How Bright Is a Welding Arc? Measurements and Eye Safety

A fabricator lifts his auto-darkening helmet for two seconds to check the fit of a bracket, gets back to work, and three hours later wakes up at 3 a.m. with the sensation of sand in both eyes. He is not imagining it: the radiation from a welding arc is so intense that researchers calculate a safe staring time of roughly one second. So how bright is a welding arc, really, and what does that brightness do to your eyes?

How Bright Is a Welding Arc? The Short Answer With Real Numbers

Here is the honest answer: a welding arc is bright enough that looking directly into it without protection becomes dangerous in about one second. Not “a bit uncomfortable” dangerous. Genuinely damaging dangerous.

Researchers who measured gas metal arc welding on mild steel found effective blue-light radiance ranging from 5.0 to 118 W/cm²/sr as the current climbed from 100 to 350 amps, and the maximum acceptable exposure durations that follow from those readings were only 0.85 to 20 seconds (Nakashima et al., Industrial Health, 2024). Twenty seconds sounds survivable. It is not, because the arc is a moving, flickering source that raises its output every time you restart a bead, and because the damage it causes often does not show up until hours later.

The Core Measurements at a Glance

Measurement	Condition	Measured range
Effective blue-light radiance	GMAW on mild steel, 100–350 A	5.0–118 W/cm²/sr
Maximum acceptable direct viewing time	Same arcs, unshielded eyes	0.85–20 seconds
UV effective irradiance	Cast iron arc, 500 mm (20 in)	0.045–2.2 mW/cm²
Maximum allowable daily UV exposure	At 500 mm	1.4–67 seconds
Maximum allowable daily UV exposure	At 5 m (16 ft)	1400–6700 seconds

What Those Units Actually Mean

Before the jargon helps, here is the plain-language version.

W/cm²/sr means watts of radiant power leaving the arc per square centimetre of arc surface, per unit of viewing direction. It is a measure of radiance, and radiance is the number that matters for eye safety because your eye focuses incoming light onto a tiny patch of retina. Two sources can emit the same total power, but the one that concentrates its light into a narrow cone is far more dangerous to look at.

mW/cm² simply means how much ultraviolet energy lands on one square centimetre of skin or cornea. That is the number behind the painful condition welders call arc eye, which is photokeratitis, a burn to the front of the eye that is invisible until the cornea has already begun to shed.

The seconds in the table are not guesses. They are exposure limits drawn from ACGIH guidance for actinic ultraviolet and blue-light hazards, the same guidance that appears in international ICNIRP recommendations. Translate them into shop language and you get: at arm’s length from a cast iron arc, your entire daily UV budget is somewhere between one and sixty-seven seconds.

“How Bright” Has Three Different Answers

Ask a welder if the arc is bright and you will get three answers, because an arc radiates three kinds of energy at once.

• Visible light. This is the glare that whites out your vision, the blue light that carries a documented retinal hazard, and the reason your eyes keep watering long after the arc is off.
• Invisible ultraviolet. You cannot see a single photon of it, which is exactly why it injures so many people. The pain arrives hours later, when the corneal epithelium sloughs off and exposes the nerve endings underneath.
• Infrared heat. The same energy that dries your face, burns forearms, and scorches a cotton shirt also loads the cornea thermally when you stare into the source.

Think about how it reaches other people. A helper on the next joint lifts the helmet a second too late and eats the flash. A neighbour two houses over complains about the blue flicker through the window. A fitter across the bay stops work because his eyes water every time the arc strikes. That reach is the practical definition of “too bright,” and it is why a shop-wide protection plan matters more than a personal one.

Why Carbon Arc and Bare Electrodes Sit at the Extreme

Not every process produces the same glare. Carbon arc welding and air carbon arc gouging are the brightest, which is why air arc welding and carbon arc gouging demand the deepest filter shades in the trade. Electrode choice moves the number as well; alloy steel welding electrode selection changes both the arc temperature and the metal vapour feeding it.

What This Article Covers

We will walk through what researchers actually measured and how, why the arc behaves differently depending on process, current and shielding gas, what the radiation does to eyes and skin, and finally the rules that decide your shade number under OSHA 29 CFR 1910.133 and 1910.252. The short version of the safety answer: never look at an arc without a rated filter lens in front of your eyes, and treat any exposure that produces gritty, burning eyes hours later as a medical event rather than something to tough out.

What ‘Bright’ Actually Means: Radiance, Irradiance and the Blue-Light Hazard

Ask five welders how bright the arc is and you get five answers, most of them about glare rather than radiation. That is not a criticism. Everyday brightness is a vision thing: how hard the light hits your eye, how hard you squint. The people who study welding arcs ask a different question: how much energy is leaving the arc, at which wavelengths, and how long can an eye survive it? The answer comes in units that look alien on a workshop wall but make the danger precise.

Lux and candela: familiar, but not for this job

The units most people already know describe visible light as the eye perceives it.

• Candela (cd) is luminous intensity, the light a source throws in one direction. One candela is one lumen per steradian, the cone of light the source fills in your field of view.
• Lux (lx) is illuminance, the visible light landing on a square metre of surface. A well-lit office desk measures around 500 lux.

Both are weighted to the peak of human vision at 555 nm, a yellow-green wavelength. That is where they fall short. A welding arc is not a yellow-green source: much of its energy sits in the ultraviolet and deep blue, exactly where the eye has no natural defences. A lux reading flatters the arc and says nothing about invisible radiation, so the arc, the sun and shop lighting are better compared by exposure time than in lux.

