Printer Friendly

Say cheese.

New sensor improves consistency in milk product manufacturing

Cheese making dates back thousands of years, perhaps to some nomadic tribes that traveled carrying milk in animal skins.

These people probably discovered that the milk swaying in bags upon the trodding pack animals soon formed a soft curd that settled out of the liquid.

Later, cheese makers described how milk from different animals and different regions affected the product's taste and texture. By about 100 B.C., science entered the cheese-making arena as producers added the enzyme rennet to control curdling. This enzyme helped retain the cheese's protein and fat, losing less in the discarded whey.

Cheese making today combines science with art. The curdling process, or milk coagulation, occurs when an enzyme is added to warm milk. The enzyme breaks down the casein and exposes potential cross-linking sites along a protein chain. The cross-linked proteins form a matrix that traps fat globules to create a gel or coagulum.

In a cheese-making facility, milk coagulation typically takes 30 minutes. After the enzyme is added to milk, the liquid flows for about 15 minutes then gradually firms up due to protein cross linking.

Determining the best cutting time - the first step in separating the curd from the whey- is required to automate the process. Inaccurate cutting time predictions can cause increased losses and decreased quality and consistency. Currently, cheese makers feel the coagulum or slice it with a spatula to determine a cutting time that minimizes protein and fat loss.

No automated technology tested to remove cutting time subjectivity has been widely accepted by the cheese industry. Producers rely on experienced cheese makers to predict cutting time.

In December 1987, a research effort began to develop an automated cutting time method. Initial tests showed that diffuse reflectance changed in a consistent pattern during coagulation. The research goal became to transform this measurable optical signal into a control technology using an optical sensor and cutting time prediction algorithm.

Tests used one optical sensor that measured light reflectance and one that measured light transmission. Although signals from both provided similar information, the reflectance configuration offered easier cleaning for dairy operations. Optical fiber technology measured changes in light reflectance during coagulation. The fibers measured light reflected off particles in the opposite direction of incidence, which is referred to as light backscatter.

Figure 1 shows how light from a light emitting diode (LED) is transmitted to particulate fluid through fibers. The light backscatter is transferred back through fibers to a detector.

Research grants and commercial interests in 1990 provided funds to determine the effects of six variables on milk coagulation: fat, protein, pH, temperature, calcium and enzyme concentration. Tests used two enzymes with response recorded at three wavelengths. The tests showed that coagulation progress was reflected in an optical response.

During coagulation, light backscatter measurement followed a consistent pattern as shown in Figure 2. A steady period observed after adding the enzyme was followed by a sigmoidal reflectance increase. Lastly, during a cross-linking period, the reflectance increased at a decreasing rate.

The time between enzyme addition and the inflection point occurrence of the sigmoidal period, referred to as Tmax, correlated with the optimal cutting time. This correlation became the basis for a process control algorithm to predict cutting time:

Tcut = Beta x Tmax.

Beta is a constant, which typically varies between 1.3 and 2.2 based on the enzyme and product type selected to replicate a cheese maker's cutting time judgment. The algorithm works when the milk's protein content is constant. But Beta must be adjusted with varied protein content.

In 1992, this cutting time technology was tested at the former Dairyman's cheese plant in Glasgow, Kentucky. Positive results prompted the company to fund a three-vat prototype designed and built at the University of Kentucky. The prototype saved the company more than $150,000 annually and reduced whey fat losses by 20%. In 1993, a similar system was installed at another of the manufacturer's plants.

The University of Kentucky Research Foundation (UKRF) in December 1992 patented the cutting time prediction technology. The next year, UKRF licensed the technology to Reflectronics, Inc. of Lexington Kentucky, for commercialization under the trade name CoAguLite.

In 1994, Reflectronics teamed with Damrow Co. of Fond du Lac, Wisconsin. Damrow installed one commercial system per year in 1995 and 1996 and two per year in 1997 and 1998 in its vertical vats.

Personal computer-based demonstration systems were tested in August 1994 at The Netherlands Institute for Dairy Research (NIZO) and at cheese plants in the Netherlands, Australia, Finland and Ireland from 1995 to 1997. Tests in Europe on horizontal vats reflected inconsistent results.

Tests in Ireland in 1997 revealed that the inconsistent results in Europe came from vat wall cooling. A new probe designed in 1997, shown in Figure 3, reduced the vat wall cooling effects. In separate tests in 1997 and 1998 at a U.S. Swiss cheese manufacturing plant, the CoAguLite system with this prototype cone-shaped probe predicted cutting time with a standard error of 36 seconds.

CoAguLite technology required designing a fiber optic sensor, a suitable programmable logic control (PLC) process control algorithm, and a probe adaptable to cheese vat conditions. Graduate students from University of Kentucky's engineering and food science programs contributed to coagulation studies and fiber optic sensor design. Project support was provided by University of Kentucky's Agricultural Experiment Station.

Technology demonstration and refinement requires industry testing. The UKRF and the College of Agriculture provided the institutional flexibility to allow startup of Reflectronics, which bridged the gap between the university and cheese industry.

CoAguLite technology is in its infant stages and future research at the University of Kentucky will be directed toward refinement. The current focus is on interpreting information from the crosslinking phase of the reflectance curve for process control. The goal is to develop an indicator of the extent of protein crosslinking.

Backscatter sensors may have other applications in the food industry. Culture monitoring and cutting time prediction for cottage cheese production is one example. Reflectronics has also developed the FiberView sensor for detecting inline transitions of food products and monitoring milk fat in dairy waste streams.

Although the basic steps for coagulation are similar to methods used thousands of years ago, technology such as CoAguLite will continue to advance the art of cheese making.

ASAE member Fred Payne is a professor in the Biosystems and Agricultural Engineering Department, University of Kentucky, Lexington, KY 40546-0276, USA; 606-257-3000, fax 606-257-5671,
COPYRIGHT 1999 American Society of Agricultural Engineers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:cheese cutting time prediction technology from the University of Kentucky Research Foundation; Technology
Author:Payne, Fred
Publication:Resource: Engineering & Technology for a Sustainable World
Date:Aug 1, 1999
Previous Article:Aerial advantage.
Next Article:100 years old.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters