GreatLight CNC Machining Factory logo
CNC Machining
Rapid Prototyping
Materials
Industries
News
About GL

Get Instant Quote

CNC Knowledge

Technical Advice to Reduce the Time of Adjustment of the Flexion Machine

Setup time on a press brake is not bending time. This page explains where the minutes actually go during adjustment of the flexion machine, which parameters are worth locking down, and when a part should move to a mill instead. Written for process engineers and shop leads who own the setup sheet.

±0.005 mm CNC tolerance12-hour DFM feedbackNo minimum order quantityISO 9001:2015
Die angle setup used to shorten the adjustment of the flexion machine
Where the minutes go

Why adjustment of the flexion machine eats hours

A flexion machine bends sheet by pushing it into a V die with a punch. The angle you get depends on the punch radius, the die opening, the material thickness, the grain direction and how far the ram travels. Change any one of those and the angle moves. Setup is the act of finding the ram depth that produces the target angle, then proving it holds across the batch.

On a typical 1-3 mm mild steel job, the first part is rarely correct. The operator runs a test strip, measures the included angle, adjusts the depth, and repeats. Each cycle costs 1-3 minutes including measurement. Six cycles is 15 minutes gone before the first good part. On stainless or high-strength steel the springback is larger, so cycles go up.

The cost is not only the operator's time. The machine sits idle, the batch waits, and the die stays clamped in a setup that may not be right. In a shop running 8 to 12 setups a day, cutting four minutes per setup returns almost an hour of machine time. That is the real target.

Everything below attacks one of three things: the number of test cycles, the time per cycle, or the number of setups that should never have been on the brake at all.

  • 1
    Test cycles dominateMost lost time is repeated try-measure-adjust loops, not the bending itself.
  • 2
    Springback scales with strengthHigher yield strength and thinner sheet push the required overbend up.
  • 3
    Wrong process choiceSome parts are cheaper to mill than to set up on a brake.
Foundations

Fix the reference edge and the backgauge first

Angle is only half the problem. Bend line position drives the flange length, and flange length is what the customer measures. If the backgauge stop is set from a rough sawn edge, every part inherits that error. Establish one reference edge per part number and deburr it before it ever reaches the brake.

Use the same reference edge across all bends in the part. Flipping the reference halfway through a sequence doubles the stack-up error and forces the operator to re-trim the backgauge. Mark the reference on the setup sheet with a sketch, not a sentence.

Backgauge fingers should contact the sheet over a flat area, not on a burr or a punched hole. A finger that sits in a hole shifts the stop by the hole clearance, typically 0.1-0.3 mm. That is enough to fail a ±0.2 mm flange callout.

Pre-position the backgauge for the first bend before the sheet is loaded. On most controllers this is a stored value in the program, so the operator only verifies it. Verification takes seconds. Re-entering it by hand takes minutes and invites typos.

  • 1
    One reference edgeSame edge for every bend in the part; deburr it first.
  • 2
    Flat finger contactNever let a backgauge finger sit inside a hole or slot.
  • 3
    Pre-store gauge valuesPull positions from the program, do not retype them.
Tooling

Match punch radius and die opening to the material

The die opening sets the inside bend radius for air bending. A common rule is that the minimum inside radius is about 0.16 times the die opening for mild steel, and the required tonnage per meter rises as the opening shrinks. Pick the opening from the material thickness, then check the tonnage against the press capacity.

A rough starting point: die opening of 6 to 8 times the sheet thickness for mild steel, 8 to 10 times for stainless, and 10 to 12 times for high-strength or 7075-grade aluminium. These are starting values. The operator still trims depth, but the starting point lands closer, so fewer cycles are needed.

Punch radius matters too. If the punch nose radius is smaller than the natural inside radius, the bend coins slightly and the angle becomes less sensitive to depth. That can help stability. If the punch radius is much larger, the bend is very sensitive to small depth changes and the setup becomes twitchy.

Keep the punch and die pair together. Mixing a worn punch with a new die, or the reverse, changes the effective geometry and pushes the operator back into trial and error. Tag tooling sets and store them as a unit.

  • 1
    Opening drives radiusInside radius tracks the die opening in air bending.
  • 2
    Check tonnageSmaller openings need more force per meter.
  • 3
    Keep sets togetherA worn punch against a new die resets the setup problem.
Springback

Control springback instead of chasing the angle

Springback is the elastic recovery of the sheet after the punch withdraws. It is why the finished angle opens up beyond the angle the ram reached. The amount depends on yield strength, thickness, die opening and the ratio of bend radius to thickness. It is predictable enough to plan for.

The standard fix is overbending: drive the ram deeper so the sheet springs back to the target angle. The overbend angle for mild steel is often 0.5-1.5°, for stainless 1.5-3°, and for titanium or high-strength steel 3-6°. Measure it once per material lot and store it in the program.

Bottoming or coining reduces springback because the sheet is forced against the die walls. The trade-off is much higher tonnage and die wear. For thin sheet with tight angle tolerance, bottoming is often faster overall than air bending with repeated depth trims.

Material lot matters. A new coil or plate lot with a different yield strength will shift the overbend. Record the lot number on the setup sheet so the next run knows whether the stored value still applies.

  • 1
    Measure overbend onceStore it per material and thickness, not per operator.
  • 2
    Consider bottomingHigher tonnage, but far fewer depth trims on tight angles.
  • 3
    Log the lotYield strength varies between lots of the same grade.
Sequence and handling

Order the bends to avoid re-clamping

Bend sequence decides how many times the part is turned and re-seated. A sequence that always references the same edge and works from the inside out usually needs fewer flips. A sequence that alternates edges forces the operator to reset the gauge and re-check the first flange each time.

