Pass Design in Roll Forming

1. Definition

Pass design (also called roll pass design in forming practice) is the engineering decision set that turns a finished cross-section drawing into a station-by-station forming plan: how many stands, how much each bend line moves per stand, in what order flanges form, where overbend and calibration occur, and how side rolls/guides support the strip. It is the process recipe behind the tooling drawings.

There is no single universal standard algorithm. Academic and industrial sources repeatedly note that flower diagrams historically grew from designer experience with material and shape—then FEA and distribution functions made the craft more systematic.

2. Pass Design vs Flower Pattern

The flower pattern (bend progression diagram) is the visual language of pass design: stacked cross-sections from flat strip to finished profile that “bloom” station by station. Pass design is the reasoning that produces that flower and the roll layouts around it.

This page focuses on the decision layer. Flower Pattern pages (when present in the wiki) focus on reading and documenting the diagram itself.

3. Design Workflow

Industry training sequences commonly follow:

1. Develop a dimensioned cross-section (radii, arc lengths, straight legs, thickness)
2. Estimate developed strip width (bend allowances / neutral-axis assumptions)
3. Produce bend progression / flower
4. Layout rolls around the flower (contours, shaft positions, clearances)
5. Add accessories: guides, side rolls, straighteners, punch dies as needed
6. Validate by experience rules, then CAE, then tryout

Skipping strip-width rigor creates chronic gap fights later. Skipping flower rigor creates chronic twist and wave.

4. How Many Stations

Station count is driven by total bending work, material grade, thickness, flange length, and whether bends can run in parallel. Vendor practice often cites rough rules such as on the order of one station per about 10–15° of total bend per bend line, with simple channels commonly in the teens of stands and complex profiles pushing into the twenties or thirties. Treat these as starting estimates, not contracts.

• More stations → gentler strain, longer mill, higher tooling cost
• Fewer stations → cheaper short mill, higher peak edge strain, more defect risk
• Asymmetric profiles may need extra stands because left and right cannot always bend equally at once

U-rail academic designs sometimes use fixed increments (e.g., many 10° flange steps plus a final springback-aware pass)—illustrating that station count and increment size are coupled decisions.

5. Bend Angle Increments

Industrial guidance often suggests smaller maximum increments for high-strength steel than for mild steel (illustrative ranges quoted in trade articles: on the order of ~20–30° per station for mild and ~10–15° for HSS on a given bend line). Exceeding sensible increments raises edge cracking risk and springback scatter; tiny increments waste capital without always improving quality.

Increments need not be constant. Designers may go aggressive early on short flanges and conservative late near finished radii, or the reverse, depending on longitudinal strain history. Always re-check against Bend Radius and Cracking pages for the R/t and elongation window.

6. Angle Distribution Methods

Classical research distributes bending so the edge path in plan view follows assumed curves (often cubic) subject to boundary conditions at entry and exit. Forming-angle methods and energy-minimization approaches (Bhattacharyya, Panton, Kiuchi, and later expert systems) estimate deformation length and station spacing from flange width, thickness, and forming angle.

Modern papers propose richer distribution functions (e.g., five-boundary-condition formulations for hat channels) and compare candidates with FEA plus experiment to pick the lowest peak longitudinal strain. The takeaway for practitioners: do not invent angles station-by-station without checking edge strain continuity.

7. Constant Radius vs Arc Length

For U-profiles and similar sections, designers choose how the curved zone evolves:

Simulation studies compare roll loads, springback, and residual stress between these families. Pick a method deliberately and stick to it across the flower; mixing philosophies mid-tooling confuses both FEA and tryout diagnosis.

8. Bend Sequence and Asymmetry

Not every bend line should move every station. Typical tactics:

• Form longer flanges first to stabilize tracking, then shorter legs
• Stage deep channels so webs stay supported
• Delay finished internal radii until edge strain is under control
• Use side rolls / cage concepts (pipe and special sections) for continuous edge support

Asymmetric C/Z and racking profiles create unbalanced flowers. Plan anti-twist stations, guide rolls, and possibly unequal left/right increments. Twisting Defect and Bow/Camber pages document what happens when sequence ignores equilibrium.

9. Overbend and Sizing Passes

Near the exit, pass design usually reserves:

• Overbend / calibration pass — form slightly past nominal so springback lands on target (see Springback Compensation)
• Sizing / finishing pass — lightly calibrate without major new plastic work
• Straightener / Turkish head — longitudinal shape, not primary section forming

Trade examples sometimes cite a few degrees of overbend then a return-to-nominal sizing concept. Exact angles belong on the setup sheet for the material family, not as a universal constant. Closed-loop systems may later vary the overbend pass dynamically; the flower must still leave that pass with enough authority.

10. Longitudinal Strain and Defects

Edge membrane strain between stands is a primary pass-design constraint. Excessive peak strain correlates with edge waviness, flare, and cracking. Insufficient total forming length (stands too few or too close for the angle jump) concentrates deformation and raises loads.

Classic deformation-length thinking: required length depends on thickness, forming angle, and flange width. Modern CAE visualizes strain bands; designers add stands or reshape the flower when peaks exceed material limits from certificates (elongation, n-value, bendability).

11. CAE Validation

Design suites (COPRA-class and peers) generate flowers and roll drawings, then FEA modules predict springback, loads, and residual stress. Pass design should iterate:

1. Draft flower from rules of thumb
2. Simulate critical coils (min/max yield, thickness)
3. Add stations or redistribute angles where strain or springback fails
4. Only then cut hardened rolls

Virtual tryout is cheaper than discovering corkscrew after carbide tooling is finished (a lesson repeatedly stressed in AHSS twist literature).

12. Design Checklist

1. CTQs and tolerances from the customer drawing are listed
2. Strip width method documented and matched to bend radii
3. Station count justified vs grade and total bend work
4. Per-station increments within agreed mild/HSS policy
5. Curvature evolution method chosen (constant ρ vs constant arc, etc.)
6. Asymmetry / anti-twist plan written
7. Overbend + sizing + straightener roles defined
8. CAE on certificate yield window, not only grade minimum
9. Tryout plan: which stands are adjustable first
10. Revision control: flower revision = roll revision = setup sheet revision

13. Stand Spacing and Mill Length

Pass design is not only angles—it is also interstand distance. Too little space for a large angle jump forces the strip into a short deformation zone and spikes longitudinal strain. Too much space without guides lets the free flange wander, especially on thin, wide legs. Shaft centers are often fixed by the purchased mill; the flower must then fit the available pitch, or the buyer must accept a longer machine when the profile demands it.

When quoting a new line, ask whether station count is driven by the profile’s strain budget or by a catalog frame length. Catalog-driven under-stationing is a common source of chronic tryout pain.

14. High-Strength and Coated Material Notes

• HSS/UHSS: prefer more stations and smaller increments; plan overbend authority early
• AHSS with low local elongation: protect finished R/t; do not dump late angle into tight radii
• Galvanized / PPGI: flower that scuffs corners will show as coating damage even when geometry passes
• Stainless: galling risk may force different roll materials and lubrication assumptions tied to the flower

Always design against the certificate window (thickness and yield scatter), not a single catalog minimum yield.

15. From Flower to Tryout

A released flower is a hypothesis. Tryout converts it into a setup sheet:

1. Thread with soft or sacrificial coil if coating/tooling risk is high
2. Establish tracking and entry guides before chasing final angles
3. Work downstream: get early stands stable, then finish radii and overbend
4. Record gap and side-roll settings that achieve drawing CTQs
5. Freeze a golden section sample and, if available, an in-line template

If tryout needs large permanent angle changes, update the flower revision—do not leave tribal knowledge only in the setup guy’s notebook.

16. Boundaries

This page covers pass-design engineering concepts. It does not sell a proprietary flower algorithm, invent machine kW or m/min ratings, or quote tooling prices/lead times. Roll Gap Adjustment covers day-to-day setup on an existing flower. Die Overview covers tooling hardware families.

17. Buyer / Engineer FAQ

Can we shorten the mill by cutting five stations?

Only if CAE and material allow higher increments. Shortening without analysis is a common root cause of edge wave and twist on HSS.

Is equal angle per station always best?

No. Equal angles are a convenient first draft. Optimized distributions often reshape early/late increments to cut peak edge strain.

Who owns pass design—machine builder or tooling house?

Clarify in the contract. Many turnkey lines include flower + rolls; many aftermarket tooling houses redesign flowers for existing shafts. Shaft spacing and stand stiffness limit what any flower can do.

How does pass design relate to in-line inspection?

Good flowers keep CTQs controllable; sensors monitor them. Bad flowers make sensors document scrap. Design first, instrument second.

Do stainless and aluminum use the same increments as mild steel?

No. Use grade-specific bendability, springback, and galling risk. Copying a mild-steel flower onto stainless is a frequent failure mode.

What is a “dummy” or idle station?

A stand with little or no intentional bend change, used for support, tracking, or future capacity. Idle stands are part of pass architecture when spacing or guidance requires them.

18. Related Encyclopedia Topics

• Flower Pattern (diagram literacy) — complement to this decision page
• Bend Radius / Cracking / Elongation — material limits on increments
• Springback Compensation / Closed-loop Control — exit strategies
• Edge Waviness / Twist / Bow / End Flare — symptoms of weak flowers
• Roll Forming CAE — validation tool
• Roll Gap Adjustment — operating the finished design

19. Summary for Specifiers

Specify pass design as an engineered flower with stated station count, increment policy, curvature method, asymmetry plan, and overbend/sizing roles—validated by CAE on the real yield window. Rules of thumb start the conversation; strain, springback, and mill length finish it. Pass design is where most dimensional quality is won or lost before the first coil hits the floor.