Start With the Duty and Split the Weld
Automated welding in pressure vessels is not one decision. It is two weld systems with different failure modes. The pressure boundary is structural containment; a planar defect or an uncontrolled heat input can become a catastrophic release. The wear package is a hardfacing or cladding surface; it can tolerate limited volumetric defects, but dilution, thickness, and cracking control decide whether the vessel lasts.
Lumping these into one RFQ line item is the fastest way to buy the wrong automation.
The service duty sets the stakes. In bulk handling, minerals processing, and slag service, abrasion drives roughly 67% of component wear. Raw mill material beds in vertical mill pressure housings can exceed 300 mm thick and carry high impact-abrasion loads.
Slag and clinker mill beds often run 5080 mm thick, concentrating contact stress and accelerating wear. Hot slag above 500C adds thermal cycling and oxidation. A hardfacing procedure that is only close enough becomes a maintenance bill. A pressure-boundary procedure that is only close enough becomes a code problem.
Before automation is discussed, the design package must state internal pressure, operating temperature, corrosion and abrasion duty, and which surfaces are pressure boundary versus wear protection. Every later decision hangs off that split. A cladding procedure for a carbon steel shell is not the same as a hardfacing procedure for slurry-handling internals, and no amount of machine repeatability lets a fabricator pretend otherwise.
The RFQ should carry two line items: pressure-boundary welds with full NDT and tight heat input, and wear-package welds with controlled dilution, thickness, and cracking acceptance. When they are priced as one, the cheaper quote usually prices only one of them.
The Qualification Decision: Standing Cell or Campaign Split
There are two workable qualification routes. Route A is a standing, dual-qualified automated cell: the shop qualifies automated procedures for both the pressure boundary and the hardfacing or cladding package, certifies operators for that cell, and maintains the WPS/PQR envelope as production changes. Route B is campaign-by-campaign split qualification: structural work is qualified and run one way, hardfacing another, and each campaign is re-qualified or extended only as needed.
Route A moves money and risk upstream. It requires more procedure qualification hours, more fixture and positioner capital, more calibration discipline, and more operator certification. It pays back when the same families of vessels repeat. A standing cell justifies itself when the shop can keep it fed with similar joint types, thickness ranges, and cladding areas.
If the cell will run fewer than 600 m of pressure-boundary weld and less than 250 m of cladding per year, the fixed cost is hard to defend. If repeated seam length exceeds 2,000 m per year and cladding area exceeds 300 m per year, the standing cell usually wins on rework avoidance and NDT consistency.
Route B keeps capital lower and suits a high-mix shop. Its weakness is that every campaign can reopen essential variables: process, position, thickness range, joint design, wire diameter, heat input, and dilution. Re-qualification can consume 40120 engineering and welding hours per campaign, and repeated coupon testing can cost more than the labour it replaces.
The deciding variable is volume, but not total shop volume. It is repeatable volume inside the qualified envelope. A shop with 40 small, dissimilar vessels per year may be better served by manual or semi-automatic welding plus targeted automation on one seam family.
A shop with 1520 similar vessels per year and a stable cladding geometry can justify a standing cell.
What Actually Changes at the Machine
Automation does not remove welding metallurgy. It converts welding variables into machine variables that must be policed. Travel speed, oscillation width, oscillation frequency, wire feed, torch standoff, torch angle, gas flow, preheat, interpass temperature, and consumable lot all become part of the procedure.
The WPS must defend each number inside the chosen code framework: ASME Section IX for procedure and performance qualification, the EN ISO 15614 series for procedure qualification, the EN ISO 15609 series for welding procedure specifications, and the PED technical file where CE marking applies.
Heat input is the variable that most often exposes a weak automation plan. Under ISO 15614-1, heat input must remain within 10% of the qualified value. That is not a paperwork tolerance. It is a machine tolerance. On a pressure-boundary weld at 280 A and 28 V, a heat input of 1.2 kJ/mm requires a travel speed near 0.39 m/min. If the carriage drifts 10% slower, heat input rises about 10% and the procedure is out of envelope. The machine needs calibrated speed control, not just a speed display.
Operator certification changes too. An automated welding operator is not certified only to start an arc. The certificate must cover machine setup, parameter control, fixture condition, consumable handling, and the ability to detect drift before a weld is buried.
