Wis. Admin. Code ATCP § ATCP 65.86

Current through December 30, 2024

Section ATCP 65.86 - Milk component test methods

(1) GENERAL. Milk component tests shall be performed using any of the following methods, subject to additional requirements under sub. (2) and (3):

(a) A method described in the American Public Health Association., Standard Methods for the Examination of "Dairy Products," 17th edition (2004).

(b) A method described in the Official Methods of Analysis of AOAC International," 18th edition (2005).

(c) A method approved in writing by the division.

Note: "A milk component test," as defined under s. ATCP 65.01(35), means a test that determines the amount of milkfat, protein, total solids, solids-not-fat, or other valuable components in milk, and that may affect the price that a dairy plant operator or milk contractor pays a milk producer for milk.

The American Public Health Association's "Standard Methods for the Examination of Dairy Products," 17th edition (2004), is on file with the division and the legislative reference bureau. Copies may be obtained from the American Public Health Association, 800 I Street, NW, Washington, D.C. 20001, telephone 202-777-2742, website www.apha.org.

The "Official Methods of Analysis of AOAC International," 18th Edition (2005), is on file with the division and the legislative reference bureau. Copies may be obtained from AOAC International, 2275 Research Blvd., Rockville, MD 20850, telephone 800-379-2622, website https://www.aoac.org.

(2) MILKFAT TEST METHODS.

(a) Milkfat tests shall be performed using the Babcock method, the ether extraction method, or another test method approved by the division. Babcock and ether extraction tests shall be conducted according to the Official Methods of Analysis of the Association of "Official Analytical Chemists (AOAC) International," 17th edition (2000), except as provided under par. (b).

(b) Each milk sample tested by the Babcock method shall be agitated for at least 3 minutes by the use of a mechanical agitator after pipetting the sample and adding sulfuric acid according to the procedure prescribed under par. (a). A reading device, such as a needlepoint divider or other mechanical divider, that accurately determines milkfat level in a test bottle shall be used in reading all Babcock tests. All Babcock test readings shall be made against a light-colored surface with adequate natural or artificial light. The Babcock test shall be read to the nearest 0.05% by weight.

(3) AUTOMATED MILK COMPONENT TESTING DEVICES.

(a)General calibration requirements. If an automated testing device is used to perform a milk component test for any milk component, that device shall be calibrated and regularly checked to ensure that it accurately tests for that milk component.

(b)Specific calibration requirements. If an automated testing device is used to test for milkfat, protein, total solids, or solids-not-fat in milk, and if the test results may affect the price paid to a milk producer, the testing device shall be calibrated according to this paragraph. The testing device shall be calibrated, for each relevant milk component, by a tester who is licensed under s. 97.17 or 98.145, Stats., to operate that device.

Note: See s. ATCP 65.78(2).

1. 'Calibration frequency'. A milk component testing device under par. (b) shall be calibrated at all of the following times:

a. Upon installation.

b. At regular 3 month intervals after installation.

c. Immediately after every significant repair or alteration to the testing device.

d. Whenever the mean difference on a daily performance check under par. (c) exceeds plus or minus 0.044% for milkfat or protein or plus or minus 0.084% for total solids or solids-not-fat.

2. 'Calibration procedure'. To calibrate a milk component testing device under par. (b), a tester shall use the device to test a set of calibration samples under subd. 3. The milk component testing device shall be adjusted, as necessary, to satisfy all of the following requirements:

a. The performance error on each calibration sample shall be as near as practicable to zero. The performance error is the difference between the known percentage content of each milk component in the calibration sample, as determined by the sample provider, and the percentage content measured by the testing device.

b. The mean difference for the entire set of calibration samples shall be as near as practicable to zero and shall not exceed plus or minus 0.044% for milkfat or protein or plus or minus 0.084% for total solids or solids-not-fat. The mean difference is the sum of the performance errors for the individual calibration samples divided by the number of samples in the set.

c. The standard deviation of test results, calculated for the set of calibration samples according to the formula set forth in the Official Methods of "Analysis of AOAC International," 18th edition (2005), section 969.16, shall not exceed 0.044% for milkfat or protein, or 0.084% for total solids or solids-not-fat.

Note: The Official Methods of "Analysis of AOAC International," 18th edition (2005), is on file with the division and the legislative reference bureau, and may be obtained from AOAC International, 2275 Research Blvd., Rockville, MD 20850, website https://www.aoac.org.