Radiance, irradiance and spectral radiance

Term	Unit	What it describes
Irradiance	W/cm²	Radiation power arriving on a surface at a given point, for example 500 mm from the arc
Radiance	W/cm²/sr	Power leaving the source per unit area per unit solid angle, as the eye receives it
Spectral radiance	W/cm²/sr/nm	Radiance broken down into individual wavelength bands

Radiance is the honest unit for an arc, because it describes what the eye receives. Irradiance is what lands on your cornea at a distance, and it falls as you step back. Spectral radiance is the raw material for any wavelength-weighted analysis.

Why wavelength weighting is not optional

Raw power tells you how much energy there is, not how much damage it does. The threshold for photochemical injury depends on wavelength, so the ACGIH blue-light hazard function applies a weight to each band, rising to a peak at 435 nm and 440 nm and weighting those blue wavelengths far above green or red. Multiply spectral radiance by that weighting, integrate across the spectrum, and you arrive at effective radiance: the number that actually predicts injury. Recent spectral analysis of welding arcs is one of several studies building this kind of weighted dataset.

The ultraviolet side has its own version. UV exposure is limited to 3 mJ/cm², so tmax (seconds) = 3 mJ/cm² divided by effective UV irradiance in mW/cm², with ACGIH’s relative spectral effectiveness for UV peaking at 270 nm, the wavelength most associated with corneal injury.

How the measurements were actually made

No instrument can sit beside a live arc, so the approach in the research literature is to stand back, narrow the field of view and filter the light to a level the sensor can survive. A spectroradiometer with a 0.1 degree viewing angle was set up 2 m from the arc behind neutral density filters, and the resulting spectral radiance was corrected for the filter’s own transmittance. Five replicate measurements were averaged for each condition, which smooths out the arc’s natural flicker and wander.

The narrow field of view was deliberate. It keeps the measurement close to point-source behaviour and matches the geometry ACGIH assumes for a worker staring at the arc from 300 mm, where the source subtends about 0.011 rad. That geometry sits behind every safe-time figure in this article.

The caveat worth keeping

An arc has no sharp edge. It is a bright core wrapped in a halo, corona and spitting spatter, and every value shifts as torch angle, arc length and shielding change. Radiance figures from these studies are estimates rather than fixed constants, best read as an order of magnitude that defines the risk. Clinical work on welding-related eye injury makes the same point from the other end: the arc is a moving, changing source, and no single measurement captures it.

Why the Arc Is So Bright: Plasma Temperature, Metal Vapour and the Spectrum

A welding arc is not a flame. It is a plasma: a column of ionized gas between electrode and workpiece, heated to roughly 10,000 °C in its core and kept conductive by the applied voltage. No fuel burns in the ordinary sense. Electrical energy becomes heat and radiation, and that radiation leaves the column in every direction at once: ultraviolet, visible and infrared simultaneously. The output splits into three bands:

Band	Role in the arc	Why it matters
Ultraviolet (below 400 nm)	Emission lines from vaporised metal	Photokeratitis (arc eye); adds to blue-light hazard
Visible (400 to 700 nm)	Intense continuum plus metal lines	Blinding white glare; most blue-light retinal risk
Infrared (above 700 nm)	Continuum radiation from plasma and work	Heat load that burns skin, clothing, eye surface

Why it reads as white instead of blue or orange

The eye has no photoreceptors for ultraviolet, so the most hazardous share of the output is invisible. You see only the visible slice, and it is intense enough to drive all three cone types to their ceiling at once, so no hue survives and the source reads flat white. The glare is a sign your eye has run out of dynamic range, so the protective filter has to take over. The molten pool and spatter glow orange and yellow because the workpiece is incandescent, not the arc.

The lines come from the metal, not from the gas

Spectroradiometer measurements show the ultraviolet emission lines come from vaporised metal in the arc column, not the shielding gas. In shielded metal arc welding of cast iron, iron emissions appeared around 240 to 260 nm and again near 275 nm, nickel showed near 280 nm, and with stainless wire, nickel and chromium appeared across 290 to 315 nm. The filler metal reads straight out of the spectrum, as this review of welding arc spectral radiance and UV exposure shows.

Shielding gas is largely inert here. Argon and CO2 do not light up in the ultraviolet in any meaningful way; they cool the column, shape the arc and protect the molten metal. The UV fingerprints belong to whatever feeds the arc.

Filler metal changes the hazard, not just the mechanical properties

Because the spectral lines are elemental, filler composition rewrites the exposure profile. In the studies above, filler elements ranked iron above nickel and chromium for UV hazard. A bare steel electrode and a stainless or nickel consumable do not give the same UV signature even at identical current, even where the schedule treats them alike.

Electrical properties feed in: iron wire has lower resistance than stainless filler, so more electrode melts and vaporises per unit current, sending more metal vapour into the column. Electrode chemistry works the same way: nickel-type coating fluxes lower welding voltage, and lower voltage means a lower arc temperature, shifting more emission toward short wavelengths. That is why the E4916 iron-based electrode with a nickel-type coating flux registered the highest effective irradiance in the study. Rod choice is a radiation question too, worth noting in any all-position rod selection guide.

Stainless and low-hydrogen work differs for the same reason: running stick welding on plain carbon steel or stainless with a stainless filler changes what the arc is made of, and what it radiates.

Shielding gas changes the shape of the arc

Carbon dioxide differs from argon blends for a reason: CO2 disassociates endothermically, so breaking the molecule apart absorbs energy rather than releasing it. Near the arc axis that draw narrows the conducting column, concentrating the electrical energy into a smaller, denser, hotter core. The result is a contracted arc with higher energy density than an argon blend, visibly more intense at the torch.

Spray transfer pushes it further: droplets are torn off the wire tip and vaporised inside the column, feeding more metal vapour into the arc so brightness climbs. Argon arcs are less contracted and spread their heat wider.

