Flux Core Welding Polarity: DCEN or DCEP? (Guide)

A fabricator loads a fresh spool of self-shielded flux core, dials in the wire feed speed, and gets a hissing, spitting mess instead of a weld — because the machine was still switched to electrode positive from the MIG job that ran earlier that morning. Swapping the leads takes five minutes and the weld changes completely. Polarity is the one setting most often inherited from the last job, and it quietly controls your penetration, spatter, bead shape and even how the arc sounds.

Flux Core Welding Polarity: The Short Answer First

A fabricator loads a fresh spool of self-shielded flux core, dials in the wire feed speed, and gets a hissing, spitting mess instead of a weld. The machine was still switched to electrode positive from the MIG job that ran earlier that morning. Swapping the leads takes five minutes and the weld changes completely. Polarity is the one setting most often inherited from the last job, and it quietly controls your penetration, spatter, bead shape and even how the arc sounds.

So here is the answer before anything else. Self-shielded flux core wire runs on DCEN, direct current electrode negative, also called straight polarity. Common examples are AWS E71T-11 and E71T-8. Most gas-shielded carbon steel flux core runs on DCEP, direct current electrode positive, also called reverse polarity, and the classic example is E71T-1. When a data sheet and a habit disagree, the manufacturer’s data sheet on the spool wins.

Here is why the terms trip people up. In flux core, the “electrode” is the continuously fed tubular wire itself, not a stick you clamp in a holder. So electrode negative simply means the wire is connected to the negative terminal of the machine, and the workpiece sits on the positive side. It does not mean anything negative is happening to your bead.

Quick Answer Table

Process	Wire Class	Typical Polarity	One-Line Reason
FCAW-S self-shielded (“gasless”)	E71T-11, E71T-8	DCEN / DC- (straight)	Core chemistry is designed around a negative electrode, so DCEN melts the wire cleanly and controls the arc
FCAW-G gas-shielded carbon steel	E71T-1 (E71T-1C), E70T-1	DCEP / DC+ (reverse)	Electropositive heat delivery gives the deep penetration this process is chosen for
Stick welding (SMAW)	E7018, E6013	Usually DCEP	Standard practice for common structural and mild steel electrodes
TIG (GTAW)	ER70S-6, ER308L	Usually DCEN	Keeps heat on the tungsten so the tungsten stays sharp and the filler melts properly
Air carbon arc gouging	Carbon gouging rod	Usually DCEP	Requires strong heat at the workpiece to melt the metal away cleanly
Gas-shielded stainless flux core	Vendor-specific, e.g. E347T-X	Check the data sheet	Stainless and metal-cored wires carry their own polarity and gas requirements

Miller Electric makes the split easy to remember: stick welding and gouging are usually electrode positive, while self-shielded flux-cored and TIG welding are typically electrode negative. That single sentence covers most of the polarity decisions a shop or field crew will face in a year.

The self-shielded side has its own authoritative confirmation. Hobart Brothers states that self-shielded T-8 and T-11 wires should be operated on DCEN, the opposite of most wire filler metals, and that using the incorrect polarity results in poor welding performance, including increased spatter levels. That last clause matters. A polarity mistake on self-shielded wire does not merely make the bead ugly, it degrades the arc and dumps metal into the air as molten droplets.

Universal Technical Institute frames the choice the same way: the polarity used for FCAW depends on the shielding method, with most gas-shielded welds performing best on DCEP and self-shielded processes running on DCEN.

If you are still building your working knowledge of the process itself, our guide to all-position welding rod types breaks down the families of electrodes and filler metals in plain language, which makes the polarity rules much easier to place.

One last framing before we go deeper. Polarity is not a quality setting you tune for feel. It is a design decision baked into the wire, the flux chemistry and the shielding system. Flip it against the wire’s design and you are asking the process to work in a way its manufacturer never qualified. That is why the answer above comes with “typical” attached, and why the spool label is the final authority.

How Welding Polarity Actually Changes the Arc

Once you know the rule, it feels arbitrary. Self-shielded wire runs electrode negative, most gas-shielded flux core runs electrode positive, and everyone just says “follow the data sheet.” That answer is technically correct and practically useless: it teaches memorization instead of understanding. Understand what polarity does to the arc and you can predict what a wire will do before you strike it.

The direct current circuit

Every direct current arc is a simple loop. One terminal of the power source feeds the electrode; the other feeds the workpiece through the ground clamp. The arc is the gap in the middle, where air has been ionised into a plasma. Polarity is simply which side of that loop is positive: with the wire negative and the workpiece positive, electrons flow from the workpiece through the arc into the wire, and reversing the wire reverses the flow.

In flux core arc welding the “electrode” is not a stick you hold but the tubular wire feeding continuously through the gun. The Fabricator’s guide to FCAW consumables makes the point: FCAW power sources are constant-voltage units, and a CV machine can be set to either straight polarity (DCEN) or reverse polarity (DCEP). Which one is correct depends on the wire formulation, not the machine.

Where the heat goes

Welding metallurgy references commonly place roughly two-thirds of DC arc heat at the anode, the positive side of the arc. That one fact explains most of what you see on the plate.

On DCEP the workpiece is the anode, so more heat lands in the steel: deeper penetration, a wider and flatter bead profile, and the ability to travel faster, since the plate absorbs more energy per inch. The trade-off is a harsher arc and, with CO2 shielding, noticeably more spatter. On DCEN the wire is the anode, so heat concentrates in the consumable: the wire melts faster, the bead is narrower and more controllable, penetration is shallower, and spatter drops. Less heat in the workpiece is also why DCEN tends to be the easier polarity out of position, where you need a stable puddle, not a deep one.

Why the flux core cares so much

The wire is a tube filled with flux that melts and decomposes into shielding gas, forms the slag on the puddle, and drives the metallurgical reactions that let the wire deposit cleanly. Change polarity and you change the temperature at which that decomposition happens. The core chemistry was designed around a specific heat pattern, and running it backwards alters how the flux melts, how fast it releases shielding gas, and how the slag forms. That is why self-shielded wires are built around a negative-electrode chemistry. Hobart Brothers’ technical article on flux core wire states that T-8 and T-11 wires should both run on DCEN, or straight polarity, opposite to most wire filler metals. Flip a self-shielded wire to DCEP and you destabilise the arc, spike the spatter rate and lose the puddle control it was built for.

