Appendix C to Subpart S of Part 431 - Enforcement for Performance Standards; Compliance Determination Procedure for Metal Halide Lamp Ballasts

DOE will determine compliance as follows:

(a) After it has determined the sample size, DOE will measure the energy performance for each unit in accordance with the following table:

Sample sizeNumber of tests for each unit41312214

(b) Compute the mean of the measured energy performance (x1) for all tests as follows:

ER09MR10.008

Where xi is the measured energy efficiency or consumption from test i, and n1 is the total number of tests.

(c) Compute the standard deviation (S1) of the measured energy performance from the n1 tests as follows:

ER09MR10.009

(d) Compute the standard error (Sx1) of the measured energy performance from the n1 tests as follows:

ER09MR10.010

(e)(1) For an energy efficiency standard, compute the lower control limit (LCL1) according to:

ER09MR10.011 or ER09MR10.012 (whichever is greater)

(2) For an energy use standard, compute the upper control limit (UCL1) according to:

ER09MR10.013 or (whichever is less) ER09MR10.014

Where EPS is the energy performance standard and t is a statistic based on a 99-percent, one-sided confidence limit and a sample size of n1.

(f)(1) Compare the sample mean to the control limit. The basic model is in compliance and testing is at an end if, for an energy efficiency standard, the sample mean is equal to or greater than the lower control limit or, for an energy consumption standard, the sample mean is equal to or less than the upper control limit. If, for an energy efficiency standard, the sample mean is less than the lower control limit or, for an energy consumption standard, the sample mean is greater than the upper control limit, compliance has not been demonstrated. Unless the manufacturer requests manufacturer-option testing and provides the additional units for such testing, the basic model is in noncompliance, and the testing is at an end.

(2) If the manufacturer does request additional testing and provides the necessary additional units, DOE will test each unit the same number of times it tested previous units. DOE will then compute a combined sample mean, standard deviation, and standard error as described above. (The “combined sample” refers to the units DOE initially tested plus the additional units DOE has tested at the manufacturer's request.) DOE will determine compliance or noncompliance from the mean and the new lower or upper control limit of the combined sample. If, for an energy efficiency standard, the combined sample mean is equal to or greater than the new lower control limit or, for an energy consumption standard, the sample mean is equal to or less than the upper control limit, the basic model is in compliance and testing is at an end. If the combined sample mean does not satisfy one of these two conditions, the basic model is not in compliance.