Weld Porosity and Cracking Root Cause: Moisture, Travel Speed and Preheat Temperature Analysis

Overview #

The welding defects that shut down production lines — porosity clusters, hot cracking in root passes, cold cracking discovered 48 hours after inspection — are almost never caused by the consumable alone. In our qualification work with Chinese welding consumable suppliers, the defect trigger is almost always an interaction between consumable condition, base metal preparation, and process parameter deviation. Procurement teams that chase electrode brand and AWS classification while ignoring moisture content, preheat compliance, and travel speed consistency are solving the wrong problem. The measurable threshold that separates acceptable from rejectable weld quality is tighter than most welding procedure specifications (WPS) acknowledge — and Chinese-sourced consumables, when not stored or handled correctly at the receiving end, will fail even when the mill certificate is clean.

Weld Porosity: Moisture as the Primary Variable #

Porosity is the most frequently reported weld defect in production environments using consumables sourced from China, and moisture is the dominant cause in over 70% of cases we have investigated. The mechanism is straightforward: hydrogen released from moisture in electrode coatings or flux decomposes in the arc and becomes trapped in the solidifying weld pool. What is less understood is how quickly this happens.

Low-hydrogen electrodes classified to AWS Welding Standards under AWS A5.1 E7018 are rated H4 or H8 — meaning ≤4 mL or ≤8 mL of diffusible hydrogen per 100g of deposited weld metal respectively. An E7018 H4 electrode exposed to ambient humidity of 70% RH for just 4 hours can absorb enough moisture to exceed the H8 threshold. At 90% RH, that threshold breach can occur in under 2 hours. These are not theoretical values — they are the numbers we use to set receiving inspection hold times at customer facilities.

The porosity pattern itself is diagnostic. Uniformly distributed fine porosity across the weld cross-section points to a systemic moisture problem — either the entire lot was improperly stored or the flux was compromised at the supplier level. Clustered porosity at weld starts and stops, by contrast, typically indicates arc initiation technique or shielding gas coverage issues rather than consumable moisture. Distinguishing these two patterns before ordering re-work saves significant time.

For incoming inspection of low-hydrogen electrodes from Chinese suppliers, we apply a spot-check protocol: 3 electrodes per lot are baked at 300°C for 1 hour, then tested for diffusible hydrogen per ASTM International E3044. Any lot returning >4 mL/100g on H4-rated product is quarantined regardless of the COA value. The COA hydrogen value is a mill measurement — it does not reflect what happened during shipping and storage.

Most Western buyers do not realize that SAC China Standards GB/T 5117 governs carbon steel covered electrodes in China, and its moisture re-drying requirements are specified differently from AWS A5.1. A Chinese supplier can be fully GB/T 5117 compliant and still deliver product that does not meet the H4 diffusible hydrogen requirement on your WPS. That gap is where porosity failures originate.

For related sealing and fluid-control consumables where moisture contamination creates analogous quality failures, see our pump valve seals category.

Hot Cracking and Cold Cracking: Preheat Temperature as the Control Variable #

Cracking failures in production welding divide into two fundamentally different mechanisms that require different corrective actions — and confusing them is the most expensive diagnostic mistake a quality team can make.

Hot cracking (solidification cracking) occurs in the weld metal while it is still at elevated temperature, typically above 1,000°C. The cause is sulfur and phosphorus segregation at grain boundaries combined with high restraint. The measurable threshold: carbon equivalent (CE) of the base metal above 0.45 using the IIW formula (CE = C + Mn/6 + (Cr+Mo+V)/5 + (Ni+Cu)/15) significantly elevates hot cracking risk. Filler metal selection matters here — ER70S-6 wire with Mn content at the high end of specification (1.65% max per AWS Welding Standards A5.18) provides better crack resistance than ER70S-3 in restrained joints on higher-CE base metals.