The heat you feel is part of the same problem

Infrared is the other half of the arc output and the part you notice first in practice. It is why an arc burns exposed forearms, neck and face and scorches clothing and leather. The eye takes the same thermal load but with no warning: skin signals discomfort and pain, the cornea has no early alarm, and the eye focuses that radiation onto a small area of retina. The thermal and photochemical consequences are reviewed in this assessment of occupational exposure to welding arc radiation.

Put together, how bright a welding arc is depends less on a single number than on three coupled variables: how hot the plasma column gets, how much metal vapour is suspended in it, and how much energy the shielding gas lets it concentrate. Current, gas, filler and process feed all three.

The Invisible Half of the Brightness: Ultraviolet Radiation and Arc Eye

Half of what makes a welding arc dangerous never reaches your eyes as visible light. Ultraviolet radiation is invisible, and it gives nothing away until hours later, when your cornea feels sanded with glass. That is why arc eye, also called welder’s flash or UV keratoconjunctivitis, is the most common injury in the trade. It is not a burn from visible light. It is a chemical injury from the part of the spectrum that eyelids and dark glasses cannot block.

Understanding the mechanism explains almost everything about how to prevent it.

What Ultraviolet Light Does to the Cornea

Your cornea is covered by a thin layer of cells called the corneal epithelium, one of the most densely innervated tissues in the human body. It lets you feel a stray eyelash instantly. It is also why corneal damage is so agonizing.

UV is absorbed by the front of the eye and does almost no good along the way. A review of UV exposure and ocular damage notes that UV-C is absorbed by the cornea and never reaches the retina, while UV-A and UV-B are absorbed mostly by the cornea and lens, with less than 1 percent reaching the retina. Below 290 nm, the corneal epithelium absorbs nearly 100 percent of the UV that hits it.

So what does the UV do in that thin layer? It kills the cells. They break down, and repair turns into a painful cycle: dying cells shed, nerve endings underneath become exposed, and the new cells growing back are tender and hypersensitive.

The Delay That Confuses Everyone

This is the part that catches new welders out. The eye can look completely normal when the exposure happens, and symptoms typically appear 30 minutes to 12 hours after exposure. Many people finish a shift, eat dinner and think they are fine. Then discomfort builds through the night, and by early morning they cannot open their eyes properly. That delay is why welders treat arc eye as random. It is not; it tracks with how much UV the eyes received and when.

How Common It Really Is

A Japan Welding Engineering Society survey found as many as 86 percent of workers in arc welding workplaces had experienced UV keratoconjunctivitis, and 45 percent reported flare-ups recurring at least once a month. Emmett and colleagues found 92 percent of welders had suffered one or more flash burns, and 40 percent had erythema on the neck. Read together, they show a routine occupational injury across much of the workforce, not a rare mishap for a few careless people.

What the Examination Shows

The clinical picture in the EyeWiki review of photokeratitis is a patient who typically presents with:

• Punctate epithelial staining on fluorescein, the diagnostic giveaway
• Conjunctival injection (red, irritated eyes) with relative sparing of the palpebral conjunctiva
• Pain, tearing and photophobia (light sensitivity)
• Eyelid swelling around the affected eyes
• A possible superficial facial burn around the eye area

A welder who spent four hours welding stainless with the shield down still turns up with this cluster of symptoms. Pain, tearing, light sensitivity and eyelid swelling, all appearing overnight, is close to unmistakable.

Recovery and Treatment

The corneal epithelium regenerates quickly, and healing is generally complete within 24 to 72 hours. Supportive care includes lubricating ointment, artificial tears, oral analgesics, and topical antibiotics such as erythromycin, often four times daily for two to three days, to reduce secondary infection. One instruction matters more than it looks: eye patching is not recommended, because it delays healing. Ophthalmology follow-up within one to two days is advised if the case is significant.

A case report in the Pan African Medical Journal follows a 35-year-old who looked at an arc for a few minutes. Ten hours later he woke with bilateral burning pain, tearing, photophobia and blepharospasm, acuity down to 8/10 in one eye and 9/10 in the other, and punctate erosive keratitis on examination. Within 72 hours, acuity had recovered fully to 10/10 in both eyes. That is the normal course of the disease, not a lucky escape.

Why Welders With Face Shields Still Get Hit

Most arc-welders experience UV keratoconjunctivitis despite wearing face shields. The protection is usually there; the exposure happens in the gaps.

• Striking the arc with the shield still up. The first fraction of a second of a new arc is the brightest part of the cycle, and that is exactly when the eyes are exposed.
• Flipping the helmet up between welds to check the bead, then lowering it late on the restart.
• Working next to another welder whose arc is running. Your protection depends on their arc-strike discipline, not just your own helmet.

Arc eye is a bystander injury as much as a personal one. The helper grinding next to the booth, the fitter holding a part in the open bay, the person walking past with a clipboard: all sit inside somebody else’s UV field. That is why shop-wide screening and shade selection for everyone in the area are non-negotiable. OSHA’s eye and face protection rules, set out in 29 CFR 1910.133, define the filter requirements that make that protection work. UV does not respect the arc you were paying attention to. It reaches every eye in the bay.

The Practical Takeaway

Treat UV as the part of the arc that punishes impatience rather than ignorance. Get the shield down before the arc strikes, not during. Lower it fully between welds instead of lifting it. Do not rely on your own helmet to protect you from a neighbour’s arc. And if your eyes feel gritty and painful hours after a shift, that is not fatigue and not dust. It is photokeratitis, it is common, it heals, and it is preventable.

The Deeper Risk: Blue Light, Retinal Injury and Welding Arc Maculopathy

Arc eye gets the attention, but the cornea is only the front line. Behind it sits the retina, where the visible brightness of a welding arc does its quietest and most serious damage. The light you can see is the light that reaches your retina, and part of it causes chemical injury rather than a sunburn-like reaction.

The key band is blue. Radiation between 400 and 500 nanometres is identified as the main cause of photochemical retinal injury known as photoretinopathy, according to the welding and eye-safety literature reviewed on PMC. Blue light sits at the short-wavelength end of what the eye focuses sharply, and the cornea, lens and aqueous humour filter it poorly, so blue photons arrive at the retina with their full energy intact.

Photochemical, Not Thermal

A welding arc does not normally cook the retina the way a laser or a hot metal slug would. Instead the damage is photochemical: blue light energy triggers a chemical reaction in the retinal pigment epithelium and photoreceptor layer. The result is sometimes called welding arc maculopathy.

That is why classic protection advice is not really about heat. A dark filter lens in a helmet does not simply block heat, it blocks the short wavelengths doing the chemical work. Shade number is a wavelength-selective decision as much as a brightness decision.

What the Symptoms Look Like

Retinal symptoms follow a recognisable pattern: decreased visual acuity, foggy vision and dark spots in the field of vision. They tend to appear either immediately or within a few hours of exposure. Vision usually recovers gradually, but some symptoms can persist for months.

That is the reason this section exists. A welder who notices foggy vision at the end of a shift usually assumes a tired eye, a dry shop or bad contact lenses. Sometimes it is. Sometimes it is a real photochemical injury to the macula, the light-sensitive centre of the retina that handles detailed straight-ahead vision. Ignoring a central blur or a persistent dark spot is how a recoverable injury turns into a slow, frustrating recovery.

• Decreased visual acuity: everything looks softer than it should, and small detail is hard to read.
• Foggy vision: a haze that does not clear when you blink or rub, and that lingers into the next morning.
• Dark spots in the visual field: a fixed or slowly fading blot, often noticed while reading or looking at a bright surface.

The Heat Side of the Story

Infrared radiation and heat travel with the visible light and ultraviolet, adding a thermal burden. Staring at the arc heats the surface of the cornea, and the tear film and blinking are the only defences the eye has against that load.

Skin handles heat differently. Skin has a rich sensory nerve network and blood supply that dilate and warn you long before damage spreads. The cornea has dense nerves too, which is why corneal injuries hurt so severely, but it cannot send a warning signal and it cannot move out of the way. It sits there absorbing what arrives while the pupil holds a fixed gaze on the brightest thing in the room.

The visible brightness of the arc is therefore not a comfort issue. The same intensity that makes you squint is delivering blue photons to a chemical target and infrared energy to a surface that cannot retreat.

Cumulative and Indirect Exposure

This is the part that gets missed in training. A welder standing beside another welder is still being irradiated by an arc they are not holding. The beam does not stop at the helmet of the person doing the work, and anyone in the bay, or catching a reflection off polished or wet metal, is accumulating dose.

The arc is also a moving, flickering source. It changes with every weld as the torch travels, the angle shifts and the puddle moves. There is no single stable exposure to measure, which is why the cumulative model matters more than the dramatic single-stare case: the risk is built from thousands of small, repeated exposures over weeks and years.

OSHA makes this a regulatory requirement rather than good manners. 29 CFR 1910.252 requires helmets or hand shields during all arc welding and arc cutting operations, with submerged arc welding excluded, and requires helpers and attendants to be provided proper eye protection.

Shade selection, auto-darkening filters, screen and curtain layout, and shop lighting that keeps the arc from being the only bright thing in the space are covered later in this article. If you work with aluminium, our guides on aluminum MIG welding and aluminum flux core welding cover the process details, but eye protection rules do not change with the base metal.

Arc Flash Versus Arc Radiation

One terminology note, because the two get confused constantly. Arc flash is the electrical event: sudden release of energy, pressure wave, molten metal ejection, UV burn and electrical shock risk. It is an arc-blast hazard that can send flying metal and cause burns without the skin ever touching the arc. Arc radiation is the light, ultraviolet and infrared energy that leaves the arc and reaches your eyes and skin.

You can be struck by arc flash without looking at the arc, and you can suffer arc radiation without an arc flash event. Both are real, both need controls, and they are controlled differently.

How Bright Compared With What? Arc Versus Sun, Gas Flame, Sparks and Shop Lighting

Comparing a welding arc to the sun or shop lighting is where popular explanations go wrong: lux numbers get quoted for sources that cannot be measured the same way, so the comparison sounds authoritative and means nothing. Exposure time is the honest unit.

Start with the measured numbers from research on welding arc radiance and UV exposure. For the GMAW arcs tested, ACGIH-based testing puts the maximum acceptable direct-viewing time between 0.85 seconds and 20 seconds. The bare arc is bright enough that a deliberate look is dangerous almost immediately.

Now add a shade 11 filter, the darkest lens commonly used. At 300 A it cut the maximum acceptable daily blue-light exposure to just 3.6 to 4.5 hours. That is why the standards push welders toward heavier shades on high-current work. Shade 11 is not all-day comfort at 300 A; it needs screens, distance and shorter exposure windows to work with it. Remember that the difference between “my helmet is on” and “my helmet is protecting me” is filter transmittance, not whether you remembered to wear it.

The exposure ladder

All four figures come from the same measurement work, so they compare honestly:

• Direct arc viewing: 0.85 to 20 seconds maximum acceptable exposure, depending on process, current and geometry.
• Shade 11 filter at 300 A: 3.6 to 4.5 hours of total daily acceptable exposure.
• UV at 500 mm (about 20 inches): only 1.4 to 67 seconds of allowable daily UV.
• UV at 5 m (about 16 feet): 1400 to 6700 seconds, roughly 23 minutes to 1 hour 50 minutes.

That pair matters most. Move the same arc from 500 mm to 5 m and allowable UV exposure time stretches by roughly two orders of magnitude. Distance is a real protective control, not a symbolic one.

Process by process: GMAW, SMAW and GTAW

Measured UV hazard ranks the processes clearly: GMAW produces more UV hazard than SMAW, which produces more than GTAW. That ordering matches earlier US Army measurements on mild steel, so it is not an artefact of one laboratory setup. A TIG job therefore gives you more headroom than a MIG or flux-cored job at the same current, and anyone using one fixed shade for every process ends up under-protected on the MIG and over-protected on the TIG.

Gas and current make a measurable difference

Two multipliers matter more than most people expect:

• Shielding gas: pure CO2 produced effective radiance roughly 1.2 to 3.0 times that of an 80 percent argon / 20 percent CO2 blend at 300 A and below. The reason is arc contraction: CO2 dissociates and cools the outer column, squeezing the arc into a denser, hotter core.
• Pulsed current: pulsed welding current produced effective radiance roughly 1.3 to 3.8 times that of steady current below 300 A. Pulse machines are often assumed safer because average current is lower, but peak arcs are brighter. Check that against the shade you selected on thin-gauge work.

Arc welding versus gas welding and brazing

OSHA treats these as separate categories for a reason. The shade guide in 1910.252 gives gas welding and oxygen cutting much lower shade numbers than arc welding, because an oxy-fuel flame puts out far less radiant load. A gas welding process does not deliver the UV punch that GMAW does, so moving between a MIG bay and an oxy-fuel station in one shift means changing shade with the process.

For context, why SMAW still wins in so many shops and TIG work in aerospace both involve a dimmer arc, part of why low-amp GTAW jobs are often called the friendliest welding task on eye protection.

The comparison table

Condition	Relative brightness or UV hazard	Practical read
Direct arc viewing	0.85 to 20 s max exposure	A look is a dose, not a glance
Shade 11 at 300 A	3.6 to 4.5 h daily	Marginal at high current
UV at 500 mm	1.4 to 67 s daily	Close quarters, worst case
UV at 5 m	1400 to 6700 s daily	Distance is a real control
GMAW / SMAW / GTAW	GMAW, then SMAW, then GTAW	Change shade with process
CO2 vs 80/20 Ar-CO2	CO2 1.2 to 3.0x brighter	Argon blends are gentler
Pulsed vs steady	Pulsed 1.3 to 3.8x brighter	Pulse is not a safer arc
Gas welding vs arc	Far lower radiant load	Match shade to the process

Why the arc can be seen from across the shop

One qualitative comparison is worth making, with the caveat that it is not a measurement: an arc flash is routinely described as comparable in severity to looking at the sun. The sun does not stare back at you and does not sit 300 mm from your face, but both deliver an intense, concentrated spectrum into a very small pupil at close range. That is also why the arc is visible from down the bay, and why the light reaches a neighbour through a window.

The takeaway is proportion, not one dramatic number. The arc sits at the top of the ladder; everything below it, from the sun to the shop fluorescent tubes, sits well below. Think in exposure times rather than brightness adjectives, and shade choice, distance and screening become concrete decisions.

Choosing the Right Shade: OSHA, ANSI and the Numbers That Govern Your Helmet

By the time you understand how bright a welding arc really is, the practical question becomes unavoidable: which shade number do you put in your helmet for the job in front of you? Get it wrong in the dark direction and you will not see the puddle. Get it wrong in the bright direction and you are burning your retina or cornea every time the arc strikes.

The good news is that you do not have to guess. Federal law hands you a table. Let us walk through it.

The OSHA Shade Number Table for Welding

29 CFR 1910.133(a)(5) requires that filter lenses have a shade number appropriate for the work being performed, and it attaches a detailed table that maps welding processes and arc currents to a specific minimum shade.

Welding process	Arc current	Minimum shade number
Shielded metal arc welding (stick)	Below 60 A	7
Shielded metal arc welding	60 to 160 A	8
Shielded metal arc welding	160 to 250 A	10
Shielded metal arc welding	250 to 550 A	11
Gas metal arc welding and flux cored arc welding	Below 60 A	7
Gas metal arc welding and flux cored arc welding	60 to 160 A	10
Gas metal arc welding and flux cored arc welding	160 to 250 A	10
Gas metal arc welding and flux cored arc welding	250 to 500 A	10
Gas tungsten arc welding (TIG)	Below 50 A	8
Gas tungsten arc welding	50 to 150 A	8
Gas tungsten arc welding	150 to 500 A	10
Air carbon arc (light)	Any	10
Air carbon arc (heavy)	Any	11
Plasma arc welding	By current	6 to 11
Plasma cutting	By current	8 to 10
Carbon arc welding	Any	14

Notice how GMAW and FCAW demand shade 10 from 60 A all the way up to 500 A. That is not a rounding error. As we covered earlier, gas-shielded arcs carry a heavy load of metal vapour, which pushes the UV output up. If you are running MIG on stainless or aluminium, the higher shade is doing real work. The aluminum MIG welding guide covers the process specifics, but the PPE side of the equation stays the same regardless of filler metal.

The Companion Guide in 1910.252

1910.252(b)(2)(ii) carries a second, older-style shade guide that many shops still use because it is organized by electrode diameter and process type rather than by current:

• Shade 10 for shielded metal arc welding with 1/16 to 5/32 inch electrodes
• Shade 12 for gas-shielded ferrous welding, and for 3/16 to 1/4 inch electrodes
• Shade 14 for 5/16 to 3/8 inch electrodes, and for carbon arc welding
• Shades 3 to 8 for gas welding and cutting ranges, where the radiant load is far lower

That gas welding range is the tell. Gas welding and brazing sit at the dim end of the scale, which is why these processes get much lighter lenses than arc welding. Carbon arc welding at shade 14 sits at the opposite extreme. Never mix these two tables up.

How to Set the Shade Without Guessing

OSHA footnotes the table with the practical rule of thumb that most shop veterans already use: start with a shade too dark to see the weld zone, then move to a lighter shade that gives you a sufficient view of the work without dropping below the minimum listed in the table.

