GTAW, GMAW and SMAW compared

Choosing a welding process is one of the engineering decisions that most directly drive the cost, quality and schedule of a fabrication project. For the same joint geometry and material, GTAW, GMAW and SMAW can each deliver a technically "acceptable" result — but usually only one of them is economical, fast and reliable for that particular job. The common field mistake is to let the machine that happens to be in the workshop, or plain habit, decide the process. Weigh material, section thickness, position, access, hydrogen sensitivity and quality class together, then pick the process that fits those constraints. This article compares the three fundamental arc welding processes, from operating principle to heat input and from deposition efficiency to positional capability, and closes with a clear selection logic.

1. How the three processes work

SMAW (stick / MMA): An arc burns between a covered consumable electrode and the workpiece. The metallic core supplies the filler material, while the outer covering melts to produce both a shielding gas atmosphere and a slag blanket over the molten metal. The equipment is the simplest of all: a power source, an electrode holder and electrodes.

GMAW (MIG/MAG): A thin, continuously fed wire electrode is pushed automatically into the arc zone, and shielding comes from an externally supplied gas. It produces no slag (only silicate islands on the bead surface, which you remove before the cap pass). MIG (inert gas — argon/helium) is used above all on aluminum, and also on copper, nickel alloys and titanium. On stainless steel, pure inert gas makes the arc unstable, so practice favors argon-based mixtures with a small active gas addition (typically about 2–2.5% CO₂ or 1–2% O₂); technically that puts the job in the MAG (active gas) class. Carbon and low-alloy steels also run MAG (CO₂ or Ar+CO₂ mixtures). Metal transfer can be short-circuiting, globular, spray or pulsed. Define shielding gas mixtures in the WPS by the EN ISO 14175 group designation (for example I1 = pure Ar, M12/M21 = Ar+CO₂ mixtures), not by free-text percentages.

GTAW (TIG): The arc burns from a non-consumable tungsten electrode; where filler is needed, feed it separately as wire or rod, manually or automatically. Shielding is by inert gas (argon, sometimes with helium). This process gives the finest control over heat input and weld pool.

Current type and polarity: This is the dimension most often skipped in process comparisons, even though it sits in the WPS as an essential variable. GTAW runs DCEN (electrode negative) on steel, stainless and nickel alloys; aluminum and magnesium need AC to break up the surface oxide film (square-wave sources let you tune the cleaning/penetration balance). GMAW in practice almost always runs DCEP (electrode positive); stable spray and pulsed transfer depend on that polarity. In SMAW, polarity follows the electrode covering: cellulosic electrodes (E6010, for example) require DCEP, some rutile electrodes also run on AC, and basic electrodes generally use DCEP. Choosing a polarity that contradicts the electrode classification gives you an unstable arc, heavy spatter and penetration that misses the target.

2. Gas shielding or slag?

This distinction largely defines the character of a process. SMAW and flux-cored arc welding (FCAW) are slag-forming processes; the slag protects the molten metal from oxidation, but you must remove it after every pass. Slag inclusion is one of the most frequent discontinuities in multi-pass work, and avoiding it demands real discipline in interpass grinding and brushing — one of the first things a welding inspector checks before radiography.

GMAW and GTAW are gas-shielded processes; they run slag-free, so the interpass cleaning burden is minimal and multi-pass joints go faster. In exchange, gas-shielded processes are sensitive to wind and drafts — disturb the gas envelope and porosity follows. It helps to put an auditable number on this: AWS D1.1 (Clause 5.12) requires that air velocity in the weld zone not exceed 5 mph (about 8 km/h ≈ 2.2 m/s) for gas-shielded processes; above that, use screens, tents or an enclosure. For an inspector this turns a subjective call like "it's windy" into an acceptance criterion you can measure with an anemometer and record in the ITP. SMAW generates its own shielding inside the arc as the electrode covering breaks down (CO₂/H₂ with cellulosic, mainly CO₂ with basic), which makes it far more tolerant outdoors and in wind. The slag covers the pool only afterwards; it is not the mechanism behind that wind tolerance.