What about AC?

Alternating current is essentially never specified for flux core wires. The arc extinguishes every half cycle, and that constant re-ignition works against the stable, focused column a continuously fed wire needs. If a data sheet does not call out AC, do not run it, and if no data sheet exists for the wire in your hand, stop guessing. Universal Technical Institute’s overview of flux cored welding makes the practical point: the consumable is designed for a specific process, shielding method and polarity, and deviating from that design produces spatter, poor bead shape and erratic penetration.

A real shift on a real jobsite

Polarity is not always a decision you make once. Miller Electric’s discussion of welding polarity switches describes the classic case: a T-6 wire used in the flat position requires electrode positive, while a T-8 wire used out of position requires electrode negative. A shift can move from flat steel to vertical-up or overhead work and need a polarity change mid-day.

Polarity mistakes are common for the same reason. The setting is inherited from the last job, and nothing on the machine tells you it changed. Miller notes that wrong polarity always degrades performance, and their TIG comparison makes it plain: run TIG on the wrong setting and the heat sits in the tungsten instead of the workpiece, and the tungsten melts.

See the arc as a heat-distribution problem, not a switch position, and the rules stop being arbitrary. Electrode negative puts the heat in the consumable, suiting self-shielded wires engineered around a negative-electrode core and a stable puddle. Electrode positive puts the heat in the plate, suiting gas-shielded wires chosen for penetration and deposition rate. The data sheet tells you which design your wire is.

Self-Shielded vs Gas-Shielded: The Polarity Rules by Wire Type

Polarity in flux core welding is not a free choice. It is tied to the shielding system of the wire, and once you see that link, the rules stop feeling arbitrary. Here is the two-rule mental model most shops run on:

1. Self-shielded flux core (FCAW-S) runs on DCEN, electrode negative.
2. Gas-shielded flux core (FCAW-G) runs on DCEP, electrode positive, for the common rutile carbon steel wires.

Those two rules cover the overwhelming majority of real work. The exceptions are worth knowing, but they are exceptions. Hobart Brothers puts it bluntly: self-shielded T-8 and T-11 wires are intended for DCEN operation, the opposite of most wire filler metals, and running them on the wrong polarity produces poor performance and increased spatter.

Side-by-side: what actually changes

Factor	FCAW-S (Self-Shielded, “Gasless”)	FCAW-G (Gas-Shielded, “Dual-Shielded”)
Shielding method	No external gas. The flux core decomposes; the arc plus a slag layer protect the puddle	External gas, usually 100% CO2 or an argon/CO2 mix, added over the flux core
Typical AWS classes	E71T-11, E71T-8, E71T-6	E71T-1, E70T-1, E71T-9, E71T-12
Typical polarity	DCEN (electrode negative)	DCEP (electrode positive)
Positions	All-position; popular for vertical-up and structural fillets	All-position with rutile wires, but some are flat and horizontal only
Spatter level	Noticeably heavy, plus more smoke and fume	Lower with mixed gas; higher with straight CO2
Portability	High. No cylinder, no flowmeter, no hose to trip over	Low in the field. The cylinder and regulator become part of the job
Typical environments	Structural erection, pipelines, outdoor sites, repair work	Shop fabrication, water tanks, shipbuilding, production welding

FCAW-S: the wire that carries its own protection

Self-shielded wire has no gas cylinder attached. The chemistry inside the tubular core generates its own shielding atmosphere as the arc burns, then leaves a slag blanket over the bead as it solidifies. Everything the wire needs to protect the molten puddle travels inside the spool.

That design dictates polarity. These wires are engineered around negative-electrode operation, and DCEN gives them the arc behaviour their chemistry expects. Flip them to DCEP and the arc destabilises, transfer turns erratic, and spatter climbs.

Two classes dominate: E71T-11, the general purpose workhorse, and E71T-8, the low-hydrogen version with better toughness. Because there is nothing to hook up but the feeder and the lead, this is the wire of choice for structural erection crews, pipeline welders, and anyone working a long way from a gas supplier.

The trade-off is mess. Universal Technical Institute notes that self-shielded wires tend to produce more smoke and spatter, while gas-shielded wires are easier to control and produce more aesthetically pleasing welds. Self-shielded wins on logistics, gas-shielded wins on finish.

FCAW-G: adding a cylinder to buy a better bead

Gas-shielded flux core is a hybrid process. It keeps the tubular wire and the flux core, then adds an external gas to replace some of the shielding the self-shielded design would have to create for itself.

The gas choice matters, and the AWS classification tells you which one the wire was designed for. According to The Fabricator, wires with a “C” designation such as E70T-1C H8 use only 100% CO2, while “M” wires such as E71T-1M require a CO2/argon mix, typically 75/25. The flux formulation is matched to the oxidising potential of that gas, which is why you cannot swap cylinders without checking the data sheet.

Rutile-style gas-shielded wires like E71T-1 are normally run DCEP, because that polarity puts more heat into the workpiece and delivers the deep penetration this process is chosen for. A community college training module spells out the same logic, describing the E71T-1 gas-shielded flux-cored electrode as a multi-pass, all-position wire run with DCEP for good penetration (PCC WLD 141).

Gas choice shifts the balance. Straight CO2 gives more penetration but noticeably more spatter. A mixed gas such as Ar-25%CO2 cuts spatter, improves out-of-position capability and Charpy impact toughness, and pushes the arc toward a near-spray transfer, at higher gas cost. If you are setting up the cylinder, our complete guide to shielding gases breaks down what that mixture actually costs you.

The exceptions worth checking

Not every gas-shielded wire is DCEP. Stainless flux-cored and metal-cored wires carry their own polarity and gas recommendations from the manufacturer, and those can differ from the carbon steel rule of thumb. Position capability matters too: an all-position wire versus a flat-and-horizontal-only wire changes your technique and often which wire you load, which changes polarity. Our position-specific filler metal guidance covers that side in more detail.