In practice, that means this: put on your shade, look at the arc from about the distance you will actually weld, and if you cannot make out the joint, the puddle, or the filler rod tip, go one shade lighter. If you can see the arc clearly enough that you are tempted to look at it directly, go one shade darker. The target is a comfortable, detailed view of the weld zone with no sense that you want to stare at the arc itself.

What the Shade Numbers Actually Mean

Shade numbers are logarithmic, and each step is roughly a factor of ten in transmission. The Japanese industrial standard figures cited in the peer-reviewed welding eye safety literature put actual visible light transmittance behind the darker end of the scale:

• Shade 11 transmits about 0.0085 percent of visible light
• Shade 12 transmits about 0.0032 percent
• Shade 13 transmits about 0.000076 percent
• Shade 14 transmits about 0.000027 percent

Now connect that to the brightness numbers from earlier in this article. Direct viewing of an arc allows roughly 0.85 to 20 seconds. Behind a shade 11 filter, the maximum acceptable exposure at 300 A stretches out to about 3.6 to 4.5 hours per day. That is the difference shade numbers buy you, and it is why shade 11 is only marginal for high-current work where you also need to see fine detail in the puddle.

Compliance: Lenses, Markings and Helmet Coverage

Three requirements from the standard matter on inspection day as much as on the bench:

1. Radiant energy transmission testing. Filter lenses must meet the radiant energy transmission test from the consensus standards listed in 1910.133(b)(1), which include ANSI/ISEA Z87.1. A lens that is a shade 14 in name only is not a shade 14.
2. Permanent marking. Lenses must bear a permanent marking identifying the source and the shade number, so an inspector can confirm the rating at a glance.
3. Helmet coverage. 1910.252(b)(2)(ii)(B) requires helmets and hand shields to be arranged to protect the face, neck and ears from direct radiant energy from the arc. A helmet that leaves the neck exposed fails, which matters because the research literature reports erythema on the neck in a large share of welders.

Passive Versus Auto-Darkening

A passive helmet is a fixed filter. You raise it to set the torch, lower it to strike, and it stays dark for the entire weld. Auto-darkening filters use a light sensor and a solar cell to switch from a clear state to the selected shade in milliseconds, so you can set up your joint with the visor up and still be protected the moment the arc strikes.

The peer-reviewed literature on welding eye injury recommends auto-darkening helmets specifically because they remove the repeated arc-strike exposure that happens when a welder raises a passive helmet after striking. The caveat is straightforward: auto-darkening helmets cost more than passive shells and lenses, and the electronics need occasional care. For anyone running a lot of short welds, or anyone who has been burned by starting an arc with the visor up, the upgrade pays for itself.

Do Not Forget the Person Standing Next to You

Arc exposure is not limited to the person holding the torch. A second welder three feet away, a fitter waiting on a fit-up, a helper grinding a bevel, a supervisor walking through the bay, all of them are receiving stray radiation. A shade 10 welding helmet does nothing for the guy beside you.

Everyone working near the arc needs their own rated eye protection. That means safety glasses with an appropriate shade rating, or goggles, or a full helmet if the position puts them in the arc’s line of sight. Screens and curtains help, but they are a barrier, not a substitute for personal protection. The air arc welding and carbon arc gouging guide covers another high-UV process where this applies just as much.

Get the table right, check the marking on the lens, and the brightness problem stops being a guessing game.

What Actually Changes Arc Brightness: Current, Gas, Process, Metal and Distance

Arc brightness isn’t random. A handful of variables control it, and knowing how they interact can help you choose the right filter shade, work safely in a shared bay, or decide whether it’s time to use a different process altogether.

Current: The Dominant Lever

Of all of these factors, welding current has the biggest impact. In measured trials, effective radiance increased linearly with welding current across the 100 A to 350 A range in both steady and pulsed gas metal arc welding, as shown by a measured arc radiance study. In practice, doubling the amperage roughly doubles the radiant output of the arc. This is why a 30 A TIG weld feels tame compared to a 250 A structural weld, and why shade requirements jump so sharply against the OSHA filter lens table.

Shielding Gas and Arc Contraction

Shielding gas changes the shape of the arc column. A tighter column means a hotter, brighter arc. When using pure CO₂, dissociation constricts the arc, producing about 1.2 to 3.0 times brighter results than an 80/20 argon-CO₂ blend at currents up to 300 A. However, that advantage flips at higher amperage. At 350 A, the argon-CO₂ blend running steady current measured about 1.2 times brighter than pure CO₂, as spray transfer throws substantially more metal vapour into the column, according to a arc radiation and gas comparison. Gas and current always interact, so neither can be used alone to predict brightness.

Pulsed and Spiking Current

Pulsed MIG and TIG may be easier on thin plate, but they are not easier on the eyes. The pulse peak runs far above the machine’s average amperage, and pulsed welding has been flagged as roughly 1.3 to 3.8 times more hazardous than steady current at the same nominal currents. A modern pulse routine won’t automatically allow a lighter shade, so it should never be assumed to be safer.

Process Differences

Some processes are simply more punishing by design. Measured UV hazard ranks GMAW and FCAW above SMAW, and SMAW above GTAW. This comes down to metal vapour and current density. A 1.2 mm GMAW wire can carry roughly 10 to 20 times the current density of a 3.2 to 5.0 mm covered electrode and is fed continuously into the arc column. GTAW has a non-consumable tungsten with no continuous filler feed, so far less filler enters the arc to intensify it.

Filler and Base Metal Composition

What goes into the arc matters just as much as what drives it. Ferrous fillers produce stronger UV emission lines than nickel or chromium under identical conditions. Stainless wire YS308, for example, has more than four times the electrical resistance of mild steel wire YGW12, meaning more of it melts and more iron enters the arc column for the same heat input, a finding detailed in a filler metal and arc emission study. By contrast, nickel-based coating fluxes tend to lower welding voltage and arc temperature, producing a cooler, less intense arc than iron-based electrodes.

