Most ice pack specifications are written once, in a meeting room, and then run unchanged for years. The fill weight that gets approved is usually the one that survived a July trial, because that is the trial that fails. Come January, the same pack goes out at 550 g into a 2 °C ambient and does two things wrong: it over-cools product that needed to stay above 0 °C, and it wastes gel that could have been sold into a heavier summer pack.
A 600 g gel pack frozen to −18 °C and placed in a 35 °C box gives up its cold in about 26 hours. The same pack in a 2 °C box gives it up in 60 to 70 hours. The cold is the same; the rate at which it leaves is not. As a rule of thumb, hold time scales roughly with the temperature difference between pack and ambient, so moving from a 35 °C summer lane to a 5 °C winter lane extends hold time by a factor of about 2.2.
That means a summer fill of 800 g is over-specified by 300 to 400 g in winter for the same 36-hour route. Trimming it is not just a cost exercise. Over-cooling is a real failure mode: produce with a 2 °C minimum, vaccines, and chilled ready meals all suffer when an oversized pack drives product below its lower limit for the first 10 hours of transit.
In an unheated warehouse at −2 °C, packs left on a loading dock will not freeze, but packs that arrive from a −18 °C freezer will not thaw either. The problem moves from insufficient cold to uncontrolled cold. Three practical adjustments matter more in winter than in summer.
Running two fills on the same line is a changeover question, not a formulation question. The cleanest approach is to keep the same SAP grade and change only the water-to-gel ratio and the fill volume, so the pump calibration stays valid and the only variable is the set point.
A workable pairing for a 20-litre box with a 36-hour target:
Label both SKUs identically except for a fill weight line printed on the film. Warehouse staff pick by weight far more reliably than by a colour code, and a scale at the packing bench catches the wrong SKU before it reaches a box.
Deciding when to switch. Do not switch on a calendar date; switch on the lane’s mean ambient. Pull 90 days of temperature logger data from the routes you actually run, and find the month where the 90th percentile ambient crosses 22 °C. That is the summer-to-shoulder switch. The shoulder-to-winter switch sits where the 10th percentile falls below 8 °C. In northern China and northern Europe those dates can be six weeks apart from the calendar solstice, and in export lanes to tropical ports summer fill is used year-round because the destination, not the origin, sets the load.
For destinations that mix climates — a February shipment from Hamburg to Lagos — size for the warmest leg and control over-cooling with placement rather than weight. Putting the pack against the carton wall rather than the product, or adding a 5 mm foam sheet between pack and product, removes the freeze risk without removing cold capacity.
The cost side of getting it wrong. Gel is cheap per kilo and expensive per shipment. A 250 g reduction on 500,000 packs a year saves 125 tonnes of gel, and more importantly 125 tonnes of frozen water that no longer has to be cooled to −18 °C — at roughly 90 kWh per tonne of gel frozen from 20 °C, that is a meaningful freezer load. Oversized winter packs also add volumetric weight, and couriers bill the outer box, not the gel.
Grades that freeze and thaw predictably across both fills are described on the SAP powder products page.
Zhengzhou Flor New Materials Co., Ltd. specializes in SAP super‑absorbent polymer products. Serving personal hygiene, agriculture & horticulture, pet supplies and other industries, we leverage supply‑chain resources to provide particle‑size selection, performance matching and customized solutions based on clients’ practical requirements.
© 2026. All Rights Reserved.