3. Efficiency and deposition rate

Deposition efficiency is the share of consumed filler material that actually ends up in the weld. Do not confuse it with the thermal efficiency factor (k) discussed later: deposition efficiency is a material efficiency, while k is the fraction of arc energy transferred into the workpiece.

SMAW: Roughly 55–65%, depending on electrode type (higher with iron-powder, high-efficiency electrodes). You throw away the stub and lose metal to spatter and slag; the stub loss alone is on the order of 15%. Deposition rate in manual work is typically 1–2 kg/h.
GMAW: Roughly 90–95%, depending on shielding gas and transfer mode: short-circuiting transfer in pure CO₂ drops to about 88% because of spatter, while low-spatter spray or pulsed transfer exceeds 95%. The wire is continuous, so there is no stub loss. Deposition rate is typically 2–5 kg/h.
GTAW: Deposition efficiency is high — roughly 90–95% in manual work, where the rod stub is the main loss, and close to 100% with automatic/orbital wire feed. Deposition rate, however, is low: typically 0.3–1.0 kg/h. GTAW is the slowest process and is not economical on its own in heavy sections.

In practice GMAW is the most productive semi-automatic process; continuous wire feed and no stops to change electrodes push arc-on time (operator factor) well above that of SMAW. As an order of magnitude: manual SMAW typically runs 15–30% arc-on time, semi-automatic GMAW/FCAW 30–50%, and mechanized SAW 50% and above. Every time a stick electrode burns down, the welder stops to change it and the operator factor drops.

4. Arc energy, heat input and metallurgical control

Heat input (Q) directly affects the properties of the weld metal and the heat-affected zone (HAZ), and follows this relationship:

Q = (η × U × I) / v

Here η (k in some standards) is the process thermal efficiency factor, U the arc voltage, I the current and v the travel speed. With U in V, I in A and v in mm/s the result is J/mm; to get the unit normally used in a WPS, divide by 1,000: Q [kJ/mm] = Q [J/mm] / 1,000.

Arc energy or heat input? This is the item most often mixed up in field records, and a silent source of WPS/PQR non-conformity:

Arc energy = U × I / v — no efficiency factor applied.
Heat input = k × arc energy — multiplied by the thermal efficiency factor.
The EN 1011-1 / EN ISO 15614-1 / EN ISO 9606-1 system applies the k factor.
ASME BPVC Section IX (QW-409.1) and AWS D1.1 define "heat input" without a factor, that is, as the same quantity Europe calls arc energy.

Apply the wrong system to the same measurement and the acceptance range shifts by 20–40% — enough to get a PQR or WPS rejected. The WPS must therefore state clearly which quantity the limit applies to (arc energy, or heat input with k) and which code system governs.

Typical thermal efficiency factors accepted in the standards (EN 1011-1:2009, Table 1; the same set also appears in ISO/TR 18491): k ≈ 0.8 for SMAW (111) and GMAW (13x); k ≈ 0.6 for GTAW (141); k ≈ 1.0 for submerged arc (12x).

A measurement warning for pulsed processes: The relationship above (ISO/TR 18491 Method A) holds only for a steady arc. In waveform-controlled processes such as pulsed GMAW, a calculation based on average voltage and average current readings can be wrong by up to 70% (usually low). In that case, measure instantaneous energy (Method B) or instantaneous power (Method C). Where a WPS carries an upper limit on heat input, pulsed values recorded by Method A are a direct non-conformity.

Another common misunderstanding treats GTAW as the "lowest heat input" process under all conditions because of its low k factor. Look at the equation again: GTAW's travel speed (v) is very low, and that pushes heat input per unit length up. In heavy multi-pass sections, slow GTAW passes often produce high total heat input and a wide HAZ. What really distinguishes GTAW is therefore not an absolute low heat input, but the fact that it controls heat input most precisely and reaches genuinely low heat input at low current in thin sections — ideal for thin material and sensitive metallurgy. GMAW, in turn, can be tuned across a wide heat input range through voltage, current and travel speed.

The heat input window by material: Treating heat input purely as an "upper limit" is a one-sided view specific to carbon steel.