Distance and Reflections

Distance is the variable most often overlooked. UV effective irradiance falls by the inverse-square law, moving allowable daily exposure from about 1.4 to 67 seconds at 500 mm to 1400 to 6700 seconds at 5 m. The arc doesn’t dim, the energy hitting your eye does. Moving back even a single pace can be a major safety gain. On top of that, long joints can create reflections off weld puddles, wet plate, or polished stainless, acting as secondary light sources that can strike from behind a shield.

Protecting Everyone Around the Arc: Screens, Curtains, Booths and the Shop Environment

The welder in the bay next door is only part of the problem. Every arc throws light into the space around it, and a grinder two booths over may be catching more stray radiation than the torch holder is. Screen placement is the only thing standing between an arc and the rest of the building.

Booths and Screens: What the Standard Actually Requires

OSHA 29 CFR 1910.252(b)(2)(iii) sets out a hierarchy. Where work allows, enclose the welder in an individual booth. Where a full booth is impractical, enclose the welder with screens. Either way, adjacent workers must be protected too: by noncombustible or flameproof screens and shields, or by wearing appropriate goggles. Bystander protection is a physical barrier or rated eye protection, and the employer decides which.

Surface finish gets attention for good reason. Booths and screens should carry a low-reflectivity finish, with zinc oxide or lamp black named as examples. These absorb ultraviolet radiation instead of bouncing it back into the bay to hit eyes and skin a second time. Bare sheet metal can act as a mirror for the arc.

The Ventilation Compromise

Screens get tricky because welding needs fume extraction and screens can block it. 1910.252(c)(1)(ii) addresses this directly: screens should be arranged so no serious restriction of ventilation exists. That means mounting them about 2 feet above the floor unless the work is performed low, when the screen should extend nearer the floor, since a screen stopping short of the ground leaks light and glare under the edge.

Ventilation is a legal requirement, not a courtesy. 1910.252(c)(1)(iii) requires local exhaust or general ventilation sufficient to keep fume, gases and dust below maximum allowable concentrations, alongside the arc-flash provisions in 1910.133 and the brazing and oxygen-fuel gas rules in 1910.253.

One wrinkle: peer-reviewed arc radiation work suggests effective arc radiance can be measured as higher when local exhaust ventilation is running, because fume removal changes the arc environment itself. That is not an argument against ventilation, but for designing fume control and optical protection as one system.

Curtains and Bystander Eye Protection

Research on welding radiation reinforces the same point from another angle. Supervisors should partition welding areas with welding curtains, and non-welding workers sharing the area should wear personal optical radiation eye protectors meeting the relevant standard. Curtains are cheap and easy to move between bays, but they need the right material and colour, since a lightweight curtain that transmits visible light may not block much UV.

They must also reach the floor on low work and hang on a proper frame rather than a scaffold post, because a curtain that flutters in the draft from an open door stops being a reliable barrier.

The Fume and Manganese Dimension

NIOSH notes welding fumes are composed of metals and most contain a small percentage of manganese. The limits are tight:

• NIOSH REL: 1 mg/m³ as an 8-hour time-weighted average, STEL 3 mg/m³
• OSHA PEL: 5 mg/m³ ceiling
• ACGIH TLVs: 0.02 mg/m³ respirable and 0.1 mg/m³ inhalable

The ACGIH numbers are the most demanding, well below the OSHA ceiling. NIOSH also notes that confined space welding can significantly increase manganese fume exposure, because there is nowhere for the fume to go. A curtained booth with no extraction is a poor combination, and in a confined space it is a hazard in its own right.

Night Work, Residential Welding and Reflections

Two situations generate complaints rather than injuries, and both are managed the same way. The first is night welding in a residential area with the window open, where the neighbour sees a flash every few minutes for hours. Curtains solve the light; a fume arm near the opening solves the fume.

The second is reflection. A car body on a rack, a stainless appliance shell, a polished table or a wet floor can throw a sharp moving image of the arc across the shop. That reflected source behaves like a second arc, and because it moves it is easier to miss than the obvious one. Aiming work away from reflective surfaces and orienting booths away from polished work is a cheap fix most shops overlook.

Booth layout, screen finish, curtain position, ventilation and reflective surfaces all feed one decision, and the last line of defence is rated eye protection for everyone sharing the space, exactly what 1910.252(b)(2)(iii) asks you to provide. Layout decisions made now cost less than retrofitting screens later. For aluminium work, this practical guide to aluminum MIG welding covers shielding and joint preparation that pair well with an enclosed bay, and the AWS safety resources are a good base for a written shop plan.

Common Mistakes, Troubleshooting and the Arc Eye First Aid Checklist

With all of the science behind arc brightness covered, the reality is simple: most injuries come down to habits, not a lack of gear. Research has found that the majority of arc welders have experienced UV keratoconjunctivitis despite wearing face shields. The common thread behind these injuries is avoidable, and breaking them is the single most effective way to prevent arc eye from ever happening.

The Mistakes That Cause Injuries

Before looking at treatment, it’s worth knowing exactly what puts welders and bystanders at risk:

• Striking too soon: Striking the arc before the shield is fully down is the leading cause of flash burns.
• Flipping between welds: Raising the helmet up between tacks or repositioning, instead of lowering it all the way, leaves the eyes exposed for just a split second.
• Unrated safety glasses: Wearing standard safety glasses with no shade rating underneath a passive helmet does nothing to block the arc’s radiation.
• One shade for everything: Using the same shade for every process and current leaves you either overexposed or unable to see, tempting unsafe workarounds.
• Working unprotected in shared bays: Without screens or curtains, welders working beside you are just as vulnerable to stray radiation.
• Assuming a quick glance is harmless: The arc’s damage is not based on comfort. Direct viewing can be dangerous within about a second, with no initial pain to warn you.