The sign-off at the machine should identify the WPS number, qualified ranges, wire or powder lot, gas and flux, preheat and interpass readings, calibration status, joint gap and mismatch, and purge condition where applicable. Records should be weld-specific.
A vessel that leaves the shop with a generic procedure file and no machine log is difficult to defend when a repair appears five years later.
Fixtures, Positioners, and Fit-Up Economics
The procedure qualification clause sets the fixture load case. A fixture that deflects under clamping force and welding heat changes the joint gap. A changing gap changes dilution, fill volume, and heat input. Under ISO 15614-1, the 10% heat input tolerance is therefore also a fixture and positioner requirement.
The positioner must hold rotation speed within a few percent, and the clamp must repeat joint alignment closely enough that the arc sees the same groove from pass to pass. For typical shell seams, a mismatch target of 1 mm or less and a root gap held within 0.5 mm is a more useful automation specification than a catalogue claim about rigidity.
Hardfacing thickness drives encoder and calibration discipline. If the wear package calls for 6 mm deposited thickness with a +0.5/0 mm tolerance, the machine must control stepover, oscillation, and wire or powder feed rate. A 5560% stepover is common for cladding beads, but the exact value belongs in the procedure.
The calibration log should cover carriage speed, wire feed, oscillation width, and gas flow. A positioner that is mechanically sound but not calibrated is a source of hidden heat-input variation.
Fit-up is the hidden multiplier. A root gap that opens from 2 mm to 4 mm can consume 3050% more filler and heat on the root side. Automation can compensate with adaptive control, but only inside a declared envelope. The fit-up sheet must show bevel angle, land, root gap, mismatch, purge, preheat, and the measurement points.
A joint audit before production is cheap. It usually pays for itself by exposing one seam family where manual fit-up cannot meet the automated cells repeatability requirement.
Heat Input, Distortion, and the Fit-Up Multiplier
For arc welding, heat input in kJ/mm is roughly proportional to volts times amps divided by travel speed, corrected for process efficiency. Push amperage up, slow the carriage down, or let a wide root gap swallow extra filler, and the vessel sees more heat per unit length.
The weld metal contracts on cooling, the adjacent base metal has already been pushed by the arc, and a restrained shell converts that thermal strain into distortion and residual stress. Automated welding does not remove the physics. It makes heat input repeatable enough to manage.
Planning windows depend on process, material, and procedure qualification, but the shop needs numbers before it can specify a cell. Orbital TIG on thin pressure-boundary tube and shell seams commonly runs 80220 A at 913 V and 60120 mm/min, giving heat inputs around 0.51.2 kJ/mm.
Hot-wire TIG cladding might run 220320 A at 1014 V and 120250 mm/min, with dilution held to roughly 515%. Plasma cladding may run 180300 A at 2432 V and 80180 mm/min, with dilution around 310%. Automated hardfacing on 5080 mm mill beds might use 300550 A at 2632 V and 0.150.4 m/min, with preheat around 150250C for crack-sensitive deposits.
These are starting points for procedure development, not universal values.
Too low a heat input risks lack of fusion and cold laps, especially on thick-section pressure boundaries. Too high a heat input buys distortion, wider HAZ, lower toughness, and higher dilution on cladding. The practical control is to write heat input as kJ/mm in the RFQ and WPS, not as amperage alone. Then require the machine to calculate and log it per weld. If the shop cannot show the calculation and the speed calibration, it cannot defend the 10% claim.
Process Choices and Where They Fail
The process name is less useful than the process envelope. Two shops can both claim automated welding and mean different things: one is orbital TIG on tube-to-tubesheet joints, the other is submerged arc on 80 mm shell seams. Buy the envelope: process, position, thickness range, joint type, heat input window, dilution limit, deposition rate, and NDT tier. The table below is a comparison for fabrication managers, not a selection shortcut.