Cold cracking (hydrogen-induced cracking, HIC) occurs hours or days after welding completes — which is why it is the defect that most often escapes in-process inspection and appears during service. The three conditions required simultaneously are: diffusible hydrogen above threshold (typically >5 mL/100g in susceptible steels), tensile stress above 50% of yield strength, and a susceptible microstructure (martensite or bainite in the HAZ). Remove any one of these three and cold cracking does not occur. Preheat is the most controllable lever because it slows cooling rate, reduces martensite formation, and allows hydrogen to diffuse out before the joint reaches ambient temperature.

Preheat temperature requirements scale with CE. For CE between 0.40–0.45, minimum preheat of 100°C is typically required. For CE 0.45–0.60, minimum preheat rises to 150°C. These values assume low-hydrogen consumables (H8 or better) — if your consumable has exceeded its moisture exposure limit, add 25–50°C to these thresholds. We have seen production facilities using the correct preheat temperature per their WPS and still experiencing cold cracking because the electrode lot had absorbed moisture during a warehouse humidity event. The preheat was right. The consumable condition was not.

In our supplier qualification program, we require three consecutive batch COAs showing CE values before recommending qualification of a Chinese structural steel filler metal supplier. CE variation of more than ±0.03 between batches is a disqualifying finding — it means the supplier is not controlling their wire rod source consistently.

Travel Speed Deviation: The Process Variable Most Often Ignored in Defect Analysis #

Travel speed is the variable that procurement and quality teams most consistently overlook when investigating weld defects, because it is a process parameter rather than a consumable specification. But travel speed directly determines heat input, and heat input determines everything: bead geometry, dilution, cooling rate, HAZ width, and susceptibility to both hot and cold cracking.

Heat input (kJ/mm) = (Amps × Volts × 60) / (Travel Speed mm/min × 1000)

A travel speed increase of 30% — easily achieved by an experienced welder trying to improve productivity — reduces heat input by 23% at constant voltage and amperage. On a 25mm thick carbon steel joint with CE of 0.42, that heat input reduction can push the cooling rate above the threshold for martensite formation in the HAZ, converting a compliant weld into a cold cracking risk without any change in consumable, preheat, or joint preparation.

The defect pattern from excessive travel speed is distinctive: undercut along the weld toes, narrow bead width relative to joint gap, and incomplete fusion at the root in multi-pass welds. Insufficient travel speed produces the opposite: excessive reinforcement, wide HAZ, and in austenitic stainless applications, sensitization risk from prolonged time in the 425–850°C carbide precipitation range.

Production Failure Scenario — Root Cause Analysis:

A fabrication facility producing pressure vessel components from SA-516 Grade 70 (CE ≈ 0.43) reported a cluster of cold cracking failures in circumferential seam welds during a three-week production run. The welds passed visual and dimensional inspection. Cracks were detected by magnetic particle testing (ASTM International E709) during final inspection, 36–72 hours after welding completion — a classic delayed cold cracking signature.

Initial investigation focused on the electrode lot, which was E7018 H4 sourced from a new Chinese supplier. COA values were within specification. Incoming hardness testing of deposited weld metal returned 185–195 HV10 — within acceptable range. Preheat records showed 120°C minimum, compliant with the WPS.

Root cause was identified at the third investigation stage: travel speed logs from the automated welding system showed a 22% increase in average travel speed compared to the qualification weld parameters, implemented by the production supervisor to recover schedule after a maintenance delay. This reduced heat input from 1.8 kJ/mm (qualification) to 1.4 kJ/mm (production). At 1.4 kJ/mm on CE 0.43 base metal with H8-boundary consumable condition, the HAZ cooling rate was sufficient to produce a partially martensitic microstructure. Combined with residual hydrogen at the upper H4 limit, cold cracking initiated at the fusion line.

Corrective action: travel speed controls locked in the automated system, electrode storage upgraded to heated cabinet at 120°C, and post-weld hydrogen release bake at 200°C for 2 hours added to the procedure. Zero cold cracking failures in the subsequent 6-month production period.