Troubleshooting: Spot the Injury Early

Not every eye irritation from welding is the same. Use this guide to tell them apart and know when to act:

Symptom	Likely Injury	Action
Gritty, burning eyes with severe pain, appearing 30 minutes to 12 hours after exposure, along with tearing and light sensitivity.	Arc eye / Photokeratitis	Cold compresses, preservative-free artificial tears, oral analgesics, and topical antibiotics such as erythromycin as directed. Never rub the eyes. Follow EyeWiki guidance, as the epithelium heals in 24-72 hours.
Dark spots, central blurring, foggy vision, or persistent haze.	Photochemical retinal injury (welding arc maculopathy)	Seek urgent ophthalmology. Recovery can take months, and symptoms may linger without follow-up.
Sharp, localized pain, watering, or the feeling of something stuck after hot work.	Foreign body or thermal burn	Do not attempt to remove embedded debris. Get a professional assessment immediately.
Redness or sunburn-like irritation on the neck, ears, or forearms.	UV skin exposure	Cover exposed skin, use appropriate flame-resistant clothing, and avoid further unprotected exposure.

Checklist For Supervisors

For any shop, prevention needs to be enforced just as much as PPE is issued:

• Shade verification: Confirm the correct shade for the actual process and current before work begins.
• Barriers: Have welding screens or booths in place and positioned to block stray radiation, especially in shared bays.
• Bystander protection: Ensure any non-welders nearby have rated eye protection for the arc they’re exposed to.
• Ventilation: Have local exhaust ventilation in place, following proper placement without unnecessarily blocking screens.
• Safety readiness: Keep an eyewash station nearby and clearly visible.
• Training: Train every welder on proper arc-strike procedure, with an emphasis on keeping the helmet down before striking.

Building a Personal Habit Set

The best first aid is never needing it. Start with three small habits that will stick for a lifetime. Keep your helmet down before striking the arc, every time. For thin-gauge work, utilize an auto-darkening helmet to eliminate the temptation of flipping up between quick tacks. Lastly, keep a laser-line or shade-reference card on hand to quickly dial in the correct shade for your process. These habits matter whether you’re doing heavy fabrication or more specialized work, such as aluminum MIG welding, where precise control is already a priority.

Stay diligent with these practices, and the arc will remain a tool to shape metal, not an injury waiting to happen.

Frequently Asked Questions

How bright is a welding arc compared to the sun?

Both deliver enough photochemical energy to injure the retina, which is why arc viewing is treated with the same seriousness as staring at the sun. Rather than rely on lux comparisons, experts measure the arc by exposure: measured arcs allowed only about 0.85 to 20 seconds of direct viewing before the blue-light exposure limit was reached, and 1.4 to 67 seconds of UV exposure at 500 mm.

How long can you look at a welding arc safely?

Without shade protection, research suggests roughly one second is all you should risk. With a correct filter the picture changes completely: a shade 11 filter still allowed only about 3.6 to 4.5 hours of daily blue-light exposure at 300 A, which is why high-current work calls for heavier shades such as 12, 13 or 14.

What shade number should I use for my welding?

Follow the OSHA 1910.133 table that matches your process and current, such as shade 10 for gas metal arc welding from 60 A up to 500 A, or shade 11 for shielded metal arc welding above 250 A. If a shade is too dark to see the weld zone, step one lighter, and never go below the listed minimum.

Can I get arc eye from a quick glance at the arc?

Yes. Photokeratitis symptoms typically appear 30 minutes to 12 hours after exposure, and the damage comes from cumulative UV, so a series of brief glances can add up. Studies also show most welders suffer arc eye despite face shields, usually because they struck the arc before pulling the shield down.

Is an auto-darkening welding helmet safer than a traditional one?

For arc-strike exposure, yes, because the filter darkens the instant the arc starts and the helmet can stay on the head at all times. Researchers specifically recommend them for this reason. The trade-off is cost, plus the need to keep the sensor clean and the battery healthy.

Why is my welding arc brighter with CO2 than with argon blend?

Pure CO2 contracts the arc because the gas separates endothermically into CO and oxygen, raising the energy density. Measured effective radiance was roughly 1.2 to 3.0 times higher than an 80/20 argon-CO2 blend at currents of 300 A and below, though the ordering flips at very high current in spray-transfer conditions.

Do people working near the arc need eye protection too?

Yes. OSHA requires adjacent workers to be protected by noncombustible or flameproof screens and shields, or by appropriate goggles, and research shows neighbouring workers suffer UV keratoconjunctivitis from someone else’s welding. Welding curtains that partition the work area are the practical control.

What should I do if I get arc eye?

Do not rub your eyes. Use cold compresses, preservative-free lubricating drops or ointment, and an oral pain reliever, and get an ophthalmology appointment within a day or two if symptoms are severe. The corneal epithelium usually heals within 24 to 72 hours, but persistent dark spots or central blurring in your vision needs urgent review because retinal photochemical injury can take months to resolve.

Does pulsed MIG welding make the arc more dangerous?

Yes, and this surprises people. Measured effective radiance with pulsed current was about 1.3 to 3.8 times higher than with steady current at the same amperage below 300 A, with the gap widest at lower currents. Thin-gauge automotive and sheet-metal work is where pulsed welding is most common.

Is a welding helmet shade number the same as sunglasses darkness?

No. Shade numbers in welding filters follow the ANSI/ISEA Z87.1 radiant energy transmission testing referenced by OSHA, and they block ultraviolet and infrared as well as visible light. Sunglasses are not rated for arc radiation, and OSHA requires proper eye protection for anyone exposed to arc light.