|
Preparation & Measurement of Meat Samples |
 Beef, Turkey, Chicken, Pork, Horse & Pony, Sausage Meats
The preparation procedures for all the above meat products are basically the same.
The usable flesh of the animal (i.e. excluding the head, bones, offal etc), is uniformly minced as required. A portion of the minced meat of approximately 300g is randomly selected from the batch, hand pressed into a solid mass and placed in a suitable glass dish or bowl.
The depth of the material should be at least 30mm flat or 60-100mm round, and the diameter or length of the sample should be greater than the dimensions of the sensor head.
The sensor head is then pressed firmly into the meat to ensure good contact over the whole of the sensor head, but not so firmly that the mass of the material is penetrated significantly. After completing such a measurement, there should be just a slight indentation in the surface of the minced meat product, a few millimeters at the most.
This procedure is repeated EIGHT TIMES, at different points on the sample to obtain the average Fatmeter reading (please refer to measurement sites section).
A sample of the minced meat of approximately 45g is then taken for analysis by Foss-Let using the normal Foss-Let methodology.
For the greatest accuracy of any calibration check, it would be advisable to perform as many Foss-Let analyses on the mince as the sample allows. This then will take account of any inhomogeneity in the fat distribution within the sample.
Conclusion
Using the techniques described above, it should be possible to achieve accuracies (95% confidence interval) within the following limits:
| Fat/Oil Reading |
Accuracy |
| 4 to 10% |
± 0.5% eg 9.5 -10.5% at 10% |
| 11 to 20% |
± 1.0% eg 18 - 20% at 19% |
| 21 to 30% |
± 1.5% eg 27.5 – 30.5% at 29% |
| 31 to 40% |
± 2.0% eg 33 – 37% at 35% |
| 41% + rising to… |
± 4.0% eg 52-60% at 56% |
|
