Doc #94 — Welding Consumable Fabrication

---

RECOMMENDED ACTIONS

First week (Phase 1)

1. Inventory all welding consumable stocks nationally — contact BOC/Linde, Air Liquide, Lincoln Electric NZ, ESAB agents, Cigweld agents, NZ Safety Blackwoods, Farmlands, PGG Wrightson. Establish total NZ stock of stick electrodes (by type and size), MIG wire, shielding gas cylinders, oxy-acetylene gas, TIG consumables, and flux-cored wire. Include this in the Category B stockpile survey (Doc #1).
2. Classify welding consumables as controlled materials. Prevent hoarding and secondary-market profiteering.
3. Verify BOC/Linde and Air Liquide NZ operational status. Confirm air separation plants are running, oxygen and argon supply is uninterrupted, and acetylene filling stations are functional.
4. Issue national guidance to welding workshops: Default to stick welding for all general fabrication. Reserve MIG for thin material, production runs, and applications requiring MIG’s ease of use. Reserve TIG for critical-quality and non-ferrous applications only.

First month (Phase 1)

5. Identify NZ metallurgists and welding engineers with knowledge of electrode flux chemistry. Universities (University of Auckland, University of Canterbury), HERA (Heavy Engineering Research Association), and NZ welding inspection firms are likely sources of expertise.
6. Secure Glenbrook EAF slag samples for laboratory analysis — determine TiO2 content, mineralogy, and suitability for electrode flux. Begin formulation experiments.
7. Assess Pacific Steel (Otahuhu) wire drawing capability for electrode core wire production. Can existing equipment draw wire in the 2.5–5.0 mm range from available feedstock? What is the wire chemistry?
8. Source and stockpile raw materials for electrode flux development: limestone (from existing quarries), cellulose (from sawmills), silica sand (from glass sand deposits).
9. Allocate welding consumables through a controlled distribution system. Essential repair activities (agricultural equipment, hydro maintenance, essential infrastructure) receive priority allocation.

First 3 months (Phase 1)

10. Produce first experimental electrode batches from NZ materials. Test systematically — arc stability, weld quality, slag behaviour. Document results and iterate.
11. Begin sodium/potassium silicate binder production development. Small-scale silicate production from silica sand and wood ash potash or soda ash.
12. Assess calcium carbide production feasibility. Identify suitable location (near limestone quarry and electricity supply), estimate power requirements, develop preliminary furnace design based on published industrial designs.
13. Establish CO2 capture from NZ fermentation sources. Contact breweries and distilleries. Assess volume, purity, and compression/bottling requirements for welding-grade CO2.
14. Begin training program for stick welding skills for welders who are primarily MIG/TIG trained. Many NZ welders under 40 have limited stick welding experience — they need to develop this skill before it becomes their primary process.

First year (Phase 1–2)

15. Achieve pilot-scale electrode production — functional electrodes in sufficient quantity for limited field use. Distribute to selected workshops for feedback and field testing.
16. Establish NZ CO2 production and cylinder filling from fermentation or lime kiln sources for MIG shielding gas.
17. Complete calcium carbide furnace design. Begin fabrication if resources are available.
18. Scale up silicate binder production to match electrode manufacturing needs.
19. Develop welding procedure specifications (WPS) for NZ-produced electrodes — document the welding parameters (current, travel speed, technique) that produce acceptable results with the specific NZ electrode formulations.
20. Coordinate with Doc #105 (Wire Drawing) development — electrode core wire is a priority product for the wire drawing program.

Years 1–3 (Phase 2)

21. Scale electrode production to national supply levels — hundreds of tonnes per year. Establish quality control procedures and batch testing.
22. Commission calcium carbide furnace. Begin acetylene production from calcium carbide and water.
23. Develop MIG wire production once wire drawing capability is established (Doc #105).
24. Expand electrode product range — develop cellulosic electrodes for pipeline and deep-penetration work; attempt basic (low-hydrogen) electrodes with available materials (without fluorspar, using increased limestone and other substitutes).
25. Establish training for electrode manufacturing workers — the factory itself needs trained production staff, quality control personnel, and a metallurgist.

Years 3–7 (Phase 3)

26. Refine electrode formulations based on accumulated production and field experience. Quality should improve steadily with iteration.
27. Develop flux-cored wire if both wire drawing and flux manufacturing are established — combines the productivity of MIG with the self-shielding advantage of stick.
28. If trade with Australia develops, import fluorspar for basic electrode production, and high-quality welding consumables for critical applications (pressure vessels, structural certification work).
29. Assess long-term oxygen production alternatives (electrolysis, PSA) as backup for ASU degradation.

---

ECONOMIC JUSTIFICATION

10.1 The cost of not having welding consumables

If welding consumable stocks deplete without domestic production being established, the consequences cascade through the entire recovery effort:

• Gasifier construction (Doc #56) stops — NZ loses a critical vehicle fuel pathway
• Agricultural equipment repair degrades — broken implements stay broken, reducing food production
• Structural fabrication becomes limited to bolting and riveting — slower, heavier, more material-intensive
• Hydro station maintenance is constrained — some repairs cannot be done without welding
• Boatbuilding (Doc #138) is severely limited — maritime recovery is delayed
• Machine shop build-up welding stops — worn components cannot be rebuilt, accelerating equipment loss

The welding capability gap would affect virtually every technical recovery activity in the library.

10.2 Investment required

Electrode production facility:

• Facility construction: 6–12 months to design and build, using NZ steel and components. Includes extrusion press, mixing equipment, drying ovens, raw material storage, testing laboratory.
• Personnel: Approximately 15–30 workers for a facility producing several hundred tonnes of electrodes per year (production workers, quality control, raw material processing, maintenance).
• Raw material supply chain: Glenbrook slag processing, limestone grinding, cellulose preparation, silicate binder production — each requires workers and equipment.
• Development cost: 3–6 person-years of skilled metallurgical work for formulation development and testing before production-quality electrodes are achieved.

Calcium carbide facility (for acetylene production):

• Facility construction: 12–18 months for furnace fabrication, electrical installation, and commissioning.
• Personnel: Approximately 10–20 workers for production and maintenance.
• Power consumption: 1–5 MW continuous during furnace operation — a significant but manageable grid allocation.

Total estimated investment: 30–60 person-years of labour for design, construction, development, and initial operation of both electrode and carbide production facilities. This is a substantial commitment but modest compared to the alternatives — the recovery cost of not having welding capability is measured in thousands of person-years of lost productivity across all welding-dependent activities.

Breakeven: Almost immediate once production begins, because the alternative (no welding) is so costly. Even low-quality NZ-produced electrodes are vastly better than no electrodes.

---

1. WHY WELDING CONSUMABLES MATTER

1.1 Welding is everywhere

A survey of the Recovery Library reveals the extent of welding dependency:

• Gasifier construction (Doc #56): Every wood gasifier body, air inlet, and gas piping assembly is welded from steel plate and pipe.
• Structural fabrication (Doc #89, #94): When NZ cannot roll structural steel sections, it must fabricate structural shapes from plate — which means welding. Welded I-beams, box sections, and plate girders replace rolled sections.
• Agricultural equipment repair: Broken chassis, worn implements, cracked castings (weld repair), and fabricated replacement parts.
• Boatbuilding (Doc #138): Steel vessel hulls, deck structures, and fittings are welded. Even wooden boat construction requires welded metal fittings, hardware, and fastenings.
• Hydro station maintenance (Doc #65): Penstock repair, gate fabrication, structural repairs — all welded.
• Machine shop build-up welding (Doc #91): Worn shafts and surfaces are rebuilt by depositing weld metal, then re-machined to dimension. This technique extends the life of irreplaceable components.
• Pipeline and tank work: Water, fuel, and gas distribution systems require welded joints for repair and new construction.
• Vehicle and rail repair: Chassis repair, body fabrication, rail joint repair.

Without welding consumables, all of these activities stop. A welder with equipment but no consumables cannot produce a weld.

1.2 Current NZ welding processes

NZ workshops use four primary welding processes, each with different consumable requirements:¹

Process	Common Name	Consumables	NZ Producibility
SMAW (Shielded Metal Arc Welding)	Stick welding	Flux-coated electrodes	Best — all ingredients available in NZ
GMAW (Gas Metal Arc Welding)	MIG welding	Bare steel wire + shielding gas (CO2/argon)	Good — wire requires drawing capability; gas producible
GTAW (Gas Tungsten Arc Welding)	TIG welding	Tungsten electrode + argon gas + filler rod	Moderate — tungsten electrodes finite; argon producible
OFW/OFC (Oxy-Fuel Welding/Cutting)	Oxy-acetylene	Oxygen + acetylene gas	Feasible — requires calcium carbide production

Flux-cored arc welding (FCAW) — a variant of MIG using tubular wire filled with flux — is also used in NZ but requires both wire drawing and flux manufacturing, making it a later-stage development.

1.3 Scale of NZ welding consumable consumption

NZ’s annual consumption of welding consumables under normal conditions is not publicly reported in aggregate. Rough estimates based on NZ’s steel consumption and fabrication activity suggest:²

• Stick electrodes: Several hundred to a few thousand tonnes per year
• MIG wire: Similar order of magnitude — MIG has increasingly replaced stick for production welding in NZ workshops
• Shielding gas: Thousands of cylinders per year, distributed across hundreds of workshops
• Oxy-acetylene gas: Declining use for welding (replaced by arc processes) but still widespread for cutting

Honest uncertainty: These are rough estimates. NZ’s welding consumable consumption data is fragmented across multiple distributors and not centrally reported. The census (Doc #8) should capture welding consumable stocks and consumption at the workshop level.

---

2. EXISTING NZ WELDING CONSUMABLE STOCKS

2.1 Who holds the stock

NZ’s welding consumable stocks are distributed across:

• Industrial gas suppliers: BOC/Linde and Air Liquide NZ are the primary suppliers of shielding gas, oxy-fuel gas, and associated welding consumables. BOC operates gas production and distribution facilities in Auckland and other locations.³
• Welding supply distributors: Companies including Welding Engineers NZ, Lincoln Electric agents, ESAB agents, Cigweld/Thermadyne agents, and general industrial suppliers (NZ Safety Blackwoods, Wurth, etc.) hold stocks of electrodes, wire, and accessories.
• Rural suppliers: Farmlands, PGG Wrightson, and rural engineering supply stores hold electrode and MIG wire stocks for the agricultural repair market. These distributed rural stocks are significant in aggregate.
• Individual workshops: Every welding workshop in NZ has stocks of consumables — typically weeks to months of normal consumption. In aggregate across thousands of workshops, this represents a substantial national inventory.