Carbon and low-alloy steel: The upper limit exists to prevent grain coarsening and loss of toughness. But there is a risk at the bottom end too: heat input governs the cooling rate (t8/5), and very low heat input means rapid cooling, a hard martensitic HAZ and susceptibility to hydrogen cracking. That is the physical bridge to the hydrogen section below.
Duplex/super duplex stainless steel: Both a lower and an upper limit apply; for ferrite/austenite balance and reaustenitization, the typical window is around 0.5–2.5 kJ/mm (the manufacturer's data and the project specification govern the exact values).
Austenitic stainless steel: Low heat input plus a low interpass temperature (commonly ≤ 150 °C) keeps sensitization and sigma phase under control. EN 1011-3 gives the detailed guidance for arc welding of stainless steels.

5. Positional capability

SMAW: Works in all positions, but only as far as the position digit in the electrode classification allows. The classification governs, not the electrode "type": in AWS A5.1 it is the third digit (E6010, E7018 → all positions; E7024 → flat and horizontal fillet only; E6048 → including vertical-down), and in EN ISO 2560 the position letters. The mechanism differs by electrode family as well: with basic and rutile electrodes, fast-freezing slag supports the molten pool overhead and in vertical-up; with cellulosic electrodes, positional capability comes not from slag (cellulosics produce very little) but from the harsh, deeply penetrating arc characteristic and the vertical-down (stovepipe) technique. That is one reason SMAW remains indispensable for field piping.
GTAW: Excellent in all positions thanks to the small, controlled pool; positional freedom on pipe root passes is one of its biggest advantages.
GMAW: Works in all positions in short-circuiting and pulsed modes. Spray transfer, however, is essentially limited to flat (PA) and horizontal fillet (PB) positions because of the high current and large fluid pool — horizontal-vertical (PC) is marginal for spray. Selecting the right transfer mode is essential for positional welding.

One final constraint: even where a process is physically suited to the position, what binds you on site is the range of qualification for position on the welder's certificate (ASME IX QW-461 tables / EN ISO 9606-1 position mapping). A 6G (H-L045) test covers all positions, while 1G/PA covers only the flat position. Putting a welder to work in a position their certificate does not cover is a non-conformity, no matter how well the process fits.

6. Hydrogen control

Diffusible hydrogen is the leading cause of cold (delayed) cracking in high-strength and heavy-section steels. It is measured in ml/100 g, the measurement method is ISO 3690, and it is declared through the H designator in the filler material classification (H5/H10/H15 in EN ISO 2560; H4/H8/H16 in AWS A5.1). Here process selection is directly a safety issue:

SMAW: Cellulosic electrodes generate high hydrogen. Basic / low-hydrogen electrodes (an H5 or H4 class basic electrode, for example — confirm the classification on the purchase documentation through the ISO 2560 / AWS A5.1 designator) deliver very low hydrogen levels when they are baked and stored correctly (holding oven, limited exposure time) — but that moisture-management discipline is critical.
GMAW and GTAW: Inherently low-hydrogen processes; the intrinsic hydrogen contribution of solid wire and inert/active gas shielding is negligible. Even so, hydrogen is never entirely absent: the usual sources are drawing lubricant or residue on the wire surface, moist shielding gas and contamination on the workpiece. On crack-sensitive materials this remains a strong argument for GMAW/GTAW.
FCAW (flux-cored wires): The generalization above does not extend to cored wires. Both gas-shielded (FCAW-G) and self-shielded (FCAW-S) wires contain flux, so the hydrogen level depends entirely on the wire class; declare and verify it through the H designator (H5/H10/H15). FCAW-S also carries toughness and code constraints: AWS D1.1 restricts certain FCAW-S classifications on thickness, multi-pass application and quenched-and-tempered (Q&T) steels. Confirm the selection against the project specification and the usage range the code permits.

Managing hydrogen risk does not end with process selection. Preheat temperature comes from three factors together, not from a single table — carbon equivalent (CE/CET) + combined thickness + hydrogen class — and EN 1011-2 gives the calculation method for ferritic steels. Control interpass temperature within the same framework. Post-heating for hydrogen removal after welding is not the same thing as PWHT. Its parameters (typically a holding temperature on the order of 200–300 °C and a holding time based on thickness) come from the project specification and the applicable fabrication code (ASME B31.3 or AWS D1.1, for example), so do not write them as if a single standard fixed them.

