, or move a harness from prototype to production. It works for automotive, robotics, medical, UAV, and general industrial cable assemblies. Adapt the standards to your risk level.

The 10-step checklist

Step 1: Freeze the drawing and revision before you look at a sample

Do not review a sample against a verbal description. Get a controlled drawing with a revision number, date, and change log. Confirm every item that matters:

  • Connector manufacturer and part number, including keying and color.
  • Terminal part number, plating, and crimp specification.
  • Wire gauge, insulation type, temperature rating, and color code.
  • Cut length, strip length, and tolerances.
  • Twist pitch, shield type, drain wire, and grounding method.
  • Label text, label material, and placement.
  • Packaging method and quantity per bag or box.

If the drawing says 'standard' anywhere, stop. Standard means nothing until someone defines it. In my first year, I made the classic specification error: assumed 'IPC-standard crimp' meant the same thing to every vendor. Cost me a $600 redo on a small robot cable order.

Step 2: Confirm the compliance path before samples ship

This is the step most buyers skip. Ask for the quality system and workmanship standard before you get samples. For cable and wire harness assemblies, IPC/WHMA-A-620 is the consensus workmanship standard. It defines crimp, solder, insulation, marking, and testing requirements, and it has classes for different end-use severity. Class 1 is general electronic, Class 2 is dedicated service, and Class 3 is high-reliability.

Depending on the application, you may also need:

  • Automotive: IATF 16949 quality management, plus customer-specific connector specs such as USCAR-2 or LV214.
  • Medical: ISO 13485 quality management. If the device touches patients, ISO 10993 biocompatibility may apply. For medical electrical equipment, IEC 60601 is common.
  • UAV / aerospace: AS9100 quality management, and often IPC/WHMA-A-620 Class 3 for critical circuits.
  • Flammability: UL 94 V-0 or the applicable wire insulation rating.
  • Environmental: RoHS and REACH declarations, with current documentation.
As of April 2026, verify the current revision of IPC/WHMA-A-620 at IPC.org and confirm which class your drawing actually requires. 'IPC standard' without a class is not a specification.

Step 3: Validate crimps with cross-sections and pull tests

Continuity testing alone will not catch a bad crimp. A crimp can pass continuity and still fail vibration, thermal cycling, or field mating. Ask for:

  1. Crimp height measurements from a calibrated micrometer or crimp tool gauge.
  2. Pull-force test data per terminal and wire combination.
  3. Cross-section photos showing strand compaction, no insulation damage, and correct bellmouth.
  4. Tooling calibration records and maintenance schedule.

Watch for over-crimping, under-crimping, nicked strands, and terminals that are not seated in the housing. Not ideal, but workable is not a crimp standard. Either it meets the spec or it does not.

Step 4: Audit the connector and terminal system, not just the assembly

Are the connectors and terminals genuine, or are they commodity equivalents? For automotive and medical work, equivalents can be a deal-breaker. Check mating cycles, keying, lock strength, and strain relief. Confirm the supplier owns or controls the applicator tooling. If they hand-crimp first articles and then switch to production tooling, your sample is not representative.

Honestly, I am not sure why some suppliers can hold tolerances on a 2,000-unit run and others drift at 200. My best guess is tooling maintenance discipline. So I ask for the maintenance log.

Step 5: Review routing, strain relief, and application stressors

A harness that passes on a bench can fail in the machine. Review the actual routing and strain relief for your application:

  • Wiring harness for car: vibration, temperature cycling, chemicals, flex fatigue, and abrasion.
  • ODM robot cable set: continuous flex, torsion, bend radius, cable chain compatibility, and shield performance.
  • Medical connector harness: cleanability, chemical compatibility, patient-contact materials, and repeated mating.
  • UAV wire harness production: weight, EMI shielding, altitude, temperature, and crash-load retention.

Ask for a bend radius calculation and a sample routed the way it will be installed. If the supplier cannot explain strain relief, they are building a cable, not a harness.

Step 6: Require electrical test data, not just a pass stamp

Get the actual test data. For a basic harness, that may include continuity, short-to-ground, and point-to-point mapping. For higher-risk builds, add:

  • Insulation resistance and dielectric withstand / hipot.
  • DC resistance limits for power and signal paths.
  • Shield continuity and shield-to-ground termination.
  • EMI or impedance testing where signal integrity matters.
  • Test fixture validation and calibration records.

Make sure the test program is locked to the drawing revision. A beautiful test report against the wrong revision is worse than no report.

Step 7: Run environmental and mechanical sampling before full release

You do not need to destroy a full order, but you do need sample evidence. Depending on risk, request thermal shock, vibration, salt spray, humidity, flex cycling, mate/de-mate cycling, and pull testing. For medical or UAV work, the validation plan should come from your risk assessment, not from the supplier's brochure.

This is where the lowest quote often stops looking cheap. The test lab bill is visible. The field failure is not.

Step 8: Lock documentation, traceability, and change control

Before you approve production, confirm the paperwork package:

  • First article inspection report with actual measurements.
  • Certificate of compliance and material declarations.
  • Lot code, date code, and traceability from wire and connector lots.
  • Work instructions and test procedures tied to the drawing revision.
  • Change notification process with a defined approval window.
  • Packaging and label approval.

If the supplier can change a terminal, wire supplier, or crimp tool without telling you, you do not have a controlled process. You have a hope.

Step 9: Calculate total cost of ownership, not unit price

My view is simple: in harness procurement, total value beats unit price. The cheapest quote has cost us more in about 60% of projects I have managed. That $180 savings on a 500-piece run turned into a $2,400 problem when the first batch failed pull testing and the line sat waiting for rework.

Build a TCO comparison that includes:

  • Unit price, tooling, and NRE.
  • Minimum order quantity and inventory carrying cost.
  • Incoming inspection time and test equipment.
  • Rework, scrap, and return freight.
  • Line downtime if a shipment fails.
  • Engineering time for approvals and corrective actions.

If the cheap supplier needs three corrective action cycles, the expensive supplier may have been the bargain. That is not a slogan. It is arithmetic.

Step 10: Red-team the pilot run before you scale

Run a pilot production order using the same tooling, fixtures, operators, and process as full production. Then red-team it. Ask: what happens if the wire lot changes? What if the connector plating changes? What if the operator is new? What if the test fixture drifts? What if the label printer goes down? A supplier that can answer those questions is a partner. One that cannot is a purchase order.

I did not fully understand change control until a vendor swapped a terminal plating without telling us. The harness worked on the bench and failed in the field after two months. That was a $22,000 redo and a delayed launch. Now every contract includes terminal, wire, and tooling change notification requirements.

Common mistakes to avoid

  • Approving a hand-built sample as if it represents production.
  • Accepting 'IPC standard' without a class and revision.
  • Testing continuity but not crimp quality.
  • Ignoring labels, packaging, and traceability until after shipment.
  • Letting the supplier own the drawing revision without your approval.
  • Choosing the lowest quote without calculating TCO.

Bottom line: a custom wire harness is not a commodity cable. It is a controlled assembly with a process behind it. If you check the process, the product usually follows. If you only check the price, you will eventually check the failure.

Mateo Alvarez

Mateo Alvarez

Mateo Alvarez is a switchboard and wiring-device analyst covering electrical panels, distribution boards, control panels, switches, industrial plugs, sockets, and outlets. He uses IEC 61439-1 and IEC 61439-2 verification evidence plus IEC 60309 ratings to examine temperature rise, short-circuit withstand, busbar capacity, rated current, creepage, clearance, and enclosure integration. He helps designers and procurement teams match assemblies and connection devices to load diversity, installation access, maintainability, and declared operating conditions.