TL;DR: Getting tea bag packaging to run cleanly on a vertical form-fill-seal or overwrap line depends on substrate tension calibration and sealing parameter validation — not just on print quality or material selection.
TL;DR: A sealing dwell time shift of as little as 0.15 seconds can move heat-seal bond strength from an acceptable 8 N/15mm to a failing 4 N/15mm on standard heat-sealable filter paper.
Heat-Seal Parameter Validation Across Substrate and Machine Combinations #
Before any production run begins, we require a sealing window study for every new substrate-machine pairing. This is not optional for us — it’s logged as Step 1 in our PV-04 process validation checklist.
The sealing window defines the intersection of temperature, dwell time, and pressure that produces a consistent bond without burning through the substrate. For heat-sealable filter paper (typically 12–18 gsm), that window is narrower than most brand partners expect. We run it across five temperature points in 5°C increments and three dwell times (0.10 s, 0.15 s, 0.20 s) for each — 15 data points minimum before we commit a substrate to production.
Sealing parameter windows by common tea bag substrate (indicative ranges from our validated trials):
| Substrate | Seal Temp Range (°C) | Dwell Time (s) | Min Bond Strength (N/15mm) |
|---|---|---|---|
| Heat-sealable filter paper, 12–16 gsm | 150–175 | 0.10–0.20 | ≥ 6 |
| Non-woven PLA pyramid mesh, 17–22 gsm | 160–185 | 0.15–0.25 | ≥ 7 |
| Foil-laminated overwrap sachet, 40–60 gsm | 130–155 | 0.20–0.35 | ≥ 10 |
| Paper-only overwrap, uncoated 60–80 gsm | 115–135 | 0.25–0.40 | ≥ 8 |
Bond strength is measured per ASTM F88 seal strength test method, pulling at 180° on a 15mm-wide specimen. Any value below the minimum threshold in our table triggers a parameter reset and re-run, not a judgment call on whether the pack “looks fine.”
One point worth stating clearly: the temperature ranges above assume the machine sealing bar has been calibrated within the past 30 days. Bar calibration drift of ±8–12°C is common after 500,000 cycles on older equipment, and that drift alone can silently push you outside your validated window. We calibrate with a surface thermocouple probe before the first production lot of any new substrate.
Where Integration Fails — Three Root Cause Scenarios #
Web tension mismatches on high-speed VFFS lines. For single-chamber flat tea bags running on vertical form-fill-seal equipment at 200–400 bags/minute, the unwind tension on the filter paper web must be held within a tight band — typically 15–35 N/m depending on paper basis weight. When a brand partner switches substrate supplier mid-run without re-tensioning (which happens more than it should), the lighter or stiffer incoming web causes skip-sealing or film telescoping at the sealing station. The consequence is incomplete side seals, which fail the hot-water immersion test we run per ISO 9022-14 environmental exposure protocols. What we check first: roll core hardness and web thickness uniformity across the reel width, measured with a contact micrometer at 10 positions.
String-and-tag attachment timing drift on double-chamber bag lines. String attachment on a double-chamber machine is synchronized to a rotary cam at 1/4-cycle increments. When the cam timing drifts beyond ±2° from the set point — which can happen after 3–4 million cycles without re-indexing — the string attaches 1.5–3 mm off center on the bag header. This looks cosmetic, but off-center string attachment shifts the crimp zone, reducing the staple or ultrasonic weld holding force by 20–35% in our pull tests. A brand partner launching an upscale loose-leaf blend with a branded tag cannot afford that failure mode in consumer hands. The check is straightforward: measure string attachment position on 30 consecutive bags, calculate Cpk; anything below 1.33 on this dimension means re-indexing before production continues.
Overwrap sachet foil registration errors at the cut-seal station. Individual tea bag overwrap sachets — typically a foil-paper or foil-PET laminate — require the print image to register within ±0.5 mm of the cut line, otherwise the brand’s front panel graphic gets clipped. On high-speed flow-wrap lines, the most common cause of registration drift is photocell eye sensitivity degrading as the foil surface develops micro-scratches from the guide rails. We’ve measured photocell detection delays of 8–14 ms on scratched foil surfaces versus 2–4 ms on fresh stock, and at 80 m/min line speed, that delay translates directly into 10–19 mm registration error. Replacing foil guide rail contact points with PTFE-coated guides, and recalibrating photocell sensitivity every 2 million cuts, controls this reliably. Our internal standard for overwrap register is ±0.3 mm, tighter than the ±0.5 mm typically cited in VFFS equipment manuals, because foil-laminate sachets are almost always used in premium positioning where print accuracy is visible.
This section is the longest because integration failures in tea bag packaging are almost always process-parameter issues — not substrate defects. Blame the material last.
Does Compostable Non-Woven Mesh Require Different Line Settings? #
Yes — and the difference is larger than most equipment specs suggest.
