Tonnage in press brake work is a force, not a capacity. What a bending chart or calculator returns is the force needed to bend a given length of a given material through a given die opening - nothing more. The machine's rating is the ceiling; the bending job sets the demand. Confusing the two is the root of most tonnage problems, because a machine well inside its rating for one tooling setup can be overloaded by the same part with a narrower die.

Press brake bending a plate with a tonnage chart and formula on the control panel for tonnage calculation
Variable | Meaning | Unit |
Tensile strength | Strength of the material being bent | N/mm2 (MPa) |
Bend length | Length actually bent, not the blank length | mm |
Thickness | Sheet thickness, used squared | mm |
V opening | Die opening width, in the denominator | mm |
Two of these deserve emphasis. The bend length is the length of material actually being bent, not the length of the blank: a part with cut-outs, or one bent in several operations, does not load the machine along its full length. The V opening sits in the denominator, so halving it doubles the force on the same job. The V openings available in a given setup are set by the press brake tooling range in use, so the calculation and the tooling choice are made together rather than in sequence.
For air bending, the force can be estimated as T = 1.42 x tensile strength x bend length x thickness squared divided by (1000 x V opening), with force in kilonewtons, tensile strength in N/mm2, and lengths and thickness in millimeters. Metric tonnage follows by dividing the result by 9.81; if short tons are used instead, divide by 8.90.
Worked example: bending 1000 mm of 3 mm mild steel with a tensile strength of 450 N/mm2 over a 24 mm V opening. The expression gives 1.42 x 450 x 1000 x 9 divided by (1000 x 24), which is 239.6 kilonewtons, or about 24 metric tons. Note where thickness sits: it is squared, so doubling thickness quadruples force. That is why a machine that handles 3 mm comfortably may not come close on 6 mm.
Two conventions around that expression are worth writing down. The 1.42 coefficient applies to air bending; bottoming and coining need substantially more force, and the multipliers quoted for them vary with tooling and material, so a part that can be air bent is usually the lighter demand on the machine. And the eight-times-thickness relationship often used to pick a V opening is a rule of thumb for air bending rather than a design rule - the tooling supplier's recommendation for a required inside radius takes precedence over it.
Published charts are built for a reference material, commonly mild steel at around 450 N/mm2, and for a specific set of tooling assumptions. Read the same case off the chart as you calculated, and the two should land close together. Where they diverge, the usual cause is that the chart is scaled for a different reference tensile strength; because the relationship is linear in tensile strength, a chart built for 450 N/mm2 can be scaled by the ratio of your material's strength to 450.
That scaling is also the most practical way to sanity-check an online calculator, which frequently hides both its reference material and its coefficient. If a calculator returns a figure well below a chart scaled to the same material, one of the two is assuming a different bending method.
The machine rating is only one of several limits in play. Tooling carries its own maximum load per meter and it is often the lower of the two: a punch and die can be overloaded well inside the machine's capacity, particularly with a narrow die. Bending off-center loads one side of the machine and the tooling unevenly, and a job split into short bends can still exceed a per-meter rating that is quoted over a full bed length.
Two habits pay for themselves. Confirm the tooling's rated load against the calculated force before the job is set, rather than after a punch has deformed. And keep the calculated figure with the setup sheet, because the next operator to run the part needs it as much as the first.
Some setups take the force up faster than the part suggests, and they are worth calculating rather than estimating. A narrow die used to produce a tight radius is the clearest case: the V opening sits in the denominator, so a die chosen for a small inside radius can demand several times the force of a general-purpose die on the same part. Short bends are another: the machine's capacity is quoted over a bed length, and a bend concentrated in the middle of a long bed loads the beam and the tooling differently from one spread across the full length.
Multi-bend parts accumulate load in a way that is easy to miss. A part bent in six operations does not load the machine with six times the force of a single bend, but each operation is a separate loading event and the tooling carries it each time. Where one tool runs a family of parts, it is the part with the shortest bend line against the narrowest die that sets the tooling requirement, not the average of the family.
Off-center bending deserves a mention of its own, because it is the case where a machine with ample tonnage still damages tooling. Loading a narrow tool at one end of the bed puts the force where the beam has least support, and the tooling sees a force per meter well above what a centered bend of the same total tonnage produces.
Unit conversion is worth stating explicitly, because bending charts travel between metric and imperial workshops. The formula returns kilonewtons; dividing by 9.81 gives metric tons and dividing by 8.90 gives short tons. A figure quoted without its unit is one of the more common causes of a machine being set up for a job it cannot do.
·Bend length taken from the drawing's overall length instead of the length actually bent.
·The squared thickness term ignored, so a small thickness change is assumed to be a small force change.
·A material factor copied from a chart built for mild steel and then applied unchanged to stainless or high-strength steel.
·The die opening changed for radius reasons without re-checking force, even though the V opening sits in the denominator.
·Air bending figures applied to a coining operation, which needs considerably more force.
·Metric tons and short tons mixed within a single calculation.
Each of these produces a plausible-looking number, which is exactly why they survive. The arithmetic is not wrong; the input is.
Once the force is known, the rest of the setup follows from it. Die opening and inside radius are chosen together, the punch tip radius follows from the radius the part needs, and the tooling is checked against its per-meter rating. This is also the point at which material choice stops being theoretical, because the tool steel grades available will limit what the setup can do; bending tooling is commonly specified from the same metallurgical blade and tooling product range as the shear and slitter blades already in the plant.
Press brake tonnage calculation is a short piece of arithmetic that answers a narrow question: how much force does this bend need? Answer it before the job is set, scale any chart you use to the material actually going into the machine, and check the result against the tooling's own rating as well as the machine's. The figure will not be exact - the formula is an estimate, and real tooling, friction and material variation all move it - but it is close enough to keep the machine and the tooling inside their limits, which is what the calculation is for.
To have a bending job checked against machine and tooling limits, send your bending parameters - material, thickness, bend length, V opening and the tooling currently in use.