For box or pan shapes, bend the short flanges first while the sheet is still flat and easy to hold. Long flanges last. If a flange would collide with the punch on a later bend, plan that collision into the sequence rather than discovering it on the machine.

Use the same tooling across a family of parts where possible. If three part numbers share a die opening and punch radius, one setup covers all three and only the program changes. That removes two setups from the day entirely.

For small parts, a simple fixture or a magnetic hold-down cuts handling time per bend. Handling is often 60-70% of the cycle on parts under 200 mm.

  • 1
    Inside out, one edgeFewer flips and fewer gauge resets.
  • 2
    Short flanges firstHold the flat sheet while it is still easy to handle.
  • 3
    Group by toolingOne die set can cover several part numbers.
When to stop

When a mill beats the brake on setup time

Not every part belongs on a flexion machine. If a part has one or two bends, tight tolerances in three axes, or features that need to be machined anyway, the brake setup may cost more than milling the whole part. A single 5-axis cycle can produce a bracket with angled faces and drilled holes in one setup.

The break-even usually sits around two to four bends with simple geometry, or any part with a tolerance tighter than ±0.1 mm across multiple features. Below that, brake setup time per part dominates. Above it, bending wins on cycle time.

When the part has bends plus machined faces, splitting the work between a brake and a mill adds a handling and re-datum step. Machining from solid removes that step but wastes material. For aluminium brackets the material cost is usually small compared with the labour saved.

At GreatLight we run both processes, so the choice is made on the drawing rather than on which machine is free. Prototypes and low-volume runs often go to 5-axis; higher volumes with simple bends go to the brake.

  • 1
    Two to four bendsRough break-even between brake setup and milling.
  • 2
    Mixed featuresBends plus machined faces add a re-datum step.
  • 3
    Prototype volumesMilling from solid avoids tooling and setup entirely.
Repeatability

Lock the setup so the next run is faster

The fastest setup is the one you have already done. Record the ram depth, die opening, backgauge positions, overbend angle and material lot for every part. Next time the operator loads values instead of discovering them. That alone can halve the first-run adjustment.

Take a photo of the tooling arrangement and the backgauge setup. A photo resolves ambiguity faster than a written note, especially for multi-bend parts where the sequence is not obvious from the drawing.

Monitor the first three parts of a run rather than only the first. Angle drift in the first few parts comes from die seating and thermal movement. Once it stabilizes, spot-check every 20-30 parts. This catches drift without adding inspection time to every cycle.

Inspection reports should be available on request for production runs. For a brake part, that means angle and flange length records, not just a pass stamp.

  • 1
    Store the numbersRam depth, gauge positions, overbend, lot number.
  • 2
    Photograph the setupFaster than reading a written sequence note.
  • 3
    Check parts 1-3Catch die seating drift before the batch runs.
Starting points

Die opening and depth sensitivity by material

Use these as first-try values, then trim depth on the test strip.

MaterialDie opening (× thickness)Springback tendencyTypical test cycles
Mild steel 1-3 mm6-8×Low2-3
Stainless 304 / 3168-10×High3-5
Aluminium 6061-T68-10×Medium3-4
Aluminium 50526-8×Low2-3
High-strength steel10-12×Very high4-6
Titanium Ti-6Al-4V10-12×Very high5-7

The trade-off in one line

If the part has one or two bends and loose flange tolerance, set up the brake and store the values. If it has tight multi-axis tolerance or bends mixed with machined features, mill it from solid and skip the adjustment entirely.

FAQs

Questions engineers ask about brake setup

How many test cycles should a first setup take?

On mild steel with a die opening of 6-8 times the thickness, two to three cycles is normal. Stainless and high-strength steel take three to five because springback is larger.

If you are past six cycles, the problem is usually the starting values, not the operator. Check the die opening and the stored ram depth before running more strips.

Does a bigger die opening reduce setup time?

It reduces tonnage and die wear, and it makes the angle less sensitive to small depth changes, which can cut cycles.

The cost is a larger inside radius. If the drawing calls a tight radius, you cannot open the die that far and should consider bottoming instead.

Why does the angle drift during the first ten parts?

Die and punch seating, plus thermal growth in the frame, both shift the effective depth early in a run.

Check parts one, two and three, then move to spot checks. Storing the stabilized depth for the next run removes most of this drift.

Can 6061-T6 be bent without cracking?

It bends, but the minimum inside radius is larger than for 5052. A die opening of 8-10 times the thickness is a reasonable start.

If the bend line runs across the grain, expect a larger radius. Bending parallel to the grain is safer for tight radii.

Does the reference edge really matter that much?

Yes. Flange length is measured from the reference edge, so any error there appears on every flange in the part.

Use one deburred reference edge for all bends and mark it on the setup sketch. It costs nothing and removes a whole class of rework.

When should a bent part be machined instead?

When it has two to four bends or fewer and also needs tight tolerances or machined features, milling from solid is often faster overall.

The brake still wins on simple, higher-volume brackets where the setup is amortized across many parts.

Send the drawing, get a process recommendation

We review the drawing and tell you whether the part should be bent or milled, with a quote and DFM notes inside 12 hours.

12-hour quote100% inspectionNo minimum order quantityNDA on request

Follow

More process notes from the shop floor

We publish setup notes, tooling trials and inspection data from the factory floor.

FacebookTikTokYouTubeLinkedInInstagramThreadsPinterest

Trusted by engineers and manufacturers worldwide

Tesla Ford Motor Company BYD Auto Denso Magna International Boeing Airbus Medtronic KUKA FANUC