Piston operated volumetric apparatus (POVA) calibration

procedure:
First determine the type of the POVA to be calibrated.
Determine the volume of calibration of the POVA under test.
Check the ambient temperature, ambient pressure and relative humidity.
Determine the temperature of the liquid that been used in calibration
If there is a high difference between ambient temperature and liquid temperature
then wait for the liquid before calibration until the temperature of the liquid
stabilizes. (Generally for 2 hours prior to calibration)
Chose the required balance information based on the table below:

Check the minimum requirements of the measuring devices:

Follow the required immersion depth and waiting time after the aspiration of the
test liquid:

For pipettes the following steps should be conducted:

8.3 Single-channel air displacement pipettes (in accordance with ISO 8655-2)
8.3.1 General
In the case of electronic motorised pipettes, aspiration and delivery of test liquid
are automatic. The remainder of the procedure is carried out following the steps
described in 8.3.2. The user should refer to the manufacturer's instructions or
operation manual for speed settings of aspiration and delivery.
NOTE More information regarding this type of piston pipette can be found in ISO
8655-2:2022, Annex B.
Forward pipetting shall always be performed.
8.3.2 Test cycle
Perform the test cycle as follows:
a) Fit the selected tip on the piston pipette.
b) In order to reach humidity equilibrium in the air-displacement piston pipette,
aspirate the test liquid five times.
c) Depress plunger.
d) Hold the pipette in a vertical position, immerse the tip in the test liquid to the
appropriate depth below the surface of the test liquid (see Table 3).
e) Release the plunger slowly, if hand operated.
f) Pause for the recommended wait time (see Table 3).
g) Withdraw tip vertically and carefully from the test liquid.
h) Record the balance indication m0 of the weighing vessel or tare the balance to
zero (m0 = 0).
i) Start the timing device; this may be omitted if using a weighing vessel with lid,
or an evaporation trap.
j) If the weighing vessel has a lid, remove it.
k) Touch the tip on the inside of the weighing vessel at an angle of approximately
30° to 45°.
l) Depress the plunger and deliver the test liquid into the weighing vessel.
m) Where applicable, use the blow-out feature of the piston pipette (second stop,
based on pipette type) to expel the last drop of liquid before drawing the delivery
end of the tip along the inner wall of the weighing vessel.
n) Draw the tip approximately 8 mm to 10 mm along the inner wall of the
weighing vessel to remove any droplets at or around the tip orifice.
o) Remove the tip from the weighing vessel.
p) Release the plunger.
q) Close the lid of the weighing vessel, if appropriate.
r) Record the balance indication mi (i = 1 to n) of the weighing vessel with the
delivered test liquid (if tare is used m¡ will be the delivered quantity of water) and
calculate the weighing value as the difference between the balance indications.
s) Repeat the test cycle starting at step c) until all measurements have been
performed and values recorded.
During the replicate measurements, the pipette tip shall be changed at least once
in order to detect the use of damaged or incorrectly manufactured tips and assess
the variability of the used tips. For n = 10 replicates, at least two tips shall be used,
and the tip shall be changed at minimum once every 5 measurements. When
replacing the tips, they shall be pre-wetted, starting the test cycle at step a) This
tip change is also applicable to positive displacement pipettes with disposable tips
(type D2).
If it is necessary to remove the weighing vessel from the balance pan to enable
delivery of the volume, avoid excessive handling and possible contamination by
the use of lint-free gloves. Return the weighing vessel to the balance pan after
delivery.
NOTE Removing the weighing vessel for delivery will increase the measurement
uncertainty of the procedure, further information can be found in ISO/TR 20461.
If a mathematical compensation of evaporation is performed, note the time to the
nearest second taken to complete the 10 test cycles. Perform one evaporation
trial at the beginning, and a second evaporation trial at the end of the series, if
relevant. Both evaporation trials should be a mimic of one dispense, but without
delivering liquid. Follow the same details and timing as an actual delivery. Record
the mass of the weighing vessel before and after each evaporation trial (see
Formula (1)).
The values obtained shall be evaluated in accordance with Clause 9.
The effect of barometric pressure on accuracy should be considered for air
displacement pipettes. See ISO/TR 20461 for further details.
8.4 Multi-channel pipettes (in accordance with ISO 8655-2)
Multi-channel pipettes are similar to single channel pipettes in that they comprise
a set of single-volume measuring and delivery units, all operated simultaneously
by a single operating mechanism. For the purposes of the volumetric performance
test, each channel shall be regarded as a single-channel and tested and reported
as such according to 8.3.
The values obtained shall be evaluated in accordance with Clause 9.
All channels of a multi-channel pipette should be tested individually to account for
the specific design and operational challenges of multi-channel pipettes.
There are generally two possible methods for testing multi-channel pipettes.
a) A multi-channel balance can be used to measure the test volume delivered
from all channels in parallel, by aspirating and emptying each channel at the same
time and analysing the results of each channel individually. The test procedure
described in 8.3.2 can be adapted accordingly. The following variation may be
applied to calibrate multi-channel pipettes with more than 12 channels or with 4,5
mm cone distance: use every second cone of the pipette to calibrate the pipette in
multiple sets. All channels shall be tested.
If the pipette has more than one row of channels, it may be tested one row at a
time.
b) Each channel can be measured individually, one after another, with a single-
channel balance. For this purpose, test liquid shall be aspirated by all channels
together and collected from one channel at a time. For the measurement of
channel 1, for example, the volume of channel 1 is delivered into the weighing
vessel, while the volumes from all other channels are discarded.
NOTE One way to accomplish this delivery method is by using a weighing vessel
with a narrow neck which extends through a dish. The liquid from the channel
under test is then delivered into the neck, while the liquid from all other channels
collects in the dish.
8.5 Positive displacement pipettes (in accordance with ISO 8655-2)
Positive displacement pipettes shall be tested in accordance with 8.3. However,
the conditioning to achieve humidity equilibrium prior to the test only needs to be
performed if required by the manufacturer.
Only change the pipette tips according to 8.3.2 when testing positive
displacement pipettes of Type D2 with disposable tips (see ISO 8655-2).
Manufacturer’s instructions shall be followed for filling the pipette tip without air
bubbles.
The values obtained shall be evaluated in accordance with Clause 9.