2.2 Depletion timelines

Estimate — these figures require verification through the census (Doc #8):

Consumable	Estimated NZ stock depth	Depletion under recovery conditions
Stick electrodes (all types)	Months to 1–2 years	Slower depletion if economy contracts; faster if repair demand surges
MIG wire	Months to 1 year	Similar to stick electrodes
Shielding gas (CO2, argon, mixed)	Weeks to months (gas); cylinders indefinite	Gas depletes; cylinders are reusable if gas can be produced
Oxy-acetylene gas	Weeks to months	Depletes with use; acetylene cylinders require periodic re-inspection
TIG tungsten electrodes	Years (low consumption rate per electrode)	Very slow depletion — tungsten electrodes last hundreds of hours
TIG filler rods	Months to 1 year	Similar to other consumable wire
Flux-cored wire	Months	Specialty product, smaller stocks

Key point: Gas stocks deplete first. Solid consumables (electrodes, wire) last longer. This shapes the transition strategy: NZ should conserve gas-dependent processes (MIG, TIG, oxy-fuel) and shift to stick welding — the process whose consumables are most producible locally.

2.3 Immediate actions

1. Inventory all welding consumable stocks nationally as part of the Category B stockpile survey (Doc #1). Include gas suppliers, welding distributors, rural suppliers, and major workshop stocks.
2. Classify welding consumables as controlled materials — prevent hoarding and ensure equitable distribution to essential repair and fabrication activities.
3. Issue guidance to workshops: Conserve MIG wire and shielding gas for applications where stick welding is inadequate (thin material, positional work requiring MIG’s ease of use). Default to stick welding for general fabrication.
4. Verify BOC/Linde air separation plant status — this plant’s continued operation is essential for argon and oxygen supply (Section 7).

---

3. STICK WELDING ELECTRODE FABRICATION

3.1 What a stick electrode is

A stick electrode (SMAW electrode, also called a “welding rod” colloquially) consists of two parts:⁴

1. Core wire: A length of mild steel wire, typically 2.5–5.0 mm in diameter and 300–450 mm long. The wire melts during welding and becomes the deposited weld metal. Core wire is low-carbon mild steel (approximately 0.06–0.15% carbon), drawn to diameter and cut to length.

2. Flux coating: A layer of mineral and organic materials bonded to the wire with a binder (typically sodium or potassium silicate — “waterglass”). The coating thickness varies by electrode type but is typically 1–3 mm. During welding, the flux coating melts and decomposes to produce:

   • Shielding gas: Protects the molten weld metal from atmospheric contamination (oxygen and nitrogen, which cause porosity and embrittlement)
   • Slag: A glassy layer that floats on top of the molten weld pool, protecting it from the atmosphere as it solidifies and slowing the cooling rate to reduce cracking
   • Deoxidisers and alloying additions: Chemical elements that react with oxygen in the weld pool to produce a sound, ductile weld
   • Arc stabilisers: Materials that produce ions in the arc, making it easier to strike and maintain

The flux coating is what makes stick welding self-shielding — unlike MIG or TIG, it does not require an external gas supply. This is why stick electrode fabrication is the highest priority for local production.

3.2 Flux coating types

Electrode flux coatings are classified by their primary mineral component:⁵

Rutile (titanium dioxide, TiO2) electrodes: The most common general-purpose electrode type (e.g., AWS E6013). Rutile-based flux produces a smooth, easy-to-control arc, moderate penetration, and a slag that peels off easily. These are the “easy to use” electrodes that most welders prefer for general fabrication. Rutile is the most producible flux type for NZ because of the country’s abundant titanium dioxide mineral resources (Section 3.3).

Cellulosic electrodes: Flux is predominantly cellulose (wood pulp, paper). Burns to produce a large volume of shielding gas (hydrogen, CO, CO2), giving deep penetration and a thin, friable slag. Used for pipeline welding and root passes where deep penetration is critical (e.g., AWS E6010, E6011). Cellulose is abundantly available in NZ from wood processing.

Basic (low-hydrogen) electrodes: Flux based on calcium carbonate (limestone) and calcium fluoride (fluorspar). Produces welds with very low hydrogen content, reducing the risk of hydrogen-induced cracking in high-strength or thick steels. These are the electrodes required for structural welding to standards (e.g., AWS E7018). Calcium carbonate is available in NZ; calcium fluoride (fluorspar) is not mined in NZ and is imported.⁶

Iron oxide electrodes: Flux based on iron oxides. Produce high deposition rates and are used for heavy fill passes. Less common in NZ practice.

For NZ local production, the priority order is:

1. Rutile type — easiest to produce, most widely useful
2. Cellulosic type — critical ingredients (cellulose, sodium or potassium carbonates) available in NZ
3. Basic type — limited by fluorspar availability; may be producible with substitutions

3.3 NZ raw materials for flux

Rutile / titanium dioxide (TiO2):

NZ has vast deposits of titanium-bearing minerals in its west coast ironsand. The same titanomagnetite deposits that supply Glenbrook steelworks (Doc #89) contain 7–8% TiO2.⁷ However, the titanium in ironsand is locked in the titanomagnetite mineral structure and is not directly usable as rutile for electrode flux. The EAF slag at Glenbrook is enriched in titanium dioxide — the slag from ironmaking concentrates the TiO2 that was originally in the ironsand. Glenbrook’s EAF slag typically contains 20–40% TiO2, which makes it a potential feedstock for rutile extraction.⁸

NZ’s west coast also has beach sand deposits containing ilmenite (FeTiO3, an iron-titanium oxide mineral) and some rutile grains, though these are present in lower concentrations than in major mineral sand deposits in Australia or South Africa. Ilmenite can be processed to produce synthetic rutile through chemical or thermal treatment, but this is not trivial industrial chemistry.⁹

Practical pathway: The most accessible NZ source of titanium dioxide for electrode flux is Glenbrook EAF slag. The slag is already produced as a waste product. Grinding this slag to a fine powder and incorporating it into electrode flux formulations is the most realistic near-term approach. The slag is not pure rutile — it contains other oxides (alumina, silica, magnesia) — and the flux chemistry will differ from imported rutile electrode formulations. Experimentation and testing are required. This is a genuine [B]-rated capability: the raw materials exist and the chemistry is understood, but developing a working formulation requires skilled metallurgical work and systematic testing.

Cellulose:

NZ has abundant forestry resources (Pinus radiata plantations cover approximately 1.7 million hectares)¹⁰ and existing pulp and paper production (e.g., Oji Fibre Solutions at Kinleith in the Waikato). Wood pulp — ground or chemically processed wood fibre — is the cellulose source for electrode flux. Sawdust from sawmills is an even simpler feedstock, though particle size and consistency control matter for electrode coating quality.

Calcium carbonate (limestone, CaCO3):

NZ has extensive limestone deposits, quarried commercially at multiple locations. Lime kilns operate in the Waikato (e.g., at Otorohanga) and elsewhere.¹¹ Finely ground limestone is a standard ingredient in both basic and rutile electrode coatings — it decomposes during welding to release CO2 (shielding gas) and forms a lime-based slag.

Sodium silicate (waterglass, Na2SiO3):

The binder that holds the flux coating on the wire. Produced by fusing silica sand (SiO2) with sodium carbonate (soda ash, Na2CO3) at approximately 1,100–1,400°C, then dissolving the resulting “glass” in water under pressure.¹² NZ has silica sand deposits. Sodium carbonate can be produced from NZ materials: wood ash yields potash (potassium carbonate, K2CO3), which can serve a similar function; or sodium carbonate can be produced from common salt (NaCl) via the Solvay process or the Leblanc process, both of which use NZ-available raw materials (salt, limestone).¹³

Potassium silicate (K2SiO3) is an alternative binder, produced from silica sand and potash. Potash from wood ash is directly available in NZ.

Manganese dioxide (MnO2):

Used in some electrode flux formulations as a deoxidiser and arc stabiliser. NZ has small manganese deposits that are not commercially mined (see Doc #89, footnote 24). Battery-grade manganese dioxide from exhausted dry cell batteries could be recycled. Manganese dioxide is a useful but not essential flux ingredient — formulations without it are feasible.

Iron powder:

Some electrode types include iron powder in the coating to increase deposition rate (the iron powder melts and adds to the weld deposit). Iron powder can be produced in NZ by grinding steel scrap or collecting steel filings from machine shops — both low-technology approaches that produce adequate if coarse powder. Atomisation from molten steel (spraying molten metal through high-pressure gas jets to form fine droplets that solidify as powder) produces superior, uniform powder but requires specialised nozzle equipment and an inert gas supply (argon or nitrogen from the ASU, Section 7).¹⁴ For electrode flux use, ground or filed iron powder is adequate — particle size can be controlled by sieving.

Mica:

Used as an extender in some flux formulations and to aid slag formation. NZ has some mica-bearing geological formations, though commercial-scale mica mining has not been practised. Mica is a useful but substitutable ingredient — other minerals can perform similar functions in the flux.

Feldspar:

A silicate mineral used in some electrode coatings to modify slag viscosity. Found in NZ granite and pegmatite formations. Can be quarried if needed.

Calcium fluoride (fluorspar, CaF2):

The critical ingredient for basic (low-hydrogen) electrodes. NZ has no known commercial fluorspar deposits.¹⁵ This is the primary barrier to producing basic-type electrodes locally. Basic electrodes without fluorspar are possible — using increased limestone and other fluxing agents — but will have somewhat different welding characteristics and may not achieve the same low-hydrogen weld deposit quality that fluorspar-containing formulations provide.

3.4 Raw materials summary

Ingredient	Function in flux	NZ availability	Source
Rutile / TiO2	Slag former, arc stabiliser	Available — from Glenbrook EAF slag or ironsand processing	Glenbrook slag (waste product)
Cellulose (wood pulp / sawdust)	Gas generation (shielding)	Abundant	Forestry, sawmills, pulp mills
Calcium carbonate (limestone)	Gas generation (CO2), slag former	Abundant	Limestone quarries (Waikato and elsewhere)
Sodium/potassium silicate	Binder	Producible from NZ materials	Silica sand + soda ash or potash
Iron powder	Deposition rate, deoxidiser	Available	Steel filings, ground scrap, atomised powder
Manganese dioxide	Deoxidiser, arc stabiliser	Limited — small NZ deposits, battery recycling	Mineral deposits, recycled batteries
Calcium fluoride (fluorspar)	Slag fluidity, hydrogen control	Not available in NZ	Import via trade if possible
Mica, feldspar, silica	Extenders, slag modifiers	Available in NZ	Mineral deposits

3.5 The manufacturing process

Electrode manufacturing is a multi-step process:¹⁶

Step 1 — Core wire preparation: Mild steel wire of the appropriate diameter (typically 2.5, 3.2, 4.0, or 5.0 mm) is straightened and cut to length (300–450 mm). One end is left bare (approximately 25–30 mm) for clamping in the electrode holder. Wire must be clean, dry, and free of rust and oil.