AWS splits its usability designators by shielding method, a handy way to sanity-check what you are holding. Self-shielded families include T-3, T-4, T-6, T-7, T-8, T-10, T-11, T-13 and T-14, while gas-shielded families include T-1, T-2, T-5, T-9 and T-12, with the G and GS designators covering metal-cored and related wire forms used with both shielding systems (AWS standards). If the designator on your spool sits in the self-shielded column, assume DCEN until the data sheet says otherwise. If it sits in the gas-shielded column, assume DCEP for carbon steel, then check the sheet for stainless and metal-cored wires.

I’ll rewrite this section, tightening it while keeping all the facts and links intact.

How to Read Your Wire Label and AWS Classification

Polarity is the easiest thing to get wrong on a wire spool and the easiest to fix, if you know where to look. Three places carry useful information: the label printed on the spool, the sticker on the wire basket, and the spec sheet in the box. They do not all carry the same information. The AWS classification tells you what kind of wire you have. The polarity comes from the manufacturer. Here is how to read the classification left to right, and where to find the setting that actually matters.

Code element	What it means	In E71T-11	In E71T-1C
E	Electrode or electrode assembly, meaning the wire plus its flux core	E	E
7	Minimum tensile strength of the deposited metal, 70 ksi	7 (70 ksi)	7 (70 ksi)
1	Electrode family, in this case carbon steel	1	1
T	Tubular, or flux-cored, construction rather than a solid rod	T	T
11 / 1	Usability designator: intended operating characteristics, which in practice points to shielding type	11 = self-shielded, “gasless”	1 = gas-shielded
C / M / H4	Optional suffixes for shielding gas and diffusible hydrogen	None, self-shielded	C = CO2, M = argon/CO2 mix

Read the last line carefully, because that is where most of the confusion lives. A self-shielded wire carries no gas suffix at all, while a gas-shielded wire announces its gas in the final character. C means the wire is designed for 100 percent carbon dioxide, M means it is designed for a mixture of argon and carbon dioxide. The flux inside is matched to the oxidising behaviour of that gas, which is why the same family number can behave very differently under CO₂ than under a blend.

What the usability designator really buys you

Family differences show up in the usability designator. Hobart Brothers explains that AWS E71T-8 wires must meet a toughness requirement of 20 foot-pounds at minus 20 degrees F, and that the optional “J” designator, written as E71T-8 J, indicates increased toughness of 20 foot-pounds at minus 40 degrees F. That is a real performance difference on a cold-weather pipeline job in the northern United States or Canada, and it is why a T-8 and a T-11 are not interchangeable even though both are self-shielded and both run on DCEN.

Those toughness numbers carry code weight. AWS D1.1, Structural Welding Code for Steel, allows E71T-11 wires for welding on material less than half an inch thick, and T-11 is generally used for stitch welds and tack welds rather than as the primary structural deposit. T-8 wires, with their higher toughness rating, are often approved for demand-critical welds under AWS D1.8/D1.8M, the seismic supplement to D1.1.

Position capability shows up in the last digit

Compare two wires that look almost identical. The Fabricator notes that E71T-1C wires are all-position while E70T-1C wires are for flat and horizontal positions only. The “1” versus “0” before the T is a tensile-strength difference of 70 ksi versus 60 ksi, and it maps to a different usability designation that limits position. If you plan to go vertical-up or overhead, that digit decides whether the wire is even legal for the job. For a broader look at position options, our all position welding rod guide breaks down how position ratings work across filler metal families.

What is inside the core

For E71T-1, the flux core makes up roughly 20 to 40 percent of the total weight of the wire, with rutile and silica as the major ingredients. That is a substantial share of what you feed, and it explains why polarity changes the arc so dramatically. The suffix H4 on E71T-1H4 indicates a diffusible hydrogen limit of no more than 4 millilitres per 100 grams of weld metal, a classification that matters for controlling hydrogen cracking in restrained joints.

Where to find the polarity answer

• The spool label: normally states recommended polarity, shielding gas and position ratings.
• The sticker on the feeder basket: often the only label left intact after a spool change.
• The machine manual: confirms what DC minus and DC plus mean on your power source and lead connections.
• The manufacturer’s product data sheet: the definitive source, listing operating ranges, polarity and shielding requirements.
• The AWS A5.20 classification table: the standard that defines each code element, in the AWS standards library.

What Goes Wrong When the Polarity Is Wrong

A wrong polarity setting rarely shows up as one clean, obvious failure. It arrives as a cluster of complaints: the arc sounds odd, the spatter is heavy, the bead looks strange, the slag is a pain to clean. That is why polarity errors survive for years in the shop. Nobody blames polarity because no single symptom belongs to it alone. Hobart Brothers puts it plainly: using the incorrect polarity can result in poor welding performance, including increased spatter levels. Here is the symptom-by-symptom map.

Symptom by symptom

• Excessive spatter and a spitting, popping arc. Bright orange droplets eject from the arc, land on the bead and plate as hard little balls, and pepper the work area. Self-shielded wire running on electrode positive is the classic cause, because the core chemistry was never designed to burn that way.
• A hissing or crackling arc instead of the steady roar. Correctly set flux core has a confident bark. Wrong polarity turns it into a hiss, a sputter, or a rapid-fire crackle. Sound is the most underrated diagnostic on the shop floor, because your ear adapts in about a minute but the machine does not lie.
• An erratic, wandering arc that will not hold a puddle. Arc length jumps, the puddle spits and runs, the wire hunts across the joint. The welder compensates by riding the gun harder, which burns hands and burns through thin material.
• Shallow penetration and weak fusion. On the wrong polarity, more heat stays in the wire and less reaches the plate, so you get a cold, stacked bead with little root fusion. This is the dangerous one, because it can look like a weld, pass a casual visual check, then fail a bend or break test later.
• A wide, flat, lifeless bead. No crown, sharp edges, a dull sooty surface, a profile that does not match what the wire should produce. Be careful, though: an over-wide, over-flat bead is also the classic signature of excess voltage, so treat this as a clue rather than a verdict.
• Undercut along the toes. A groove cut into the base metal beside the bead where it meets the plate, caused by an arc that wanders and throws heat sideways instead of into the joint. It is a fatigue crack waiting to happen.
• Slag inclusions and worm tracking. Slag trapped inside the bead, plus worm-like trails of spatter along the surface. Before blaming polarity, check stick-out: The Fabricator notes that improper stick-out can produce burnback, worm tracking, incomplete slag coverage and difficult slag removal, a textbook case of one symptom with several possible causes.
• Porosity. Hollow, pitted bubbles, often clustered at the start or in a cool section. Wrong polarity destabilises shielding and pushes gas into the puddle. So does a poor power source: FCAW needs a constant-voltage (CV) machine, and Hobart Brothers warns that a constant-current (CC) source fluctuates too much with these wires and can cause inconsistent quality and defects such as porosity.
• Poor wetting and slag that will not release. The bead sits on the plate like thick paint instead of flowing into it, and the slag is crusty and stubborn. Hobart Brothers reports that self-shielded T-8 wires produce very heavy slag that can interfere with the arc and prevent the bead from wetting out properly, so confirm what wire you are actually running before chasing polarity.
• Wire burnback and stubbing at the feeder. The wire melts back into the contact tip, the gun runs hot, and the feeder stalls or stutters. On a machine left on electrode positive from the previous stick or MIG job, this is one of the first symptoms to show up.

