Chapter

Contents

7 Milk Meter Calibration ................................................................................................. 3

7.1 Calibrating Electrodes ........................................................................................... 3
7.1.1 Milk Meter Electrodes Calibration tool ............................................................ 3
7.1.1.1 Electrodes Calibrating Procedure .............................................................. 4
7.1.1.2 Reviewing Calibrated Values of Electrodes ............................................... 5
7.2 Air intake compensation....................................................................................... 5
7.2.1 Checks before Calibration ................................................................................ 6
7.2.2 Measuring Claw Air Admission ......................................................................... 7
7.2.3 Default G and H parameters per air intake ...................................................... 8
7.2.4 Adjusting G and/or H parameter value ............................................................ 8
7.2.5 Adjusting the G and/or H parameter value via communication ...................... 9
7.3 Accuracy Adjustment .......................................................................................... 10
7.3.1 Use of bulk tank and fine adjustment and references by water test ............ 11
7.3.1.1 Equipment and water test preparation and description ........................ 12
7.3.1.2 Water test adjustment and setting reference ........................................ 15
7.3.2 Milk test for calibration and references by water test .................................. 17
7.3.2.1 Collecting the Milk into a Milk Bucket ..................................................... 17
7.3.2.2 Statistical Accuracy Calculation ............................................................... 18
7.3.2.3 Setting the reference value by water test ............................................... 19
7.3.3 Milk test for calibration (two sessions) and references by water test .......... 21
7.3.3.1 Collecting the Milk into a Milk Bucket ..................................................... 21
7.3.3.2 Statistical Accuracy Calculation ............................................................... 22
7.3.3.3 Setting the reference value by water test ............................................... 23

2 AfiMilk MPC Installation Manual

7 Milk Meter Calibration
Afimilk MPC is calibrated in the factory to ensure milk measurement results within the
required accuracy ranges. However, since installation environment may affect milk
measurement accuracy, Afimilk MPC requires calibration at installation.
Three calibration procedures are described in this chapter:
1. Calibrating Electrodes
2. Air intake compensation
3. Accuracy adjustments
(on farm calibration)

7.1 Calibrating Electrodes

Three electrodes in the milk meter
body are the measuring probes of the
meter. Although electrodes are
calibrated at the factory, cable and
connection resistance may affect their
accuracy. Electrodes calibration
compensates for resistance of the lines
and therefore, it is necessary to
calibrate the milk meter after
installation.
Electrodes calibration procedure is
performed with a 100Ω (±0.1%)
resistor. During calibration, the
resistor is connected between
electrodes to test and calibrate them.
Afimilk offers a calibration tool for the
technician incorporating the resistor
and probes for hooking to the
electrodes (See below).

7.1.1 Milk Meter Electrodes Calibration tool

The easy to use calibration tool is built of an encased resistor
connected to two socket connectors used for plugging on the
calibrated electrodes.

3 AfiMilk MPC Installation Manual

7.1.1.1 Electrodes Calibrating Procedure
At each meter, calibrate the holding and operating electrodes in one single procedure, as
described below:
Holding (middle) electrode:
Calibrate electrodes as follows:
1. Carefully unlock the bottom clasp of the body and remove the lower cover with the
inner valve cylinder connected to it.

2. Connect the 100Ω resistor between the Common (lower) and the holding (middle)
electrodes.

3. Ensure that the meter is in wash mode (illuminated cleaning key indicator ).

4. Press and to access electrode calibration mode.

AdJt appears on the display panel.

5. Press .

H### is displayed (### represents the calibration value). The range of calibration
values for Holding electrode is typically between H112 and H118.

WARNING

Typically, calibration values displayed should be stable or vary

by only one digit. Out of range or unstable figures indicates an
electrical problem. Refer to the section “Searching for an
Electrical Problem,” for solving the problem.

6. Press again, to calibrate the Holding electrode and save the value.

AdJt is displayed.

4 AfiMilk MPC Installation Manual

Operating (top) electrode:
Connect the 100Ω resistor between the Operating (top) electrode and the Common (lower)
electrode.
1. Press .

2. H### is displayed (### represents the calibration value).

3. Press a second time to calibrate the electrode.

AdJt appears.
4. The electrodes have been calibrated.

5. Press to exit editing mode.

7.1.1.2 Reviewing Calibrated Values of Electrodes

If a milk meter fault leads to a suspected calibration problem, the technician may check
calibration value at the display:

1. Ensure that Afimilk MPC is in wash mode (wash LED illuminated ).

2. Type 2580 and press to access programming mode.

3. Press Group access number - 4 to enter Calibration Group
4. Scroll down to oA (Operating Electrode).

5. oA## is displayed. (## represents the calibration value.) If the resulting value is less
than 17 or more than 34, electrode calibration is required.

6. Press to scroll to calibration value oH (Holding Electrode).

oH## is displayed. (## represents the calibration value.) If the resulting value is less
than 17 or more than 33, electrode calibration is required.

