TL;DR: Tolerance stackup in water-soluble pouch design is the leading cause of seal failures that don’t appear until humidity-controlled storage — geometry must be validated before tooling is cut.
TL;DR: PVA film at 25µm nominal requires a ±3µm dimensional tolerance band in your CAD model to prevent seal jaw misalignment exceeding 0.5mm, which our inline vision system flags as a Category B defect.
Film Geometry and CAD Setup for Dissolvable Formats #
Water-soluble and edible packaging materials behave differently from conventional PE or PET films in every dimension that matters for CAD — elastic modulus, hygroscopic expansion, and thermal response under sealing conditions. If you’re building a die-cut or pouch geometry in CAD using standard film constraints, the output will be wrong before the first prototype is made.
PVA (polyvinyl alcohol) film, which covers the majority of water-soluble pouch applications, has an elastic modulus in the range of 1.8–3.2 GPa depending on grade and relative humidity. Compare that to LDPE at approximately 0.2 GPa. This means PVA is substantially stiffer in ambient conditions but loses dimensional stability rapidly above 60% RH. In our CAD templates for PVA formats, we apply a moisture expansion coefficient of 0.8–1.2% per 10% RH increase as a simulation input — a value derived from tensile testing on 25µm and 35µm Monosol M8630-equivalent grades across our production sampling program (roughly 40 film lots over two years).
For edible substrates such as rice paper and seaweed-derived films, the modulus variation is wider and less predictable. These materials typically range from 0.4–1.1 GPa in controlled conditions. CAD geometry for rice paper sachets requires larger tolerance buffers and our team flags these jobs in our internal MP-09 material classification form, which routes them to manual seal inspection rather than camera-only QC.
When setting up part geometry in CAD, the first decision is whether your format is a pre-formed pouch (three-side seal or pillow), a unit-dose sachet, or an open-top dip bag. Each has a different tolerance stackup profile.
| Format Type | Primary Tolerance Risk | Recommended CAD Clearance at Seal Zone | Typical Film Grade Thickness |
|---|---|---|---|
| Three-side seal pouch | Longitudinal seal width variation | +0.8 / -0.3mm from nominal jaw width | 25–40µm PVA |
| Unit-dose sachet | Corner fold geometry, fill volume displacement | ±1.0mm on all panel edges | 25–30µm PVA |
| Dip bag / open-top | Hang hole position relative to seal edge | Min. 8mm land from hole centre to seal | 35–76µm PVA or seaweed film |
The clearance values above are not conservative guesses. They come from our tolerance stackup analysis on 12 active pouch SKUs and reflect the cumulative effect of film reel camber, jaw alignment drift, and material expansion under process heat.
The Misdiagnosed Root Cause: Thermal Simulation Inputs Set for Conventional Film #
This is where most designs go wrong, and it’s rarely caught until the first production trial.
Water-soluble PVA film has a sealing initiation temperature of 120–140°C depending on grade, which is lower than typical BOPP or PET laminate seal thresholds (150–180°C). The practical consequence for design is that your heat seal jaw geometry and dwell time assumptions must change, and those changes propagate into the thermal model. When engineers run finite element or thermal simulation on a new sachet format, they frequently pull seal jaw temperature and contact pressure inputs from a previous project that ran on conventional film. The result is a simulated weld that looks structurally sound but was modeled at 160°C and 0.35 MPa — conditions that, when actually applied to PVA, cause gel degradation at the seal boundary and a measurable reduction in peel strength (confirmed by ASTM F88 seal strength testing).
The correct simulation inputs for 25µm PVA in our process are: jaw temperature 130 ± 5°C, contact pressure 0.25–0.30 MPa, and dwell time 0.8–1.2 seconds. At these parameters, the seal zone achieves a minimum peel strength of 8 N/25mm, which is our internal pass threshold and aligns with the lower bound specified in ISO 11607-1 for hermetic seals in unit-dose formats — even though ISO 11607 targets medical packaging, the structural seal strength criterion maps cleanly to soluble pouch applications.
The failure mode when these inputs are wrong is specific: the seal appears intact visually and passes a 48-hour ambient storage check, but shows delamination under humidity cycling above 65% RH. The weld zone has been thermally stressed at the polymer chain level without visible evidence. Confirmation test: run a cross-section of the seal under polarized light and look for crystallinity disruption bands within 0.2mm of the seal edge.
Edible films behave differently again. Seaweed-based film (typically sodium alginate or carrageenan composite) has no conventional thermoplastic seal mechanism — adhesive bonding or compression sealing is used instead, and thermal simulation is largely irrelevant. Design constraints shift to adhesive coverage area, minimum bond width of 6mm, and substrate porosity.