The spatter trap

Here is where most self-diagnosis goes sideways. Pure CO2 gas-shielded flux core spatters far more than an argon/CO2 mix, even when polarity is perfectly correct. Spatter on its own is not proof of a polarity error. Read the whole picture instead: the arc sound, the bead width and crown, the depth of fusion, and how the puddle behaves. If the sound is a steady roar, the bead has proper reinforcement, and penetration is where it should be, then the spatter is your gas, not your polarity.

Contrast that with TIG, where the same mistake is unmistakable. Miller Electric notes that in TIG the tungsten melts because all the heat is in the tungsten and not in the workpiece. Nobody has to diagnose that one; the tungsten is gone. The logic runs in reverse for stick welding, where electrode positive is standard, a fair part of the reason SMAW still wins for field repair and farm-shop work.

Finally, expect the misdiagnosis. A polarity mistake is very often blamed on voltage or wire feed speed, and chasing it there burns hours of a shift on parameter tweaking that can never work, because the settings are not the problem. If your arc sound is still wrong after three parameter changes, stop adjusting and go check the leads. Changing polarity is the fastest fix on the machine, and it costs five minutes.

How to Set Flux Core Polarity on Your Machine, Step by Step

Changing polarity takes about five minutes, and it is the most common mistake carried from one job to the next. A MIG job runs all day, the machine is left on electrode positive, then someone loads self-shielded flux core and hits the trigger. The arc spits and an hour disappears into parameter tweaking that will never fix it, because the problem was never voltage or wire feed speed.

Know Your Two Leads

The electrode lead is the heavy cable ending in the welding gun or holder; in flux core it feeds the continuously fed wire. The work lead ends in the ground clamp on the workpiece. Which of those is negative and which is positive is the definition of polarity.

• DCEN (DC-): electrode lead negative, clamp positive. “Straight polarity.”
• DCEP (DC+): electrode lead positive, clamp negative. “Reverse polarity.”

Trace the cables physically to the studs. On some machines the electrode stud is labelled “Work” and the clamp stud “Electrode,” which confuses people on its own.

Step 1: Kill the Power First

Turn the machine off, unplug it or trip the breaker, and let the capacitors discharge before touching a connector. Never swap leads under load. An energized stud with a loose lead arcs across the connector, burning operators and pitting the lugs.

Step 2: Machines With a Polarity Switch

Most modern multiprocess sources have a polarity toggle or a menu setting. Set DC- for self-shielded wire such as E71T-11 or E71T-8, and DC+ for most gas-shielded carbon steel flux core such as E71T-1C. Check the manual; on some units the setting applies only when a particular process is selected.

Step 3: Machines Without a Polarity Switch

On older single-process and field units, polarity is set physically. Unplug the electrode lead from its stud, move it to the other stud, and tighten the connector firmly with a wrench. Then move the work lead to the stud the electrode lead just vacated. Both connections must change; moving only one leaves the machine unable to run correctly.

Step 4: Multi-Process Machines and Remote Switching

On a multiprocess welder, check whether process and polarity are handled at the feeder or a remote pendant rather than at the power source. Miller’s ArcReach polarity-reversing systems, offered on machines such as the XMT 650 ArcReach and XMT 350 FieldPro, let operators change process and polarity from the wire feeder or from a stick and TIG remote, with the change made inside the power source and no cable swapping.

Miller

How to Set Flux Core Polarity on Your Machine, Step by Step

Changing polarity takes about five minutes, and it is the most common mistake carried from one job to the next. A MIG job runs all day, the machine is left on electrode positive, then someone loads self-shielded flux core and hits the trigger. The arc spits and an hour disappears into parameter tweaking that will never fix it, because the problem was never voltage or wire feed speed.

Know Your Two Leads

The electrode lead is the heavy cable ending in the welding gun or holder; in flux core it feeds the continuously fed wire. The work lead ends in the ground clamp on the workpiece. Which of those is negative and which is positive is the definition of polarity.

• DCEN (DC-): electrode lead negative, clamp positive. “Straight polarity.”
• DCEP (DC+): electrode lead positive, clamp negative. “Reverse polarity.”

Trace the cables physically to the studs. On some machines the electrode stud is labelled “Work” and the clamp stud “Electrode,” which confuses people on its own.

Step 1: Kill the Power First

Turn the machine off, unplug it or trip the breaker, and let the capacitors discharge before touching a connector. Never swap leads under load. An energized stud with a loose lead arcs across the connector, burning operators and pitting the lugs.

Step 2: Machines With a Polarity Switch

Most modern multiprocess sources have a polarity toggle or a menu setting. Set DC- for self-shielded wire such as E71T-11 or E71T-8, and DC+ for most gas-shielded carbon steel flux core such as E71T-1C. Check the manual; on some units the setting applies only when a particular process is selected.