3. 'Calibration samples'. A set of calibration samples shall be obtained from a sample provider approved by the division. A set of calibration samples shall consist of at least 12 individual samples, each of which complies with all of the following requirements:

a. Each sample shall be not more than 21 days old.

b. Each sample shall be a fresh milk sample preserved with bronopol (2-bromo-2-nitro-1, 3-propanediol) or another approved preservative. Preservative methods, formulations, and concentrations shall be approved by the division.

c. Each sample shall have a known percentage content of each relevant milk component determined by the sample provider under pars. (e) to (h).

(c)Daily performance check.

1. If an automated testing device is used to test for milkfat, protein, total solids, or solids-not-fat in milk, and if the test results may affect the price paid to a milk producer, the device shall be subjected to a daily performance check before each day's testing. The daily performance check shall be conducted, for each relevant milk component, by a tester who is licensed under s. 97.17 or 98.145, Stats., to operate the testing device.

2. To conduct a daily performance check under subd. 1., a tester shall test a set of daily performance check samples under subd. 4. Based on the daily performance check, the tester shall do all of the following:

a. Determine the performance error of the testing device with respect to each daily performance check sample. The performance error is the difference between the known percentage content of each milk component in that sample, as determined by the sample provider, and the percentage content measured by the testing device.

b. Based on the performance errors for the individual samples under subdivision paragraph a, calculate the mean difference for the set of daily performance check samples. The mean difference is the sum of the performance errors for the individual samples, divided by the number of samples in the set.

3. If, on a daily performance check under subd. 1., the mean difference calculated under sub. (2) (b) exceeds plus or minus 0.044% for milkfat or protein or plus or minus 0.084% for total solids or solids-not-fat the testing device shall not be used until it is recalibrated under par. (b).

4. A set of daily performance check samples shall be obtained from a sample provider approved by the division. A set shall consist of at least 5 individual samples, each of which complies with all of the following requirements:

a. Each sample shall be not more than 21 days old.

c. Each sample shall have a known percentage content of each relevant milk component, determined by the sample provider under pars. (e) to (h).

(d)Reference checks.

2. A daily reference check required under subd. 1 shall be done in accordance with all of the following requirements:

a. A daily reference check shall be conducted before each day's testing, at the same time that the dairy plant operator conducts the daily performance check under par. (c). The daily reference check shall be conducted, for each relevant milk component by a tester who is licensed under s. 97.17 or 98.145, Stats., to operate the testing device.

b. To perform a daily reference check, a tester shall perform 10 tests on a reference sample. The reference sample may be a homogenized milk sample prepared by the dairy plant operator, or it may be a daily performance check sample obtained from a sample provider approved by the department under par. (c) 4. The 10 test results shall be averaged and the average result shall be used as a comparison value for the hourly reference checks under subd. 3.

3. An hourly reference check required under subd. 1 shall be done in accordance with all of the following requirements:

a. An hourly reference check shall be conducted for each milk component before each hour's testing for that component. To conduct an hourly reference check, a tester shall test the same reference sample used for the daily reference check under subd. 2.

b. For each relevant milk component the hourly reference check result shall be compared to the average result obtained on the daily reference check under subd. 2. If an hourly reference check result differs from the average result on the daily reference check by more than 0.034 % for milkfat or protein or 0.064% for total solids or solids-not-fat, the testing device shall not be used until the condition causing the difference is found and corrected. Test results obtained before the device is corrected, and after the last previous conforming reference check, shall not be used in determining the amount paid to milk producers.

(e)Calibration and daily performance check samples; milkfat contents.

1. The provider of a calibration sample under par. (b) or a daily performance check sample under par. (c) shall determine the known percentage content of milkfat in that sample by averaging the results of 3 milkfat tests using a method specified under subd. 4. The percentage milkfat results from those 3 milkfat tests shall not vary by more than 0.034 percentage points.

2. The known milkfat content of a calibration sample, expressed as a percentage of the sample weight, shall be at least 2.5%. Within a set of calibration samples, the difference in known milkfat content between the lowest milkfat sample and the highest milkfat sample, expressed as a percentage of average sample weight, shall be at least 2.5%.

