User Info (Unlocked Cells)

Sample
Location Descripton 1 2 3 Mean Std Dev
1 99.0 100.0 101.0 100.0 1.0
2 99.0 100.0 101.0 100.0 1.0
3 99.0 100.0 101.0 100.0 1.0
4 99.0 100.0 101.0 100.0 1.0
5 99.0 100.0 101.0 100.0 1.0
6 99.0 100.0 101.0 100.0 1.0
7 99.0 100.0 101.0 100.0 1.0
8 99.0 100.0 101.0 100.0 1.0
9 99.0 100.0 101.0 100.0 1.0
10 99.0 100.0 101.0 100.0 1.0
11 99.0 100.0 101.0 100.0 1.0
12 99.0 100.0 101.0 100.0 1.0
13 99.0 100.0 101.0 100.0 1.0
14 99.0 100.0 101.0 100.0 1.0
15 99.0 100.0 101.0 100.0 1.0
15 45
OVERALL
Mean 100.0
Std Dev 0.83
Min 99.0
Max 101.0

Variance Components
Source Std Dev % Total
Between Location 0.00 0.0
Within Location 1.00 100.0
Total 1.00 100.0

125
120
115
110
Assay (% Target)

105
100
95
90
85
80
75
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Location
85
80
75
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Location
User Info (Unlocked Cells)
Uncorrected | Weight Corrected
Location 1 2 3 Mean Std Dev | Location 1 2
1 99.0 100.0 101.0 100.0 1.0 | 1 99.0 100.0
2 99.0 100.0 101.0 100.0 1.0 | 2 99.0 100.0
3 99.0 100.0 101.0 100.0 1.0 | 3 99.0 100.0
4 99.0 100.0 101.0 100.0 1.0 | 4 99.0 100.0
5 99.0 100.0 101.0 100.0 1.0 | 5 99.0 100.0
6 99.0 100.0 101.0 100.0 1.0 | 6 99.0 100.0
7 99.0 100.0 101.0 100.0 1.0 | 7 99.0 100.0
8 99.0 100.0 101.0 100.0 1.0 | 8 99.0 100.0
9 99.0 100.0 101.0 100.0 1.0 | 9 99.0 100.0
10 99.0 100.0 101.0 100.0 1.0 | 10 99.0 100.0
11 99.0 100.0 101.0 100.0 1.0 | 11 99.0 100.0
12 99.0 100.0 101.0 100.0 1.0 | 12 99.0 100.0
13 99.0 100.0 101.0 100.0 1.0 | 13 99.0 100.0
14 99.0 100.0 101.0 100.0 1.0 | 14 99.0 100.0
15 99.0 100.0 101.0 100.0 1.0 | 15 99.0 100.0
16 99.0 100.0 101.0 100.0 1.0 | 16 99.0 100.0
17 99.0 100.0 101.0 100.0 1.0 | 17 99.0 100.0
18 99.0 100.0 101.0 100.0 1.0 | 18 99.0 100.0
19 99.0 100.0 101.0 100.0 1.0 | 19 99.0 100.0
20 99.0 100.0 101.0 100.0 1.0 | 20 99.0 100.0
|
Min 99.0 | Min
Max 101.0 | Max
|
ASTM E2810 Statistics | ASTM E2810 Stati
Overall Mean 100.00 | Overall Mean
Between Location STD DEV 0.00 | Between Location STD DEV
Within Location STD DEV 1.00 | Within Location STD DEV
|
Acceptance Limits |
1 Lower Err:502 | 1
0 Upper 1.0 | 0
Pass? Err:502 |
|
|
Variance Components | Variance Compon
Source Std Dev % Total | Source
Between Location Err:502 Err:502 | Between Location
Within Location 1.00 Err:502 | Within Location
Total Err:502 Err:502 | Total
 | 125
125 | 120
120 | 115
115 | 110
110
| 105
105

CU(%LC)
CU(%LC)

| 100
100
| 95
95
| 90
90
| 85
85
| 80
80
| 75
75 0 5 10 15
0 5 10 15 20 |
Location | Location
|
Weight Corrected
3 Mean Std Dev
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0
101.0 100.0 1.0