| Process | Typical pressure-vessel use | Planning values in context | Main failure mode | What to specify |
|---|---|---|---|---|
| Orbital TIG | Tube-to-tubesheet, thin shell and nozzle seams, clean pressure boundary | 80220 A, 913 V, 60120 mm/min, heat input 0.51.2 kJ/mm | Lack of fusion, tungsten inclusion, poor purge | Electrode type and diameter, purge oxygen limit, arc voltage control, travel speed calibration |
| Hot-wire TIG cladding | Corrosion-resistant internal cladding on shell and heads | 220320 A, 1014 V, 120250 mm/min, dilution 515%, stepover 5560% | Dilution above limit, porosity, lack of fusion between beads | Wire feed accuracy, oscillation width, stepover, interpass temperature, dilution test method |
| Plasma cladding | Large cladding areas where deposition rate matters | 180300 A, 2432 V, 80180 mm/min, dilution 310% | Keyhole instability, powder feed drift, dilution variation | Powder feed calibration, plasma gas flow, standoff, torch cooling, bead overlap |
| Automated hardfacing | Wear package on mill beds, slurry internals, 5080 mm thick sections | 300550 A, 2632 V, 0.150.4 m/min, preheat 150250C where required | Cracking, spalling, excessive dilution, uneven thickness | Deposit thickness tolerance, hardness range, preheat and cool-down, crack acceptance, stepover |
| Submerged arc | Thick shell seams, long circumferential and longitudinal welds | 500800 A, 2834 V, 0.30.8 m/min, heat input 1.53.5 kJ/mm | Slag entrapment, lack of fusion, distortion | Flux handling, joint prep, wire/flux combination, heat input per pass, back gouging |
NDT, Acceptance, and the RFQ Language That Protects Price
Two quotes for the same vessel can differ by 30% because the RFQ did not settle acceptance. The pressure boundary should be priced with the full code-required NDT tier, typically volumetric and surface examination as applicable to the joint and material.
The wear package should be priced with its own acceptance: visual and surface examination for cracks, thickness measurement, dilution checks, and limited volumetric acceptance where the procedure allows. Lumping them hides the fact that one weld can tolerate a small pore and the other cannot tolerate a single planar indication.
Specifications that come back wrong often share the same faults. Heat input is written as amperage, not kJ/mm. Dilution is left as a word, not a maximum percentage. Cladding thickness is written as nominal, not as a tolerance. The process is named but the position, thickness range, and joint type are omitted.
The qualification basis is implied rather than named. The NDT tier is described as good workmanship rather than a code acceptance level. Each omission gives the welding house room to price what it read instead of what the vessel needs.
Cost per metre is also frequently misread. A 6 mm cladding pass deposited at 2.5 kg/h with 55% stepover may consume $25$45 per metre in wire, gas, and power before labour and equipment. A pressure-boundary TIG seam may look cheap per metre in consumables but expensive per metre in qualification, purging, and NDT.
Automation changes the balance: it reduces rework and variability, but it adds fixture cost, calibration labour, and engineering hours. The RFQ should ask for those fixed costs separately so the fabrication manager can see the real break-even volume.
Operator Certification and Records That Survive the Vessel
Automated welding operators need certification for the process and machine arrangement they run. The certificate should cover the essential variables that the operator can affect: setup, parameter entry, wire or powder feed, gas flow, oscillation, travel speed, preheat, interpass, and fixture condition.
It should also cover what the operator must do when the coupon disagrees with the procedure. Stop, quarantine, and escalate is the correct response; adjusting the machine to make the coupon pass is not.
Re-qualification is required when an essential variable moves outside the qualified range. That includes a process change, a position change, a joint design change, a thickness range change, or a heat input shift beyond 10% of the qualified value. The shop should maintain a register that links each vessel to the WPS, PQR, operator certificate, machine calibration, consumable certificates, preheat and interpass records, heat input calculations, and NDT reports.
These records must outlive the vessel. A procedure file that cannot be reconstructed after a maintenance weld is a liability, not a qualification.
Where Automation Does Not Pay
Automation is not worth it on one-off vessels with irregular joint access, frequent design changes, and fit-up that varies more than the cell can correct. If a shop builds fewer than 15 similar vessels per year, or runs less than 400 m of pressure-boundary weld and 150 m of cladding per year, a standing cell is usually a capital trap. The qualification hours, fixture design, calibration, and operator certification are real costs, and they do not disappear when the cell is idle.
It is also not worth it where manual skill is doing the tolerance correction. A vessel with root gaps varying by 23 mm, mismatched shell courses, and heavy tack welds that must be blended will defeat a rigid automated cell unless the shop first fixes fit-up.
