Guide
What minimum bend radius stops sheet metal and stamped parts cracking, by material, temper and grain direction?
The minimum bend radius is the smallest inside radius a sheet can be bent to without cracking or critically weakening the metal on the outside of the bend, and it is usually written as a multiple of the sheet thickness (t), such as 1t or 3t. It is not one number per material: it rises as the temper gets harder, as the sheet gets thicker, and when the bend line runs parallel to the rolling direction instead of across it.
By Saola Manufacturing engineering team · Published · Updated
This guide is for design engineers, quality engineers and buyers who are looking at a bracket, clip or cover with a crack along the outside of a bend, or who want the next drawing to avoid one. It explains how to tell a radius problem from an edge, burr or relief problem, ranks the usual causes, gives starting R/t values from published government and university sources for aluminum, steel and stainless sheet, and ends with a drawing and RFQ checklist. The numbers here are published guidance and typical starting points, not the capability of any press shop; the mill and the die still decide the final radius for a given coil.
On this page
What does a bend crack look like, and when is it a minimum bend radius problem?
A minimum bend radius problem shows up as cracks or tears running along the outside (convex) face of the bend, often across most of its length, sometimes with a rough “orange peel” surface around them. A crack that starts at a burr or at the end of a bend line where two bends meet points first to an edge or relief problem; a crack at the edge of a narrow tab can be either, so check the edge condition before blaming the radius.
When sheet is bent, the fibers on the outside of the bend are stretched and the fibers on the inside are compressed. The tighter the radius relative to the thickness, the higher the tensile strain on the outside, and past a certain ratio the outer surface starts to crack; that ratio is what lecture notes from India’s national engineering education program define as the minimum bend radius (NPTEL, sheet metal bending). The FAA’s airframe handbook describes the same limit as the sharpest bend the sheet can take without critically weakening the metal, and warns that a radius that is too small may result in cracking (FAA-H-8083-31A).
Not every bend defect is a crack you can see at a glance. Look at the outside of the bend under a low-power magnifier and a raking light. Fine hairline cracks, a coating that has crazed only on the bend, or a matte, lumpy “orange peel” surface (a roughening that is more pronounced in coarse-grained sheet) are warning signs that the outer fibers have been strained hard. Parts at that stage may pass a visual check at the press and then crack in service under vibration, or after plating or painting makes the surface defects easier to see. Cracks that appear only after electroplating a hardened steel part have a second suspect, hydrogen embrittlement, covered in the guide on baking after zinc plating.
Where the crack sits is the most useful clue. The same NPTEL notes explain that narrow strips tend to crack at the edges, while wide sheets tend to crack at the center of the bend, because the center of a wide sheet is in a biaxial stress state that lowers the strain the metal can take before fracture; the notes put the onset of that biaxial state at a width-to-thickness ratio (w/t) of about 8. So measure the flange width as well as the thickness. A crack along the middle of a flange wider than about 8t is a classic minimum bend radius failure. A crack at the edge of a narrow tab can come from the radius too, but look first at the edge: a burr, a torn shear zone or a punched notch on the tension side will start a crack at a radius that would otherwise pass.
What causes cracking on the outside of a bend, and in what order should you check the causes?
Check the causes in this order, from the quickest to rule in or out: outside-of-bend cracks usually come from a radius too tight for the alloy and temper, a bend line running parallel to the grain, a rough or burred sheared edge on the outside of the bend, missing bend relief at corners, and material that is harder or less ductile than the drawing assumed.
- Radius too tight for the temper and thickness. The FAA advisory circular on aircraft repair states that the minimum radius is set by the composition of the metal, its temper and its thickness, that thinner sheet takes a smaller radius, and that the recommended radius increases as the material gets harder (FAA AC 43.13-1B, paragraph 4-55). A radius copied from a soft-temper drawing onto a hard-temper part is a common way to end up here.
- Bend line parallel to the rolling direction. Rolling elongates grains and inclusions along the strip, so a bend line that runs with the grain puts those stringers across the tensile direction on the outside of the bend. Both FAA documents advise laying bend lines at an angle to the grain, preferably at right angles to it.
- Edge condition. Rough edges, burrs and edges that were cold worked by shearing or punching reduce bendability and crack first, and inclusions or rolling anisotropy can make edge cracking worse, according to the NPTEL notes. A burr left on the outside of the bend is a ready-made crack starter.
- No relief where bends meet. The FAA advisory circular tells mechanics to remove burrs and drill relief holes at the ends of bend lines and at corners to prevent cracks from starting. Without relief, the corner metal is torn rather than bent.
- Material harder or less ductile than specified. A coil at the top of its strength range, a different temper from the one on the drawing, or an alloy substitution all shift the minimum radius upward. Ductility is the underlying property: the NPTEL notes relate minimum R/t to the percent reduction of area in a tensile test as Rmin/t = 50/r − 1 (an approximation for lower-ductility material), so a less ductile sheet needs a larger radius.
One less common cause is worth ruling out. A part that is bent, then re-bent or straightened during die setup or rework, has already used up part of its ductility at that bend. If only the first parts of a run cracked, ask whether they were the setup parts.
How do you diagnose a cracked bend before deciding on a fix?
Record where the crack sits, which way the bend runs relative to the grain, which side the burr is on, and what the mill certificate says about temper and properties, then compare the drawn radius with the published R/t for that material. Those five facts identify the cause in most cases without any lab work.
Start with the part, not the paperwork. Mark the crack location on a sketch: middle of the bend, sheared edge, or bend-line end. Find the rolling direction from the surface texture or the strip layout and note whether the cracked bend runs with it or across it. Check which face carries the punch burr; the companion guide on burr direction on a stamping drawing explains how to read it. Then measure the actual inside radius with radius gauges or an optical comparator, because the punch nose may not match the drawing, and measure the flange width so you know whether the bend behaves as a narrow strip or a wide sheet.
If the cause is still unclear, bend coupons cut from the same coil. Shear two strips, deburr the edges, and bend one with the bend line across the rolling direction and one along it, both at the drawing radius, then look at the convex surfaces. If the deburred coupons crack, the radius is too tight for that material and orientation; if they pass while production parts crack, the problem is in the part (edge, burr side, relief, setup or re-bending) rather than in the coil.
Next, read the material certificate. Confirm the alloy, temper or condition and the thickness, and compare the reported properties against the specification. Be careful about what a passing mill bend test proves. Under ASTM E290, the convex surface of a bend specimen is examined for cracks, but cracks within one thickness of the specimen edge are not counted as a failure, so a certificate that shows a passed bend test says little about whether a sheared edge on your part will crack.
| What you see | Most likely cause | How to confirm | First fix |
|---|---|---|---|
| Crack along the middle of the outside of the bend | Radius too tight for alloy, temper and thickness | Measured R/t below the published value for the temper on the certificate | Increase inside radius, or form in a softer temper |
| Cracks on bends parallel to the grain only | Bend line running with the rolling direction | Bends across the grain on the same part are sound | Rotate the blank in the strip layout, or enlarge that radius |
| Crack starting at a sheared edge | Burr or work-hardened edge on the tension side | Burr on the outside of the bend; crack origin at the edge | Put the burr on the inside of the bend, deburr, or improve die clearance |
| Tear at the end of a bend line or at a corner | No bend relief | Crack runs from the corner where two bends or a bend and an edge meet | Add a relief notch or hole at the bend-line end |
| Cracks on one lot, not on earlier lots | Harder coil, wrong temper or substitute alloy | Certificate or hardness check differs from earlier lots | Quarantine the lot; hold material to the specified temper |
| Cracks visible only after plating or painting | Outer fibers already cracked or near their limit at forming; on hardened steel that was electroplated, also hydrogen embrittlement | Hairlines under magnification on unfinished parts from the same lot; for plated hardened steel, check whether a post-plate bake was specified and done | Fix the radius if unfinished parts show it; otherwise review the plating and bake requirement |
What minimum bend radius should you use by material, temper and grain direction?
Use published R/t values for the exact alloy, temper and thickness as a floor, not a target, and add margin when the bend runs parallel to the grain. Annealed sheet can typically be bent to about one thickness, hard aluminum tempers need several thicknesses, and stainless is singled out by the FAA as needing a fairly large radius.
For aluminum, the most widely cited public table is Table 4-6 of the FAA advisory circular, which gives recommended radii for 90-degree bends by alloy, temper and thickness. Selected rows are reproduced below; read across a row to see the radius grow with thickness, and down a column to see it grow with temper. The circular also notes that alclad sheet may be bent over slightly smaller radii than uncoated sheet of the same temper, and that 2024 formed immediately after quenching can take appreciably smaller radii.
| Alloy and temper | 0.032 in sheet | 0.064 in sheet | 0.128 in sheet | 0.258 in sheet |
|---|---|---|---|---|
| 5052-O | 0 | 0 to 1t | 0 to 1t | 0 to 1t |
| 5052-H32 | 0 | 0.5t to 1t | 0.5t to 1.5t | 0.5t to 1.5t |
| 5052-H34 | 0 | 0.5t to 1.5t | 1.5t to 2.5t | 2t to 3t |
| 5052-H38 | 1t to 2t | 1.5t to 3t | 2t to 4t | 4t to 6t |
| 6061-O | 0 to 1t | 0 to 1t | 0 to 1t | 0 to 1t |
| 6061-T4 | 0 to 1t | 0.5t to 1.5t | 1t to 2t | 2.5t to 4t |
| 6061-T6 | 0.5t to 1.5t | 1t to 2t | 1.5t to 3t | 3t to 4t |
| 2024-T3 | 2t to 4t | 3t to 5t | 4t to 6t | 5t to 7t |
| 7075-T6 | 3t to 5t | 4t to 6t | 5t to 7t | 6t to 10t |
Steel and stainless. There is no single public table for steel sheet comparable to the FAA aluminum table, so treat the following as typical ranges and confirm against the mill or the material specification. The NPTEL notes give a minimum radius of about 0.5t for soft low-carbon steel and about 4t for hardened low-carbon steel. Formable commercial and drawing grades of cold-rolled sheet typically sit at the low end, and high-strength low-alloy grades typically move toward the high end as yield strength rises. For stainless, the FAA handbook notes in general that annealed sheet can be bent to a radius approximately equal to its thickness, and singles out stainless steel as a material that requires a fairly large bend radius. ASTM A666, the specification for annealed and cold-worked austenitic stainless sheet, strip, plate and flat bar, includes bend testing (free-bend and controlled-bend or V-block tests) among its mechanical requirements. Ask for the condition by name, read the bend requirement for that grade and condition in the full specification, and note whether it is stated as a bend radius or a bend diameter before you compare it with your drawing; a diameter is twice the radius.
Grain direction. Every published table above assumes a sensible bend orientation. When a bend must run parallel to the rolling direction, add margin to the radius, move to a softer temper, or both. On a progressive die, the strip layout fixes the grain direction for every bend on the part. For a bracket with bends at right angles to each other, either one bend runs with the grain and must be sized for that worse case, or the part is laid out at an angle in the strip so both bend lines cross the grain at an angle, which the FAA advisory circular accepts when 90 degrees is not possible, at some cost in material use.
Thickness ratio, not absolute radius. Because the limit is an R/t ratio, a radius that works on thin sheet can crack the same alloy in a thicker gauge. If a part family changes gauge, recheck every bend radius on every member of the family rather than carrying one dimension across.
How do you fix a part that cracks at the bend without a full redesign?
The cheapest fixes, in order, are a larger inside radius, a different bend orientation in the strip layout, the burr moved to the inside of the bend, a relief cut, and a softer temper formed before heat treatment. Change the alloy only when none of those fits the part.
- Open the radius. If the function allows it, raising the inside radius to the published value for the actual temper is the direct fix. Check what else moves: the bend tangent line shifts, so holes near the bend may need to move, as described in the hole-to-bend distance guide.
- Turn the blank. On a progressive die, rotating the part in the strip so the critical bend runs across the grain can solve the problem with no change to the part drawing, at the cost of a new strip layout and sometimes more scrap.
- Move the burr. Flip the blanking direction or the forming direction so the burr sits on the compression side of the bend. Industry guidance is to keep burrs off the outside of a tight bend.
- Add relief. A small notch or hole at the end of each bend line lets the metal move. The FAA advisory circular treats relief holes at bend-line ends and corners as standard crack prevention.
- Form soft, then harden. For heat-treatable aluminum, forming in O or T4 and then heat treating to T6 lets a tighter radius pass, as the T4 and T6 rows of the table show. The trade-offs are distortion from heat treatment and an extra step that has to be written into the specification.
- Change the alloy or temper. A bracket that only needs moderate strength can often move from 6061-T6 to 5052-H32, which the FAA table shows taking a much tighter radius at the same thickness; the 5052 vs 6061 guide covers what else changes.
Whatever the fix, do not accept a cracked lot because the crack is “only cosmetic”. A crack on the tension side of a bend is a stress concentration in the most highly strained part of the bracket. Disposition belongs with the design owner, and the drawing should change so the next lot is not decided by argument.
What should the drawing and RFQ say so bends do not crack?
State the material by specification with its temper or condition, the inside bend radius with a tolerance, the bend orientation relative to the rolling direction where it matters, the burr side, and the acceptance criterion for the outside of the bend. A note that only says “bend radius to suit” leaves the crack risk with whoever builds the die.
- Material and temper. For example, “Aluminum 5052-H32 sheet per ASTM B209” or “Stainless 304, annealed, per ASTM A666”. Name the temper, not only the alloy, and state whether forming happens before or after heat treatment.
- Inside radius. Dimension the inside radius on every bend, or give one general note such as “all inside bend radii 2t unless otherwise specified”, and make sure it meets the published value for the specified temper and thickness.
- Grain direction. Where a bend is close to its limit, add “bend line perpendicular to rolling direction” or show the rolling direction with an arrow on the flat pattern.
- Burr side. “Burr side to be inside of bends”, or mark the burr side on the view.
- Relief. Show bend reliefs on the flat pattern rather than leaving them to the die designer.
- Acceptance. “No cracks on outside of bends when examined at low magnification”, with the magnification stated, so receiving inspection and the supplier apply the same test.
On the RFQ, add the annual volume, whether the part is cosmetic, any finish that follows forming, and whether the radius can change if the supplier proposes it. A radius that can open by half a thickness is often the difference between a cheap die and a difficult one, and the supplier can only offer it if you say it is acceptable.
How are bend radius and cracking questions handled on a Saola order?
On a Saola order, stamped and formed parts are made at partner stamping plants we manage, and the quote names the production site and country of origin. The bend notes on your drawing, meaning radius, temper, grain direction and burr side, are what the forming plant works to, so the clearer they are, the fewer bend questions reach the receiving dock.
Blanking, piercing and forming run on progressive and compound dies at partner plants, as described on the stamping page; partner plants are not named and are not necessarily in Vietnam, which is why each quote states where the work is done. Any machined features on the same assembly are cut at our lead CNC plant, T.K Viet Nam Precision Company Limited, in operation since 2017. Plating and other finishes are applied by partner finishing plants we manage. The quote comes back with a short manufacturability note, which is the place to look for a comment on a bend that seems tight for its temper; if you are willing to accept a larger radius or a different grain orientation, say so on the RFQ, and ask for any proposed radius change to be confirmed in writing before tooling starts.
If a delivered lot shows cracking at the bends, send photos of the crack location, the lot identity and the drawing revision. We take it up with the plant that formed the parts and work through the diagnosis table above, starting with the material certificate, the measured radius, the burr side and the strip orientation. ISO 9001 certification is in progress, not held. Send the drawing with the material, temper and bend notes through the quote form.
Frequently asked
Is the minimum bend radius measured on the inside or the outside of the bend?
On the inside. Both FAA references define bend radius as the radius measured to the inside surface of the bend, and published R/t tables use that convention. The outside radius is the inside radius plus the thickness.
Can sheet metal be bent with a zero inside radius?
Some soft tempers can. The FAA table lists a zero radius for thin annealed 5052 and 6061, and the NPTEL notes describe a zero minimum radius as a sheet that can be folded on itself. Hard tempers and thicker gauges cannot, because a sharp bend puts the highest strain on the outer fibers.
Why does the same radius crack in one direction but not the other?
Because rolling makes sheet anisotropic. Grains and inclusions are stretched along the rolling direction, so a bend line parallel to it puts them across the tension on the outside of the bend. Bend across the grain where possible, or give the parallel bend a larger radius.
Does a passed mill bend test mean my part will not crack?
Not necessarily. The mill test uses a specimen, radius and orientation set by the product specification, and ASTM E290 does not count cracks close to the specimen edge. Your part may use a tighter radius, a different orientation, or a sheared edge with a burr.
Will annealing a cracked part fix it?
No. Annealing can restore ductility before forming or between forming steps, but it does not close a crack that has already opened. Scrap cracked parts and fix the cause for the next lot.
Sources
- Advisory Circular AC 43.13-1B, Acceptable Methods, Techniques, and Practices: Aircraft Inspection and Repair, paragraph 4-55 and Table 4-6, Federal Aviation Administration, 1998 (Change 1, 2001)
- Aviation Maintenance Technician Handbook — Airframe, Volume 1 (FAA-H-8083-31A), Chapter 4: layout and forming, minimum bend radius, grain of the metal, Federal Aviation Administration, 2018
- Sheet metal operations: bending and related processes (R. Chandramouli, SASTRA University), Module 7 Lecture 2, NPTEL, Joint Initiative of IITs and IISc, accessed October 2026
- ASTM E290-22, Standard Test Methods for Bend Testing of Material for Ductility (scope), ASTM International, 2022
- ASTM A666/A666M-24, Standard Specification for Annealed or Cold-Worked Austenitic Stainless Steel Sheet, Strip, Plate, and Flat Bar (scope), ASTM International, 2024
- ASTM B209/B209M-21a, Standard Specification for Aluminum and Aluminum-Alloy Sheet and Plate (scope), ASTM International, 2021