7. Press + (double press) to return to wash mode.

7.2 Air intake compensation

Milking clusters are manufactured with orifices allowing admission of air for improving
milk flow inside the milking system tubing. Air intake of these orifices generally varies

5 AfiMilk MPC Installation Manual

from 6 to 12 liters per minute. Afimilk MPC calibration neutralizes air intake influence.
Calibration is required whenever clusters are changed and/or when air vented liners
are used.

Calibration procedure adjusts the value of parameters "G" (Air admission Adjustment)
and "H" according to the air admission of the claws orifice.

7.2.1 Checks before Calibration

A thorough visual check of the milking parlor is necessary before calibrating. Complete
the following procedure:
1. Check the milking equipment and pipe connections between the milk clusters and the
milk lines. At this point, unless all milking tubing is new, AfiMilk recommends
replacing rubber liners, as well as rubber gaskets and cracked (usually plastic
transparent) parts of used milking clusters.

2. Ensure an open air orifice in each cluster.

3. Check the vacuum lines. Clean and drain.

4. Check the milk meter power supply voltage at the last milking point of each parlor
side. A minimum of 23Vac is required while operating at full capacity.
NOTE

Steps 5 through 9 below are maintenance steps and may be skipped at initial
calibration (new milk meters).

5. Replace all milk-meter rubber parts if not new.

6. Use a spirit level to check the levelness of the length and width of the top rim of the
milk meter body mid-section.

7. Remove the body's solenoid diaphragm and ensure that it is complete and flawless
(replace it if there is any minor difference or flaw). Check for moisture on the
underside of the diaphragm. If moist, dry it.

8. Blow pressurized air into the solenoid from its diaphragm end. Check for clean and
free airflow to the bottom opening of the solenoid.

9. Check all parts of the milk meter body, hoses, and connectors.

6 AfiMilk MPC Installation Manual

7.2.2 Measuring Claw Air Admission
For calibration purposes an exact claw air admission measurement is required. Although
claw manufacturers specify products' air admission, it is highly recommended to verify
actual performance of each claw on-site.

This method incorporates the use of an air gauge to measure Air Admission. Afimilk
recommends an air gauge manufactured by Dwyer Instruments Inc., the RATEMASTER®
FLOWMETER RMA-22-SSV.
Follow the procedure described below if an air gauge is available to ensure accurate
results.
1. Remove the milk and pulsation hoses from the cluster as illustrated in Figure 12
below.

2. Insert four plugs into the four milking cups.

3. Clean the air orifice in the cluster.

4. Connect the air gauge as illustrated in Figure 12 below. The lower port of the air
gauge is connected to the claw via a short 5/8 inch hose connector. The top outlet
port of the air gauge is inserted into the milk hose leading to the milk meter body.

Figure 12: Air Gauge Installation

5. When the gauge is securely connected as described, press on the display panel in
order to open vacuum to the cluster.

6. Check and seal any vacuum leaks at connections in the air gauge and cluster.

7. Block the air orifice with your finger to verify that the only air intake allowed is via the
air orifice. The ball in the gauge should fall to zero if no air leakage occurs.
8. Hold the air gauge in a vertical position and allow air admission via claw orifice. Write
down the air intake the gage measures. This measurement will be used for calibrating
the “G” parameter value as described below.

7 AfiMilk MPC Installation Manual

Repeat measurement procedure at each milking point.

7.2.3 Default G and H parameters per air intake

The table shows the default G and H parameters according to measured air admission in the
cluster.

Air Admission in the G value for calibration H value for

Cluster calibration
(Liters per minute)

6–7 40 30

7–8 46 37

8–10 49 (default) 40

11–13 55 46

14–17 60 50

18–20 65 55

To adjust G and H parameters follow the procedure in paragraph Adjusting G and/or H

parameter value or via the AfiControl interface mentioned in paragraph Adjusting the G
and/or H parameter value via communication.

7.2.4 Adjusting G and/or H parameter value

o Ensure that Afimilk MPC is in wash mode (wash LED illuminated ).

o Type 2580 and press to access programming mode.

o Press Group access number 4 to enter Calibration Group. G is the first
parameter in the group and its value will appear after a short delay.

o Press -Gflashes indicating that editing is enabled. Type the desired

value.

o Press to confirm.
o Scroll down to the second parameter in the group – H and change its value in
the same manner as explained above.

8 AfiMilk MPC Installation Manual

 o Press to confirm.

o When done, press twice to exit editing mode.

7.2.5 Adjusting the G and/or H parameter value via communication

The easiest way to adjust G and H parameters values is via the AfiControl interface. As
explained below, you may change the value of each milking point individually OR set a
uniform value to ALL the milk meters.

To change the G and H values via AfiFarm PC enter Parameters➢Tools➢Parlor and select
the stall to adjust as shown below.

On the right hand side section of the screen, adjust the G and H values for the selected
stalls.