99.0
101.0

ASTM E2810 Statistics

Overall Mean 100.00
Between Location STD DEV 0.00
Within Location STD DEV 1.00

Acceptance Limits
Lower Err:502
Upper 1.0
Pass? Err:502

Variance Components
Source Std Dev % Total
Between Location Err:502 Err:502
Within Location 1.00 Err:502
Total Err:502 Err:502
5 10 15 20
Location
User Info (Unlocked Cells)
Uncorrected |
Location 1 2 3 4 5 6 7 Mean Std Dev | Location
1 | 1
2 | 2
3 | 3
4 | 4
5 | 5
6 | 6
7 | 7
8 | 8
9 | 9
10 | 10
11 | 11
12 | 12
13 | 13
14 | 14
15 | 15
16 | 16
17 | 17
18 | 18
19 | 19
20 | 20
|
Min |
Max |
|
ASTM E2810 Statistics |
Overall Mean |
Between Location STD DEV |
Within Location STD DEV |
|
Acceptance Limits |
### 0 Lower |
### 0 Upper |
Pass? |
|
|
Variance Components |
Source Std Dev % Total |
Between Location |
Within Location |
Total |
 |
125 | 125
120 | 120
115 | 115
110
| 110
105
CU(%LC)

| 105

CU(%LC)
100
| 100
95
| 95
90
| 90
85
| 85
80
| 80
75
0 5 10 15 20 | 75
Location |
|
 Weight Corrected
1 2 3 4 5 6 7 Mean Std Dev

Min
Max

ASTM E2810 Statistics

Overall Mean
Between Location STD DEV
Within Location STD DEV

Acceptance Limits
### 0 Lower
### 0 Upper
Pass?

Variance Components
Source Std Dev % Total
Between Location
Within Location
Total
 125
120
115
110
105
CU(%LC)

100
95
90
85
80
75
0 5 10 15 20
Location
User Info (Unlocked Cells)
Uncorrected | Weight Corrected
Location 1 2 3 Mean Std Dev | Location 1 2
1 | 1
2 | 2
3 | 3
4 | 4
5 | 5
6 | 6
7 | 7
8 | 8
9 | 9
10 | 10
11 | 11
12 | 12
13 | 13
14 | 14
15 | 15
16 | 16
17 | 17
18 | 18
19 | 19
20 | 20
21 | 21
22 | 22
23 | 23
24 | 24
25 | 25
26 | 26
27 | 27
28 | 28
29 | 29
30 | 30
31 | 31
32 | 32
33 | 33
34 | 34
35 | 35
36 | 36
37 | 37
38 | 38
39 | 39
40 | 40
 |
Min | Min
Max | Max
|
ASTM E2810 Statistics | ASTM E2810 Sta
Overall Mean | Overall Mean
Between Location STD DEV | Between Location STD DEV
Within Location STD DEV | Within Location STD DEV
|
Acceptance Limits |
#VALUE! Lower | #VALUE!
#VALUE! Upper | #VALUE!
Pass? |
|
|
Variance Components | Variance Compo
Source Std Dev % Total | Source
Between Location | Between Location
Within Location | Within Location
Total | Total
|
125
125 |
120
120 |
115
115 | 110
110 | 105
105
CU(%LC)

|
CU(%LC)

100
100
| 95
95
| 90
90
| 85
85
| 80
80
| 75
75
| 0 5 10 15 20 25 3
0 5 10 15 20 25 30 35 40
Location | Location
|
Weight Corrected
3 Mean Std Dev
 ASTM E2810 Statistics
Overall Mean
Between Location STD DEV
Within Location STD DEV

Acceptance Limits
Lower
Upper
Pass?