For burets the following steps should be conducted:

Perform the test cycle as follows:
a) Load the burette, bubble free, in accordance with the manufacturer's
instructions.
b) Record the balance indication m0 of the weighing vessel or tare the balance to
zero (m0 = 0).
c) Deliver the test liquid from the burette into the weighing vessel, until the
selected volume is reached. If the burette is automatically controlled, deliver test
liquid until the volume pre-set is reached and no further delivery occurs.
d) Record the balance indication mi (i = 1 to n) of the weighing vessel with the
delivered test liquid (if tare is used mi will be the delivered quantity of water) and
calculate the weighing value as the difference between the balance indications.
e) Repeat the test cycle until all measurements have been performed and values
recorded.
When testing partial volumes of the nominal volume of the burette, the piston
shall not be reset to the initial position (zero) prior to the next measurement.
Ensure that the upper volume limit of the piston, and thus the nominal volume of
the burette, are not exceeded when dispensing a partial volume.
The values obtained shall be evaluated in accordance with Clause 9.

For dispensers the following steps should be conducted:

8.8 Dispensers (in accordance with ISO 8655-5)
Perform the test cycle as follows:
a) Load the dispenser, bubble free, in accordance with the manufacturer's
instructions.
b) Record the balance indication m0 of the weighing vessel or tare the balance to
zero (m0 = 0).
c) Deliver the test liquid from the dispenser into the weighing vessel.
d) Record the balance indication mi (i = 1 to n) of the weighing vessel with the
delivered test liquid (if tare is used mi will be the delivered quantity of water) and
calculate the weighing value as the difference between the balance indications.
e) Repeat the test cycle until all measurements have been performed and values
recorded.
Due to the large effect of piston speed on the measuring result, any information
contained in the manufacturer's instructions or operation manual regarding piston
speed is particularly important (e.g. selection of the speed appropriate for water
with electronic motorized apparatus).
During operation, ensure that the piston does not hit the stroke limits.
For multiple delivery dispensers (see ISO 8655-5), do not reset the piston between
each of the n test cycles to its initial position if there is sufficient test liquid
remaining to deliver the next aliquot.
The values obtained shall be evaluated in accordance with Clause 9.

9 Evaluation
9.1 Calculation of evaporation loss
It is possible to determine the evaporation loss during each test cycle by
calculating the difference between the starting mass (m00) and the mass after the
first cycle time without making any delivery of liquid (m0). This can be repeated
again at the end of the test, to obtain mn and mn+1.
The average evaporation loss per test cycle can be determined by using either
Formula (1) or by another appropriate method or formula.