9 AfiMilk MPC Installation Manual

7.3 Accuracy Adjustment
AfiMilk MPC milk meters are supplied with default adjustment calibration values. After the
initial setup (see adjustment per air intake explained above) inaccuracies of milk
measurements vs. bulk tank OR in specific milk meters can be found; therefore G and H
parameters values may need further adjustment. Initial calibration (after installation) is
performed by adjusting the G parameter value. If additional – fine tune calibration – is
required for individual milk meters, update the H parameter value according to the
procedure explained here.

Local authorities may require an initial calibration and periodic accuracy checks (usually
yearly) of the milk meters, mainly for herd book approval. This can be done by three
different methods. These methods will be described here. The periodic accuracy check or
periodic routine test is usually done as a water test which will be compared to water test
reference values set during or after the initial calibration (see document Afimilk MPC
Periodic routine test).
o Compaire between bulk tank vs. milk measurements with fine adjustment with a
water test (minimal required for official recording).
Based on the comparison, the G parameter is set equal for all meters. A water test is
fulfilled to check differences (i.e. faulty meters) and set a farm avarage as the
reference value. Based on the individual meter results (minimum of two
measurements) a G parameter adjustment for each meter is done. If adjustment is
needed for more then 5 points (2%), a minimum two new water tests are required
and are then used as the reference value for this meter.
o Milk test for calibration and meter adjustment followed by a water test for future
reference values (recommanded by Afimilk).
Based on a milk test of at least three valid readings the G parameter will be adjusted.
After all milk meters are calibrated, a water test is performed for each meter. This will
be used as reference value.
o Milk test for calibration and meter adjustment (advanced calibration with H-
parameter).
Based on a first milk test of at least three valid readings the G parameter will be
adjusted. A second milk test, which might include milk speed data, is used to adjust
the H parameter. After all milk meters are calibrated, a water test is performed for
each meter. This will be used as the reference value.

10 AfiMilk MPC Installation Manual

7.3.1 Use of bulk tank and fine adjustment and references by water test
Bulk tank calibration is used for adjusting average accuracy of all the milk meters together.
This procedure is performed by calculating the parlor difference between the milk measured
by the milk meters and the bulk tank total. This results in an average G parameter at which
the bulk tank and the total of the milk recorded by the Afikim MPC meters is more or less
equal.
Before performing the test the following parlor tests and checks should be carried
out:
o Milk meters are cleaned properly. Verify functionality and operation of each unit.
o Consumable rubber parts (upper and lower gaskets, diaphragm and valve gaskets)
are replaced as recommended, and seem flawless and clean.
o Clean air line is drained of water, milk meter solenoid air inlet is properly connected
to the clean air line and to the vacuum supply.
o Test the vacuum level and stability. Ensure vacuum level above 30 kPa.

Below an example of using data of bulk tank and total milk yield recorded

Description Amount
The amount collected by the milk tanker 12.480 Kg
Milk consumption locally (not loaded in tanker) 100 Kg
Thru milk meter but not in bulk tank (i.e. bucket collection) -150 Kg
Total milk amount 12.430 Kg

The milk amount recorded by all AfiMilk MPC in the same period 12.315 Kg

Previous G parameter value (i.e. based on air blend) 50

New G parameter (50 +(12.430/12.315) x 2,5)) 1) 53
1) milk meters show almost 1% less than bulk tank collection
So modify the G parameter: -1 x (-2,5)=+2,5 to 53 (rounded)

Note - Use the same measurement unit

- Make sure to be accurate in adding and subtracting
- For further expenation calculating difference and adjustment (see Statistical Accuracy Calculation)

11 AfiMilk MPC Installation Manual

 IMPORTANT

Make sure you work accurate with collecting all the data you need.
This includes i.e. missing data due to missing cow identification, milk thru
the meter but not in the bulk tank and milk substraction
. out of bulk tank.

To adjust the G parameter valua follow the procedure in paragraph Adjusting G and/or H
parameter value or via the AfiControl interface mentioned in paragraph Adjusting the G
and/or H parameter value via communication.

7.3.1.1 Equipment and water test preparation and description

7.3.1.1.1 Equipment required

For performing the AfiMilk MPC water test, the following
equipment is used:

Milking bucket equipped with 7/8” inlet and vacuum ports

Manual vacuum shut off valve placed between the milk line
and the hose connected to the collection bucket.

Suction pipe with an inlet hole of 3.5mm, which allows a 4 ± 1 LPM (Liter per Minute)
flow of water. The entire set should be
tested and verified for the defined flow
rate (see testing the equipment
below). This part will replace the
milking claw for the purpose of the
test.

12 AfiMilk MPC Installation Manual

A nipple with air inlet 1mm in diameter should provide
air intake of 10 ± 1 LPM.
Air inlet nipple should be connected no closer than 50cm
from the water sucking port.

Hoses and connectors to set-up the test system – the

milking bucket should have an inlet port of 20mm or
more. Hence, the internal diameter of piping connecting
between milk meter and bucket must be at least 20mm.
Water pail containing at least 20 liters water volume.
Electronic weighing scales.