PLA-based non-woven pyramid mesh is more temperature-sensitive than conventional nylon or PET mesh. Its sealing range sits 15–25°C lower than equivalent nylon grades, and its melt-flow behavior is sharper, meaning the process window is narrower. On our lines we drop bar temperature to 160–170°C for PLA mesh versus 185–200°C for nylon, and we reduce line speed by 10–15% to maintain dwell time. The machine’s cooling station also needs to be active for PLA — passive cooling causes weld creep that shows up as bag deformation during downstream collation. For brands moving toward TÜV Austria OK Compost HOME certified mesh, line re-commissioning with a full PV-04 parameter re-validation is the correct approach, not a simple temperature dial-down.
Specification Notes for Brand Partners #
When you brief us on a tea bag packaging integration project, the single most useful thing you can provide is your target filling line speed and the model of your sealing equipment — or your co-packer’s equipment. Substrate qualification is always machine-specific. A heat-sealable filter paper validated at 300 bags/minute on a Teepack Constanta may need re-validation at 400 bags/minute on an IMA C23, even if the paper grade is identical.
The gap we see most often in incoming briefs is the absence of a filled-bag weight specification. Dwell time, sealing pressure, and even bag geometry tolerances depend on how much product is inside the bag — a 2 g chamomile bag and a 3.5 g rooibos blend behave differently under sealing compression. Missing this number means at least one additional sampling iteration.
Our standard line-integration sampling timeline runs 15–20 working days from receipt of confirmed substrate and filling parameters to first sealed samples with QC data. That timeline extends to 25–30 working days if compostable substrates are involved, because the PLA sealing window study requires additional thermal profiling steps.
Frequently Asked Questions #
Can we use our existing foil sachet tooling if we switch from a 45 gsm laminate to a 60 gsm laminate?
It depends on your current seal bar gap and pressure setting. A 15 gsm increase in laminate caliper — roughly 18–22 µm additional thickness depending on foil gauge — will require the sealing bar gap to be re-set, and seal pressure typically needs to increase by 10–20% to maintain the same bond strength. If your tooling is fixed-gap rather than spring-loaded, you may need new jaw inserts.
What AQL level do you apply to sealed tea bag samples before approving a production run?
We apply AQL 1.0 for seal integrity defects (open seals, incomplete side welds) and AQL 2.5 for cosmetic defects (registration, tag position) per ANSI/ASQ Z1.4 sampling tables. For a standard 500,000-bag order, that means inspecting 200 units for seal defects with an acceptance number of zero.
How long does commissioning take if we’re introducing a new substrate that has never run on your line?
New substrates that have no prior sealing data on our equipment require a minimum of 3 working days for sealing window mapping before any sealed samples are produced. If the substrate also requires a new reel format or core diameter, add 2 working days for unwinder changeover and tension calibration.
Planning a packaging project? Contact our team to request a complimentary specification review and sample quote.
On the PLA pyramid mesh — are you sealing against itself or onto a paper outer, and does that affect where in the 160–185°C window you actually land during production runs?
The PLA pyramid mesh needing a 160–185°C range versus filter paper at 150–175°C seems manageable on paper, but the real pain is that PLA’s dwell tolerance is less forgiving when your line speed fluctuates even slightly — we had a run where ambient temp in the facility dropped about 4°C overnight and the pyramid mesh started failing ASTM F88 pulls at what should’ve been a validated dwell. Filter paper at 12–14 gsm recovered faster when we nudged temperature; the PLA didn’t have the same forgiveness.
The 15-data-point minimum for sealing window studies is real — we also found that on PLA pyramid mesh specifically, ambient humidity in the filling hall (we run at roughly 55–60% RH) shifts the validated dwell time by 0.03–0.05 s, enough to drop you below that ≥7 N/15mm threshold if you locked parameters in winter and didn’t revalidate before summer production.
The PLA pyramid mesh validation cost us more than the material switch itself — running 15-point sealing studies across three machine lines at our Utah co-packer added roughly $2,200 in technician time and substrate waste per SKU before we even hit production. We’ve since batched new substrate validations to run concurrent with scheduled PM windows, which cut that overhead by about 40%.
For the foil-laminated overwrap sachet, that 130–155°C range sitting *below* the filter paper window surprised me — does the lower seal temp mean you’re seeing more delamination risk at the foil-sealant interface when bond strength is tested at 180° per F88, or does the laminate construction actually make it more forgiving there?
The ≥10 N/15mm minimum on foil-laminated overwrap is where we’ve consistently had to push dwell time toward the high end of that 0.20–0.35 s range — on our Hayssen Ultima line we couldn’t hit repeatable bond strength below 0.28 s without seeing peel failures on roughly 1 in 40 sachets during ASTM F88 pulls.
The 5°C increment spacing across five temperature points works well for most substrates, but on paper-only overwrap (especially at the lower end of that 115–135°C window) we’ve found you need tighter spacing — 3°C increments — because the bond strength curve flattens out and then drops off faster than the filter paper data would lead you to expect. We missed a narrow failure zone running at 119°C on an uncoated 70 gsm kraft before we tightened the study protocol.