Variance Components
Source Std Dev % Total
Between Location
Within Location
Total

5 10 15 20 25 30 35 40

Location
Version Date Update Sheet Cell

01_02 1/12/2015 Initial N/A N/A

11/11/2015 1st CU 20X7 U41

1/30/2016 2nd Blend E29

10/14/2016 3rd 90CI-95Cov 20x3 Table BV3:CC33
10/14/2016 3rd 90CI-95Cov 40X3 Table CN3:CO19
4/22/2018 4th 90CI-95Cov 20x3 Table B3:E4, B5:C5
4/22/2018 4th 90CI-95Cov 40X3 Table B3:E4, B5:C5
 Change
Old
N/A
=IF(W36 = " "," ",SQRT(MAX(((W30*W30*3-W31*W31)/3), 0)))

=IF($B$20=0," ",IF($C$20=3*$B$20,MAX(SQRT((3*VAR(G5:G19)-E30*E30)/3),0)," "))

See N6:V38 on this sheet
See N44:O60 on this sheet
See N73:V80 on this sheet for changes
See N82:V89 on this sheet for changes
Change
New Comment
N/A N/A
=IF(W36 = " "," ",SQRT(MAX(((W30*W30*7-W31*W31)/7), 0))) Changed multiplier/divisor fr
Changed order of MAX and SQRT to co
between locationvariance component
=IF($B$20=0," ",IF($C$20=3*$B$20,SQRT(MAX(((3*VAR(G5:G19)-E30*E30)/3),0))," ")) correct in the other sheets)
See X6:AF36 on this sheet Removed Duplicate row and 100 100 li
See W44:X59 on this sheet Removed Duplicate row and 100 100 li
There are no changes don't to the Blen
CU20X7, or CU40X3 spreadsheets. See
explanation of change.
 Comment
N/A
Changed multiplier/divisor from 3 to 7 UPDATE 3
order of MAX and SQRT to correct for negative
ocationvariance component (This calculation is Old
the other sheets)
Duplicate row and 100 100 limits(not needed). 3.7 3.8
Duplicate row and 100 100 limits(not needed). SE LL UL LL UL
no changes don't to the Blend, CU20x3, 0.1 98.5 101.5 98.9 101.1
r CU40X3 spreadsheets. See lines 63-182 for
n of change. 0.2 98.5 101.5 98.9 101.1
0.3 98.5 101.5 98.9 101.1
0.4 98.5 101.5 98.9 101.1
0.5 98.5 101.5 98.9 101.1
0.6 98.5 101.5 99.0 101.0
0.7 98.6 101.4 99.0 101.0
0.8 98.6 101.4 99.0 101.0
0.9 98.6 101.4 99.0 101.0
1.0 98.6 101.4 99.0 101.0
1.1 98.7 101.3 99.1 100.9
1.2 98.7 101.3 99.1 100.9
1.3 98.7 101.3 99.1 100.9
1.4 98.8 101.2 99.2 100.8
1.5 98.8 101.2 99.2 100.8
1.6 98.9 101.1 99.3 100.7
1.7 98.9 101.1 99.3 100.7
1.8 99.0 101.0 99.4 100.6
1.9 99.0 101.0 99.4 100.6
2.0 99.1 100.9 99.5 100.5
2.1 99.2 100.8 99.6 100.4
2.2 99.2 100.8 99.6 100.4
2.3 99.3 100.7 99.7 100.3
2.4 99.4 100.6 99.8 100.2
2.5 99.5 100.5 99.9 100.1
2.6 99.6 100.4 100.0 100.0
2.7 99.7 100.3 100.0 100.0
2.8 99.7 100.3 . .
2.9 99.8 100.2 . .
3.0 99.9 100.1 . .
3.1 100.0 100.0 . .

Old
4.6
LL UL
 99.6 100.4
99.6 100.4
99.6 100.4
99.6 100.4
99.6 100.4
99.6 100.4
99.7 100.3
99.7 100.3
99.7 100.3
99.7 100.3
99.8 100.2
99.8 100.2
99.8 100.2
99.9 100.1
99.9 100.1
100.0 100.0
100.0 100.0

UPDATE 4
1) The original SAS program used a range of 84.9 to 115.1 to establish limits f
Due to the 98.5 -101.5 indifference zone in the USP 905 test to determine M,
2) The lower limit calculation started at 84.9 and increased by 0.1 until the pr
Then the upper limit calculation started at 115.1 and decreased by 0.1 until th
This caused the program to generate values outside 85 to 115 that were not c
The starting points should be 83.4 and 116.6 instead of 84.9 and 115.1.