Wire source under recovery conditions: NZ does not currently produce wire rod (Doc #89, Section 7). Wire drawing capability must be developed (Doc #105). In the interim, core wire can be drawn from existing NZ wire stocks — fencing wire, tie wire, and other mild steel wire products can potentially be re-drawn to electrode core dimensions, though chemical composition must be verified. Wire with excessive carbon, sulfur, or phosphorus produces poor welds.¹⁷

Step 2 — Flux preparation: Mineral ingredients are dried, crushed, and ground to a fine, consistent powder. Particle size matters — coarse particles produce an uneven coating that disrupts arc stability. Industrial electrode manufacturers use ball mills and vibrating screens to achieve controlled particle sizes (typically 100–250 mesh for most ingredients).¹⁸

Under NZ recovery conditions, grinding can be achieved using existing industrial ball mills, hammer mills, or rod mills. Fine grinding of minerals is standard practice in NZ’s mining and cement industries (Doc #97). The quality of the grinding directly affects electrode quality — this is not a step where “close enough” is adequate.

Step 3 — Dry mixing: The ground dry ingredients are blended in controlled proportions. A representative starting formulation for a rutile-type electrode (analogous to E6013):¹⁹

• Rutile (TiO2) or Glenbrook slag: 30–45% by weight of dry flux
• Calcium carbonate (limestone): 15–25%
• Cellulose (fine wood pulp): 5–15%
• Feldspar or silica: 5–15%
• Iron powder: 5–15%
• Manganese dioxide: 2–5% (if available)
• Mica: 2–5% (if available)

These proportions are approximate starting points. Actual formulations must be developed through systematic experimentation and weld testing. Published electrode flux recipes exist in welding metallurgy literature, but translating them to NZ-available materials — particularly using Glenbrook slag rather than pure rutile — requires adaptation.²⁰

Step 4 — Wet mixing: Sodium or potassium silicate solution (the binder) is added to the dry mix to form a stiff, plastic paste. The consistency must be correct — too wet and the coating slumps off the wire; too dry and it cracks. The silicate solution concentration (typically 35–45% solids in water) and the ratio of silicate to dry ingredients are critical variables that must be determined experimentally.

Step 5 — Application to wire: The flux paste is applied to the core wire. In industrial production, this is done by extrusion — the wire is fed through a die that applies a concentric coating of flux paste, producing uniform thickness. Under NZ recovery conditions, extrusion equipment can be fabricated (it is a metal die, a hydraulic press or screw press, and a wire feed mechanism — within NZ’s machine shop capability, Doc #91). The hydraulic press requires hydraulic fluid, which is an imported consumable with finite stock; a screw-driven mechanical press avoids this dependency.²¹ For initial development and small-scale production, hand-dipping or trough-coating methods can produce functional if less uniform electrodes.²²

Step 6 — Drying: Coated electrodes must be dried to remove moisture from the silicate binder. This is done in two stages:

• Air drying for several hours to set the coating
• Oven baking at 150–350°C (depending on electrode type) for 1–2 hours to drive off residual moisture and cure the silicate binder

For basic (low-hydrogen) electrodes, higher baking temperatures (350–450°C) are required and moisture control is critical — these electrodes absorb atmospheric moisture and must be stored in heated ovens or sealed containers before use.²³

Under NZ conditions, drying ovens can be constructed from steel plate and lined with NZ-produced fireclay refractory (see Doc #89, Section 5.3). Heat source: electric elements (grid power is available) or wood-fired.

Step 7 — Quality testing: This is the step most likely to be underestimated. Producing an electrode that looks like a welding electrode requires only the physical steps above. Producing one that welds reliably — stable arc, adequate penetration, sound weld metal, manageable slag — requires systematic testing and iteration that industrial manufacturers treat as a distinct quality-engineering discipline.

Testing should include:

• Arc stability: Does the electrode strike easily and maintain a stable arc? Or does it sputter, go out, or produce an erratic arc?
• Weld bead appearance: Is the bead uniform, with consistent width and reinforcement?
• Slag removal: Does the slag peel off cleanly, or does it stick and require chipping?
• Weld soundness: Are there visible defects (porosity, undercut, cold lap)? If NZ has ultrasonic or radiographic testing capability, internal soundness should be checked.
• Mechanical testing: Bend tests (can the weld bend without cracking?), tensile tests (does the weld have adequate strength?). These require a tensile testing machine — NZ has these in materials testing laboratories at universities and some industrial facilities.

Honest assessment: Developing a working electrode formulation from NZ materials will take weeks to months of systematic experimentation, not days. The first attempts will produce electrodes that are difficult to use and produce marginal welds. Iterating toward a reliable product requires metallurgical understanding, welding skill, and patient testing. This is a genuine [B]-rated capability — it is feasible, the materials exist, the process is understood, but it requires significant effort and will not produce electrodes equivalent to imported products.

3.6 Expected performance gaps

NZ-produced stick electrodes will be inferior to imported products in several respects:

• Arc stability: Imported electrodes are formulated and manufactured with precise quality control. NZ-produced electrodes, particularly early production runs, will likely have less stable arcs — harder to strike, more spatter, less forgiving of poor technique.
• Slag behaviour: Imported rutile electrodes produce self-peeling slag. NZ electrodes may produce slag that is harder to remove, stickier, or less uniform in coverage.
• Weld quality: Weld metal purity, ductility, and strength may be lower. For non-critical applications (general fabrication, agricultural repair), this is acceptable. For structural or pressure-vessel welding, lower-quality electrodes are a significant concern.
• Consistency: Batch-to-batch variation will be greater than imported products. Each batch of NZ-produced electrodes should be tested before being issued for critical work.
• Moisture sensitivity: Without precise moisture control during manufacturing and storage, NZ electrodes may absorb more moisture, leading to porosity in welds and, for basic electrodes, hydrogen cracking risk.

These performance gaps are real and important but do not make the electrodes useless. For the vast majority of recovery welding — repair fabrication, structural brackets, agricultural equipment, gasifier construction, tank work — a functional if imperfect electrode produces a functional if imperfect weld, which is vastly better than no weld at all.

3.7 Production scale

NZ’s recovery welding demand is difficult to estimate but will be substantial. A rough order-of-magnitude calculation:

• If NZ has 2,000–5,000 active welders (an estimate requiring census verification — Careers NZ estimated approximately 6,700 welders employed in 2022, but recovery-conditions activity may reduce active numbers).²⁴
• Each welder uses approximately 2–10 kg of electrode per working day, depending on the intensity of the work — lighter repair work uses less; heavy structural fabrication can exceed 10 kg.²⁵
• Annual electrode consumption could range from approximately 1,500 to 18,000 tonnes per year across the full range of assumptions, with the most likely recovery scenario (2,000–3,000 active welders at moderate intensity) suggesting 2,000–6,000 tonnes per year.

Electrode production at this scale requires:

• Wire: Several hundred to several thousand tonnes of core wire per year
• Flux materials: Similar tonnage of mineral and organic ingredients
• Manufacturing equipment: Extrusion press, mixing equipment, drying ovens
• Quality control: Testing facility and skilled metallurgist

Assessment: This is not a cottage industry. An electrode factory serving NZ’s needs would be a significant manufacturing operation requiring tens of workers, purpose-built equipment, and reliable supply chains for raw materials. But it is within NZ’s capability — the equipment is fabricable from NZ steel, the materials are available, and the process is well-documented.

---

4. MIG WIRE AND SHIELDING GAS

4.1 MIG wire

MIG (GMAW) welding uses bare steel wire, typically 0.8 mm, 0.9 mm, or 1.2 mm diameter, fed continuously from a spool through a welding gun into the arc. The wire melts and becomes the weld deposit. Standard MIG wire is mild steel with a thin copper coating (a few micrometres thick) that provides corrosion resistance during storage and improves electrical contact in the welding gun.²⁶

Production requirements:

• Wire drawing: Steel wire must be drawn through progressively smaller dies to achieve the required diameter. NZ does not currently have wire drawing capability at the fine diameters required for MIG wire (0.8–1.2 mm), though wire drawing is addressed in Doc #105. Dies for fine wire drawing are precision-made items (Doc #91). The wire drawing process is feasible using NZ-produced steel from Glenbrook (once wire rod production is established) but represents a significant development effort.
• Wire chemistry: MIG wire chemistry must be controlled to produce sound welds. Standard MIG wire (e.g., AWS ER70S-6) contains small additions of silicon (~0.8–1.0%) and manganese (~1.4–1.6%) as deoxidisers.²⁷ Without these additions, the weld metal absorbs oxygen from the arc atmosphere, producing porosity. Achieving this chemistry requires either starting with appropriately alloyed steel from Glenbrook (which would need to develop wire rod of the right composition — a non-trivial metallurgical task) or adding deoxidisers by other means.
• Copper coating: The copper coating on MIG wire is applied by electroplating or chemical displacement from a copper sulfate solution. The dependency chain: copper sulfate requires copper (NZ has limited supply — small historical mines and recycled material) and sulfuric acid (not currently produced in NZ at scale; see Doc #113 for production pathway). Uncoated MIG wire works — it rusts more quickly in storage (requiring dry, sealed packaging) and feeds less smoothly through the gun liner (increasing the risk of birdnesting and feed jams) — but produces functional welds.²⁸
• Spooling: Wire must be wound onto spools in a way that feeds smoothly through the MIG gun without tangling. This requires a winding machine, which can be fabricated but requires precision.

Assessment: MIG wire production is feasible [B] but more demanding than stick electrode production because of the wire drawing requirement and the tighter chemistry control needed for deoxidiser content. It is a Phase 3 capability — dependent on wire drawing infrastructure (Doc #105) being in place.

4.2 Shielding gas for MIG welding

MIG welding requires a shielding gas to protect the molten weld pool from atmospheric contamination. Common shielding gas mixtures for mild steel welding:²⁹

• CO2 (100%): Cheapest option. Produces a hotter, more penetrating arc with more spatter than argon-containing mixtures. Adequate for most structural and general fabrication.
• Argon + CO2 (typically 75–80% Ar, 20–25% CO2): The standard mixture for general-purpose MIG welding. Produces a smoother arc, less spatter, and better-looking welds than pure CO2.
• Argon (100%): Used for non-ferrous metals (aluminium, stainless). Not typically used for mild steel.