7. Typical application: root pass, heavy section, field and shop

In real fabrication we usually see the processes used together:

Pipe root pass: GTAW is superior. It gives a slag-free, fully penetrated, smooth internal root, and it is the standard on critical lines and on the stainless piping of the food and pharmaceutical sectors. On stainless and nickel alloys, though, a GTAW root only makes sense with a back purge: without it the internal surface oxidizes (sugaring), chromium depletion destroys corrosion resistance, and the inside of the root will not pass visual or RT acceptance. In practice you purge the root volume with argon and measure the residual oxygen; common acceptance before striking the arc is ≤ 0.1% O₂, and ≤ 0.05% O₂ on critical lines (the exact limit is subject to the specification). On titanium, add a trailing shield and a low interpass temperature to the back purge; inadequate shielding causes oxygen/nitrogen pickup and embrittlement, and the surface discoloration (blue/gray/white) is judged against the color acceptance criteria in the specification. Cellulosic SMAW is the widespread alternative for field pipe roots — but a cellulosic root generates high hydrogen, so its use on hydrogen-sensitive high-strength or heavy steels comes with conditions: keep cellulosic to the root pass, run fill and cap with low-hydrogen electrodes or FCAW, apply preheat, and put in the hot pass without delay. On field pipelines this arrangement is qualified to API 1104; as the strength grade rises (X80 and above), mechanized GMAW becomes the preference.
Fill and cap passes (heavy section): Fill with SMAW or GMAW at high deposition rate. A common combination is "GTAW root + SMAW/MAG fill".
Shop series production: GMAW/MAG dominates — fast, automation-friendly, low operator fatigue.
Field and erection work: SMAW is still king; the equipment is portable, it tolerates wind, and it needs no external gas. GMAW on site without a wind screen is a gamble.

8. A brief look at FCAW and SAW

FCAW (flux-cored wire): Runs on GMAW equipment but with flux inside the wire, so it produces slag. It comes in gas-shielded (FCAW-G) and self-shielded (FCAW-S) types. It combines high deposition rate with good positional capability, and it is very common in heavy steel structures and field steel fabrication. Two records are essential when you select it: verify the hydrogen class per wire through the H designator, and check the toughness and code constraints for FCAW-S (thickness, multi-pass use, Q&T steel). One more distinction matters here: flux-cored wire (FCAW) produces slag, whereas metal-cored wire (MCAW) produces almost none and behaves in practice like GMAW — so do not lump the two together.

SAW (submerged arc): The arc burns hidden beneath granular flux, which gives k ≈ 1.0 thermal efficiency and very high deposition rates. But it is essentially limited to flat and horizontal positions and to long, straight seams (boilers, pressure vessels, longitudinal seams in heavy plate), and it needs a mechanized setup.

From the field

On a refinery project, a stainless process line with a 12 mm pipe wall thickness tempted the crew to run the whole joint with MAG and save time. They put the root in with short-circuiting MAG, and the inside surface came out with irregular penetration and patches of lack of fusion and lack of penetration. On radiography the root indications exceeded the acceptance criteria, and the welds were rejected.

The fix was to split the process again: GTAW root, MAG fill and cap. Moving the root to GTAW brought a second condition with it — nobody struck an arc until the line had been purged with argon and the residual oxygen measured (≤ 0.1% O₂). An unpurged GTAW root leaves an oxidized, sugared inside surface even when the outside looks flawless, and on stainless that is the end of the corrosion resistance. Once the purge discipline was in place, the inside surface came out smooth and fully penetrated, and the fill passes still got the benefit of MAG's deposition rate. The lesson is plain: choose a process on speed alone and you pay for it in quality at the root, where it matters most.

The other classic on that site was running the GMAW torch in the open and getting porosity for it. For gas-shielded processes a wind screen is not negotiable, and the measurable limit is the 5 mph (≈ 2.2 m/s) in AWS D1.1 — take the anemometer reading and write it down. An inspector who catches this early saves meters of weld from the scrap pile.

