ISO 6722, Crimp Validation and What Automotive Buyers Put on a Harness Spec

The gap between a harness that works on the bench and a harness that survives four years under a bonnet is written down, and it is written down in standards you cite rather than in adjectives you use. An automotive harness specification that names ISO 6722 for the wire, a recognised crimp validation regime for the terminations, and a workmanship class for the build is a document a factory can quote against and an inspector can judge against. One that says "automotive grade" is a document that means whatever the cheapest interpretation says it means.

This piece is written from the buyer's side of the table. It covers what the main regimes actually govern, what crimp validation involves as a physical activity, and how to write the requirement so a supplier that has never seen your programme knows exactly what it is being asked to do — and so you can tell, from the evidence that comes back, whether it did.

Three layers, three different documents

Buyers who get burned usually specified one layer and assumed the other two. An automotive harness requirement resolves into three separate technical questions, each with its own standards family and its own evidence.

The wire is a purchased commodity with a published construction. Its temperature class, wall thickness, conductor stranding, abrasion resistance and fluid resistance come from the cable maker, not the harness shop, and are evidenced by the cable maker's test report against a cable standard.

The termination and the connector system are where the harness shop's own process lives. Crimp geometry, crimp strength, contact resistance and the behaviour of the mated pair under vibration and temperature belong here, and the evidence is test data from a defined validation sequence.

The build — routing, branch lengths, taping, sleeving, labelling, strain relief, splice location — is workmanship, and it is judged against an acceptance standard with defined classes rather than against a photograph.

Write all three, or you have written none. The drawing and wire-list structure that carries them is set out in the companion piece on what goes in a wire harness RFQ, and the covering choices in tape wrap, split conduit and braided sleeve.

The wire: what ISO 6722 actually pins down

ISO 6722 is the reference most automotive buyers cite for single-core road-vehicle cable at 60 V and 600 V ratings. Its usefulness is that it turns several vague preferences into selectable grades.

• Temperature class. Cables are classified by the maximum continuous conductor temperature they are rated for, with the classes stepping up from around 85 °C into the 150 °C-plus region. A class is not a marketing tier; it is tied to a long-term ageing test at that temperature, and picking one class higher than the environment needs is one of the cheapest reliability decisions available on a vehicle harness.
• Wall thickness grade. The same conductor size is offered in thin-wall and thicker-wall constructions. Thin wall saves diameter and mass in a dense loom; thicker wall buys abrasion and cut-through margin. The standard defines the dimensional bands so "thin wall" stops being a sales word.
• Abrasion and cut-through. Scrape-abrasion cycles and pressure-at-temperature tests are what separate a jacket that survives a chafe point from one that does not. If any part of your harness runs against a bracket edge, this is the clause that matters and the clause a substitution will quietly drop.
• Fluid and flame behaviour. Resistance to the fluids a vehicle actually contains, and a flame-propagation requirement, are both in scope.

Two practical notes. First, a newer road-vehicle cable series, ISO 19642, is in use alongside or in place of ISO 6722 on some programmes; ask the cable maker which one its report is written against rather than assuming. Second, confirm current editions and clause numbering before you cite anything — a spec that references a superseded edition invites an argument you will lose at the worst moment.

Choosing between the workmanship classes is a cost decision as much as a quality one — that trade-off is worked through in IPC/WHMA-A-620 Class 1, 2 or 3.

Crimp validation: what a buyer is actually asking for

"Validated crimp" sounds like a certificate. It is a sequence of physical measurements, and knowing the sequence is what lets you read a report critically.

Geometry: cross-section and crimp height

A crimp is judged by cutting it open. The terminated barrel is potted, sectioned, polished and examined under a measuring microscope, and the section is read for a defined list of features: how far the strands have been compacted, whether every strand is inside the barrel with none cut or trapped outside, whether the barrel seam is closed, whether a bell mouth is present at both ends of the barrel, and whether the insulation grip holds the jacket without cutting into it. Each of those has an acceptance window that comes from the terminal maker's application specification for that specific terminal and wire combination — which is the detail most buyers miss. There is no universal crimp acceptance table. There is a table per terminal.

Crimp height and crimp width are the shop-floor proxies for the same thing, measured with a blade micrometer in seconds rather than by sectioning in an hour. The relationship between them is what a setup approval establishes, and it holds only for that terminal, that wire, that applicator and that press.

Strength: pull-force and the electrical check

Pull-force testing pulls the wire out of the terminal on a tensile tester and records the peak load. The acceptance minimum is a function of conductor cross-section and comes from the crimp performance specification and the terminal maker's data, not from a general rule. The useful discipline for a buyer is to ask for the failure mode alongside the number: a crimp that fails by strands breaking outside the barrel is behaving differently from one where the bundle slides out intact, and the second is the one that gets worse in the field.

Alongside strength sits the electrical measurement — millivolt drop or contact resistance across the crimp at a defined current — which is the parameter that actually predicts heating in service. The mechanics of reading these three together are unpacked in how harness crimps are actually judged.