Corrective Actions When CAD Geometry and Process Outputs Diverge #
-
Rebuild tolerance stackup from film lot data, not spec sheet nominals. Spec sheet thickness for PVA is a nominal value. Actual lot-to-lot variation for 25µm film can span 22–28µm, and that 6µm range translates directly into seal jaw gap variation. Pull caliper data from at least 5 reel samples before locking CAD dimensions. This fixes the largest proportion of misalignment defects at the lowest cost.
-
Re-run thermal simulation with process-specific material properties. Input the measured glass transition temperature (Tg) of the actual film grade rather than a literature value. PVA Tg ranges from 58–85°C depending on hydrolysis degree and plasticiser content — a 27°C spread that significantly affects simulation output.
-
Adjust CAD seal land width for humidity expansion. If the design spec calls for a 6mm seal land, the CAD nominal should be 6.8mm to absorb moisture-induced expansion during storage before use. This adds material cost at a marginal level but eliminates the most common field complaint: apparent seal narrowing.
-
Validate corner geometry under simulated fill weight. Unit-dose sachets filled to 20–30g with powdered detergent or agricultural chemical experience 0.3–0.8mm bottom panel distortion under static load. Model this in simulation before approving the fold die. Failure to account for it results in corner stress concentrations that accelerate pinhole formation — a defect governed by ASTM F1140 burst testing criteria.
-
Run design-for-manufacturing review with your sealing machine OEM. For formats below 60mm × 60mm, machine jaw clearance and registration timing become constraints that override ideal CAD geometry. This step is expensive in time (typically 2–3 weeks with the machine OEM) but eliminates the most costly failure path: a validated design that simply cannot be run at speed on available equipment.
Prevention — What to Lock Down Before Tooling #
Specify film thickness tolerance (not just nominal), roll camber maximum (we accept ≤2mm/m), and Tg range in your purchase specification. For PVA, reference GB/T 12027 for dimensional stability test method. For edible films intended for food contact, confirm compliance with FDA 21 CFR 176.170 or EU 10/2011 (for the base polymer) before finalizing the material in your BOM.
The document to request from your film supplier before any tooling is cut: the film’s moisture sorption isotherm across 20–80% RH at 23°C. Without it, your CAD expansion coefficients are unanchored.
Specification Notes for Brand Partners #
When you brief us on a soluble or edible packaging project, the single most useful thing you can provide upfront is the fill material’s bulk density and particle size distribution. These two values determine feasible sachet volume, corner geometry, and whether your format needs a gusset. Without them, our first CAD draft will require at least one revision cycle after the fill trial.
The brief gap we see most often: fill weight is specified but fill volume is not. A 25g powder fill can occupy anywhere from 35ml to 90ml depending on density, and that range determines whether your target pouch size is feasible at all. Resolving this before we draw tooling saves three to five weeks.
Our standard sampling timeline for PVA pouch formats is 18–22 working days from approved brief to first sealed prototype. Formats using edible film (rice paper, seaweed, starch) run 25–30 working days due to the adhesive curing qualification step in our MP-09 sampling workflow. Timeline extends when the fill material must be customer-supplied for the trial batch.
What fill-weight tolerance should I specify in my brief for a unit-dose sachet?
For powdered fills, specify ±2% of target fill weight at minimum, and confirm this with your filling line OEM before briefing us. Tighter than ±1.5% is generally not achievable on high-speed rotary fillers without a gravimetric checkweigher. The sachet geometry we design must accommodate the volume at +2% fill, not nominal.
Can the same CAD geometry work for both PVA and a starch-based edible film?
Not directly. PVA and starch-based edible films differ in seal mechanism (thermal vs. adhesive), expansion coefficient, and minimum bend radius. A geometry optimized for PVA will need revised corner radii and seal land width to run on edible film substrates — typically a 15–20% increase in seal land area and tighter dimensional tolerances on the die. The base pouch dimensions can be preserved; it’s the tooling and process parameters that change.
Is 25µm PVA strong enough for a 30g detergent pod, or do I need 35µm?
It depends on the fill state (liquid vs. powder) and the drop height your supply chain must handle. For liquid or gel fills at 30g, we specify 35µm minimum per our internal packaging assessment; 25µm passes ASTM F1140 burst at ambient but shows a measurable failure rate above 1.5% in drop testing at 1.0m onto a hard surface. For dry powder at the same weight, 25µm is typically sufficient if corner geometry is correct.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