Step 3: Machines Without a Polarity Switch

On older single-process and field units, polarity is set physically. Unplug the electrode lead from its stud, move it to the other stud, and tighten the connector firmly with a wrench. Then move the work lead to the stud the electrode lead just vacated. Both connections must change; moving only one leaves the machine unable to run correctly.

Step 4: Multi-Process Machines and Remote Switching

On a multiprocess welder, check whether process and polarity are handled at the feeder or a remote pendant rather than at the power source. Miller’s ArcReach polarity-reversing systems, offered on machines such as the XMT 650 ArcReach and XMT 350 FieldPro, let operators change process and polarity from the wire feeder or from a stick and TIG remote, with the change made inside the power source and no cable swapping.

Miller notes that swapping a cable at the power source wastes operator time, increases slip, trip and fall risk on crowded jobsites, and wears the cable connections. OSHA reinforces the rule: 29 CFR 1910.252(b)(1)(ii) requires welders to place welding cable and other equipment so it stays clear of passageways, ladders and stairways.

Step 5: Verify Before Production

1. Label it. Tape a tag on the lead reading FLUX CORE DC- (or DC+) and mark the machine to match.
2. Run a pull test. Feed the wire and confirm smooth, consistent travel with no hesitation or wire damage.
3. Make a test bead on scrap. Match the thickness and position of the production joint, then compare bead width, penetration, arc sound and spatter against a known-good weld.
4. Listen. Correct self-shielded wire on DCEN gives a steady, deep roar. Hissing or crackling means something is wrong, and polarity is the first thing to check.

Two More Checks While You Are at the Machine

Constant voltage, not constant current. Self-shielded T-8 and T-11 wires must run on a constant-voltage (CV) source rather than a constant-current (CC) source, because the machine must manage the wire’s melting characteristics itself. A CC machine set for stick, as on many stick welding setups, will stub and burn back when switched to flux core. The Hobart Brothers guide to flux core wire covers this in more detail.

Feeder choice for field work. Hobart recommends suitcase wire feeders for self-shielded wires because they encase and protect the wire and improve portability on the job site. Weight that feature heavily when buying, along with drive roll type, since flux core wire is softer than solid wire and needs V-knurled rolls.

If you are shopping rather than troubleshooting, the deciding factor is how often your work changes process mid-day. Moving between stick, MIG and flux core daily? A unit with feeder-based polarity switching saves time and removes the cable swap. One process in one environment? A simpler inverter with a front-panel switch is fine. More equipment context is in our guide to air arc welding and carbon arc gouging. (Affiliate disclosure: weldgallery may earn a commission from equipment links.)

Troubleshooting: Is It Really the Polarity?

Most people who blame polarity are actually looking at a voltage, wire feed speed, or moisture problem. Polarity matters, but it is also the setting every welder checks first, because it is easy to check and easy to get wrong.

Work through the symptoms in order before touching any knobs. The table separates failures that point squarely at polarity from the setting errors that are far more common.

Symptom	Likely cause	What to check
Heavy spatter, spitting arc, unstable arc sound	Wrong polarity for the wire, voltage too high, or 100% CO2 where a mix is specified	Confirm the spool label and the polarity switch. If polarity is right, step voltage down 1 to 2 volts at the same feed speed.
Harsh, erratic arc that will not hold a puddle	Wrong polarity, wire feed speed too high, damp wire, long stick-out	Verify polarity first, then check feed speed against the data sheet, wire moisture and gun cable length.
Wire stubs, erratic feed, burns back into the contact tip	Wire feed speed too high, stick-out too long, wrong drive rolls	Use V-knurled drive rolls for flux core wire. It is softer than solid wire, so knurled rollers crush or deform it, causing erratic feed and burnback. The Fabricator explains how the right roll profile keeps the feed smooth.
Undercut along the toes of the bead	Voltage too high, travel speed too fast, long stick-out	Drop voltage, slow travel slightly, hold the gun closer. On self-shielded wire, fix the polarity first.
Slag inclusions and worm tracking	Travel speed too slow, poor slag control, damp wire	Hobart Brothers notes that moisture pickup in self-shielded FCAW wire causes worm tracking, a porosity-like discontinuity under the slag on top of the bead. Check wire storage and travel speed.
Porosity, pinholes, dull oxidation on the bead	Travel speed too slow, gas flow too low, damp wire	For gas-shielded wire, confirm flow rate and check for leaks or a kinked hose. For self-shielded wire, check storage moisture.
Wide, flat, lifeless bead with poor penetration	Voltage too high, travel speed too fast, or DCEN on a DCEP wire	Compare bead width and profile on scrap with the polarity flipped. If the bead improves, you have the answer.
Poor wetting, slag will not release or clean up	Voltage too low, slow travel, or wrong polarity for a rutile gas-shielded wire	Raise voltage slightly, speed up travel, and confirm the polarity specified for the wire class.

Start with the settings, not the switch

Look at how many rows point at the same handful of causes: voltage, wire feed speed, travel speed, stick-out, gas flow and moisture. They outrank polarity in frequency, so if the arc is erratic rather than merely ugly, check the setup first.

Stick-out. The length of wire between the contact tip and the work changes arc resistance and preheats the wire, raising the deposition rate. Too much and the wire burns back into the contact tip before the feeder can push it out. Too little and you lose the preheating benefit. Self-shielded wires with a heavier slag system, such as E71T-8, generally work with a longer extension than the smoother E71T-11.

Gun angle and drag angle. The recommended drag angle for flat, horizontal and overhead positions is generally 10 to 30 degrees, and for vertical-up welds a gun angle of 5 to 15 degrees is typical. Running the gun nearly straight down, or far ahead of the puddle, fights the process and changes how slag forms.

Storage habits that prevent half of these problems

Moisture is the quiet killer of self-shielded flux core. It causes worm tracking, porosity, erratic arc behaviour and poor slag release, and none of those symptoms point back at the machine.

• Keep wire in its original packaging until you are ready to weld with it.
• Store it between 40 °F and 120 °F, with maximum humidity held at 80 percent, according to Hobart Brothers guidance.
• Remove the spool from the wire feeder at night and store it in a plastic bag. The Fabricator recommends this simple step to reduce moisture problems.
• Let wire acclimate between cold and warm spaces. Condensation on a cold spool hitting warm air can wet the flux core.