3) The SAS program was corrected and rerun using wider limits that included
Below are the only cells in the entire 20x3 and 40x3 tables that change (The 2

Old
LL
Sheet SE SM Cell Value
0.1 B3 84.8
0.2 0.1 B4 84.7
20X3 Table 0.3 B5 84.6
0.1 D3 84.7
0.2
0.2 D4 84.6

Old
LL
Sheet SE SM Cell Value
0.1 B3 84.8
0.2 0.1 B4 84.7
40X3 Table 0.3 B5 84.6
0.1 D3 84.7
0.2
 40X3 Table

0.2
0.2 D4 84.6

4) Since the New values provide a wider range for the overall mean than the o
The changes could have an effect on a validation conclusion if the standard de
the within location standard deviation was less than or equal to 0.3%LC and th
and even if this were the case, the overall mean fell between the old and new

5) For users that use SAS to generate limits, the following minor edits (shown

%MACRO CALCUSP2;
DATA TABC;
D=&D1;
Z = PROBIT((1 + SQRT(&CILEVEL/100))/2);
NN = #
L = &LOC;
N = NN*L;
CALL SYMPUT("TOT",PUT(N, 5.0));
CHIERR = CINV(1 - SQRT(&CILEVEL / 100), L*(NN - 1));
CHILOC = CINV(1 - SQRT(&CILEVEL / 100),L-1);
SEBOUND = 9.2;
SMLIM = 9.2;
NEXTL = 83.4;
NEXTU = 116.6;
DO SE = D TO SEBOUND BY D;
MEANL = NEXTL;
MEANU = NEXTU;
SMBOUND = SMLIM;
SE2 = SE * SE;
H2 = L * (NN - 1) / CHIERR - 1;
SEC = ((1 - 1/NN)*H2*SE2)**2;
DO SM = D TO SMBOUND BY D;
IF MEANL = . THEN GOTO OVER;
SL2 = SM * SM * NN;
SL2UB = (L - 1) * SL2 / CHILOC;
H1 = (L - 1) / CHILOC - 1;
FIRST = ((1 / NN)*H1*SL2)**2;
PTEST = (1 / NN) * SL2 + (1 - 1/NN) * SE2;
VAR = PTEST + SQRT(FIRST + SEC);
MVAR = SL2UB;
SIGMA = SQRT(VAR);
DO MEAN = MEANL - D TO 116.6 BY D;
%cullu
IF OVERBDL > &LBOUND/100 THEN DO;
MEANL = MEAN;
 GOTO UPPER;
END;
END;
MEANL = .;
MEANU = .;
IF SE=D THEN DO;
SMLIM = SM - D;
OUTPUT;
SM=10;
GOTO OVER;
END;
IF SM=D THEN DO; SE = 10; GOTO OVER; END;
GOTO SKIP;
UPPER:

DO MEAN = MEANU + D TO 83.4 BY -D;

%cuulu
IF OVERBDU > &LBOUND/100 THEN DO;
MEANU = MEAN;
GOTO OUT;
END;
END;
OUT:
IF MEANU <= MEANL OR MEAN <= MEANL THEN DO;
MEANL = .;
MEANU = .;
IF SE=D THEN DO;
SMLIM = SM - D;
OUTPUT;
SM=10;
GOTO OVER;
END;
IF SM=D THEN DO; SE = 10; GOTO OVER; END;
END;

SKIP: OUTPUT;
IF SM = D THEN DO;
NEXTL = MEANL;
NEXTU = MEANU;
END;
OVER:
END;
END;
KEEP N NN L D MEAN SE SM MEANL MEANU OVERBDL OVERBDU;
Old New