Hardfacing on worn, irregular surfaces is another poor first target: the wear pattern is not the drawing, and the machine may spend more time avoiding defects than depositing metal. Repair and rework are the same story. If more than 8% of production requires rework, automation may simply make the rework more repeatable.
Fix the upstream process first.
Capital payback should be calculated against the weld family, not the whole shop. A cell costing $350,000$900,000 needs a stable production stream to amortize. If the payback exceeds five years on the repeatable volume, the honest answer is targeted semi-automation, better fixtures, and manual welding on the low-volume work. A trade audience should trust that answer more than a claim that every pressure vessel shop needs a fully automated cell.
Specification Set for a Fabrication Manager
The specification should force the welding house to price the same scope you intend to buy. Use the following set as the minimum commercial and technical control.
- Split the RFQ into pressure-boundary welds and wear-package welds, with separate NDT tiers, acceptance criteria, and heat input limits.
- Name the qualification basis and code edition, and require the WPS/PQR register to be submitted before production.
- Specify process, position, joint type, thickness range, and whether cladding or hardfacing is included in the same cell.
- Write heat input as a kJ/mm range with the calculation method and the 10% tolerance; require per-weld machine logs.
- Set dilution limits for cladding, deposit thickness tolerances for hardfacing, and the test method used to verify both.
- Define fit-up tolerances: root gap, mismatch, bevel angle, land, and measurement points. Require a pre-production joint audit.
- Require positioner speed calibration, wire or powder feed calibration, gas flow calibration, and a calibration sticker on the machine.
- Specify operator certification scope, re-qualification triggers, and records retention for the vessel life.
- Require consumable certificates, preheat and interpass records, purge records where applicable, and NDT reports linked to weld identification.
- Include a volume trigger: if annual repeatable seam length or cladding area falls below the stated threshold, the automation route is reviewed rather than assumed.
Automated welding earns its place on pressure vessels when the duty is split correctly, the procedure envelope is qualified before the cell is bought, the fixture and positioner are treated as heat-input controls, and the RFQ prices the NDT and wear package honestly. It fails when the shop automates a fit-up problem, hides dilution behind a process name, or buys a standing cell for campaign work.
The fabrication managers job is not to choose automation. It is to choose the weld family where repeatability has value, then specify the repeatability tightly enough that the machine can be defended in the code file and on the floor.
Automated welding in other sectors
If you work across more than one sector, these guides use the same engineering logic with the numbers that differ by industry.
Self-Aligning Welding Rotator (SAR Series)
Technical Specifications:
| Model | Max Load (tons) | Vessel Dia (mm) | Wheel DiaWidth (mm) | Motor Power (kW) | Wheel Speed | Overall Dim (mm) |
|---|---|---|---|---|---|---|
| SAR5 | 5 | 250-2300 | 250140 | 0.75 | 0.1-1 m/min | 1620690700 |
| SAR10 | 10 | 320-2800 | 300170 | 1.5 | 0.1-1 m/min | 1880780820 |
| SAR20 | 20 | 500-3500 | 350180 | 2.2 | 0.1-1 m/min | 2250930950 |
| SAR40 | 40 | 600-4200 | 400200 | 3.0 | 0.1-1 m/min | 250010151050 |
| SAR60 | 60 | 750-4800 | 450220 | 4.0 | 0.1-1 m/min | 300012001100 |
| SAR80 | 80 | 850-5000 | 500240 | 4.0 | 0.1-1 m/min | 320012001250 |
| SAR100 | 100 | 1000-5500 | 50080 (PU) | 5.5 | 0.1-1 m/min | 350014101350 |
| SAR150 | 150 | 1000-6000 | 500220 (Metal) | 6.0 | 0.1-1 m/min | 42009001450 |
| SAR200 | 200 | 1800-8000 | 550260 | 7.5 | 0.1-1 m/min | 48009001500 |
| SAR250 | 250 | 1800-8000 | 660260 | 11 | 0.1-1 m/min | 501011131600 |
| SAR500 | 500 | 1800-8000 | 800300 | 15 | 0.1-1 m/min | 560013702100 |
Note: Variable rotation speed (VFD). Wheels padded with heavy-duty rubber (SAR100 & smaller). Remote control with digital speed display.
For detailed specifications and pricing, contact us at [email protected] or visit https://biqee.cn.
Conclusion
Choosing the right welding automation solution is key to improving productivity and quality in your fabrication operations.