The part most quotations skip: the environmental sequence

A crimp validation regime does not stop at a fresh sample. It measures resistance, then subjects the terminated sample to a defined stress sequence — thermal shock or temperature cycling, humidity, vibration, and in many programmes a salt-fog or mixed-flowing-gas exposure — and measures resistance again, looking for drift rather than for a pass mark. That drift is the whole point: a crimp with a marginal gas-tight interface reads fine on day one and reads badly after cycling. When a supplier quotes "crimp tested", ask which of those steps were in the sequence. Very often the answer is the pull test alone, which is the cheapest and least predictive of the set.

Writing it so a supplier can price it — and verifying what comes back

Naming six standards on a drawing does not make a small harness shop capable of running six test regimes, and a buyer who conflates the two ends up with a quote priced on hope. The productive split is this: the wire and connector evidence comes from the component makers; the crimp and workmanship evidence comes from the assembly shop; the assembly-level environmental testing, if your programme needs it, is a third-party or buyer-side activity that someone has to pay for by name.

Say which of the three you are buying, in the purchase order, in those terms. Then verify:

1. Ask for reports, not certificates. A certificate says a scheme was audited. A report says a sample was measured. For crimp questions the report is the only useful document.
2. On any certificate you are shown, check three things — the issuing body, the holder name and the validity date on the document itself. A framed document on a supplier's website can be issued to a differently named legal entity, or can have expired; both are common and both are visible in ninety seconds. Working through that is the subject of reading a harness supplier's certificates properly.
3. Separate equipment from process. A published equipment roster tells you what machines are in the building. It does not tell you the sampling frequency, the reaction plan when a crimp height drifts, or who signs off an applicator change. Ask for the control plan, not the machine list.
4. Buy a first article against the spec you wrote. Sectioned crimps from the actual production applicator, a pull-force set, a continuity and withstand-voltage record, and photographs of the build against the workmanship class. The general discipline is covered in what documents to ask a China supplier for before the first order, and the spec-writing mechanics in how to write a product spec sheet for a Chinese factory.

Automotive electronics buyers already run this playbook one level up the assembly, where component qualification and assembly class are specified explicitly rather than assumed — the parallel is set out in automotive PCBA manufacturing and class selection.

Common questions

Do I need ISO 6722 wire for an aftermarket accessory lead?

Often not, and specifying it anyway raises your cost without buying anything. The question is the environment, not the market: an in-cabin accessory lead that never sees engine-bay temperature or a chafe point can be built on a general-purpose wire to a named construction. What you should refuse is a lead built on an unnamed wire. Specify some standard and grade, and get the cable maker's report, even if the standard is not an automotive one.

The factory lists a cross-section analyser and a tensile tester. Is that crimp validation?

Those are the right instruments, and having them is better than not having them. Validation is what you do with them: a defined sequence, defined acceptance criteria per terminal, a sampling plan, and records that tie a measurement to a lot. Ask for the sampling frequency, the reaction plan and a sample report before you treat an equipment list as evidence of a programme.

Who pays for assembly-level environmental testing?

Whoever wants it needs to name it and fund it. Assembly-level thermal cycling, vibration and ingress testing to a road-vehicle regime is a laboratory activity with a real price, and it is not inside a per-piece harness quote. Decide early whether your programme genuinely requires it, whether a documented read-across from component-level qualification is acceptable to your customer, and who commissions the lab. Leaving it unstated is how it silently does not happen.

What to ask the supplier next

1. Which cable standard and edition each wire on the bill of materials is supplied against, and a copy of the cable maker's report plus the per-lot certificate.
2. The terminal maker's application specification for every terminal in the assembly, with the crimp height and width windows it defines.
3. The crimp setup record: press, applicator, target crimp height, first-article sectioned images and the approval signature.
4. The in-process sampling plan — how often crimp height is measured, how often a section is cut, what triggers a re-check after an applicator or reel change, and the documented reaction plan when a reading drifts.
5. A sample pull-force report showing conductor size, acceptance basis, measured values and failure mode.
6. Whether any environmental sequence has been run on terminations, by whom, to what protocol, and whether the report can be shared.
7. The workmanship class the shop will build to, and a sample inspection record against it.
8. A live video walkthrough of the crimping bench and test bench during production of your first article, with the camera on the press setup sheet and the measurement, not on the machine badge.

Huisheng Electronics is a usable case to run those questions against. Its published range includes wrapped vehicle harness assemblies, adapter and ignition leads, and cigarette-lighter and battery-clip cables added to the automotive line in a 2024 catalogue update, alongside multi-circuit vehicle harnesses built to drawing. Its supplier-published equipment roster lists more than twenty production and test machines, including automatic cut-strip-crimp units, servo crimping presses, a terminal cross-section analyser, tensile and insertion-force testers, withstand-voltage and insulation testers, a salt-spray chamber and continuity testers. That is a bench, and a bench is a starting point rather than a conclusion: the sampling plans, acceptance criteria, control plan and test evidence behind any of the standards named in this article are not published details, and capacity figures are the company's own. Ask for the records, check any document you are shown against issuer, holder and validity date, and confirm the process in an audit or a live walkthrough before you commit a programme to a schedule. If the same programme is also crossing a border, the duty and origin side is covered in duty and origin on imported wiring harnesses.