CO2 production:

CO2 is the easier shielding gas to produce in NZ:

• Fermentation: Alcoholic fermentation produces CO2 as a byproduct. NZ has breweries (DB Breweries, Lion NZ, numerous craft breweries) and distilleries that produce CO2 during fermentation. Under normal conditions, some NZ breweries already capture and sell CO2 for carbonation and industrial use.³⁰ Under recovery conditions, alcohol production for fuel (ethanol) and morale purposes will continue and potentially increase, providing a CO2 source. The CO2 must be captured, compressed, and purified for welding use — moisture and organic contaminants must be removed.
• Combustion with capture: Burning carbon-containing fuel (coal, charcoal, natural gas) produces CO2 and water vapour. The CO2 can be separated from the combustion gases by absorption in water or alkaline solution (e.g., lime water) and then released by heating the solution. This is more complex than fermentation capture but produces larger volumes.
• Limestone calcination: Heating limestone (CaCO3) produces lime (CaO) and CO2. Lime kilns already operate in NZ. The CO2 can be captured from the kiln exhaust and purified. This is a dual-purpose process — producing lime (needed for steelmaking, construction, and other applications) and CO2 simultaneously.³¹

CO2 purification and compression:

Welding-grade CO2 requires moisture content below approximately 200 ppm to avoid porosity in welds.³² Captured CO2 must be dried (using desiccants or refrigeration) and compressed into high-pressure cylinders (typically 5,000–6,000 kPa). Compression requires a suitable compressor. High-pressure cylinder filling requires cylinders rated for the pressure and filling equipment. NZ’s existing BOC/Linde and Air Liquide infrastructure includes cylinder filling stations.

Argon production:

Argon constitutes approximately 0.93% of Earth’s atmosphere by volume. It is separated from air by fractional distillation in an air separation unit (ASU) — the same process that produces industrial oxygen and nitrogen.³³ BOC/Linde operates air separation facilities in NZ that produce argon as a product.³⁴ As long as these facilities operate (they require electricity, which is available under baseline grid assumptions), argon supply continues.

However: The ASU is imported equipment. If it fails and cannot be repaired from NZ-available parts, argon production stops. Building a new ASU from scratch is a major industrial project requiring precision heat exchangers, compressors, and cryogenic technology. Argon should be treated as a finite resource in the long term — available while the existing ASU operates, uncertain thereafter.

Mixed gas production:

Producing argon-CO2 mixtures for MIG welding requires mixing the two gases in controlled proportions and filling cylinders with the mixture. The dependency chain: calibrated gas mixing manifolds, mass-flow controllers or rotameters for proportioning, high-pressure cylinder filling equipment (rated for the mixture pressure), and test equipment to verify the fill composition. NZ’s industrial gas suppliers already have this infrastructure; it continues to function as long as the component gases are available and the filling equipment is maintained.

4.3 MIG welding under recovery conditions

Near term (Phase 1–2): Continue using existing MIG wire and shielding gas stocks. Conserve by using MIG only where its productivity advantage over stick welding justifies the consumable cost — production fabrication, thin material, and applications where weld quality is critical.

Medium term (Phase 2–3): Transition shielding gas to 100% CO2 from NZ production (fermentation capture, lime kiln capture). Compared to standard 75/25 argon-CO2 mixtures, pure CO2 produces noticeably more spatter (requiring additional post-weld cleanup), deeper but less controlled penetration, a harsher arc that is less forgiving of poor technique, and higher fume levels requiring better ventilation.³⁵ These are real but manageable performance gaps — for general fabrication and repair work, pure CO2 shielding produces sound welds. Reserve argon-containing mixtures for thin material, out-of-position work, and critical applications where weld appearance and minimal rework matter.

Long term (Phase 3+): Develop MIG wire production once wire drawing capability is established. MIG with NZ-produced wire and CO2 shielding becomes a sustainable welding process.

---

5. OXY-ACETYLENE GAS

5.1 Current use

Oxy-acetylene (oxy-fuel) welding and cutting uses a flame produced by burning acetylene (C2H2) in oxygen (O2). The flame temperature — approximately 3,100–3,300°C — is sufficient to melt steel.³⁶

In NZ practice, oxy-acetylene is used for:

• Cutting: The primary current use. Oxy-acetylene cutting (torch cutting) is the standard method for cutting steel plate and structural sections in fabrication shops.
• Welding: Less common than arc welding for steel but still used for thin sheet, pipe, and repair work. Gas welding produces a slower, more controllable heat input than arc welding, which is an advantage for some applications.
• Brazing and soldering: The oxy-acetylene flame is used for brazing (joining with brass filler — temperatures ~800–900°C) and silver soldering.
• Heating: Pre-heating steel before welding (required for some materials and thicknesses), bending, and straightening.

5.2 Oxygen production

Oxygen is produced by air separation — the same process described in Section 4.2 for argon. BOC/Linde’s NZ air separation facilities produce industrial oxygen. NZ Steel’s Glenbrook works also has an air separation unit producing oxygen for the Kaldo steelmaking converter (Doc #89, Section 3.4).³⁷

Alternative small-scale oxygen production: Oxygen can be produced by electrolysis of water — passing electric current through water splits it into hydrogen and oxygen.³⁸ This is energy-intensive (roughly 5–10 times the energy cost per cubic metre of oxygen compared to cryogenic air separation), slow (a workshop-scale electrolyser might produce 1–5 cubic metres of oxygen per hour, compared to hundreds or thousands from an ASU), and limited in purity (typically 99.5–99.8%, adequate for cutting and welding but below medical-grade requirements without further purification). However, it uses only water and electricity (both available in NZ) and can be fabricated from NZ materials (steel or nickel electrodes, a lye electrolyte from wood ash, and a gas collection system). The oxygen must be dried and compressed for torch use.

Pressure swing adsorption (PSA) is another method for producing oxygen from air, using zeolite molecular sieves to selectively adsorb nitrogen. PSA units are simpler than cryogenic ASUs and could potentially be fabricated in NZ, though the zeolite adsorbent is an imported material with finite stock.

5.3 Acetylene production

Acetylene is produced by the reaction of calcium carbide (CaC2) with water:³⁹

CaC2 + 2H2O → C2H2 + Ca(OH)2

The reaction proceeds vigorously and is well-characterised. However, the engineering required to use it safely is not trivial: the reaction rate must be controlled (excess water prevents runaway; insufficient water causes hot calcium carbide to crust over and block the generator), the acetylene produced contains impurities (hydrogen sulfide, phosphine, ammonia from calcium carbide contaminants) that must be scrubbed before the gas is used for welding or stored in cylinders, and acetylene’s instability at pressures above approximately 200 kPa requires careful pressure management throughout.⁴⁰ Calcium carbide also reacts violently with water if it becomes wet in storage, creating a fire and explosion hazard. The result is calcium hydroxide (slaked lime) as a useful byproduct.⁴¹

The dependency chain is: calcium carbide requires lime and carbon in an electric arc furnace.

5.4 Calcium carbide production

Calcium carbide is produced by heating lime (CaO) and carbon (coke or charcoal) in an electric arc furnace at approximately 2,000–2,200°C:⁴²

CaO + 3C → CaC2 + CO

This is an energy-intensive process that requires sustained high temperatures. The electric arc furnace used for carbide production is similar in principle to the EAFs used for steelmaking but smaller and operated differently.

Raw materials — all NZ-available:

• Lime (CaO): Produced by calcining limestone (CaCO3) in a lime kiln at approximately 900–1,000°C. NZ has limestone quarries and operating lime kilns.⁴³
• Carbon: Coke (from coal coking — NZ has suitable bituminous coal from the West Coast, South Island) or charcoal (from wood — NZ has abundant forestry). Anthracite coal can also serve, though NZ does not mine anthracite. Charcoal is the most accessible carbon source for NZ and was historically used for carbide production before petroleum coke became dominant.⁴⁴
• Electricity: A carbide furnace of production scale (1–5 MW) draws significant power. Under recovery conditions, where electricity is allocated rather than purchased at market rates, this represents a meaningful claim on grid capacity — roughly 0.01–0.06% of NZ’s total installed generation capacity (approximately 9,700 MW as of 2024).⁴⁵ This must be weighed against competing demands. Given that the alternative is no acetylene (and therefore no oxy-fuel cutting), this is a justifiable allocation.

Scale of the project:

Calcium carbide production is not a backyard operation. A functional carbide furnace requires:

• An electric arc furnace (smaller than Glenbrook’s EAFs but still a substantial piece of equipment — fabricable from NZ steel)
• Graphite or carbon electrodes (the carbide furnace itself consumes electrodes, creating a dependency on the same graphite electrode supply chain discussed in Doc #89 — though at much lower rates than steelmaking EAFs; Soderberg-type self-baking electrodes, which use a continuously fed paste of calcined coke and coal-tar pitch formed in a metal casing, are more practical at carbide-furnace scale because they eliminate the need for precision graphite electrode fabrication)⁴⁶
• Refractory lining (NZ-produced fireclay and dolomite refractories are adequate for carbide furnaces, which operate at lower temperatures than steelmaking EAFs)
• Feed preparation equipment (crushers, conveyors)
• Gas handling equipment (the CO generated is acutely toxic — NIOSH IDLH 1,200 ppm; occupational exposure limit 25 ppm — and must be safely vented to height or burned off in a flare; worker exclusion zones around the furnace are required during operation)⁴⁷
• Carbide handling and storage (calcium carbide reacts with moisture and must be stored in sealed, dry containers — atmospheric humidity is sufficient to initiate slow acetylene generation)

Historical precedent: Calcium carbide production was widespread globally in the early 20th century. Carbide factories operated in many countries with access to hydroelectric power and limestone. NZ’s combination of cheap renewable electricity and limestone makes it a natural location for carbide production. No commercial carbide production in NZ has been confirmed from available historical sources; GNS Science and the Ministry of Culture and Heritage hold records that should be checked for any historical operations.⁴⁸

Development timeline: Designing, fabricating, and commissioning a small calcium carbide furnace would take 6–18 months from the decision to proceed, assuming NZ steel is available for fabrication, experienced electrical engineers can design the power supply, and the furnace design is adapted from published industrial designs. This is a Phase 2–3 project.

Assessment: Calcium carbide production is feasible [B] for NZ. The raw materials (lime, carbon, electricity) are all domestically available. The technology is well-understood and documented. The main challenges are the scale of the furnace construction project, the electrode consumption problem (which is less severe than for steelmaking EAFs), and the need for skilled operators. The byproduct — calcium hydroxide (slaked lime) — is itself useful for construction and chemical processes.