One more thing while you have the machine apart: spatter and fume are not just nuisance problems. OSHA 29 CFR 1910.252(c)(1)(iii) requires local exhaust or general ventilating systems to be arranged so that toxic fumes, gases and dust stay below the maximum allowable concentration. A spitting wire throws metal and fume into your breathing zone.

The rule of thumb

If the symptom is erratic, suspect polarity, feed setup or moisture before you touch any parameter. If it is simply ugly but steady, suspect voltage, travel speed and technique. Polarity deserves a check on every machine change, because it is inherited silently from the last job, and its symptoms, spatter and a wide flat bead, look a lot like ordinary setup errors.

The reliable test is a side-by-side coupon. Run the same joint in both polarities on scrap of the same thickness and compare penetration, bead width, arc sound and spatter level. That comparison takes five minutes and ends the argument faster than any troubleshooting chart.

Polarity Is Only One Setting: Voltage, Wire Feed Speed and Heat Input

There is a familiar pattern in almost every shop. The weld looks wrong, so the welder turns the voltage up. Still wrong, so the wire feed speed comes down. Still wrong, so somebody blames the gas, the gun, or the polarity. Hours later the real cause turns out to be a reversed lead or a spatter-filled feeder head.

Polarity matters, but it sits inside a cluster of settings that behave like a system. Flip it and every other number you dialled in needs re-examining, because the same machine settings produce a different arc, a different bead and a different spatter level on the opposite setting.

The Heat Input Equation

Heat input is the simplest way to describe how much energy you are putting into the metal, and it is the number a welding engineer looks at when judging a procedure.

Heat input (kJ/in) = (Volts × Amperes × 60) ÷ Travel speed (in/min)

The 60 is a unit conversion factor; everything else comes off the machine and your actual travel speed.

Polarity decides where that energy lands. On reverse polarity more of the arc heat lands on the workpiece, so the same calculated heat input produces deeper penetration and a wider, flatter bead. On straight polarity more heat concentrates in the wire, which melts faster, deposits faster and gives you a narrower, more controllable bead with less heat soaking into the base metal.

So identical numbers on the dial can mean two entirely different joints. A setting that produced a sound weld at DCEN can undercut or burn through on DCEP, and the fix may be a voltage adjustment rather than a polarity change. That is why the polarity decision belongs in the procedure, not the wiring bay.

Voltage and Wire Feed Speed Move Together

Wire feed speed sets how much metal you ask the arc to melt. Voltage sets the length of the arc and how that melting energy is distributed along the wire and the puddle. Push wire feed speed up without adding voltage and the arc pushes the wire back, the puddle gets disturbed and the arc turns harsh and erratic. Raise voltage without raising wire feed speed and you get a long, lazy arc with a wide, flat bead that lacks fusion.

The common shop rule is that voltage and wire feed speed should always move together. Practical set-up sequences for getting a wire to feed smoothly exist, with voltage generally adjusted last, but these two settings are a matched pair, not independent knobs. Check every adjustment against the manufacturer’s data sheet, because the correct numbers change with wire diameter, thickness, position, gas and polarity. The ranges below are starting points from Hobart Brothers, one of the largest flux core wire manufacturers in North America.

Baseline Settings for Self-Shielded Wire

Parameter	E71T-11 (self-shielded, structural)	E71T-8 (self-shielded, all-position)
Voltage	17 to 22 volts	18 to 23 volts
Amperage	50 to 275 amps (varies with wire diameter)	200 to 275 amps for all-position welding
Stick-out	1/2 to 3/4 inch	1 inch to 1-1/4 inches, depending on diameter
Drag angle	10 to 20 degrees	Typically similar, adjusted for position control
Slag	Moderate, easy to remove in flat and horizontal	Heavier, and it can interfere with arc and bead wetting

The T-11 drag angle of 10 to 20 degrees is not cosmetic. Pushing the gun at an angle helps the slag peel back cleanly rather than folding into the puddle, which minimises slag inclusions and maximises joint penetration in the same motion. And the wide 50 to 275 amp spread for T-11 exists because the usable range depends heavily on wire diameter: a 3/32 inch wire and a 5/32 inch wire with a 3/32 inch nozzle are not the same animal. Start from the data sheet chart for your exact diameter and thickness, then trim for position.

Why T-8 Uses a Longer Stick-Out

T-8 wire carries a higher fill percentage, meaning more of the core is powder rather than metal. That is what gives it the toughness and impact properties it is specified for, and it is also why it produces a heavier slag and runs at a higher amperage band.

Its stick-out recommendation runs longer, 1 inch to 1-1/4 inches depending on diameter, and the reason is mechanical. As The Fabricator explains, stick-out provides resistive heating of the wire, and that preheating helps raise the deposition rate. More stick-out gives the wire more time to heat before it reaches the arc, so more of it melts per unit of wire feed speed. The same principle applies elsewhere: recommended stick-out for some gas-shielded flux core wires can exceed 2 inches. One limit matters for T-8 users: the smallest commonly available T-8 diameter is 1/16 inch, because high powder content makes fine diameters hard to draw through the feeder and the gun.

Make the Comparison Yourself

Charts get you close, but your machine and your joint close the gap, and a scrap coupon is cheap. Weld the same joint, thickness and position in both polarities at identical settings, then break or bend the pieces and compare. Look at four things: how far penetration reached, whether the bead is narrow or flat, how much spatter cleaned up, and what the arc sounded like. That test settles arguments the shop has been having for weeks and gives you a personal baseline for every wire and machine in the shop.

Modern multi-process machines, covered in more detail in the polarity setup section, change how some of this gets dialled in. When polarity is switched internally at the feeder or remote rather than at the work, the voltage and feed settings that worked in one mode may need a fresh look in the other, because the machine is compensating for the shift in heat distribution behind the scenes.

To see what the wire itself is doing while it melts, read our guide on aluminum flux core welding, which covers why aluminum filler behaves so differently from steel wire and why most aluminum work still ends up on solid wire processes.

Bottom line: polarity sets the direction the heat flows, but voltage, amperage, wire feed speed and stick-out set how much heat there is and where it lands. Get the polarity right, set the rest from the data sheet, then prove it on scrap.

Choosing Polarity by Metal, Position and Job

Polarity rules make sense once you treat them as a job decision rather than a wire label. The same spool that lays a clean flat bead on a tank floor can spit badly overhead, and the reason is rarely the machine. It is the wire, the position and the metal.

Carbon and Low-Alloy Steel: The Default Assumptions

For most structural and fabrication work in carbon and low-alloy steel, the rules you already know cover nearly every case. Self-shielded E71T-11 and E71T-8 run on DCEN, and gas-shielded E71T-1 runs on DCEP.

Field erection crews load their trucks with self-shielded wire for more than convenience: no gas cylinders to haul, no contamination worries on dirty steel, and a bead that tolerates wind and open-air work. That is the logic behind Hobart Brothers’ self-shielded wire guidance, and why site crews rarely run FCAW-G. Shops run the opposite way, with gas on the wall and screens to catch spatter. Gas-shielded E71T-1 on DCEP delivers deeper penetration, a wider, flatter bead and faster travel speeds, which is exactly why the process is chosen.

Strength options widen the picture. The Fabricator points out, flux-cored wires for steel come in 70 ksi tensile for mild steel and 80 to 120 ksi for high-strength, low-alloy steels. High-strength steel changes heat input expectations, and how sensitive you are to polarity and parameters.

Position Changes the Wire, and the Wire Changes the Polarity

Here the simple rules break down. Flat and horizontal work favours flat and horizontal wire families, which deposit faster and lay a wider, more forgiving bead. Vertical-up and overhead work demand all-position wire, tighter technique and slower travel to hold the puddle.

All-position wires and flat and horizontal wires are different products with different designators, so a position change can mean a wire change, and a wire change can mean a polarity change. Miller Electric uses this scenario: flat-position work with a T6 wire needs electrode positive, while out-of-position work with a T8 wire needs electrode negative. One welder, one shift, two polarities.

Position limits are embedded in the classification. E71T-1C is rated all-position, while E70T-1C is limited to flat and horizontal. Using the wrong family for the position is not a polarity error, but it looks like one, because the arc behaves poorly and spatters. Temperament matters too: Hobart Brothers calls T-11 forgiving, with a wide parameter window that helps newer operators find a stable arc, while T-8 is less forgiving and demands tighter technique. T-8 is the standard for structural and seismic-critical work, so it exposes bad setup fastest.

Stainless Steel: Read the Specific Data Sheet

Stainless is where generalising from carbon steel gets expensive. Gas-shielded stainless flux-cored wires carry their own polarity and shielding gas recommendations, which may not match your memory. Some want DCEP and 100% CO2, others a tri-mix or argon-heavy blend.

Heat input is the other factor. Stainless is prone to distortion and, in the wrong hands, to sensitisation and intergranular corrosion. Lower heat input often means shorter stick-out, a different drag angle and tighter voltage control. The Fabricator also warns that changing shielding gas may require new procedures and testing.

UTI lists stainless among the materials FCAW suits, but the wire on your spool carries the polarity answer. Read its data sheet rather than assuming DCEP because you are in a shop.

Aluminium: A Different Process Entirely

Flux-cored carbon steel wire is not an aluminium process. Aluminium filler is normally deposited with solid wire GMAW or GTAW using argon-based shielding, because aluminium’s oxide layer and thermal conductivity demand a different approach. Polarity rules for aluminium are separate, so if aluminium is on your work order, read our guide to aluminium flux core welding before reaching for FCAW wire.

Typical Jobs, and Why the Process Splits

UTI lists FCAW applications from shipbuilding and construction to water tank repair, structural erection and field repairs.

• Structural steel erection: self-shielded E71T-8 on DCEN, because the site has no gas and the wire handles out-of-position work.
• Water tanks and shop fabrication: gas-shielded E71T-1 on DCEP, for penetration and a flat bead matching a qualified procedure.
• Shipbuilding: both, depending on shop versus yard, with position changes driving wire and polarity changes.
• Pipe and heavy equipment repair: self-shielded, for the same reasons as field erection.

Polarity is a consequence of the job. Match the wire to the metal and position first, set the polarity the spool calls for, then verify with a scrap bead.

Codes, Certification, Safety and Your Polarity Checklist

Polarity is not a preference like travel speed or bead placement. On coded work it is a qualified parameter, and flipping it on the fly without paperwork puts you outside the procedure you were hired to weld to. Here is the compliance and safety picture, plus a two-minute checklist.

Why Polarity Lives Inside the Qualified Procedure

The American Welding Society has authored more than 350 standards covering welding practices, procedures, inspection and testing. For structural flux core work in the United States, the code of record is AWS D1.1/D1.1M, listed as AWS D1.1/D1.1M:2025-AMD1, with the seismic supplement for demand-critical welds in AWS D1.8/D1.8M. E71T-8 class self-shielded wire is commonly approved for those seismic applications because of its toughness, another reason the T-8 and T-11 labels matter.

A WPS is qualified on a specific combination of polarity, wire, gas, voltage, amperage and travel speed. Switch from DCEP to DCEN and you have changed a qualified variable: heat distribution, penetration and bead profile all shift, so the qualification no longer covers the weld. Your quality group has to re-qualify the procedure, and you on it, before the work counts. Treat polarity as part of the work package, not a shop-floor setting.

What OSHA Actually Requires

The federal welding rules live in 29 CFR 1910.252. The mistake that gets you in trouble is rarely the welding itself. It is the paperwork around it.

• Face protection. Section 1910.252(b)(2)(i)(A) requires helmets or hand shields during all arc welding operations, and the shade guide lists shade 12 for gas-shielded arc welding of ferrous material.
• Ventilation. Local exhaust and general ventilation must keep fume concentrations below the maximum allowable concentration. Fume-heavy self-shielded wire in a poorly ventilated bay is a real exposure problem.
• Hazard communication. Section 1910.252(c)(1)(iv) requires hazardous materials in fluxes, coatings, coverings and filler metals introduced by the employer to be included in the Hazard Communication program, so safety data sheets and labels are on hand for your wire.
• Fire watch. Section 1910.252(a)(2)(iii) requires a fire watch where appreciable combustible material is within 35 feet of the point of operation, kept up half an hour after hot work ends.
• Confined space. Under 1910.253, cylinders and the welder’s power source stay outside the space, with an attendant and rescue provisions in place.

