Multi-pass weld pass-sequence documentation on a WPS
Most WPS forms capture a single parameter row for all passes combined. Here's why pass-by-pass documentation matters, what AWS D1.1:2025 requires, and how to structure a passes table on your WPS.
A WPS that lists a single row of filler metal, current range, and voltage range is technically a WPS—but for a multi-pass CJP groove weld, it may not actually control what happens at the arc. A root pass on a CJP joint looks nothing like a fill pass in the same groove: smaller electrode, lower amperage, different travel speed, different heat input. Documenting them with one row creates a qualification gap that surfaces during audits, or worse, during post-failure analysis.
Root, hot, fill, and cap: four different welding events
In a typical multi-pass CJP groove weld, four distinct pass types appear:
Root pass. Least forgiving. A fusion defect or lack-of-penetration at the root propagates through the joint under fatigue loading. Root passes typically use a smaller electrode diameter (SMAW) or reduced wire feed speed (GMAW/FCAW) and lower heat input to control penetration. Burn-through is the primary risk on open-root joints; cold lap on backing.
Hot pass. Deposited immediately after the root, the hot pass ties the root bead into the fill and burns out any slag trapped at the root toe. It runs slightly hotter than the root to ensure fusion but the interpass window is narrow—too cold and you get cold lap; too hot and you push the interpass temperature toward its maximum.
Fill passes. The bulk of the weld volume. Higher current, larger wire or electrode, faster travel, maximum deposition rate. Parameters are optimized for productivity. This is where most shops document their WPS parameters because it's the longest part of the welding sequence.
Cap passes. The visible surface. Cap passes often run at slightly reduced heat input and current to control crown height and avoid undercutting the toes. Profile defects—undercut, overlap, excessive reinforcement—are most visible here and most likely to be flagged on visual inspection.
Each pass type has its own optimal parameters. Documenting them all with one row leaves the CWI, the welder, and any future auditor without a reference point for what parameters were actually qualified.
Heat input as the controlling variable
Heat input is an essential variable under AWS D1.1:2025 Table 6.6. The standard formula:
Root pass: 3/32 in E7018, 90 A / 22 V / 5 ipm → heat input ≈ 23.8 kJ/in
Fill pass: 1/8 in E7018, 145 A / 23 V / 8 ipm → heat input ≈ 25.0 kJ/in
Both values must fall within the PQR-qualified heat input range. If the PQR was run at 24–27 kJ/in and the root pass produces 23.8 kJ/in, the root pass is outside the qualified range—even though a lower heat input might seem conservative. The essential variable is a window with both a minimum and a maximum, not just a ceiling.
Rule library based on AWS D1.1:2025; verify against your governing edition.
When CVN testing is required, AWS D1.1:2025 Table 6.8 supplementary essential variables add tighter constraints on heat input changes. For CVN-required work, even modest pass-to-pass variation in heat input can cross a Table 6.8 threshold. See CVN impact testing and AWS D1.1:2025 Table 6.8 for those supplementary requirements.
The WPS specifies a maximum interpass temperature. Production welding must stay within it. For pass-sequence work, this means:
Preheat is established before the root pass and maintained at or above the minimum between all subsequent passes.
Interpass temperature is measured on the base metal near the joint immediately before starting each new pass.
If the measured temperature exceeds the WPS maximum, welding stops. The joint cools to within range before the next pass begins.
Production logs on CVN-required work, seismic demand-critical welds, and high-strength steel joints typically require the inspector or welder to record the interpass reading for each pass.
The connection between interpass temperature and pass sequence is direct: a welder moving fast between passes in a thick joint may not allow enough cooling time, pushing interpass above the maximum. A WPS that documents the pass sequence with expected pass times gives both the welder and the inspector a realistic reference.
When the PQR was run, the test welder should have recorded actual parameters for each pass type. That is the evidence that the qualified range actually covers real production practice. A PQR with a single data row—likely from the fill passes where the welder spent the most time—may not have qualified root-pass parameters at all.
A well-structured PQR records for each pass:
Pass designation (R, H, F1, F2, ..., Cap)
Electrode or wire classification and diameter
Amperage and polarity
Voltage
Travel speed
Heat input calculated
Interpass temperature measured before starting the pass
That is the audit trail a WPS can point back to when production parameters are questioned.
The problem with legacy WPS templates
Most WPS templates circulating in fab shops are single-row forms inherited from earlier editions. They have one line for process, one for filler metal, a range for amperage, a range for voltage. They were designed around single-pass fillet welds or were never updated when the shop moved to heavier sections.
Applied to a 1-inch CJP column splice with eight or ten passes, the single-row form is inadequate. The listed range might technically span root and fill parameters, but that only holds if the PQR was run across that full range—which is uncommon. More often the range was copied from a previous WPS or estimated, and no one compared it to the actual PQR heat input data.
Digital WPS forms with a structured passes table address this directly. Each pass type gets its own row: electrode diameter, current range, voltage range, travel speed range, and computed heat input range. The system calculates heat input from the entered values and flags any row where the resulting range falls outside the supporting PQR envelope. That moves the check from the audit to the WPS drafting step.
What inspectors check at pass hold points
A CWI on a multi-pass structural joint typically works from three hold points:
Fit-up inspection. Before any welding. Joint geometry, root opening, alignment, cleanliness.
Root inspection. After the root pass and before any fill. The CWI checks the root bead visually, and on complete-penetration joints with a back-gouge requirement, inspects the back-gouge for complete removal of the root prior to filling from the second side.
Final visual. After the cap pass. Profile, undercut, crater fill, arc strikes.
The pass sequence on the WPS tells the inspector what to expect at each hold point: what pass is being inspected, what electrode size and process were used, and what heat input range is expected. Without that reference, the inspection is disconnected from the documented procedure.
For documentation practices that keep the WPS-to-production chain intact across all pass types, see Weld map and WPS traceability in production. A full digital workflow that structures pass-by-pass documentation is available at pricing.
FAQ
Frequently asked.
Do I need a separate row on the WPS for each individual weld pass?
Not necessarily each individual pass, but each distinct pass type—root, hot, fill, cap—needs its own parameter range. A single range lumping root and fill passes together is often invalid because the root pass uses a smaller electrode and lower heat input than fill passes, and both ranges must be covered by the PQR.
What is the difference between a layer and a pass?
A pass is one travel of the arc across the joint. A layer is a set of passes deposited at approximately the same depth before building up the next layer. A three-run layer has three side-by-side passes. Most WPS forms track passes, not layers, because the essential variable heat input applies to individual arc travel runs.
Does heat input need to be calculated separately for each pass type?
Yes. Heat input is an essential variable under AWS D1.1:2025 Table 6.6. If the root pass runs at 22 kJ/in and fill passes run at 30 kJ/in, both ranges must be qualified by the PQR. A PQR run at a single mid-range value may not support either extreme.
Can I list a wide heat input range on the WPS to cover all passes at once?
Only if the PQR was qualified across that full range. A PQR test plate run at a single heat input validates that heat input—it doesn't validate the full width of whatever range you write on the WPS. The WPS range cannot exceed what the PQR actually demonstrated.