The cross-section is the whole calculation
Weld cost estimating has a reputation for being a dark art and it is not. Every figure comes from one number: the cross-sectional area of the weld. Multiply it by the length to get a volume, multiply by density to get the deposited weight, and every other line on the estimate is that weight divided by a rate or multiplied by a price.
For a fillet weld with equal legs and a flat face, the cross-section is a right triangle: leg squared over two. A quarter inch fillet is 0.0625 over 2, which is 0.03125 square inches. Twelve inches of it is 0.375 cubic inches, and at 0.2836 pounds per cubic inch that is 0.106 pounds of deposited steel. That is the entire fillet weld calculation. Real beads are usually a little convex, which is why there is a convexity field, but it defaults to zero so the number on screen is the one you can check on the back of a drawing.
A single-V groove has three pieces. The vee itself is a triangle whose depth is the plate thickness minus the root face and whose half-width is that depth times the tangent of half the included angle, giving depth squared times tan(angle/2). The root opening contributes a rectangle of the gap times the full thickness. The reinforcement cap on top is approximated here as the top width times the crown height times two thirds, which treats it as a shallow parabolic segment rather than a rectangle. That last term is an approximation and is stated as one.
Deposition efficiency: the number that moves the answer
Deposited metal is not purchased metal. Some of what comes off the spool leaves as spatter, some burns off, some ends up as slag, and with stick electrodes a substantial stub goes in the bin. Deposition efficiency is the ratio between them, and it is process-dependent and operator-dependent in roughly equal measure.
| Process | Deposition efficiency | Why |
|---|---|---|
| MIG (GMAW), solid wire | 90-95% | Spatter and a little burn-off. Spray transfer runs at the top of the band, short circuit at the bottom. |
| Flux-cored, gas shielded | 80-85% | The flux core becomes slag rather than weld metal |
| Flux-cored, self shielded | 78-82% | More flux, more slag, more spatter |
| TIG (GTAW) | 85-95% | Almost no spatter, but every rod leaves a stub |
| Stick (SMAW) | 55-65% | Coating becomes slag, and about two inches of every electrode is thrown away |
These are working bands, not constants. A tidy TIG welder who runs rods down to an inch will beat a careless one by five points. A MIG machine set badly will spatter away a tenth of the wire onto the bench. If you buy filler by the drum and weigh what you deposit, you can measure your own figure over a month, and it will be more useful than any table.
Where the arc time and the labor come from
Arc time is deposited weight divided by deposition rate. Deposition rate is a machine and process figure: short circuit MIG runs about 3 to 6 pounds an hour, spray transfer 8 to 12, gas-shielded flux core often more than that, stick around 3 to 5, and TIG under 2. Those rates are for the arc actually burning.
Operating factor is the fraction of paid time the arc is burning, and it is where most weld estimates go wrong. A welder working manually on varied parts, positioning, tacking, changing position, chipping slag and inspecting, is at arc for something like 20 to 30 percent of the shift. Repetitive work in a fixture might reach 40 or 50. Automated and robotic work goes much higher. If you take an arc time of six minutes and bill six minutes of labor, you have underquoted the job by a factor of three or four, and the wire price you argued about with the supplier was never going to close that gap.
Shielding gas, and why the meter lies
Gas consumption here is flow rate times arc time, which is the honest arithmetic and is also an underestimate. Real consumption includes pre-flow and post-flow on every start and stop, purge when the line has been standing, and the surge that comes out of a long hose the instant the trigger is pulled, which on a badly plumbed setup can be several times the set flow for the first second. On short welds with many starts, actual gas use can run well above the calculated figure.
Flow rate itself is worth checking. Most MIG work is adequately shielded at 25 to 35 CFH, and turning it up to 50 does not improve coverage. It creates turbulence that pulls air into the arc and makes porosity more likely while emptying the cylinder faster. To convert a cylinder price into the per-cubic-foot figure this page wants, divide the refill cost by the cylinder capacity: a 125 cubic foot bottle at $55 is $0.44 per cubic foot.
Using the number in a quote
Take the total here and treat it as the arc portion, then build the rest of the quote around it. Material preparation, cutting and bevelling, fit-up and tacking, fixture time, cleaning before and grinding after, distortion correction, inspection, consumables like contact tips and nozzles and cups, machine depreciation, and shop overhead are all real and none of them are in this figure. Across most fabrication work the arc itself accounts for well under half the cost of a welded assembly. What this page is genuinely good for is comparing two options against each other: a 3/16 fillet against a 1/4, a single-V against a double-V on thick plate, stick against flux core on the same joint. In those comparisons the parts it leaves out mostly cancel, and the parts it includes are exactly the ones that differ.
Questions people ask
How much weld metal is in a 1/4 inch fillet?
About 0.106 pounds per foot in steel. The cross-section of an equal-leg fillet with a flat face is the leg squared divided by two, so 0.25 squared over 2 is 0.03125 square inches. Twelve inches of that is 0.375 cubic inches, times 0.2836 pounds per cubic inch is 0.1064 pounds. A useful consequence of the squared term: doubling the leg size quadruples the metal. A 1/2 inch fillet is not twice the work of a 1/4, it is four times, which is the strongest argument there is for not oversizing fillet welds on a drawing.
Why is the filler I buy more than the metal I deposit?
Spatter, slag, burn-off and stubs. With solid wire MIG you lose 5 to 10 percent, mostly as spatter stuck to the work and the nozzle. With flux-cored wire the flux itself is 15 to 20 percent of the wire by weight and it ends up as slag on the floor. With stick electrodes the coating becomes slag and the last two inches of every rod goes in the bin, which is why 60 percent is a realistic figure and why stick looks cheap per pound of electrode and is not cheap per pound of weld. TIG loses least at the arc and most at the end of the rod.
What operating factor should I use?
For manual welding on varied work, 20 to 30 percent. For repetitive work in a fixture with parts staged, 35 to 50. For mechanised or robotic welding, 60 to 90. If you have never measured yours, start at 25 and check it: pick a typical job, note the total hours booked to it, and compare against the arc time this page predicts for the welds on it. Most people are surprised how low the real number is, and most underquoting in small fabrication shops traces directly back to assuming it is higher than it is.
Does the calculation handle multi-pass welds?
The volume, yes. A joint that takes four passes contains the same deposited metal as the same joint hypothetically done in one, so the weight and the filler figures are right. What it does not capture is the time cost of the extra passes: stopping, cleaning between passes, restarting, checking interpass temperature, and repositioning. Deposition rate stays roughly constant while the arc burns, but the operating factor drops noticeably on multi-pass work. If you are estimating a heavy groove weld, drop the operating factor rather than adjusting anything else.
Is this accurate enough to quote from?
For the arc portion, yes, within the accuracy of your deposition efficiency and operating factor figures, which are the two soft numbers in the whole calculation. What it is not is a quote for a welded assembly, because cutting, prep, fit-up, fixturing, cleaning, grinding, inspection, consumables and overhead are all excluded and together they usually exceed the arc cost. Use it to price the welding, use it to compare joint designs against each other, and build the rest of the estimate separately.