Process selection logic — a summary decision framework

Ask these questions in order when you choose a process:

Material and thickness: Thin or sensitive sections and reactive metals (stainless root, titanium, thin aluminum) → GTAW; use AC on aluminum, and DCEN plus a back purge on stainless and titanium (plus a trailing shield on titanium). Medium-to-heavy carbon steel in series production → GMAW/MAG. Heavy section in the field → SMAW/FCAW.
Position: All-position work on fixed pipe → GTAW root + SMAW/FCAW fill. Flat position in the shop → GMAW spray/SAW. Confirm that the chosen position falls within the range of qualification on the welder's certificate.
Environment: Windy open site (measured air velocity > 2.2 m/s) → SMAW/FCAW-S; for FCAW-S, check the toughness/thickness and code constraints of the wire class. Enclosed shop → GMAW/GTAW.
Quality class: Radiographic full penetration and a low-hydrogen requirement → GTAW root, H5/H4 class low-hydrogen SMAW, or GMAW.
Productivity and cost: Highest deposition rate and operator factor → GMAW/SAW; lowest equipment cost and best portability → SMAW.

This framework ties the process to the real constraints of the job rather than to habit.

Related standards

ASME BPVC Section IX — welding and brazing procedure (WPS/PQR) qualification and welder performance qualification; defines heat input without an efficiency factor, that is, as arc energy.
EN ISO 15609 (particularly -1) — preparation of the welding procedure specification (WPS).
EN ISO 15614 (particularly -1) — welding procedure qualification by test (PQR).
EN ISO 9606 (particularly -1) — welder qualification testing and ranges of qualification for position.
EN 1011 (particularly -1 and -2; -3 for stainless steels) — welding recommendations; thermal efficiency factors (EN 1011-1:2009, Table 1), preheat and interpass temperature calculation for ferritic steels.
ISO/TR 18491 — measurement of heat input/arc energy in welding (Methods A/B/C; B or C for pulsed processes).
ISO 3690 — determination of diffusible hydrogen in weld metal.
EN ISO 14175 — gases and gas mixtures for welding and allied processes (group designations I1, M12, M21 and so on).
EN ISO 2560 (covered electrodes), EN ISO 14341 (GMAW solid wire), EN ISO 636 (GTAW rods), EN ISO 17632 (flux-cored wire) — filler material classifications and H designators.
EN ISO 5817 — quality levels for imperfections in fusion-welded joints in steel, nickel, titanium and their alloys.
EN ISO 6520-1 — classification and definitions of imperfections in welds.
EN ISO 3834 — quality requirements for fusion welding.
EN ISO 2553 / AWS A2.4 — welding symbols and their representation on drawings.
AWS D1.1 — structural welding code for steel; wind velocity limit for gas-shielded processes (Clause 5.12) and FCAW-S usage restrictions.
API 1104 — welding of pipelines and related facilities; qualification of field pipe procedures and acceptance criteria.
EN ISO 4063:2023 — welding process numbering: SMAW = 111, MIG = 131, MAG = 135, GTAW = 141; submerged arc main group 12 (121 with solid wire electrode); gas-shielded flux-cored FCAW-G = 136 (active gas) / 137 (inert gas); self-shielded tubular-cored arc welding FCAW-S = 114; metal-cored MCAW = 138 (active gas) / 133 (inert gas) — a separate process, distinct from flux-cored wire. The numbering was revised in the 2023 edition, so always cite the edition year.

This article is for educational purposes and does not replace the official standard. Always work from the current edition of the applicable code or standard; in every application the governing project specification, that current edition and the approved WPS are binding.

Take it to the field: To keep these welding standards, procedure logic and field control steps in your pocket — completely offline and free — take a look at the Doawise Welding Eng Guide app.

At DoaWise we carry out welding engineering and inspection services to international standards such as ASME Section IX, EN ISO 15614 and EN ISO 3834; from process selection to WPS/PQR approval, and from welder qualification to non-destructive testing, we document every step with recordable, auditable and traceable results. Our aim is to match the right process to the right job, safeguarding both quality and productivity, and to hand over every weld on site with the confidence the standards demand.