Carbon arc gouging and air arc work follow the same DC logic. Read up on air arc welding and carbon arc gouging if gouging is part of cleanup, and on why SMAW still wins in shops where stick and gouging are familiar ground.

Fume and Manganese: The Number to Remember

The NIOSH fume and manganese hazard page is blunt about the numbers. Welding fumes are composed of metals, and most contain manganese. The OSHA permissible exposure limit is 5 mg/m³ as a ceiling, NIOSH recommends a REL of 1 mg/m³ as an 8-hour TWA and 3 mg/m³ as a short-term limit, the IDLH value is 500 mg/m³, and the ACGIH TLV is 0.02 mg/m³ respirable TWA. NIOSH also notes that confined space welding can significantly increase manganese fume exposure.

That ties back to wire choice. Self-shielded flux core makes more fume and more spatter than a gas-shielded run, and spatter is cleanup cost as much as fume source. Wrong polarity adds more of both, so polarity is a ventilation and housekeeping issue before it is a bead shape issue.

Before You Strike an Arc

1. Wire class. Read the AWS classification off the spool label: E71T-11, E71T-8 or E71T-1.
2. Shielding type. Self-shielded (FCAW-S) or gas-shielded (FCAW-G)? This sets the default polarity.
3. Polarity. Self-shielded runs DCEN (DC-); most gas-shielded carbon steel flux core runs DCEP (DC+). Confirm on the data sheet.
4. Power source. Constant voltage for flux core. In constant current stick mode the arc behaves nothing like the wire expects.
5. Shielding gas. 100% CO2 or an argon/CO2 mix for FCAW-G, matched to the wire. Check flow and cylinder condition.
6. Stick-out. Per the data sheet: roughly 1/2 to 3/4 inch for common T-11 work, 1 to 1-1/4 inch for T-8 all-position work.
7. Drag angle. A 10 to 20 degree pull keeps the slag behind the puddle.
8. Scrap test. Bead on scrap of the same thickness. Listen to the arc, check bead width, profile, penetration and spatter.

Conclusion

Two rules carry the whole topic. Self-shielded flux core runs DCEN, and most gas-shielded carbon steel flux core runs DCEP, with the spool data sheet as the final word. Polarity changes how heat splits between wire and workpiece, so it changes penetration, bead shape, spatter and the sound of the arc. Get it right before you start, document it when the code requires it, and respect the fume, fire watch and confined space rules. Ten minutes of setup saves a shift of guessing.

Frequently Asked Questions

What polarity should I use for flux core welding?

For self-shielded flux core (FCAW-S, such as E71T-11 and E71T-8), run DCEN — direct current electrode negative, also called straight polarity. For most gas-shielded carbon steel flux core (FCAW-G, such as E71T-1), run DCEP — reverse polarity — because it delivers the penetration that process is chosen for. Always confirm with the manufacturer’s data sheet on the spool, since stainless and metal-cored wires can differ.

Is flux core welding DCEN or DCEP?

It depends entirely on the shielding type, not on the process name. Self-shielded wire is designed around DCEN operation, which is the opposite of most other wire filler metals, while gas-shielded rutile flux core is normally run DCEP. If your machine only has one polarity, match the wire you bought to the setup you have — or swap the leads, which takes a couple of minutes.

What happens if I use the wrong polarity on flux core?

Expect a spitting, unstable arc, poor penetration, a wide flat bead, excessive spatter and possibly slag inclusions. Filler metal manufacturers warn that incorrect polarity results in poor welding performance including increased spatter. The quickest test is a short bead on scrap: the arc sound, bead width and penetration tell you immediately whether the machine is set up right.

Can you weld flux core with AC?

In practice, almost never — flux core wires and power sources are specified for direct current operation, and AC gives an unstable arc with excessive spatter for this process. If a manufacturer ever lists AC for a specific wire, follow the data sheet; otherwise use DC and set the polarity the wire calls for.

Does polarity change how deep flux core welds penetrate?

Yes. With DCEP more arc heat lands on the workpiece, giving deeper penetration and a wider bead, which is why gas-shielded E71T-1 is normally run electrode positive. With DCEN more heat concentrates in the wire, producing a narrower, more controllable bead and less spatter — the behaviour self-shielded wires are built around.

Why does my self-shielded flux core wire spatter so much?

Spatter is normal with self-shielded wire because the flux core generates its own shielding gases and slag, but a machine still set to electrode positive makes it dramatically worse. Check polarity first, then check voltage, wire feed speed, stick-out and drag angle, and confirm the wire is dry and the drive rolls are V-knurled. A large fume and spatter load also means you need real ventilation.

Can I use the same welder for MIG and flux core?

Yes, but only if the machine can reverse polarity — MIG solid wire is normally DCEN and most gas-shielded flux core is DCEP, so the leads or a polarity switch must change between processes. Multi-process machines handle this with a switch or with internal polarity-reversing systems that choose DCEN or DCEP from the feeder or remote.

Is DCEN always better than DCEP for flux core?

No. DCEN is the correct choice for self-shielded wires and gives less spatter and a narrower bead; DCEP is the normal choice for gas-shielded rutile flux core because it produces the deeper penetration that process is used for. The wire’s classification and data sheet decide, not personal preference — and code-governed work must match the qualified procedure.

What polarity do I use for stainless flux core wire?

Read that wire’s data sheet rather than generalising from carbon steel. Stainless flux-cored wires are gas-shielded and carry their own polarity and shielding gas requirements, and lower heat input on stainless is usually desirable, so match polarity and parameters to the manufacturer’s chart for the specific alloy.

Do I have to requalify if I change polarity on a coded job?

Yes, in most cases. A welding procedure specification is qualified with specific polarity settings, so changing from DCEP to DCEN means the qualified procedure no longer covers the work and the welder needs requalification for the new settings. Always check the code of record — AWS D1.1 for structural steel or D1.8 for seismic applications — and the inspector before changing polarity mid-job.