Cooling of jarred cheese spreads

Kraft Foods Ltd

The solution of a heat flow equation showed that a cooling tunnel for jars of cheese spreads is operating near the limit of its capacity. If export demand led to the use of a larger sized jar, then slower operation of the cooling line or an increase in the cooling capacity of the line would be needed.

The investigation also uncovered a possible mechanism for the occasional appearance of some tiny bubbles in the jars. Some remedies to counteract this mechanism were found.

Cheese spread is a significant export. More than 60 per cent of the cheese spread Kraft makes in Australia is shipped overseas. To produce its cheese spread products, Kraft uses about ten million kilograms of cheese a year. This cheese is blended with cream and emulsifiers and then cooked under pressure using steam at high temperature.

Cheese spread is a remarkably good insulator, taking a long time to cool after it is cooked. If the temperature of the spread is not reduced from 85šC sufficiently quickly, it could spoil. In order to achieve the correct cooling, Kraft Foods passes filled jars of cheese spread through a cooling tunnel that lowers the temperature to an acceptable level within about 50 minutes.

After cooling, the jars of cheese spread are packed and ready for shipping within 48 hours.

Forced cooling is energy intensive. Any reduction in the use of chilled water saves money.

Cheese spread can also boil in the jar, leaving unsightly bubbles that discolour as fat migrates into them. The resulting product is unsuitable for sale.

MISG was asked to develop a model for the current operation of the cooler. The model would allow the company to determine how changing the operation of the cooler would affect the temperature within the jar as time passes, and thereby suggest the best conditions for cooling the cheese spread.

The cooling process was modelled, and the resulting heatflow equation solved, to find the time needed for the temperature of the spread in the jar to drop sufficiently to permit progress to the next cooling zone.

This information allows more efficient operation of the cooling process and reduces the costs of producing spread that was heat damaged or which needed reprocessing. The model can also be used to test proposals such as the introduction of a larger size of jar.

There are several constraints on the cooling process. For instance, the difference in temperature between the inside and the outside of the jar could not exceed 43šC, as this would crack the glass. Also, condensation would form outside the jar if the temperature of the exterior of the glass fell too low, and the time which jars spent in the cooling tunnel could be lengthened arbitrarily by delays in other parts of the production line.

One group considered the overall heat balance of the cooling system including the heat gained by the water and the heat lost by the jars, which should balance. This simple observation allowed the group to construct an equation which was in good agreement with the data supplied for the system.

Using this equation it can be shown that the cooler's 300kW heat exchanger was working at the limit of its capacity.

Another group studied the problem of why the cheese boils near the center of the jar as it cools. It came to the conclusion that when cheese near the edge of the jar cools too rapidly it begins to shrink excessively creating a reduction of pressure in the jar. As the pressure drops so does the boiling temperature, and so it is possible that spread near the center of the jar actually starts to boil as it is cooled, because of the very substantial reduction in pressure under some circumstances.

If this argument is correct then the solution to this boiling problem is to cool the jar less rapidly at the beginning of the process. This is done by using a warmer water spray for longer at the start of the cooling process. Leaving more air in the top of the jar would also help.

A company representative noted that the recommendation that you could retard boiling by starting with a warmer spray was almost counter intuitive, and demonstrated the value of mathematics.

Through the process of mathematical modelling, the study group had identified several important features including the constraints and limitations of the unit, and the sensitivity of the process parameters. It had also developed a model that could be used for analysis thus reducing the costly exercise of carefully controlled field trials.