7.3.1.1.2 Testing the equipment

Prior to the water test, the equipment (sucking pipe and air inlet) needs to be tested
and confirmed for running the desired water flow rate for the test (4 ± 1 LPM) and for
precise air intake of the air hole (10 ± 1 LPM).

Testing the water flow

To test the sucking pipe, a water flow gauge is used.

When testing the equipment, the water flow gauge is

connected between the sucking pipe and air inlet nipple,
before the milk meter inlet. When the vacuum is opened,
water starts running through the system and the water flow is
read (as shown in picture on the right).

Once confirmed for flow rate in the range of 4 ± 1 liters per

minute, the water flow gauge is no longer used during the
actual water test.

13 AfiMilk MPC Installation Manual

Testing the air intake
To test the air inlet an air flow gauge is used

To perform the test, place the air flow gauge in place of the water
flow gauge.

Block the water inlet to allow air flow through the air inlet alone.
Open the vacuum valve and register the air intake measured.
o Verify that the air intake measured is in the range of 10 ± 1
liters per minute.

7.3.1.1.3 The water test procedure

Preparing the testing solution
Prepare the test solution as follows:

Measure a quantity of 13-20 liter of water for the

test solution. Use lukewarm water at temperature as
close as possible to the ambient temperature.
Calculate amount of salt and dissolve 4.5 grams of
table salt in each liter of water. Stir carefully to
ensure that all salt is dissolved.
Measure the conductivity of the solution using a
conductivity meter (see picture on the right). The
target is to achieve conductivity between 10 and 12
Mili-Siemens. If required, add water or salt to reach
the desired conductivity.

Testing procedure
The basic steps of the water test
o Produce 20 kg of testing solution in a pail.
o Using the sucking pipe connected to the milk meter, run testing solution through the
unit and into the milking bucket until the display shows 12 ± 0.5 kg.
o Stop the vacuum using the manual valve and press the “removal” button of the milk-
meter.
o Wait until the milk-meter valve opens to release the water accumulated during the
last flow.
o The collected quantity of test solution should now be weighed and recorded against
the figure shown on the display of the milk meter.

14 AfiMilk MPC Installation Manual

7.3.1.2 Water test adjustment and setting reference

7.3.1.2.1 Global Test description

o Perform one (or direct two consecutive) water tests procedures as described above
(same amount of water) for each milk meter in the parlor.
o Check if on or more meters deviate in result against the other meters (result
difference > 0,25 kg on average difference of all meters). If so, re-check this meter(s)
and approve result or perform a re-test.

Example form to store and analyse the collected data

Afimilk MPC Setting reference values by water test
Farm: 1265897 42,0 Parlor: Harringbone (2x5)
Technical check Vacuum level (kPa)
Farm tesing Hygienic check Date: 30-7-2018
Cow road Performer: Test person Joe
two measurements with mutual difference max. 0,1 kg
Cow city three within 0,2 ( else replace faulty measurements)
Pre watertest Conduc- 2e water test Conduc-
Meter G-value Display Unster Diff. tivity afwijking
Display Unster Diff. tivity
1 53 12,20 11,44 0,76 11,7 12,32 11,48 0,84 11,8
2 53 12,30 11,56 0,74 11,6 12,26 11,46 0,80 11,8
3 53 12,25 11,55 0,70 11,6 12,20 11,48 0,72 11,7
4 53 12,28 11,20 1,08 11,6 12,32 11,20 1,12 11,6
5 53 12,30 11,48 0,82 11,5 12,40 11,60 0,80 11,6
6 53 12,44 11,68 0,76 11,8 12,30 11,48 0,82 11,9
7 53 12,18 11,26 0,92 11,7 12,46 11,66 0,80 11,8
8 53 12,20 11,40 0,80 11,7 12,28 11,46 0,82 11,7
9 53 12,24 11,46 0,78 11,6 12,26 11,50 0,76 11,6
10 53 12,16 11,36 0,80 11,5 12,20 11,34 0,86 11,6
Avarage water test diff. 0,82

In this example, two measurements are done. Meter no. 4 shows a bigger difference as
all other meters. This meter is re-checked and considered to be Ok. This value is
accepted.

o If within the milk meter the two water tests are differ no more then 0,1 kg, that
meter is Ok. If the difference is more then 0,1 kg a third measurement should be
performed. The three measurements should differ no more then 0,2 kg. If so, re-test
and replace one or more existing measurements.

15 AfiMilk MPC Installation Manual

Afimilk MPC Setting reference values by water test
Farm: 1265897 42,0 Parlor: Harringbone (2x5)
Technical check Vacuum level (kPa)
Farm tesing Hygienic check Date: 30-7-2018
Cow road Performer: Test person Joe
two measurements with mutual difference max. 0,1 kg
Cow city three within 0,2 ( else replace faulty measurements)
Pre water test Conduc- 2e water test Conduc- 3e water test Conduc- Avarage Calculated
Meter G-value Display Unster Diff. tivity afwijking
Display Unster Diff. tivity Display Unster Diff. tivity Diff Deviation% G- value
1 53 12,20 11,44 0,76 11,7 12,32 11,48 0,84 11,8 0,80 -0,2% 53
2 53 12,30 11,56 0,74 11,6 12,26 11,46 0,80 11,8 0,77 -0,5% 54
3 53 12,25 11,55 0,70 11,6 12,20 11,48 0,72 11,7 0,71 -0,9% 55
4 53 12,28 11,20 1,08 11,6 12,32 11,20 1,12 11,6 1,10 2,2% 47
5 53 12,30 11,48 0,82 11,5 12,40 11,60 0,80 11,6 0,81 -0,1% 53
6 53 12,44 11,68 0,76 11,8 12,30 11,48 0,82 11,9 0,79 -0,3% 53
7 53 12,18 11,26 0,92 11,7 12,46 11,66 0,80 11,8 12,40 11,52 0,88 11,7 0,87 0,3% 53
8 53 12,20 11,40 0,80 11,7 12,28 11,46 0,82 11,7 0,81 -0,1% 53
9 53 12,24 11,46 0,78 11,6 12,26 11,50 0,76 11,6 0,77 -0,5% 54
10 53 12,16 11,36 0,80 11,5 12,20 11,34 0,86 11,6 0,83 0,0% 53
Avarage water test diff. 0,82 Farm reference wil be: 0,83

In this example, the two measurements for meter no. 7 differ more then 0,1. A third water
test is performed. All three differences are within the 0,2 limit, so no futher action needed.

o Based on de water test differences a new G parameter can be calculated (Statistical

Adjustment Calculation). If the new G parameter differs more then 5 (meaning >2%
deviation and adjustment), two additional water tests for that specific meter are
required after the adjustment of G in order to set references (in this example meter
4).
o Set de new G parameter in the individual meter using the G adjustment procedure
(see links below)

For meter no. 4 the adjustment is more than 2% and two new water tests are needed.
In this example the new water test looks like:

1e water test Conduc- 2e water test Conduc- Meter

Meter G-value Display Unster Diff. tivity afwijking
Display Unster Diff. tivity Reference
4 47 12,10 11,40 0,70 11,8 12,16 11,48 0,68 11,8 0,69

To adjust the G parameter valua follow the procedure in paragraph Adjusting “G” and/or
“H” parameter value or via the AfiControl interface mentioned in paragraph Adjusting the
“G” and/or “H” parameter value via communication.

For the periodic routine test, the following results, following this example, have to be
handed over to the persons/organizations involved in routine testing (reference values for
the next annual routine test :

Form with the calculated reference values for the annual routine test based on the
calibration test.

16 AfiMilk MPC Installation Manual

Afimilk MPC Periodic Routine test
Farm: 1265897 Technical check Vacuum level (kPa) Parlor: Harringbone (2x5)
Farm tesing Hygienic check - at reference determine 42,0 Date:
Cow road G-/H-value check - this routine test Performer:
Cow city Against ref. val.: One test: limit 0,1 kg; two tests: avg 0,2 kg
Nr. G-value H-value Air intake Ref.value Cond. Display Unster Diff. Display Unster Diff. Conclusion Remarks
1 53 40 9 0,83 11,8 0,00 0,00
2 54 40 9 0,83 11,7 0,00 0,00
3 55 40 9 0,83 11,7 0,00 0,00
4 47 40 9 0,69 11,8 0,00 0,00
5 53 40 9 0,83 11,6 0,00 0,00
6 53 40 9 0,83 11,9 0,00 0,00
7 53 40 9 0,83 11,7 0,00 0,00
8 53 40 9 0,83 11,7 0,00 0,00
9 54 40 9 0,83 11,6 0,00 0,00
10 53 40 9 0,83 11,6 0,00 0,00

7.3.2 Milk test for calibration and reference values by water test
This test is performed by collecting milk of individual cows into a bucket, measuring it and
comparing the amount with the reading of the milk meter. Then, statistical deviation
calculations are done to indicate the accuracy level of the device.
It is very important to carry out a proper scaling procedure to ensure accurate scaling in
accordance with regulations.
To carry out the scaling procedure, an electronic weighing scale and a few milk buckets are
required. This chapter describes the correct procedure of scaling and statistical calculations.
If deviations in measurements are found, correct those by adjusting G&H values as
described above.

Before performing the test the following parlor tests and checks should be carried
out:
o Milk meters are cleaned properly. Verify functionality and operation of each unit.
o Consumable rubber parts (upper and lower gaskets, diaphragm and valve gaskets)
are replaced as recommended, and seem flawless and clean.
o Clean air line is drained of water, milk meter solenoid air inlet is properly connected
to the clean air line and to the vacuum supply.
o Test the vacuum level and stability. Ensure vacuum level above 30 kPa.

7.3.2.1 Collecting the Milk into a Milk Bucket

Per international regulations the following rules are adopted for scaling:
o Multiple attachments measurements are not included in the statistics.
o Only milking claw air admission is allowed. Tests may not be valid if liners cracks or
piping holes allow additional air intake by the system. Furthermore, tests where
liner slippage or large udder causing squawks (liner slip) occurs are also rejected.

17 AfiMilk MPC Installation Manual

 o The Milk Meter reading should be taken after the last dump of milk (milk dump after
removal).
o For each reading, list milk meter reading vs. actual milk weighed.
o At least three valid readings are required for calculating milk meter accuracy.
To ensure proper measurements, see that the following restrictions are kept:
o The collecting bucket must be placed below the meter outlet port.
o The hose connecting the Milk Meter to the bucket is situated correctly in a
downward slope towards the bucket.
o The hose connecting the Milk Meter to the bucket has no pulling or pushing
pressure on the outlet of the Milk Meter.

7.3.2.2 Statistical Accuracy Calculation

After scaling milk weights for all the milk meters in the parlor, statistical accuracy
calculation is performed for each milk meter.
Statistical calculation is done for each reading and corresponding weigh scale measurement
for calculating the difference as a percentage value between the reading of the milk meter
and the actual milk weighed.
The following formula is used for calculating difference between the milk meter reading and
the milk weighed.

Average all of the DIF% deviations for each Milk Meter.

Change the value of "G" parameter per calculated average deviation. If the result is negative
(Milk meter reading is lower than the actual weight), the calculated adjustment value needs
to be added to the value of G. If the adjustment is positive, the calculated value needs to be
subtracted from the value of G.
General adjustment calculation →%Dev X (–2.5) = Adjustment value
Example: if the actual value of G in the milk meter is 55 and the measured deviation -4%
(Milk meters show 4% less than actual weight), modify the “G” parameter value
by:
-4 X (-2.5) = +10 to 65

18 AfiMilk MPC Installation Manual

To adjust the G parameter valua follow the procedure in paragraph Adjusting “G” and/or
“H” parameter value or via the AfiControl interface mentioned in paragraph Adjusting the
“G” and/or “H” parameter value via communication.
During initial calibration it is the aim to adjust each meter around 0% deviation, while this is
the base for setting the reference values with water.

Example form for data collection of milk meter data compared with collected milk in a
bucket

Afimilk MPC Initial milk test

Farm: 1265897 Technical check Vacuum level (kPa) 42,0 Parlor: Harringbone (2x5)
Farm tesing Hygienic check Date: 30-7-2018
Cow road Minimal three measurements Performer:
Test person Joe
Cow city Aim for 0% deviation as base for water test
Nr. Display Unster Diff. G-value Concl. Remarks Nr. Display Unster Diff. G-value Concl. Remarks
1 8,35 8,46 -1,3% 50 6 13,26 13,65 -2,9% 50
15,73 15,55 1,2% 50 16,90 17,30 -2,3% 50
20,22 20,20 0,1% 50 Ok 16,15 16,20 -0,3% 56 After adjustment
17,78 17,70 0,5% 56 Ok

2 14,86 15,20 -2,2% 50 7 19,36 19,10 1,4% 50

18,92 19,30 -2,0% 50 18,42 18,30 0,7% 50
15,24 15,15 0,6% 55 After adjustment 15,96 16,00 -0,2% 52 After adjustment
17,40 17,36 0,2% 55 Ok 14,56 14,50 0,4% 52 Ok

7.3.2.3 Setting the reference value by water test

An initial water test is performed after installation of the milk meters, after calibration of
the system or during installation of the entire system. The results of this water test are
the “reference values” for the periodic water tests performed for each milk meter
annually or more frequently.

Please note that the initial water test is performed to establish a baseline and the
following water tests at the farm are conducted to validate stability of the system. Since
the device is calibrated for measuring milk rather than water, the deviations between
water measurements and the scaling of the fluid are the baseline tested, rather than
ultimate accuracy of measurements (as calculated in milk tests).

Preparation of the needed equipment and how to perform the water test is described in
paragraph 7.3.1.1 Equipment - and water test preparation and description.

At least two measurements per meter with a mutual difference of max. 0,1 kg are recorded.
The reference value for each meter is the average of these two measurements.

19 AfiMilk MPC Installation Manual

Example form for collecting reference values for the annual water test
Afimilk MPC Setting reference values by water test
Farm: 1265897 42,0 Parlor: Harringbone (2x5)
Technical check Vacuum level (kPa)
Farm tesing Hygienic check Date: 30-7-2018
Cow road Performer: Test person Joe
two measurements with mutual difference max. 0,1 kg
Cow city three within 0,2 ( else replace faulty measurements)
Nr. G-value H-value Display Unster Diff. Cond. Display Unster Diff. Cond. Ref. value Cond. Remarks
1 50 40 12,10 11,44 0,66 11,7 12,22 11,48 0,74 11,8 0,70 11,75
2 55 40 12,20 11,36 0,84 11,6 12,16 11,36 0,80 11,8 0,82 11,70
3 53 40 12,15 11,55 0,60 11,6 12,10 11,48 0,62 11,7 0,61 11,65 More measurements needed
4 58 40 12,18 11,20 0,98 11,6 12,22 11,20 1,02 11,6 1,00 11,60
5 52 40 12,20 11,48 0,72 11,5 12,30 11,60 0,70 11,6 0,71 11,55
6 56 40 12,34 11,48 0,86 11,8 12,20 11,38 0,82 11,9 0,84 11,85
7 52 40 12,08 11,46 0,62 11,7 12,36 11,76 0,60 11,8 0,61 11,75
8 50 40 12,10 11,40 0,70 11,7 12,18 11,46 0,72 11,7 0,71 11,70
9 52 40 12,14 11,46 0,68 11,6 12,16 11,50 0,66 11,6 0,67 11,60
10 51 40 12,06 11,36 0,70 11,5 12,10 11,34 0,76 11,6 0,73 11,55

For the periodic routine test the following results, following this example, would be handed
over to the persons/organizations involved in routine testing:

Form with the calculated reference values for the annual routine test based on the
calibration test.

Afimilk MPC Periodic Routine test

Farm: 1265897 Technical check Vacuum level (kPa) Parlor: Harringbone (2x5)
Farm tesing Hygienic check - at reference determine 42,0 Date:
Cow road G-/H-value check - this routine test Performer:
Cow city Against ref. val.: One test: limit 0,1 kg; two tests: avg 0,2 kg
Nr. G-value H-value Air intake Ref.value Cond. Display Unster Diff. Display Unster Diff. Conclusion Remarks
1 50 40 9 0,70 11,8 0,00 0,00
2 55 40 9 0,82 11,7 0,00 0,00
3 53 40 9 0,61 11,7 0,00 0,00
4 58 40 9 1,00 11,6 0,00 0,00
5 52 40 9 0,71 11,6 0,00 0,00
6 56 40 9 0,84 11,9 0,00 0,00
7 52 40 9 0,61 11,8 0,00 0,00
8 50 40 9 0,71 11,7 0,00 0,00
9 52 40 9 0,67 11,6 0,00 0,00
10 51 40 9 0,73 11,6 0,00 0,00

20 AfiMilk MPC Installation Manual

7.3.3 Milk test for calibration (two sessions) and references by water test
This test is performed by collecting the milk each cow gives into a bucket, measuring it and
comparing it to the reading of the milk meter. Then, statistical deviation calculations are
done to indicate the accuracy level of the device.

It is very important to carry out a proper scaling procedure to ensure accurate scaling in
accordance with regulations.
To carry out the scaling procedure, an electronic weigh scale and a few milk buckets are
required. This chapter describes the correct procedure of scaling and statistical calculations.
If deviations in measurements are found, correct those by adjusting G and H values as
described below.

Before performing the test, the following parlor tests and checks should be carried
out:
o Clean milk meters properly. Verify functionality and operation of each unit.
o Replace consumable rubber parts (upper and lower gaskets, diaphragm and valve
gaskets) as recommended, and flawless and clean.
o Drain the clean air line of water; milk meter solenoid air inlet is properly connected
to the clean air line and to the vacuum supply.
o Test the vacuum level and stability. Ensure vacuum level above 30 kPa.

7.3.3.1 Collecting the Milk into a Milk Bucket

Per international regulations the following rules are adopted for scaling:
o Multiple attachments measurements are not included in the statistics.
o Only milking claw air admission is allowed. Tests may not be valid if liners cracks or
piping holes allow additional air intake by the system. Furthermore, tests where
liner slippage or large udder causing squawks (liner slip) occurs are also rejected.
o The Milk Meter reading should be taken after the last dump of milk (milk dump after
removal).
o For each reading, list milk meter reading vs. actual milk weighed.
o At least three valid readings are required for calculating milk meter accuracy.
To ensure proper measurements, see that the following restrictions are kept:
o The collecting bucket must be placed below the meter outlet port.

21 AfiMilk MPC Installation Manual

 o The hose connecting the Milk Meter to the bucket is situated correctly in a
downward slope towards the bucket.
o The hose connecting the Milk Meter to the bucket has no pulling or pushing
pressure on the outlet of the Milk Meter.

7.3.3.2 Statistical Accuracy Calculation

After scaling milk weights for all the milk meters in the parlor, statistical accuracy
calculation is performed for each milk meter.
Statistical calculation is done for each reading and corresponding weigh scale measurement
for calculating the difference as a percentage value between the reading of the milk meter
and the actual milk weighed.
The following formula is used for calculating difference between the milk meter reading and
the milk weight.

Average all of the DIF% deviations for each Milk Meter.

Change the value of "G" parameter per calculated average deviation. If the result is negative
(Milk meter reading is lower than the actual weight), the calculated adjustment value needs
to be added to the value of G. If the adjustment is positive, the calculated value needs to be
subtracted from the value of G.
General adjustment calculation →%Dev X (–2.5) = Adjustment value
Example: if the actual value of G in the milk meter is 55 and the measured deviation -4%
(Milk meters show 4% less than actual weight), modify the “G” parameter value
by:
-4 X (-2.5) = +10 to 65

Fine Tune Calibration

After basic calibration of G parameter, it might happen that specific meters are deviating
and need further adjustment. In that case the H parameter can be used.
The procedure of calibrating with H parameter is basically the same procedure as calibrating
with G parameter. In this case change the e value of H parameter per calculated average
deviation.

