Almost every buying decision in welding comes down to three things you have to fix before you spend anything: the process you are actually going to run, the input power and gas you can actually supply, and the duty cycle you actually need at the amperage you actually weld at. Machines are marketed on their headline amps and their feature lists. Neither of those is the number that stops a project. The number that stops a project is the one printed on the data plate at the bottom of the duty cycle chart, and whether your shop can feed the machine the gas and power it wants.
Start with the process, not the machine
If you are welding thin stainless, aluminum, or anything that will be inspected or polished, you are buying TIG. Gas Tungsten Arc Welding uses a nonconsumable tungsten electrode and an inert shielding gas — typically argon — to protect the molten pool from the atmosphere, and that is what gives it the control and cleanliness that stick and MIG cannot match on thin, precise work. If you are building trailers, gates, fence, or repair work on mild steel from roughly an eighth of an inch up, you are buying MIG or flux-core, and a TIG machine will cost you more and slow you down. If you are cutting plate rather than joining it, none of the above applies and you are shopping plasma.
The reason to settle this first is that the accessories do not transfer. A TIG torch, a foot pedal, a gas solenoid, and a bottle of high-purity argon (welding grade is around 99.99% or better) are a separate economy from a MIG gun and a spool of wire. Buying a multiprocess machine to "cover everything" sounds like the efficient answer and often is not: the TIG side of a cheap multiprocess box is usually the weakest part of it, and you end up buying the standalone TIG anyway.
The honest cost of choosing TIG is speed. It is slower per inch on thick material, it demands clean metal, and it demands that you learn to feed filler with one hand while the other holds the torch. If your work is structural mild steel and nobody is going to look at the bead, TIG is the wrong purchase no matter how good the machine is.
Input power decides more purchases than the welder does
Check the outlet before you check the spec sheet. A machine that wants a dedicated 240-volt circuit is not a machine you can run off a garage outlet with an extension cord, and the cost of adding the circuit — an electrician, a breaker, the run of wire — can rival the price of the welder itself. Dual-voltage machines that run on 120 and 240 are popular for exactly this reason, but understand what you are buying: on 120 volts the machine is a lower-output tool with a lower duty cycle, and the 240-volt rating on the box is the one the marketing leans on.
Plasma cutters have their own power story. A common DC voltage range required to start and maintain a plasma cutting arc is 200–400 volts DC, which tells you the internal supply is doing real work, but the number that constrains you in the shop is the air. A 40 A air plasma cutter wants a compressed air supply at 80–120 PSI and roughly 5.3 cfm at that pressure. A small homeowner compressor that cycles constantly will starve the cut, and you will see it as a ragged kerf, dross on the bottom edge, and a cutter that seems to quit mid-cut. Size the compressor for continuous flow at the cutter's rated cfm, not for its tank size.
For TIG, gas supply is the other half of the input question. Argon flow for TIG commonly runs 10–35 cfh depending on cup size, joint configuration, and whether you are welding outdoors where the breeze eats your coverage. Some power sources specify tighter windows — a recommended argon range of 10–15 L/min on a given machine — and you should follow the manufacturer's number when they give one, because too much flow causes turbulence that pulls air into the plume just as surely as too little flow leaves the pool exposed. Either way, you need a regulator and flowmeter, and you need a gas supplier within reasonable driving distance, because a bottle that runs dry on a Saturday ends the weekend.
The trade-off here is real: a bigger bottle costs more up front and less per cubic foot over time, and a small bottle is easier to haul. Buy the largest bottle you can physically move and afford if you weld regularly; buy small if you weld rarely and hate rental fees.
Duty cycle is the number that actually limits you
Duty cycle is the percentage of a ten-minute window a machine can run at a given amperage before it needs to cool. It is always quoted at a specific amperage, and the amperage matters more than the percentage. A machine rated at a high duty cycle at low amps and a low duty cycle at its maximum is normal; the spec sheet is telling you where the machine is comfortable. Read the whole chart, not the headline.
What this costs you in practice: if you are welding long beads on thick aluminum, you will hit the thermal cutout and stand there waiting. That is not a defect, it is the machine telling you it is the wrong class of tool for the job. The fix is either a bigger machine with a higher duty cycle at your working amperage or a work rhythm that lets the machine cool between passes. Buying a small machine and planning to "just take breaks" works fine for hobby work and falls apart the moment someone is paying you by the hour.
Duty cycle is also where the honest trade-off between price and capability shows up most clearly. Two machines with the same maximum amperage can differ substantially in what they will do continuously at that amperage, and the cheaper one is cheaper for a reason that is printed on the label.
TIG features worth paying for, and the ones that are not
Arc starting is the first feature to care about. High-frequency start is convenient and lets you start without touching the work, but the high-frequency noise can interfere with nearby electronics and radios. Lift TIG starting touches the tungsten to the workpiece at low open-circuit voltage and then lifts it to strike the arc, which reduces that interference risk and is the friendlier choice in a shared space or near sensitive equipment. If you are welding in a garage with a router and a couple of screens nearby, lift start is not a gimmick.
After starting, the features that change your results are pulse, which helps control heat input on thin material and gives a stacked-dime look on aluminum; adjustable pre-flow and post-flow, so the gas is already covering the pool before the arc starts and keeps covering it while the tungsten cools; and amperage control at the torch or via a foot pedal. Foot pedal control is close to mandatory for bench TIG on thin work and close to useless when you are lying on your back under a car, where a torch-mounted thumbwheel or a fingertip control earns its price.
AC/DC is the decision most buyers get wrong in the other direction. Aluminum and magnesium need AC; steel and stainless run on DC. If aluminum is on your list at all, buy AC/DC once rather than buying DC and upgrading later. If aluminum is genuinely not on your list, a DC-only machine is cheaper, simpler, and lighter, and there is no shame in it.
What you can skip: synergic programs on a TIG machine, most digital displays, and the highest-end waveform menus. They are not harmful, but they are not where the money should go first. Spend it on duty cycle, arc start quality, and a torch you can actually hold for an hour.
Gas, consumables, and the cost that outlives the purchase
The welder is the cheapest part of the hobby over a few years. Shielding gas is a recurring bill, and the choice of gas affects both results and cost. Pure argon is the standard for TIG on steel, stainless, and aluminum. Argon-helium mixes run hotter and can help on thick aluminum, and they cost more per bottle. For MIG on mild steel, a mix of argon and CO2 is the norm and straight CO2 is cheaper and runs hotter and dirtier; the trade-off is bead appearance and spatter, and most people doing visible work pay for the mix.
Consumables are where beginners get surprised. TIG consumes tungsten (and you will dip it, and you will regrind it, and eventually you will get good at not dipping it), filler rod, collets, collet bodies, gas lenses, and cups. Buy a spare parts kit up front with the machine rather than discovering on a Sunday that your only collet is crushed. MIG consumes contact tips, nozzles, and wire, and the wire itself is the biggest ongoing cost — buy it in the largest spool your machine will hold, because small spools cost more per pound and run out at the worst time.
Plasma cutting consumables — electrodes and nozzles — wear out faster than most buyers expect, and their life depends heavily on air quality. Moisture and oil in the air supply destroy electrodes. A decent filter and a dryer on the compressor is money well spent and is routinely skipped, and the symptom is consumables that die early and cuts that wander.
What to check before you pay, and when to walk away
Confirm four things before any purchase: the duty cycle at the amperage you will actually run, the input voltage and circuit size, the gas or air supply the machine requires, and whether the consumables are commonly available locally or only from the manufacturer. That last one is the quiet killer. A machine that only takes proprietary consumables is a machine you are locked into, and if the line is discontinued you own a paperweight.
Then check the return policy and the warranty terms on the torch and the internal boards, not just the chassis. Torches and leads take the abuse, and a warranty that excludes them is worth less than it looks.
Walk away when the spec sheet will not tell you the duty cycle at a stated amperage, when the machine's rated input cannot be met by your shop without an electrician's visit you did not budget for, or when the TIG side of a multiprocess machine is described only in terms of maximum amperage with no arc-start detail. Those are not small omissions; they are the parts of the machine you will notice every time you use it.
The purchase that works is the boring one: the process that matches the work, the power and gas you can supply, a duty cycle that survives your longest bead, and consumables you can buy on a Tuesday. Everything else is preference, and preference is the part you can afford to get a little wrong.
What the gas flow number actually buys you
Set argon flow with a flowmeter at the torch, not a pressure gauge at the regulator, and set it while gas is flowing through the cup you will actually weld with. A gauge reading taken with the solenoid closed tells you bottle pressure, not coverage. The 10–35 cfh band covers most TIG work, and where you sit in it depends on cup size and joint type: a large gas lens cup on an outside corner wants more than a small cup on a flat butt weld. Too much flow is the failure beginners do not recognize, because the arc still looks fine while the turbulence siphons air into the plume and you get porosity you blame on the metal. If a machine manual hands you a tight window — 10–15 L/min on a given power source — that number wins over the general range, because the manufacturer tested that torch and cup combination.
Air supply is a plasma cutter purchase, not an accessory
Budget the compressor and dryer into the cutter's price before you compare cutter models, because a 40 A air plasma cutter needs 80–120 PSI at roughly 5.3 cfm, and a compressor rated for that cfm only in bursts will run the cutter at the edge of its cut quality. Continuous flow at the rated cfm is the spec to match, and tank size does not substitute for it. Add a filter and a dryer: moisture and oil in the line eat electrodes and nozzles, and you will notice it as consumables dying early and a kerf that wanders instead of a clean, square edge.








