TL;DR: The geometry of a crash-lock base is only as reliable as the crease rule depth and lock-panel taper angle — and both depend on board caliper, not just substrate grade.
TL;DR: A tolerance stackup across four folded panels can accumulate up to ±1.6mm of dimensional error if crease-to-cut register is not held within ±0.2mm at the die-cutting stage.
Crease Geometry as the Primary Design Constraint #
The specification parameter that drives crash-lock performance is crease channel depth relative to board caliper — not burst strength, not ECT rating, and not substrate grade in isolation. We specify crease depth at 65–70% of finished board caliper for SBS substrates in the 280–350 gsm range. Drop below 60% and the panel hinge doesn’t open cleanly under assembly force; exceed 75% and you fracture the surface fiber, which creates a stress riser that leads to crease failure after 15–20 carton open-close cycles in retail environments.
This matters because most CAD briefs we receive specify board grade and print spec in detail, then leave crease geometry to the die-cutter’s defaults. That gap is where dimensional inconsistency originates.
Two standard references govern this systematically. ISO 2759 (burst strength of board) gives you the substrate input, but it’s TAPPI T809 — the crease stiffness method — that tells you whether the crease will function as a hinge under the assembly force your auto-bottom geometry demands. We require TAPPI T809 data on all crease qualification runs for retail carton programs above 50,000 units per quarter.
For auto-bottom panels specifically, the lock-tab engagement depth is a second critical parameter. We design lock tabs at 6.5–8.0mm insertion depth for boards in the 320–400 gsm range. Below 6mm, the tab disengages under 3–4 kg vertical load — a failure mode that appears in transit, not on the assembly floor, and is nearly impossible to attribute without instrumented drop testing per ISTA 2A.
What to Request From a Tooling Supplier — and What the Response Tells You #
When a new auto-bottom or crash-lock project enters our pre-production workflow, we run what we internally call an MDR-3 die specification check before approving any rotary or flatbed tool. MDR-3 covers three parameters: crease rule height, crease channel width, and cutting clearance at panel intersections. We ask external tooling suppliers for the same data in writing before tool manufacture begins.
Ask your tooling supplier for crease rule specification in millimeters, including whether they’re using 2pt, 3pt, or 4pt matrix channel, matched to your submitted board caliper. A competent tooling house responds within 24 hours with specific values. A vague response — “standard crease for that board” — is a signal that they’re defaulting to a universal setup, which works adequately for standard SBS at 300 gsm but creates assembly problems at the extremes: thin FBB at 230–250 gsm or thick coated duplex above 420 gsm.
Also request the angular specification for the crash-lock taper. The lock panels that splay outward during assembly need to be designed at a 3°–5° inward taper from the base panel edge. Some tooling houses default to 0° (straight cut) and rely entirely on the adhesive bond to hold the base geometry. That approach is acceptable for low-weight contents up to around 400g, but for products above 800g we require the taper-cut geometry to be confirmed before first steel.
One more thing to ask: panel intersection clearance. At the corners where the base lock panels meet the side walls, we specify 0.3–0.4mm cutting clearance. Tighter than 0.3mm and fiber drag creates a notch that initiates tearing during high-speed erection on packing lines running above 30 cartons per minute.
Cost-Performance Trade-offs in Crash-Lock Base Engineering #
The main cost lever in auto-bottom and crash-lock carton tooling is flatbed versus rotary die cutting. Flatbed tooling costs roughly 3–5× more per setup than rotary tooling, but delivers ±0.2mm register accuracy versus ±0.4–0.5mm for rotary. For crash-lock geometry, that difference is material — at ±0.5mm, tolerance stackup across four lock panels can push the assembled base out of square by 0.8–1.2mm, which causes auto-erection failures on tight-tolerance filling equipment.