3. The known milkfat content of a daily performance check sample, expressed as a percentage of the sample weight, shall be at least 2.8%. Within a set of daily performance check samples, the difference in known milkfat content between the lowest milkfat sample and the highest milkfat sample, expressed as a percentage of average sample weight, shall be at least 1.5%.

4. To determine the milkfat content of a calibration sample or daily performance check sample, the sample provider shall use either a manual or robotic version of the Modified Mojonnier method as described in the "Official Methods of Analysis of AOAC International," 18th edition (2005), section 989.05.

Note: The Official Methods of Analysis of AOAC International," 18th edition (2005), is on file with the division and the legislative reference bureau, and may be obtained from AOAC International, 2275 Research Blvd., Rockville, MD 20850, website https://www.aoac.org.

(f)Calibration and daily performance check samples; protein contents.

1. The provider of a calibration sample under par. (b) or a daily performance check sample under par. (c) shall determine the known percentage content of protein in that sample by averaging the results of 3 protein tests using the method specified under subd. 3. The percentage protein results from those 3 tests shall not vary by more than 0.034 percentage points.

2. The known protein content of a calibration or daily performance check sample, expressed as a percentage of sample weight, shall be at least 2.7%. Within a set of calibration samples, the difference in known protein content between the lowest protein sample and the highest protein sample, expressed as a percentage of average sample weight, shall be at least 0.7%. Within a set of daily performance check samples, the difference in known protein content between the lowest protein sample and the highest protein sample, expressed as a percentage of average sample weight, shall be at least 0.5%.

3. To determine the protein content of a calibration sample or daily performance check sample, the sample provider shall use the traditional or block digester/steam distillation Kjeldahl method as described in the Official Methods of Analysis of AOAC International," 18th edition (2005), section 991.20.

(g)Calibration and daily performance check samples; total solids.

1. The provider of a calibration sample under par. (b) or a daily performance check sample under par. (c) shall determine the known percentage content of total solids in that sample by averaging the results of 3 total solids tests using the method specified under subd. 3. The percentage total solids results from those 3 tests shall not vary by more than 0.054 percentage points.

2. The known total solids content of a calibration or daily performance check sample, expressed as a percentage of sample weight, shall be at least 11%. Within a set of calibration samples, the difference in known total solids content between the lowest total solids sample and the highest total solids sample, expressed as a percentage of average sample weight, shall be at least 2.00%. Within a set of daily performance check samples, the difference in known total solids content between the lowest total solids sample and the highest total solids sample, expressed as a percentage of average sample weight, shall be at least 1.5%.

3. To determine the total solids content of a calibration sample or daily performance check sample, the sample provider shall use the direct forced air oven drying method as described in the Official Methods of Analysis of AOAC International," 18th edition (2005), section 990.20.

(h)Calibration and daily performance check samples; solids-not-fat. The provider of a calibration sample under par. b or a daily performance check sample under par. (c) shall calculate the known percentage content of solids-not-fat in that sample by subtracting the percent milkfat as determined under par. (e) from the total solids for that sample as determined under par. (g). The calculation method shall be that described in the Official Methods of Analysis of AOAC International," 18th edition (2005), section 990.21.

(i)Automated testing devices; constant voltage. A constant voltage regulator shall be connected to, or form a part of, every milk component testing device that is in line with a single phase 115 or 220-volt power supply.

(j)Records related to calibrations, daily performance checks, and reference checks.

1. A dairy plant operator shall keep a record of every calibration, performance check, or reference check conducted on a milk component testing device under this section.

2. Every record required under subd. 1 shall be signed by the licensed tester who made the record. Calibration records shall be kept separate from performance check and reference check records.

(k)Accuracy of devices; division audit. The division may audit the accuracy of milk component testing devices using test samples prepared by the division under pars. (e) to (h).

Wis. Admin. Code Department of Agriculture, Trade and Consumer Protection ATCP 65.86

Adopted by, CR 14-073: cr. Register August 2016 No. 728, eff. 9-1-16; correction in (3) (c) 3. made under s. 35.17, Stats., Register August 2016 No. 728, eff. 9/1/2016

Amended by, correction in (3) (b) 2. (Note), (3) (e) 4. (Note), (3) (f) 3. (Note), (3) (g) 3. (Note), (3) (h) (Note) made under s. 13.92(4) (b) 7, Stats., Register January 2017 No. 733, eff. 2/1/2017