22 AfiMilk MPC Installation Manual

If the result is negative (Milk meter reading is lower than the actual weight), the calculated
adjustment value needs to be added to the value of H. If the adjustment is positive, the
calculated value needs to be subtracted from the value of H.

General adjustment calculation →%Dev X (–5) = Adjustment value

Example: if the actual value of H in the milk meter is 37 and the measured deviation -4%
(Milk meters show 4% less than actual weight), modify the “H” parameter value
by:
-4 X (-5) = +20 to 57

To adjust the G and H parameter valua follow the procedure in paragraph Adjusting G
and/or H parameter value or via the AfiControl interface mentioned in paragraph Adjusting
the G and/or H parameter value via communication.
During initial calibration it is the aim to adjust each meter around 0% deviation, while this is
the base for setting the a reference values with water. If a single meter after both G & H
calibration still is deviating, the meter has to be replaced.

7.3.3.3 Setting the reference value by water test

An initial water test is performed after installation of the milk meters, after calibration of
the system or during installation of the entire system. The results of this water test are
the “reference values” for the periodic water tests performed for each milk meter
annually or more frequently.

Please note that the initial water test is performed to establish a baseline and the
following water tests at the farm are conducted to validate the stability of the system.
Since the device is calibrated for measuring milk rather than water, the deviations
between water measurements and the scaling of the fluid are the baseline tested, rather
than ultimate accuracy of measurements (as calculated in milk tests).

The preparation like needed equipment and how to perform the water test is described
in paragraph 7.3.1.1 Equipment - and water test preparation and description.

At least two measurements per meter with a mutual difference of max. 0,1 kg are recorded.
The reference value for each meter is the average of these two measurements.

23 AfiMilk MPC Installation Manual

Form calibration test with collected data resulting in the calculation of reference values
Afimilk MPC Setting reference values by water test
Farm: 1265897 Technical check Vacuum level (kPa) 42,0 Parlor: Harringbone (2x5)
Farm tesing Hygienic check Date: 30-7-2018
Cow road two measurements with mutual difference max. 0,1 kg Performer: Test person Joe
Cow city three within 0,2 ( else replace faulty measurements)
Nr. G-value H-value Display Unster Diff. Cond. Display Unster Diff. Cond. Ref. value Cond. Remarks
1 50 46 12,10 11,44 0,66 11,7 12,22 11,48 0,74 11,8 0,70 11,75
2 52 38 12,20 11,56 0,64 11,6 12,16 11,46 0,70 11,8 0,67 11,70
3 53 50 12,15 11,55 0,60 11,6 12,10 11,48 0,62 11,7 0,61 11,65 More measurements needed
4 58 42 12,18 11,20 0,98 11,6 12,22 11,20 1,02 11,6 1,00 11,60
5 52 40 12,20 11,48 0,72 11,5 12,30 11,60 0,70 11,6 0,71 11,55
6 51 36 12,34 11,68 0,66 11,8 12,20 11,48 0,72 11,9 0,69 11,85
7 57 46 12,08 11,26 0,82 11,7 12,36 11,56 0,80 11,8 0,81 11,75
8 50 58 12,10 11,40 0,70 11,7 12,18 11,46 0,72 11,7 0,71 11,70
9 52 60 12,14 11,46 0,68 11,6 12,16 11,50 0,66 11,6 0,67 11,60
10 51 42 12,06 11,36 0,70 11,5 12,10 11,34 0,76 11,6 0,73 11,55

For the periodic routine test, the following results, following this example, would be
handed over to the persons/organizations involved in routine testing:

Form with the calculated reference values for the annual routine test based on the
calibration test.
Afimilk MPC Periodic Routine test
Farm: 1265897 Technical check Vacuum level (kPa) Parlor: Harringbone (2x5)
Farm tesing Hygienic check - at reference determine 42,0 Date:
Cow road G-/H-value check - this routine test Performer:
Cow city Against ref. val.: One test: limit 0,1 kg; two tests: avg 0,2 kg
Nr. G-value H-value Air intake Ref.value Cond. Display Unster Diff. Display Unster Diff. Conclusion Remarks
1 50 46 9 0,70 11,8 0,00 0,00
2 52 38 9 0,67 11,7 0,00 0,00
3 53 50 9 0,61 11,7 0,00 0,00
4 58 42 9 1,00 11,6 0,00 0,00
5 52 40 9 0,71 11,6 0,00 0,00
6 51 36 9 0,69 11,9 0,00 0,00
7 57 46 9 0,81 11,8 0,00 0,00
8 50 58 9 0,71 11,7 0,00 0,00
9 52 60 9 0,67 11,6 0,00 0,00
10 51 42 9 0,73 11,6 0,00 0,00

24 AfiMilk MPC Installation Manual