WELDING EQUIPMENT

Welding sample trials: what to test and how to accept the result

How to choose representative parts, write acceptance criteria before the trial and keep a record that still means something after delivery.

By WeldOrigin Editorial · Published · Updated

A demonstration comes before acceptance

Almost every equipment page on this site ends with the same advice: run a trial on your own parts before you commit. A showroom weld proves that the machine can join something under conditions the supplier chose. Acceptance means the offered configuration produces parts that meet criteria you wrote down in advance, at a rate and with a level of operator help you can live with.

Treat the trial as a small engineering project with three outputs: a defined sample set, written acceptance criteria and a record tying the result to the exact configuration. That record later becomes the yardstick when you are comparing welding equipment quotations from several suppliers, because two offers are only comparable if they were judged on the same parts.

Pick samples that carry production variation

The most common weakness is an ideal coupon. Clean, tightly fitted plate tells you little about a part that arrives with gaps, coatings or dimensional scatter. Build the sample set from your hardest cases as well as the typical one.

The published pages show what that means across machine types. For the GWC-3200-C mesh welding line, Xinzhou's advertised row rate only becomes useful when the trial uses your mesh drawing and wire combination and counts acceptable panels rather than rows. Try Chengyu's seam and positioning stations on the hardest part you make: the one with the tightest access and the most variable fit-up. For stud welding, Wintures equipment is best tested with the thinnest substrate and the most awkward position. On stator and battery work, Mingtai Laser needs samples spanning the real variation in stack dimensions and surface condition.

Plastic joining is no different. A trial at Dongho or Yuyao Mingfeng should use moulded production parts in the actual polymer, because unrelated coupons say nothing about the joint design that decides an ultrasonic, hot-plate or vibration result.

Write the acceptance criteria first

Agree the criteria with your responsible engineer before the first part is welded, or appearance becomes the only test. For fusion-welded joints in steel, nickel, titanium and their alloys from 0.5 mm thickness, ISO 5817 sets three quality levels for imperfections, B, C and D, with B the most stringent. It excludes beam welding and does not cover metallurgical properties such as hardness; beam welds in steel are covered by ISO 13919-1. Name the level and the inspection method, for example visual examination, macro sections or radiography, in the trial plan.

For laser welding, TWI notes that the destructive tests used on arc welds generally apply, but very narrow, hard welds can complicate Charpy and fracture toughness results, so choose tests that reflect the service loads. Ultrasonic joints have no comparable imperfection table among these sources; the criteria come from the product itself: pull or peel strength, leak tightness, dimensions and cosmetic limits. The ultrasonic plastic welding guide explains why joint design and material dominate those results. For thin stainless parts made on Eweld systems, a leak test may belong in the plan next to weld appearance.

A good trial weld is not a qualified procedure

Arc buyers often expect the trial to settle the welding procedure. It does not. Procedure qualification is a separate, standard-driven step in which a preliminary procedure is welded under controlled conditions, tested and documented in a qualification record. ISO 15614-1 does this for arc and gas welding of steels and nickel alloys, either at Level 1 with lighter testing and wider ranges or at Level 2 with more testing and narrower ranges; TWI notes that Level 2 applies when the contract or application standard names neither. ISO 15614-11 is the corresponding test for laser and electron beam joints. Under the codes TWI discusses, the organisation doing production welding is responsible for its procedures, so a supplier's demonstration settings are input to your procedure, not a substitute for it.

In heavy fabrication, a thick-plate sample on the Dabu Split MIG500 SiC, or spray-transfer runs on Linlong's MIG400GFW / MIG500GFW, shows whether the combination of source, feeder and torch handles your joint; the welding duty cycle ratings guide explains why a trial at your real arc-on time adds to the published 60% and 100% points. Robot cells from Kaierda add the integrator's programme and fixture to the list of things a qualification depends on. If your customer requires a certified welding quality system, ISO 3834 defines levels of quality requirements for the fusion welding manufacturer, which again is your organisation rather than the equipment seller.

Record the run so it can be repeated

A useful trial record names the model and revision, source and head, tooling and fixture identities, settings, consumables, material lots, operator interventions and who inspected what. Keep rejected parts and their faults next to the accepted ones; a file of successes alone usually means failures were discarded. HCSONIC's metal cable-to-terminal work and RPS-SONIC's plastic and nonwoven applications both rely on the condition of the horn and anvil, so tool identity belongs in the record. Where a cell logs process data, as Penta states for the IPW 0503S, ask for an export from your own trial parts and compare it with the physical results.

Count accepted parts over a defined period, including loading, clamping and changeovers, instead of quoting a single weld time. Fixture design often decides both, so read the welding fixtures and tooling guide before the trial date is fixed.

Two acceptances, and when to repeat them

Plan a factory acceptance at the supplier and a site acceptance after installation, each with its own record, configuration list and run definition. A laser cell such as the Penta RW 1510P also needs its enclosure, interlocks and extraction checked on site; the laser welding cells sources and safety guide covers what that check involves. Before shipment, confirm that nameplates and documents match the configuration that passed, because verifying welding equipment certificates is a separate task from weld acceptance.

Finally, agree what triggers re-acceptance: a repaired horn, a new fixture, a different laser source or a revised programme. For resistance welding lines from Yuren, replaced electrodes and control updates are typical triggers. Put the trigger list into the contract so the accepted condition stays defined long after commissioning.

Related sourcing information

Sources

About our editorial team · How we check information

Report a correction to this guide