5.5 Oxy-acetylene under recovery conditions

Near term (Phase 1–2): Use existing acetylene and oxygen cylinder stocks. Conserve acetylene for cutting (its primary value) rather than welding (where stick welding can substitute). Existing stocks of dissolved acetylene in cylinders will last months, depending on use rate.

Medium term (Phase 2–3): Develop calcium carbide production. Once operational, acetylene supply becomes sustainable. Oxygen supply depends on continued ASU operation (or development of alternative oxygen production).

Long term (Phase 3+): Oxy-acetylene cutting becomes a sustainable capability if both carbide production and oxygen production continue. For welding, stick welding with NZ-produced electrodes is more practical for most applications — oxy-acetylene welding should be reserved for speciality work (thin material, pipe, brazing).

5.6 Acetylene cylinder considerations

Acetylene cylinders are unlike other gas cylinders. Free (undissolved) acetylene is unstable at pressures above approximately 100–200 kPa (compared to 15,000–20,000 kPa for oxygen) and can decompose explosively.⁴⁹ Commercial acetylene cylinders are filled with a porous mass (originally calcium silicate, now typically calcium silicate or charcoal-based material) saturated with acetone, in which the acetylene dissolves. This allows safe storage at a fill pressure of approximately 1,500–1,800 kPa.⁵⁰

Refilling acetylene cylinders requires:

• An acetylene generator (calcium carbide + water)
• Gas purification (removing impurities — hydrogen sulfide, phosphine — from the crude acetylene)
• A compressor rated for acetylene service (oil-free or acetylene-compatible)
• Filling equipment calibrated for dissolved-acetylene cylinders

Existing BOC/Linde acetylene filling stations in NZ have this equipment. If carbide production is established, these stations can resume filling cylinders. If the filling equipment degrades, fabricating replacement acetylene handling equipment is within NZ’s capability but requires careful engineering — acetylene’s explosive properties make safety margins critical.

---

6. TIG WELDING CONSUMABLES

6.1 Tungsten electrodes

TIG (GTAW) welding uses a non-consumable tungsten electrode to create the arc. The electrode does not melt (tungsten’s melting point — approximately 3,422°C — is higher than any metal being welded). It erodes slowly during use and is periodically resharpened by grinding.

NZ tungsten stocks: Tungsten is imported. NZ has no tungsten deposits. However, tungsten electrode consumption is low — a single electrode can last tens to hundreds of hours of welding, depending on current levels and technique. NZ’s existing stock of tungsten electrodes (held at welding suppliers, workshops, and TIG welding stations throughout the country) will last years under reduced consumption — a rough estimate of 3–10 years is plausible given that TIG consumption would contract sharply as welders shift to stick process, but this requires census verification of actual stock quantities.⁵¹

Conservation: Tungsten electrodes should be inventoried and allocated to skilled TIG welders for applications where TIG is genuinely required (thin stainless steel, aluminium, critical-quality welds). Using TIG electrodes for work that stick welding could accomplish wastes a finite resource.

6.2 Argon

TIG welding requires high-purity argon shielding gas (typically 99.995%+ purity). This is available from NZ air separation (Section 4.2). Supply continues as long as the ASU operates.

6.3 Filler rods

TIG filler rods are bare wire of the appropriate alloy, typically 1.6–3.2 mm diameter, supplied in straight lengths. For mild steel, these are similar to MIG wire but uncoated. Producible from NZ materials once wire drawing is established. For stainless steel and aluminium — filler rods require alloys (chromium, nickel for stainless; aluminium-silicon for aluminium) that are limited or unavailable in NZ.

6.4 TIG under recovery conditions

TIG becomes a niche process under recovery conditions — reserved for thin material, aluminium (while stocks last), stainless steel (while stocks last), and critical-quality applications. Tungsten electrode stocks and argon availability define the envelope. For most recovery welding, stick welding replaces TIG.

---

7. AIR SEPARATION: THE SHARED DEPENDENCY

7.1 What the ASU produces

Air separation is the technology that produces oxygen, nitrogen, and argon from atmospheric air. It is fundamental to multiple welding processes (MIG shielding gas, oxy-acetylene cutting, TIG shielding) and to steelmaking (Doc #89). NZ’s air separation capability is concentrated at BOC/Linde facilities, with an on-site ASU at Glenbrook steelworks.⁵²⁵³

The cryogenic air separation process:

1. Air is filtered and compressed to approximately 5–10 atmospheres
2. Water vapour and CO2 are removed by molecular sieve adsorption
3. The compressed, clean air is cooled to cryogenic temperatures (-170 to -200°C) using expansion cooling
4. The liquefied air is fractionally distilled in a “cold box” column, separating it into:
   • Nitrogen (boiling point -196°C): ~78% of air
   • Oxygen (boiling point -183°C): ~21% of air
   • Argon (boiling point -186°C): ~0.93% of air

7.2 ASU continuity

Current status: BOC/Linde operates ASUs in NZ that supply industrial oxygen, nitrogen, and argon. These facilities require:⁵⁴

• Electricity (major consumer — compression is energy-intensive)
• Compressor maintenance (bearings, seals, valves)
• Instrumentation and control systems (imported electronics, finite life)
• Heat exchanger integrity (aluminium brazed plate-fin exchangers in the cold box — specialised, imported, irreplaceable if they fail)

Risk: ASU failure cannot be repaired from NZ-manufactured parts. The cold box heat exchangers are precision-manufactured aluminium brazed assemblies that cannot be fabricated locally. If a cold box leak develops, that ASU may be unrepairable.

Mitigation: NZ has multiple ASUs (BOC facilities plus Glenbrook’s). Failure of one does not mean loss of all air separation capability. But the total number of ASUs in NZ is small, and each failure reduces capacity. Long-term, NZ may need to develop alternative oxygen production (electrolysis — Section 5.2) as a backup.

Assessment: Air separation is an [A]-rated capability while existing ASUs function. It becomes a critical uncertainty in Phase 4–5 (years 7–30) as equipment ages without access to replacement parts. This document assumes ASU availability through Phase 2–3.

---

8. EXISTING NZ WELDING CONSUMABLE MANUFACTURERS

8.1 Current landscape

NZ does not have any domestic welding electrode manufacturers of significant scale. NZ’s welding consumables are imported, primarily from:

• Australia: Cigweld (now part of ESAB/Illinois Tool Works), Lincoln Electric Australia
• Europe: ESAB (Sweden/global), Bohler Welding (Austria)
• Asia: Various manufacturers supplying through NZ distributors
• USA: Lincoln Electric, Hobart Brothers

NZ has welding supply distributors — companies that import, warehouse, and distribute welding consumables — but not manufacturers.⁵⁵

8.2 NZ manufacturing capability for electrode production

While NZ lacks dedicated electrode manufacturers, it has relevant industrial capability:

• Wire drawing: NZ has some wire drawing and nail manufacturing capability (e.g., Pacific Steel, a subsidiary of Fletcher Building, operates a wire mill in Otahuhu, Auckland, producing reinforcing mesh, nails, and wire products from steel billet).⁵⁶ This existing wire drawing capability could potentially be adapted for electrode core wire production — the diameters are in the right range, though wire chemistry requirements for electrode core differ from construction wire.
• Mineral processing: NZ has mining, crushing, and grinding capability in its extractive industries. Processing Glenbrook slag, limestone, and other mineral ingredients into electrode flux powders is within existing NZ industrial capability.
• Chemical processing: Silicate binder production (sodium or potassium silicate) requires basic chemical processing — heating silica sand with alkali, dissolving the product in water. NZ has chemical processing capability in its industrial sector.
• Metal fabrication: The extrusion press and associated equipment for electrode manufacturing can be designed and fabricated in NZ machine shops (Doc #91).

Assessment: NZ has the component industrial capabilities needed for electrode manufacturing. What it lacks is the specific integration of these capabilities into an electrode production facility. Establishing this integration is the core challenge — and it is a [B]-rated challenge, not a [C] or [D].

---

9. WELDING WITHOUT CONSUMABLES: ALTERNATIVES

9.1 Forge welding

Before arc welding existed, metals were joined by forge welding — heating two pieces of iron or steel to near their melting point (white heat, approximately 1,300–1,400°C for mild steel) and hammering them together. The oxide layer on the surface is disrupted by the hammering, and the two clean metal surfaces bond under pressure.⁵⁷

Forge welding requires:

• A forge capable of reaching welding heat (charcoal or coal-fired, with adequate air blast — see Doc #92, Blacksmithing)
• A flux (borax is the traditional forge welding flux — it dissolves the oxide layer, promoting a clean bond. Borax is imported but NZ has some borate mineral occurrences)
• Skill — forge welding is a traditional blacksmithing skill that requires practice and judgment to achieve consistent results. The metal must be at the right temperature (too cold and it won’t bond; too hot and it burns — the surface oxidises and crumbles), the surfaces must be properly shaped (scarfed) for the joint, and the hammering must be done quickly before the metal cools.

Limitations: Forge welding is limited to simple butt and lap joints in iron and low-carbon steel. It cannot join complex shapes, thick sections, or dissimilar metals. It is slow, requires a forge and an anvil, and depends on the blacksmith’s skill. It is not a replacement for arc welding in any modern sense.

Value: Forge welding [A] is a backup capability that uses no imported consumables — only fuel (charcoal from NZ wood), flux (borax if available, or fine silica sand as a less effective substitute), and skill. For simple agricultural repairs, fence work, and basic joining tasks, it provides a consumable-free option.

9.2 Riveting and bolting

Before welding became dominant in the 20th century, metal structures were joined by riveting (inserting hot steel rivets through aligned holes and upsetting the ends to form a permanent joint) and bolting.⁵⁸ NZ’s existing bolt and rivet stocks are finite but substantial. Bolts can be produced by machining or forging from NZ steel (Doc #91). Rivets are short headed steel blanks — a blacksmith can produce them by forging from steel rod, though rivet-heading equipment (a heading die and press or hammer) is required for consistent output at any scale.⁵⁹

Riveted and bolted joints are heavier and less efficient than welded joints — a riveted connection typically requires 30–50% more steel than the equivalent welded connection to achieve the same strength — and they leak (making them unsuitable for tanks and pressure vessels without additional caulking). Pre-drilled holes must be accurately aligned and punched or drilled, requiring drill bits or a punch press. But they are a functional joining method that reduces welding consumable demand for structural applications.

Feasibility: [A] — requires only NZ-available steel, drill bits or punch press, and blacksmithing skill. Rivets are a fully sustainable joining method once steel production is established.

9.3 Brazing and soldering

Brazing (joining with a brass or bronze filler metal at temperatures above 450°C but below the melting point of the base metal) and soldering (joining with tin-lead or other soft solder below 450°C) require filler metals and flux but not welding electrodes. NZ has limited copper and tin stocks; these joining methods are available for specialised applications — electrical connections, refrigeration tubing, plumbing, instrument repair — but not at the scale needed to replace welding for structural work.⁶⁰

Feasibility: [B] for brazing (requires copper alloy filler, which NZ can produce in limited quantities from recycled copper); [C] for tin-based soldering at scale (tin must be imported — NZ has no tin deposits).

---

CRITICAL UNCERTAINTIES

Uncertainty	Impact if Wrong	Resolution Method
NZ welding consumable stock levels	If less than estimated, the urgency of local production increases dramatically	National inventory — census (Doc #8) and distributor audit (Doc #1)
Glenbrook EAF slag suitability as rutile substitute in electrode flux	If unsuitable, alternative TiO2 sources must be found (ilmenite processing — more complex)	Laboratory testing — grind slag, formulate test electrodes, weld and test
Wire rod availability from Glenbrook	If wire rod production is not developed, core wire must come from reworked existing wire stocks (limited)	Doc #89 wire rod development — coordinate timelines
Pacific Steel wire drawing adaptability for electrode core wire	If existing wire drawing can produce suitable electrode core wire, this accelerates production significantly	Technical assessment with Pacific Steel — first 3 months
Sodium silicate production feasibility in NZ	If silicate binder cannot be produced, electrode manufacturing is blocked	Chemical processing development — soda ash + silica sand is well-understood chemistry
BOC/Linde ASU remaining operational life	If ASU fails early, argon and oxygen supply stops — affects MIG, TIG, and oxy-acetylene	Condition assessment of ASU equipment — coordinate with Doc #162
NZ borax availability for forge welding flux	If NZ borate deposits are insufficient, forge welding effectiveness is reduced	Geological survey of NZ borate occurrences
Electrode formulation development time	If longer than estimated, the gap between stock depletion and local production widens	Begin R&D immediately — do not wait for stock depletion
Calcium carbide furnace electrode consumption	If high, creates competition with steelmaking for scarce graphite electrodes	Engineering assessment — Soderberg electrodes more feasible at carbide scale

---

CROSS-REFERENCES

• Doc #1 — National Emergency Stockpile Strategy (welding consumable inventory and controlled distribution)
• Doc #8 — National Skills and Asset Census (welding workforce and workshop inventory)
• Doc #56 — Wood Gasification (gasifier construction requires welding)
• Doc #65 — Hydroelectric Station Maintenance (maintenance welding demand)
• Doc #89 — NZ Steel Glenbrook (steel and wire rod supply; EAF slag as TiO2 source; air separation)
• Doc #91 — Machine Shop Operations (complementary capability; electrode extrusion press fabrication; build-up welding)
• Doc #92 — Blacksmithing (forge welding as backup joining method)
• Doc #93 — Foundry Operations (carbide furnace casting; complementary manufacturing)
• Doc #97 — Cement Production (limestone supply chain overlap)
• Doc #102 — Charcoal Production (carbon source for calcium carbide)
• Doc #105 — Wire Drawing (electrode core wire and MIG wire production)
• Doc #106 — Small-Scale Electric Arc Furnaces (carbide furnace parallels)
• Doc #113 — Sulfuric Acid (industrial chemistry for copper sulfate — MIG wire coating)
• Doc #138 — Sailing Vessel Design (welded hull construction; maritime trade for consumable imports)
• Doc #157 — Trade Training Priorities (welding as priority trade skill)
• Doc #162 — University Reorientation (air separation plant status; materials testing for electrode development)

---

---

1. Welding process classifications follow the American Welding Society (AWS) and ISO 4063 standard nomenclature. SMAW (AWS designation), MMA (ISO designation) = stick welding. GMAW = MIG. GTAW = TIG. OFW = oxy-fuel welding. See: AWS A3.0, “Standard Welding Terms and Definitions.”↩︎

2. NZ welding consumable consumption is not centrally reported. The estimates here are rough order-of-magnitude figures based on NZ’s steel fabrication activity, construction sector size, and comparison with per-capita consumption data from comparable economies. Actual figures would need to be established through distributor and workshop surveys.↩︎

3. BOC (now part of Linde plc) operates gas production and distribution facilities in New Zealand, including air separation units and gas filling stations. https://www.boc.co.nz — Air Liquide also operates in NZ. Between them, these companies supply the majority of NZ’s industrial gas market.↩︎

4. Stick electrode construction and function: See Lancaster, J.F., “Metallurgy of Welding,” 6th edition, Abington Publishing, 1999; also AWS “Welding Handbook,” Vol. 2, “Welding Processes.” The flux coating functions described are well-established welding metallurgy.↩︎

5. Electrode flux coating classifications: AWS A5.1 (carbon steel electrodes) and AWS A5.5 (low-alloy steel electrodes) define electrode classifications based on flux type, mechanical properties, and welding position. The rutile, cellulosic, basic, and iron oxide categories correspond approximately to AWS E6013, E6010/E6011, E7018, and E6020 respectively.↩︎

6. NZ calcium fluoride (fluorspar) deposits: No commercially significant fluorspar deposits are known in NZ. Fluorspar is used in basic electrode coatings and as a fluxing agent in steelmaking. NZ would need to import fluorspar or develop alternative flux formulations. See: Crown Minerals NZ; GNS Science mineral occurrence databases.↩︎

7. NZ ironsand titanium content: See Doc #89, footnotes 10 and 11. NZ ironsand concentrate contains approximately 7–8% TiO2, present as titanium substitution in the magnetite crystal structure.↩︎

8. Glenbrook EAF slag titanium content: EAF slag from the Glenbrook process is enriched in TiO2 because the iron is reduced and removed from the ironsand while the titanium oxide remains in the slag. Published analyses of NZ Steel slag suggest TiO2 content in the range of 20–40%, though specific figures vary with operating conditions. See: NZ Steel technical publications; GNS Science; HERA technical documentation. This slag is currently treated as waste, though its potential value as a TiO2 source has been recognised. Exact TiO2 content should be verified through laboratory analysis for electrode flux development.↩︎

9. Ilmenite processing: Ilmenite (FeTiO3) can be converted to synthetic rutile (>90% TiO2) through several established industrial processes including the Becher process (reduction roasting followed by aeration in ammonium chloride solution to remove iron) and the Benilite process (acid leaching). These are mature industrial processes practised in Australia and elsewhere, but represent significant chemical engineering projects. See: Habashi, F., “Handbook of Extractive Metallurgy,” Wiley.↩︎

10. NZ forestry statistics: Ministry for Primary Industries (MPI) National Exotic Forest Description (NEFD). Pinus radiata planted production forest area approximately 1.7 million hectares. https://www.mpi.govt.nz/ — Forestry provides abundant cellulose for electrode flux and charcoal for calcium carbide production.↩︎

11. NZ limestone and lime production: NZ has multiple limestone quarries and lime kilns. McDonald’s Lime (now part of Graymont) at Otorohanga in the Waikato is one of the largest. Other limestone operations exist in the Wairarapa, Nelson, Canterbury, and Southland. See: Doc #89, footnote 23; Crown Minerals; NZ Minerals Industry Association.↩︎

12. Sodium silicate (waterglass) production: Produced industrially by fusing silica sand (SiO2) with soda ash (Na2CO3) at 1,100–1,400°C in a furnace, producing sodium silicate glass, which is then dissolved in water under pressure to produce liquid sodium silicate of the desired concentration. The process is well-established industrial chemistry dating to the early 19th century. See: Vail, J.G., “Soluble Silicates: Their Properties and Uses,” Reinhold, 1952.↩︎

13. Soda ash (sodium carbonate, Na2CO3) production: The Solvay process reacts salt (NaCl), limestone (CaCO3), and ammonia to produce soda ash. All three feedstocks are available in NZ (NZ has salt from seawater evaporation or Marlborough salt deposits, limestone, and ammonia can be produced from nitrogen and hydrogen). The Leblanc process is an older, simpler process using salt, limestone, and sulfuric acid. Wood ash is a direct source of potassium carbonate (potash, K2CO3), which can substitute for sodium carbonate in some applications including silicate binder production. See: general industrial chemistry references.↩︎

14. Iron powder atomisation: Gas atomisation of molten steel involves delivering a stream of molten metal through a nozzle into a high-pressure gas jet (argon or nitrogen), which disperses it into fine droplets that solidify as powder. The nozzle and refractory components must withstand molten steel temperatures (~1,600°C). The gas supply (argon or nitrogen) is available from the ASU. Particle size distribution is controlled by gas pressure and nozzle geometry. For electrode flux applications, where fine uniform powder is desirable, atomisation produces superior results to grinding. Dependency chain: ASU for gas supply (Section 7); a suitable high-temperature nozzle assembly (fabricable in NZ from refractory ceramics and steel); a melting furnace; and collection/sieving equipment. See: Neikov, O.D. et al., “Handbook of Non-Ferrous Metal Powders,” Elsevier, 2009.↩︎

15. NZ calcium fluoride (fluorspar) deposits: No commercially significant fluorspar deposits are known in NZ. Fluorspar is used in basic electrode coatings and as a fluxing agent in steelmaking. NZ would need to import fluorspar or develop alternative flux formulations. See: Crown Minerals NZ; GNS Science mineral occurrence databases.↩︎

16. Welding electrode manufacturing process: Described in AWS “Welding Handbook,” Vol. 2; also Nadkarni, S.V., “Modern Arc Welding Technology,” Oxford & IBH Publishing. The extrusion process (flux paste applied concentrically to wire through a die) is the standard industrial method. Small-scale alternatives (hand-dipping, trough coating) are documented in older welding literature.↩︎

17. Electrode core wire chemistry: Core wire for general-purpose mild steel electrodes is typically AISI/SAE 1010 or similar low-carbon steel (C 0.06–0.15%, Mn 0.3–0.6%, Si <0.1%, S <0.03%, P <0.03%). Higher carbon, sulfur, or phosphorus content in the wire degrades weld quality — carbon increases hardness and cracking risk; sulfur and phosphorus cause hot cracking. Wire chemistry must be verified before use as electrode core. See: AWS A5.18 (wire chemistry specifications).↩︎

18. Welding electrode manufacturing process: Described in AWS “Welding Handbook,” Vol. 2; also Nadkarni, S.V., “Modern Arc Welding Technology,” Oxford & IBH Publishing. The extrusion process (flux paste applied concentrically to wire through a die) is the standard industrial method. Small-scale alternatives (hand-dipping, trough coating) are documented in older welding literature.↩︎

19. Electrode flux formulations: The composition ranges given are representative of rutile-type electrode coatings based on published welding metallurgy literature. Exact commercial formulations are proprietary. The ranges provided are suitable as starting points for experimental development. Specific NZ formulations using Glenbrook slag rather than pure rutile will require adaptation. See: Lancaster (footnote 4); Nadkarni (footnote 14).↩︎

20. Adapting electrode formulations to NZ materials: Because NZ will be using Glenbrook EAF slag rather than pure rutile, and NZ-sourced cellulose, limestone, and other minerals rather than commercial electrode-grade ingredients, the formulations must be developed through systematic experimentation. Published formulations provide starting points, but the specific mineralogy and chemistry of NZ materials will produce different flux behaviour. This development work requires a welding metallurgist and a systematic testing program. Historical precedent exists — during WWII, several nations developed electrode manufacturing using locally available materials when imports were disrupted. See: general WWII industrial history; AWS welding history publications.↩︎

21. Screw-driven extrusion press: A screw-driven press for electrode coating extrusion avoids dependence on hydraulic fluid (an imported petroleum product with finite NZ stock). The screw press uses a rotating screw thread to advance the flux paste through the extrusion die, driven by an electric motor through a gearbox. NZ machine shops can fabricate the screw, barrel, and die from NZ steel; the gearbox can be adapted from existing industrial equipment or fabricated. The performance gap compared to hydraulic presses is modest for electrode extrusion — throughput is lower but coating quality is comparable for the pressures involved (typically 5–20 MPa). See: general extrusion engineering references; Nadkarni (footnote 14).↩︎

22. Hand-dipping and trough-coating for initial electrode production: In the absence of extrusion equipment, a flux paste of appropriate consistency can be applied to wire by drawing the wire through a trough of paste (the “dipping” or “pull-through” method). The wire is drawn through one or more times to build up coating thickness, then straightened and dried. The resulting coating is less concentric and less uniform than extruded electrodes — wall thickness varies, which affects arc stability and weld quality. For development and small-scale testing, hand-dipped electrodes are adequate to evaluate flux formulations. For production use, they produce more variable results but may still be acceptable for non-critical repair welding. See: older welding literature; Lincoln Electric historical publications on electrode manufacturing.↩︎

23. Electrode moisture control: Basic (low-hydrogen) electrodes are particularly sensitive to moisture absorption. AWS D1.1 (Structural Welding Code — Steel) specifies storage conditions for low-hydrogen electrodes: either in hermetically sealed containers until use, or in holding ovens at 120–150°C. Moisture in electrode coatings introduces hydrogen into the weld metal, causing hydrogen-induced cracking (cold cracking) in susceptible steels. Even rutile and cellulosic electrodes perform better when properly stored dry. See: AWS D1.1; Lincoln Electric “The Procedure Handbook of Arc Welding.”↩︎

24. NZ welder workforce: Careers NZ / Te Pūkenga estimated approximately 6,700 welders employed in New Zealand as of 2022, across manufacturing, construction, engineering, and agricultural sectors. Actual recovery-period active numbers would depend on which sectors remain active and whether recovery fabrication demand draws in welders from reduced sectors. See: Careers NZ occupational data. https://www.careers.govt.nz/ — Figure requires verification from MBIE labour market statistics.↩︎

25. Electrode consumption per welder-day: AWS welding engineering practice estimates typical production welding at approximately 5–10 kg of electrode deposited per eight-hour shift for a continuously productive welder in a fabrication shop setting; repair welding and field work use less — approximately 2–5 kg — due to setup time, fit-up, and intermittent work. Upper-bound estimates of 10–15 kg/day apply only to high-intensity production welding with multiple welding stations and support staff. See: AWS “Welding Handbook,” Vol. 1; Lincoln Electric “The Procedure Handbook of Arc Welding.”↩︎

26. MIG wire construction: Bare steel wire with a thin copper coating (approximately 0.1–0.5 μm thickness) applied by electroplating or chemical displacement (immersion in copper sulfate solution — the iron in the wire reduces the copper from solution, depositing a thin copper layer). See: AWS A5.18, “Specification for Carbon Steel Electrodes and Rods for Gas Shielded Arc Welding.”↩︎

27. MIG wire chemistry: AWS ER70S-6 (one of the most common MIG wire specifications) requires approximately C 0.06–0.15%, Mn 1.40–1.85%, Si 0.80–1.15%, plus trace element limits. The silicon and manganese are deoxidisers — they react preferentially with oxygen in the arc atmosphere, protecting the weld metal from porosity. Without adequate deoxidiser content, weld quality degrades significantly. See: AWS A5.18.↩︎

28. Uncoated (copper-free) MIG wire performance: Standard MIG wire has a copper coating approximately 0.1–0.5 μm thick (see footnote 19). Without this coating, the wire surface is bare steel, which oxidises (rusts) in storage and feeds less smoothly through the copper-lined conduit in the MIG gun. Performance penalties: (1) rust on the wire surface can cause porosity in welds if the rust layer is thick — storage in sealed, dry packaging mitigates this; (2) increased friction in the conduit liner raises the risk of birdnesting (wire jamming and coiling in the drive mechanism) — using a Teflon-lined conduit (rather than the standard copper conduit) reduces friction significantly; (3) contact tip wear increases, as the bare steel wire is harder against the copper tip than copper-coated wire. None of these penalties makes uncoated wire unusable — birdnesting can be managed by reducing wire feed speed and keeping the conduit clean and short; tip wear is managed by more frequent tip replacement. Weld quality is not significantly affected if the wire surface is clean. See: Miller Electric “MIG welding guide”; AWS A5.18 (notes on wire surface condition).↩︎

29. Shielding gas selection for MIG welding: The choice of CO2, argon-CO2 mix, or argon affects arc behaviour, penetration, spatter, and weld appearance. Pure CO2 produces the most spatter and deepest penetration; higher argon content produces smoother arcs and less spatter. For mild steel welding under recovery conditions, pure CO2 is the most practical choice because it can be produced locally. See: Lincoln Electric “The Procedure Handbook of Arc Welding”; AWS “Welding Handbook,” Vol. 1.↩︎

30. NZ CO2 production from fermentation: NZ breweries produce CO2 as a byproduct of beer fermentation. Some larger breweries (DB Breweries, Lion NZ) capture and purify this CO2 for carbonation. The volumes available from fermentation alone are modest relative to potential welding demand, but contribute to overall supply. See: NZ brewing industry publications.↩︎

31. CO2 from limestone calcination: The calcination reaction CaCO3 → CaO + CO2 releases approximately 44% of the limestone’s mass as CO2. At NZ’s existing lime production scale (tens of thousands of tonnes per year), CO2 availability from this source alone would be significant. Capturing and purifying kiln exhaust CO2 requires additional equipment but is well-established technology in the cement and lime industries. See: general chemical engineering references.↩︎

32. Welding-grade CO2 purity: AWS C5.10/C5.10M specifies shielding gas purity requirements. CO2 for welding should have moisture content below 0.25% (2,500 ppm) at a minimum; better quality (below 200 ppm moisture) is recommended for critical work. Higher moisture causes porosity in welds.↩︎

33. Air separation process: Standard cryogenic air separation is described in general chemical engineering references. See: Smith, A.R. and Klosek, J., “A review of air separation technologies and their integration with energy conversion processes,” Fuel Processing Technology, 2001. Argon’s boiling point (-185.8°C) is between nitrogen (-195.8°C) and oxygen (-183°C), making it separable by fractional distillation but requiring a dedicated argon column in the ASU for high-purity production.↩︎

34. BOC (now part of Linde plc) operates gas production and distribution facilities in New Zealand, including air separation units and gas filling stations. https://www.boc.co.nz — Air Liquide also operates in NZ. Between them, these companies supply the majority of NZ’s industrial gas market.↩︎

35. Performance of pure CO2 shielding versus argon-CO2 mixtures: Pure CO2 shielding for mild steel MIG welding produces welds with acceptable mechanical properties but noticeably different operating characteristics from Ar/CO2 mixtures. Typical differences: spatter rate increases 3–5× (requiring additional post-weld cleanup); penetration profile deepens and widens (can cause burn-through on thin material below approximately 3 mm); arc is harsher and less forgiving of parameter variation; fume generation increases (CO2 dissociates to CO in the arc — adequate ventilation is critical). These effects are well-documented in AWS and Lincoln Electric welding engineering guides. For structural and repair welding on material above 5 mm, pure CO2 shielding produces sound welds with no significant mechanical property penalty. See: Lincoln Electric “The Procedure Handbook of Arc Welding”; AWS C5.6, “Recommended Practices for Gas Metal Arc Welding.”↩︎

36. Oxy-acetylene flame temperature: Approximately 3,100–3,300°C, depending on oxygen-to-acetylene ratio. This is the hottest commonly used industrial flame. See: AWS “Welding Handbook,” Vol. 2; general oxy-fuel welding references.↩︎

37. NZ Steel Glenbrook air separation unit: See Doc #89, Section 3.4 and footnote relating to oxygen production for the Kaldo converter. Glenbrook’s ASU produces oxygen for steelmaking and is a significant industrial-scale air separation facility.↩︎

38. Water electrolysis for oxygen production: The electrolysis reaction 2H2O → 2H2 + O2 requires approximately 4–5 kWh per cubic metre of oxygen at practical electrode efficiency (theoretical minimum is approximately 2.96 kWh/m³; real alkaline electrolysers operate at 4.5–6 kWh/m³). By comparison, cryogenic air separation produces oxygen at approximately 0.35–0.5 kWh/m³. The energy cost ratio is roughly 10:1. A workshop-scale alkaline electrolyser producing 2 m³/hour of oxygen requires approximately 10 kW of electrical input, sufficient for one oxy-fuel cutting station. The electrodes are steel or nickel; the electrolyte is potassium hydroxide (KOH) solution at approximately 25–30% concentration. KOH can be produced in NZ by electrolysis of potassium chloride solution or from wood ash. Oxygen purity from alkaline electrolysis is typically 99.3–99.8%, adequate for cutting and welding. See: Zeng, K. and Zhang, D., “Recent progress in alkaline water electrolysis for hydrogen production and applications,” Progress in Energy and Combustion Science, 2010.↩︎

39. Calcium carbide and acetylene chemistry: The reaction CaC2 + 2H2O → C2H2 + Ca(OH)2 is exothermic and proceeds vigorously when water contacts calcium carbide. Acetylene generators that control the water-carbide contact rate to produce gas at a manageable rate are well-documented engineering. See: general industrial chemistry references; Morehead, J.T. and de Chalmot, G., “The Manufacture of Calcium Carbide,” Journal of the American Chemical Society, 1896.↩︎

40. Calcium carbide and acetylene chemistry: The reaction CaC2 + 2H2O → C2H2 + Ca(OH)2 is exothermic and proceeds vigorously when water contacts calcium carbide. Acetylene generators that control the water-carbide contact rate to produce gas at a manageable rate are well-documented engineering. See: general industrial chemistry references; Morehead, J.T. and de Chalmot, G., “The Manufacture of Calcium Carbide,” Journal of the American Chemical Society, 1896.↩︎

41. Calcium carbide and acetylene chemistry: The reaction CaC2 + 2H2O → C2H2 + Ca(OH)2 is exothermic and proceeds vigorously when water contacts calcium carbide. Acetylene generators that control the water-carbide contact rate to produce gas at a manageable rate are well-documented engineering. See: general industrial chemistry references; Morehead, J.T. and de Chalmot, G., “The Manufacture of Calcium Carbide,” Journal of the American Chemical Society, 1896.↩︎

42. Calcium carbide production: The electric arc furnace reaction CaO + 3C → CaC2 + CO requires sustained temperatures above 2,000°C. Power consumption is approximately 2,500–3,500 kWh per tonne of calcium carbide. This makes calcium carbide production energy-intensive but feasible where cheap electricity is available — historically, carbide production was concentrated in countries with hydroelectric power (Norway, Switzerland, Canada). NZ’s hydroelectric generation makes it a natural fit. See: Mantell, C.L., “Electrochemical Engineering,” McGraw-Hill; general industrial chemistry references.↩︎

43. NZ limestone and lime production: NZ has multiple limestone quarries and lime kilns. McDonald’s Lime (now part of Graymont) at Otorohanga in the Waikato is one of the largest. Other limestone operations exist in the Wairarapa, Nelson, Canterbury, and Southland. See: Doc #89, footnote 23; Crown Minerals; NZ Minerals Industry Association.↩︎

44. Charcoal as a carbon source for calcium carbide: Early calcium carbide production (late 19th and early 20th century) used charcoal before petroleum coke became the dominant carbon source. Charcoal-based carbide production is feasible but less efficient than coke-based production due to charcoal’s lower density and carbon content. NZ’s abundant forestry makes charcoal available at scale (Doc #102). See: Mantell (footnote 29).↩︎

45. NZ installed electricity generation capacity: As of 2024, NZ’s total installed electricity generation capacity is approximately 9,700 MW, comprising hydro (~5,600 MW), wind (~1,000 MW), geothermal (~1,100 MW), thermal gas and coal (~1,700 MW), and other sources. See: Electricity Authority / Transpower NZ generation register; Ministry of Business, Innovation and Employment (MBIE) Energy in New Zealand annual publication. https://www.mbie.govt.nz/energy-and-natural-resources/ene... — Figure requires verification from current MBIE data as capacity changes with decommissioning and new builds.↩︎

46. Soderberg electrodes for electric arc furnaces: Soderberg (or self-baking) electrodes consist of a steel casing filled with a paste of calcined petroleum coke (or charcoal) and coal-tar pitch binder. As the electrode advances into the furnace, the paste is baked by furnace heat into a solid carbon electrode. This eliminates the need for pre-baked graphite electrodes (which require a separate high-temperature furnace and graphite precursor materials). Soderberg electrodes are used in ferroalloy furnaces and some carbide furnaces. Their limitations include higher electrical resistivity than graphite electrodes (lower efficiency) and the coal-tar pitch binder, which may be difficult to source in NZ (though wood-derived tars are a possible substitute). Charcoal-based paste is feasible if coal-tar pitch is unavailable. See: Mantell, C.L., “Electrochemical Engineering,” McGraw-Hill; Tveit, H. et al., “The Søderberg Electrode,” SINTEF Materials and Chemistry, general literature.↩︎

47. Carbon monoxide hazards in calcium carbide production: CO is produced as a direct byproduct of the carbide reaction (CaO + 3C → CaC2 + CO). CO is colourless, odourless, and acutely toxic — NIOSH IDLH (immediately dangerous to life and health) 1,200 ppm; NZ workplace exposure standard (WES-TWA) 20 ppm (WorkSafe NZ). In an enclosed building, CO from a running carbide furnace can reach lethal concentrations rapidly. Mitigation requires: forced ventilation of the furnace enclosure, CO monitoring at worker level, exclusion zones during furnace operation, and flaring or combustion of the CO offgas rather than direct venting near ground level. See: NIOSH Pocket Guide to Chemical Hazards; WorkSafe NZ Workplace Exposure Standards (2024). https://www.worksafe.govt.nz/↩︎

48. NZ hydroelectric development and electrochemical industry: NZ’s early hydroelectric schemes (from the 1880s onwards) attracted energy-intensive industries. Whether calcium carbide was ever produced commercially in NZ has not been confirmed from available sources. NZ’s aluminium smelter at Tiwai Point (1971) is the most prominent example of an energy-intensive electrochemical industry located in NZ for its cheap hydroelectricity. See: NZ engineering history publications; Ministry for Culture and Heritage. https://nzhistory.govt.nz/↩︎

49. Acetylene cylinder construction: Dissolved acetylene cylinders contain a porous mass (calcium silicate, charcoal, or other approved porous material) saturated with acetone. Acetylene dissolves in acetone under pressure, allowing safe storage at approximately 1,500 kPa (free acetylene is dangerous above approximately 200 kPa). This cylinder technology dates to the early 20th century and is standardised internationally. See: AS/NZS 2030 (gas cylinder standards); CGA (Compressed Gas Association) pamphlets.↩︎

50. Acetylene cylinder construction: Dissolved acetylene cylinders contain a porous mass (calcium silicate, charcoal, or other approved porous material) saturated with acetone. Acetylene dissolves in acetone under pressure, allowing safe storage at approximately 1,500 kPa (free acetylene is dangerous above approximately 200 kPa). This cylinder technology dates to the early 20th century and is standardised internationally. See: AS/NZS 2030 (gas cylinder standards); CGA (Compressed Gas Association) pamphlets.↩︎

51. Tungsten electrode consumption rate: A tungsten TIG electrode (typically 1.6–3.2 mm diameter, 150–175 mm long) erodes slowly during welding — the electrode tip wears and must be periodically resharpened on a dedicated grinding wheel, but the electrode itself lasts tens to hundreds of hours of arc time depending on current level and technique. NZ’s stock of tungsten electrodes, distributed across welding supply chains, represents years of supply at reduced consumption rates. Tungsten is not available from NZ sources — all tungsten is imported.↩︎

52. BOC (now part of Linde plc) operates gas production and distribution facilities in New Zealand, including air separation units and gas filling stations. https://www.boc.co.nz — Air Liquide also operates in NZ. Between them, these companies supply the majority of NZ’s industrial gas market.↩︎

53. NZ Steel Glenbrook air separation unit: See Doc #89, Section 3.4 and footnote relating to oxygen production for the Kaldo converter. Glenbrook’s ASU produces oxygen for steelmaking and is a significant industrial-scale air separation facility.↩︎

54. Air separation unit maintenance requirements: ASUs are complex plants requiring ongoing maintenance of main air compressors, pre-purification systems (molecular sieve beds), cold box components, product compressors, and instrumentation. The cold box — an insulated assembly containing precision aluminium brazed plate-fin heat exchangers and distillation columns — is the component most difficult to repair from local materials. Compressor and valve maintenance is more routine and partially achievable with NZ machining capability (Doc #91). See: general cryogenic engineering references; Smith and Klosek (footnote 25).↩︎

55. NZ welding consumable supply chain: NZ’s welding consumables are supplied through distributor networks of international manufacturers. No electrode, wire, or flux manufacturing of industrial scale is conducted in NZ. Major brands sold in NZ include Lincoln Electric, ESAB, Cigweld (ESAB), Bohler Welding, and various Asian brands. See: NZ welding supplier catalogues; HERA member directories.↩︎

56. Pacific Steel (Otahuhu, Auckland): A subsidiary of Fletcher Building, Pacific Steel operates an electric arc furnace steelmaking facility and a wire mill producing reinforcing steel products (rebar, mesh, wire) from scrap steel. The wire mill draws steel rod into wire products and could potentially produce electrode core wire, though the existing product range (reinforcing wire, tie wire) differs from electrode core wire in chemistry and dimension requirements. See: Pacific Steel / Fletcher Building company information. https://www.pacificsteel.co.nz/↩︎

57. Forge welding: The oldest metal joining technique, predating arc welding by millennia. The process requires the metal to be heated to “welding heat” (white heat, approximately 1,300–1,400°C for mild steel, lower for wrought iron) and hammered together. Borax (sodium tetraborate, Na2B4O7) is the traditional flux, dissolving iron oxide at the joint interface and promoting bonding. Forge welding is limited to butt and lap joints in ferrous metals and requires significant skill. See: Bealer, A.W., “The Art of Blacksmithing,” Castle Books; general blacksmithing references.↩︎

58. Riveting: Hot riveting was the standard method for joining structural steel and boiler/ship plate from the mid-19th century until welding replaced it in the mid-20th century. Riveted joints are strong in shear but not gas-tight without additional caulking. NZ’s riveted structures (bridges, older buildings, some ships) demonstrate the durability of the technique. Rivet production requires steel rod, a heading die to form the rivet head, and either hammer work or a press — within the capability of a machine shop (Doc #91) or experienced blacksmith. Cold riveting (driving rivets at room temperature in smaller diameters) simplifies the process. See: general structural engineering history; Blodgett, O.W., “Design of Welded Structures,” Lincoln Electric, 1966 (structural efficiency comparisons).↩︎

59. Riveting: Hot riveting was the standard method for joining structural steel and boiler/ship plate from the mid-19th century until welding replaced it in the mid-20th century. Riveted joints are strong in shear but not gas-tight without additional caulking. NZ’s riveted structures (bridges, older buildings, some ships) demonstrate the durability of the technique. Rivet production requires steel rod, a heading die to form the rivet head, and either hammer work or a press — within the capability of a machine shop (Doc #91) or experienced blacksmith. Cold riveting (driving rivets at room temperature in smaller diameters) simplifies the process. See: general structural engineering history; Blodgett, O.W., “Design of Welded Structures,” Lincoln Electric, 1966 (structural efficiency comparisons).↩︎

60. Brazing and soldering: Brazing uses copper-zinc (brass) or copper-silver filler metals at 600–900°C; soldering uses tin-lead or tin-silver fillers at 180–350°C. Both require filler metals that depend on materials NZ has in limited supply (copper, zinc, tin, silver). These joining methods are valuable for specific applications (electrical connections, plumbing, refrigeration repair) but cannot substitute for welding at structural scale. See: AWS “Brazing Handbook”; general joining references.↩︎