3.9 4.0 3.7 3.8

LL UL LL UL SE LL UL LL UL
99.3 100.7 99.7 100.3 0.1 98.5 101.5 98.9 101.1
99.3 100.7 99.7 100.3 0.2 98.5 101.5 98.9 101.1
99.3 100.7 99.7 100.3 0.3 98.5 101.5 98.9 101.1
99.3 100.7 99.7 100.3 0.4 98.5 101.5 98.9 101.1
99.3 100.7 99.8 100.2 0.5 98.5 101.5 99.0 101.0
99.4 100.6 99.8 100.2 0.6 98.6 101.4 99.0 101.0
99.4 100.6 99.8 100.2 0.7 98.6 101.4 99.0 101.0
99.4 100.6 99.8 100.2 0.8 98.6 101.4 99.0 101.0
99.4 100.6 99.8 100.2 0.9 98.6 101.4 99.0 101.0
99.5 100.5 99.9 100.1 1.0 98.7 101.3 99.1 100.9
99.5 100.5 99.9 100.1 1.1 98.7 101.3 99.1 100.9
99.5 100.5 99.9 100.1 1.2 98.7 101.3 99.1 100.9
99.6 100.4 100.0 100.0 1.3 98.8 101.2 99.2 100.8
99.6 100.4 100.0 100.0 1.4 98.8 101.2 99.2 100.8
99.6 100.4 100.0 100.0 1.5 98.9 101.1 99.3 100.7
99.7 100.3 . . 1.6 98.9 101.1 99.3 100.7
99.7 100.3 . . 1.7 99.0 101.0 99.4 100.6
99.8 100.2 . . 1.8 99.0 101.0 99.4 100.6
99.8 100.2 . . 1.9 99.1 100.9 99.5 100.5
99.9 100.1 . . 2.0 99.2 100.8 99.6 100.4
100.0 100.0 . . 2.1 99.2 100.8 99.6 100.4
100.0 100.0 . . 2.2 99.3 100.7 99.7 100.3
. . . . 2.3 99.4 100.6 99.8 100.2
. . . . 2.4 99.5 100.5 99.9 100.1
. . . . 2.5 99.6 100.4 . .
. . . . 2.6 99.7 100.3 . .
. . . . 2.7 99.7 100.3 . .
. . . . 2.8 99.8 100.2 . .
. . . . 2.9 99.9 100.1 . .
. . . . 3.0 . . . .
. . . .

New
4.6
LL UL
 99.6 100.4
99.6 100.4
99.6 100.4
99.6 100.4
99.6 100.4
99.7 100.3
99.7 100.3
99.7 100.3
99.7 100.3
99.8 100.2
99.8 100.2
99.8 100.2
99.9 100.1
99.9 100.1

.9 to 115.1 to establish limits for the overall CU mean.

USP 905 test to determine M, range should have been 83.5 to 116.5.
nd increased by 0.1 until the probability of passing the USP UDU test was greater than 95%.
1 and decreased by 0.1 until the probability of passing the USP UDU test was geater than 95%
tside 85 to 115 that were not correct.
stead of 84.9 and 115.1.

sing wider limits that included 83.5 to 116.5.

40x3 tables that change (The 20X7 Table Entrees remained the same).

Old New
UL LL UL
Cell Value Value Value
C3 115.2 84.0 116.0
C4 115.3 84.2 115.8
C5 115.4 84.4 115.6
E3 115.3 84.4 115.6
E4 115.4 84.5 115.5

Old New
UL LL UL
Cell Value Value Value
C3 115.2 84.0 116.0
C4 115.3 84.1 115.9
C5 115.4 84.3 115.7
E3 115.3 84.3 115.7
 E4 115.4 84.4 115.6

for the overall mean than the old values, any batch or validation that met the old limits would meet the new limits.
n conclusion if the standard deviation of location means was less than or equal to 0.2%Label Claim,
than or equal to 0.3%LC and the mean of all the CU results was less than 85.0% or greater than 115.0%LC
n fell between the old and new limits which are less than 85 or greater than 115 which most likely would be considered unacceptable.

e following minor edits (shown in red) in the CALCUSP2 macro are required to obtain the new limits are shown below in red:
VERBDL OVERBDU;
New

3.9 4.0
LL UL LL UL
99.3 100.7 99.7 100.3
99.3 100.7 99.7 100.3
99.3 100.7 99.7 100.3
99.3 100.7 99.8 100.2
99.4 100.6 99.8 100.2
99.4 100.6 99.8 100.2
99.4 100.6 99.8 100.2
99.4 100.6 99.8 100.2
99.5 100.5 99.9 100.1
99.5 100.5 99.9 100.1
99.5 100.5 99.9 100.1
99.6 100.4 . .
99.6 100.4 . .
99.6 100.4 . .
99.7 100.3
99.7 100.3 . .
99.8 100.2 . .
99.8 100.2 . .
99.9 100.1 . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
. . . .
nsidered unacceptable.

below in red: