Ninety Minutes Before the Filler Runs: An Incoming SAP Gate for Ice Pack Plants

A pack maker in Vietnam changed SAP suppliers to save 180 dollars a tonne and discovered the problem on day nine of a production run: 60,000 packs filled, sealed and frozen, and the gel in every one short-bodied and slightly cloudy. The new lot hydrated to 340 g/g instead of 390, inside the supplier’s minimum, but milled finer, so the fines set before the mixer could disperse them. The run was scrapped at a cost of about 14,000 dollars.

Incoming inspection at a gel pack plant is usually a moisture check and a look at the certificate. That is not enough to protect a filler, because the properties that break a production run are process properties, not lab properties. Here is a 90-minute gate that catches the four failures that actually stop a line.

Test One: Free Swell in Your Water, Not Deionised

Weigh 1.0 g of powder into a 1 litre beaker of the water your line actually uses, at the temperature it arrives. Stir at 300 rpm for 60 seconds and leave for 20 minutes, then pour through a 100 mesh sieve for 5 minutes and weigh the drained gel. Compare against your own historical baseline, not the certificate. Most plants run 380 to 420 g/g in deionised water and 300 to 350 in plant water at 250 microsiemens hardness, and a 10 percent drop on a new lot is a warning even when the certificate says the powder is fine.

Run the same test in the water you will use for the next production week. Hardness varies with the season and the municipal source, and a lot that passed in April can fail in August without the powder changing at all.

Test Two: Dispersion Time, Because Your Mixer Has a Deadline

Drop 5 g of powder into 500 ml of water at 25 degrees without stirring and time how long it takes to wet out completely, judged by the disappearance of dry islands. Good lots take 8 to 20 seconds. Fines-heavy lots take 3 to 6 seconds and then form fisheyes, which is worse: fast wetting creates a gel skin around each lump and the inside stays dry for hours.

Then repeat with stirring at your line speed and check for lumps after 10 minutes by pouring through a 2 mm screen. More than a dozen lumps on a 5 g sample means the lot will not disperse in a production batch, and no amount of mixing time will fix it.

Test Three: Retention After a Freeze, Which Is What the Customer Sees

Make 200 g of gel at production concentration, freeze it at minus 18 for 16 hours, thaw at 25 degrees for 8 hours, and weigh the free water that separates. More than 2 g of free water on a 200 g sample predicts a pack that weeps in a stacked carton. Repeat twice more to simulate a reuse cycle if your product is sold as reusable.

This is the test that correlates best with field complaints, and it is the one most plants skip because it takes a day. Run it in parallel with the first production batch rather than before it, and hold the release of that batch until the result is in.

Test Four, Moisture, and the Paperwork That Makes the Gate Stick

Moisture above 6 percent causes two problems: the powder weighs heavy, so you under-dose, and it bridges in the hopper. A 10 g sample in a moisture balance at 105 degrees for 20 minutes gives an answer in the time it takes to run the dispersion test. Flow is simpler still: pour 100 g through a 10 mm funnel and time it. A lot that takes more than 25 seconds to empty has absorbed moisture in transit or has an off-spec particle size distribution.

Keep a single sheet per lot: supplier, lot number, arrival date, the four results, the operator, and a pass or hold decision. Hold lots that fail two or more tests and run a 200 kg trial before releasing the rest. The sheet is also the document that gets you paid when a supplier disputes a claim, because it shows a consistent method and a baseline built over months.

If you want the baseline in the first place, ask for grade data you can test against rather than a generic certificate. Our SAP powder for ice pack production ships with lot-level free swell, retention and sieve data, which turns the gate above from a guess into a comparison.