The counterargument: for carton sizes above 200mm in the longest panel dimension, panel-to-panel tolerance contribution decreases proportionally, and rotary die cutting is often entirely adequate. A 250mm × 180mm crash-lock base in 350 gsm SBS at 100,000 units per run — rotary is the right call on cost grounds, and the geometry risk is low.
The real cost exposure in this category isn’t tooling. It’s sample iterations caused by crease specification not being locked before first die. Each physical sample iteration at a typical development timeline adds 7–10 working days and 1–2 rounds of tooling modification cost. Locking the MDR-3 crease spec before tool manufacture cuts average sample rounds from 3.2 to 1.4 on our folding carton programs, based on internal data across 48 projects in 2023–2024.
Where we do see brands over-specify: requesting SBS 400 gsm for a crash-lock base on a product weighing under 300g. FBB at 300–320 gsm with a correctly specified crease achieves equivalent base rigidity at lower material cost and better crease recovery angle. The board choice matters less than the crease execution.
Tolerance Stackup in Multi-Panel Crash-Lock Geometry — A CAD Integration Note #
This is the section most CAD briefs skip, and skipping it is the direct cause of first-sample failures in crash-lock base geometry.
A standard crash-lock base involves four folded panels: two side lock panels, one front glue flap, and the base itself. Each panel contributes a dimensional tolerance derived from the die-cutting accuracy and the board thickness variation. If we assume:
- Die-cutting register tolerance: ±0.20mm (flatbed, per our production spec)
- Board caliper variation: ±0.05mm per layer (typical for SBS 350 gsm, per GB/T 22819 incoming inspection)
- Crease displacement under assembly force: ±0.10mm
…then a linear stackup across all four panel fold axes gives a worst-case accumulated error of ±1.40mm in the assembled base footprint dimension. Under statistical (RSS) stackup, the 3-sigma value is approximately ±0.55mm — which is the number we use when matching crash-lock carton tolerances to filling line guide-rail specifications.
Crash-Lock Base Tolerance Stackup by Die-Cut Method
| Die-Cut Method | Per-Panel Register Accuracy | 4-Panel Linear Stackup | RSS 3σ Estimate | Recommended Filling Line Clearance |
|---|---|---|---|---|
| Flatbed (precision) | ±0.20mm | ±0.80mm | ±0.35mm | +1.0mm over carton nominal |
| Flatbed (standard) | ±0.30mm | ±1.20mm | ±0.52mm | +1.5mm over carton nominal |
| Rotary | ±0.45mm | ±1.80mm | ±0.78mm | +2.0mm over carton nominal |
These values are computed from our 2023–2024 dimensional audit data across 31 crash-lock programs. They assume consistent board caliper within ±0.05mm — which requires incoming caliper testing per TAPPI T411 at a minimum 5% lot sampling rate.
One open question we’re still tracking: thermal expansion behavior in crash-lock adhesive bonds when cartons are stored at 35–40°C before filling (common in Southeast Asian distribution). Our current data suggests bond shear strength drops approximately 12–18% in that temperature range for standard EVA hot-melt, but our dataset only covers three adhesive grades from two suppliers. We’ll have a cleaner picture after the Q3 2025 thermal conditioning program we’re running on 8 adhesive variants.
Specification Notes for Brand Partners #
When you brief us on an auto-bottom or crash-lock carton project, the four inputs that determine whether we can quote accurately on first submission are: finished carton dimensions (L×W×D in millimeters, all internal), product weight and any fragile content that sets a minimum base strength, substrate preference or SDS/regulatory constraint if applicable, and target erection method — hand assembly or automated filling line.
The brief gap that causes the most sample iterations is filling line tolerance. Brand partners frequently provide carton dimensions without specifying the guide-rail clearance their filling equipment requires. When that’s missing, we design to our standard +1.5mm clearance, which is conservative for precision flatbed work but may be too tight or too loose for the client’s actual line setup. A single email from your operations team with the filling machine model or the guide-rail opening dimension eliminates this iteration entirely.
Our standard sampling timeline for crash-lock carton development is 18–22 working days from approved structural brief to first physical sample. Print-ready artwork adds 5–7 working days if supplied concurrently. The main variable that extends this timeline is tooling queue depth — during peak Q4 periods, flatbed tool manufacture alone can add 5 working days. For time-sensitive launches, briefing the structural spec 3–4 weeks before artwork finalisation gives us runway to start tooling manufacture in parallel.
What board caliper range is appropriate for a 500g product in a crash-lock base carton?
For a 500g product, we’d typically specify SBS or FBB in the 350–400 gsm range, which delivers a finished caliper of approximately 0.42–0.48mm. The more important variable is lock tab insertion depth — we design that at 7.0–7.5mm for this weight class to ensure the base holds under the 4–5 kg dynamic load typical in secondary packaging and transit conditions.
Does crash-lock geometry work reliably on automated filling lines running above 40 cartons per minute?
It depends on the erection mechanism. Vacuum-suction erectors handle crash-lock geometry reliably up to 60 cartons per minute if the carton’s glue bond shear strength is above 180 N/m and panel intersection clearance is correctly specified. Push-plate erectors are more sensitive to base geometry variance and generally require flatbed die-cutting at ±0.2mm register to avoid jamming at speeds above 35 cpm.
Can auto-bottom and crash-lock be combined in the same carton structure?
Yes, and we produce this configuration regularly — auto-bottom base with a crash-lock reinforced rear panel. The CAD constraint is that the auto-bottom fold sequence must be validated before the crash-lock panel geometry is finalised, because the auto-bottom fold creates a small panel overlap that shifts the effective carton footprint by 0.5–1.0mm depending on board caliper. If the crash-lock taper is dimensioned off the nominal footprint without accounting for this shift, the assembled base can be slightly non-rectangular.
What’s the minimum order quantity for a custom crash-lock carton with bespoke die?
For flatbed-tooled crash-lock cartons with custom dimensions, our practical MOQ is 5,000 units per SKU. Below that threshold, tooling amortisation makes unit cost unworkable for most brand programs. For standard footprint sizes where existing tooling applies, we’ve run programs at 2,000 units, though substrate and print setup cost makes the per-unit economics less favourable.
How does humidity affect crash-lock base performance in export shipping?
Meaningfully. SBS board at 350 gsm loses approximately 15–20% of its short-span compressive strength (SCT) when conditioned at 85% relative humidity for 24 hours, per our internal conditioning data aligned with ISO 7263. For export programs targeting Southeast Asia or coastal US markets, we recommend either a moisture-barrier coating on the base panels or a minimum board upgrade to 400 gsm to maintain adequate base rigidity through the humidity exposure window in transit.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
Ran into exactly this on a 280gsm SBS program with a Shenzhen converter last year — their default crease depth was sitting at 58% of caliper across the board, which their die shop had been running for years on straight-tuck cartons without issue. Took TAPPI T809 data from two qualification runs to actually show them where the hinge failure was coming from. Once we got them to 67% the auto-bottom assembly force dropped noticeably and the lock-tab engagement held consistently through 4kg drop simulation, but getting that change signed off through their tooling department added six weeks to the program.
Curious whether the 65–70% crease depth spec holds when you’re running SBS at the lower end of that 280gsm range on a rotary die — we’ve had hinge opening issues that didn’t show up on the TAPPI T809 qualification sheets but started failing around the 8,000–10,000 carton mark on a high-speed filling line.
The 65–70% crease depth spec works for SBS on a flatbed, but we’ve had to push to 72–74% on our 300gsm folding boxboard when running through a rotary die on the Bobst 1628 — the anvil pressure isn’t as consistent and you lose effective depth off the crease rule faster than the spec assumes. Never hit fiber fracture at 74% on FBB the way we do on SBS, so that 75% ceiling isn’t universal across substrate types.