The lesson here is not about the Chinese supplier. The COA was accurate. The failure was a process deviation that the WPS did not adequately control — and that a standard incoming inspection would never have caught.

Practical Guidance for Buyers #

When sourcing welding consumables from China, the first specification to request is not the tensile strength of deposited weld metal — every supplier can meet AWS minimum tensile requirements. The parameter that actually determines defect rate in production is diffusible hydrogen classification (H4, H8, H16) combined with the supplier’s documented re-drying protocol and shelf-life data under defined storage conditions. Most buyers ask for the mill certificate. Ask instead for the hydrogen test data and the storage condition validation.

The sourcing mistake with the most direct production consequence is accepting a Chinese electrode lot without verifying storage conditions during transit. A lot shipped in a container through a humid port environment in summer can absorb enough moisture to shift from H4 to H8 classification before it reaches your receiving dock. The COA will still show H4. Your weld will show porosity.

Before committing to volume order from any Chinese welding consumable supplier, require: three consecutive batch COAs with CE and hydrogen classification data, a re-drying procedure document with temperature and time specifications, and a sample lot for incoming diffusible hydrogen testing per ASTM International E3044 at your facility or a third-party lab. If the supplier cannot provide consecutive batch COAs, that is your answer.

For buyers also sourcing abrasives and cutting consumables used in weld preparation and post-weld finishing, lot-to-lot consistency in grit specification directly affects surface cleanliness and therefore porosity risk at the base metal interface.

Frequently Asked Questions #

Q1: What is the most reliable incoming inspection test for moisture-related porosity risk in low-hydrogen electrodes?

A: Diffusible hydrogen testing per ASTM International E3044 after a 300°C/1-hour bake. Any H4-rated lot returning above 4 mL/100g is quarantined — COA values are irrelevant once the lot has been exposed to ambient humidity.

Q2: How do I select between E7018 H4 and E7018 H8 for structural carbon steel applications?

A: Use H4 for any base metal with CE above 0.40 or joint thickness above 25mm. H8 is acceptable for lower-CE, lower-restraint applications. The AWS Welding Standards A5.1 classification table defines the hydrogen limits — H4 means ≤4 mL/100g, H8 means ≤8 mL/100g deposited weld metal. The difference sounds marginal. In restrained joints on higher-CE steel, it determines whether you get cold cracking 48 hours after inspection.

Q3: What is the most common sourcing failure pattern for Chinese welding electrodes that passes initial qualification but fails in production volume?

A: Raw material substitution at the wire rod or flux compounder level between the qualification sample and production batches. We have seen suppliers pass initial sample approval with H4-compliant product and then deliver H8-boundary material at volume because their flux raw material source changed. The trigger is almost always a cost reduction at the compounder level. Three consecutive batch COAs before qualification is the minimum screen — it will not catch every substitution, but it catches the most common pattern.

Q4: What certification documentation should I require for welding consumables used in pressure vessel fabrication?

A: Require a PED-compliant test certificate if supplying into European pressure equipment, or an ASME Section II Part C material certification for ASME-coded work. Both require third-party witnessed testing — a supplier self-declaration is not sufficient. Also request the SAC China Standards GB/T 5117 compliance certificate and cross-check the CE values against your base metal specification before approving the lot.

Q5: Does increasing preheat temperature always reduce cold cracking risk?

A: Yes, up to the point where you introduce HAZ softening or distortion problems — but for most carbon and low-alloy steels below CE 0.60, the practical preheat range of 100–150°C is well below that threshold. The more common error is applying the correct preheat temperature to a consumable that has already exceeded its moisture exposure limit. Preheat compensates for microstructure susceptibility. It does not remove hydrogen that is already in the electrode.

Published by sinoraw.com Technical Team | Request a sourcing consultation