Where:
mevap the estimated evaporated mass within a test cycle;
m00 is the balance indication at the first reading (starting mass);
m0 is the balance indication after the cycle time;
mn is the balance indication after the nth replicate delivered;
mn+1 is the balance indication after the nth replicate delivered after the cycle
time.
9.2 Calculation of the corrected weighing value of each quantity delivered
If the tare of the balance has not been used, calculate the weighing value of each
delivered quantity mi by subtraction m1 − m0, m2 − m1, m10 − m9. Add the
evaporation loss per test cycle calculated in 9.1 to each delivered quantity mi.
9.3 Conversion of the corrected weighing values to volume
9.3.1 General
The corrected weighing values shall be converted to volume using one of the
methods described in 9.3.2 or 9.3.3.
9.3.2 Calculation of volume using the general formula
9.3.2.1 General formula for volume
The general formula for calculation of the volume at the reference temperature of
20 °C, V20 (at a reference temperature of 27 °C, V27), from the weighing value of
the delivered water is described in ISO 4787 [13] and is the basis for Formula (2)
in this document. The values mi obtained in accordance with 9.2 are balance
indications. A correction taking into account water density and air buoyancy is
necessary for the conversion of the weighing value of each delivered quantity mi
to volume Vi. If the cubic thermal expansion coefficient γ of POVA is known, as
specified by the manufacturer or measured with sufficient reliability by other
means, Vref,i can be calculated according to Formula (2).

Where:
Vi,ref is the calculated volume at the reference temperature, in ml;
mL is the balance indication of the weighing vessel after liquid delivery, in g;
mE is the balance indication of the weighing vessel before liquid delivery, in g (mE
= 0 in case the balance was tared with the weighing vessel);
mevap the estimated evaporated mass within a test cycle, in g;
ρA is the density of air, in g/ml, at the temperature, humidity and atmospheric
pressure of the
test, see Formula (3);
ρB is the reference density of the weights (typically 8 g/ml);
ρW is the density of water at the test temperature (in units of °C), in g/ml,
calculated with the
“Tanaka” Formula (4);
γ is the combined cubic thermal expansion coefficient of the POVA under test, in
°C−1
tW temperature of the POVA, assumed to be equal to the temperature of the
liquid used in the
test, in °C;
tref is the reference temperature of the POVA (20 °C or 27 °C).

9.3.2.2 Calculation of air density

Formula (3) for the air density ρA can be used at temperatures between 15 °C and
27 °C, barometric pressure between 600 hPa and 1 100 hPa, and relative humidity
between 20 % and 80 %:

where
ρΑ is the air density, in g/ml;
tA is the ambient temperature, in °C;
p is the barometric pressure, in hPa;
hr is the relative humidity, in %.
The uncertainty of this formula can be calculated according to OIML R111 -1: 2004
[12] , section C.6.3.6. At other environmental conditions, Formula (3) shall be
replaced with the current air density calculations.
NOTE At the time of publication of this document, the current CIPM air density
calculations are described in Reference [5].
9.3.2.3 Calculation of water density
The density of pure water ρW is normally provided from formulae given in the
literature. Formula (4) given by Tanaka [6] can be used:

Where:
ρW is the density of water, in g/ml
tW is the water temperature, in °C;
a1 = −3 983 035 °C;
a2 = 301 797 °C;
a3 = 522 528 9 (°C)2;
a4 = 69 348 81 °C;
a5 = 0,999 974 950 g/ml.
9.3.3 Calculation of volume using the Z correction factor
Another possible procedure for conversion of weighing values into volume is using
the Z correction factors specified in Table A.1. The Z correction factors given in
Table A.1 take into account water density and air pressure during weighing at the
corresponding test temperature.
Convert each weighing value mi obtained from 9.2 to the delivered volume Vi, by
applying the Z correction factors at the mean temperature and barometric
pressure measured according to 8.1.4 and using Formula (5):

If the test temperature is different from the temperature of adjustment (which is

20 °C or 27 °C, see ISO 8655-2 to ISO 8655-5 and ISO 8655-9) and if the cubic
thermal expansion coefficient γ of the pistonoperated volumetric apparatus is
known, as specified by the manufacturer or measured with sufficient reliability by
other means Formula (5) may be replaced by Formula (6):

9.3.4 Mean delivered volume

Calculate the mean delivered volume V as shown in Formula (7).

Where:
V is the mean delivered volume;
Vi,ref is each delivered reference volume;
N is the number of replicates.
When the cubic thermal expansion coefficient is not applied, Vi,ref is replaced by
Vi when using
Formula (7).
9.4 Systematic error of measurement
Calculate the systematic error of measurement es of the POVA using Formula (8):

where
eS is the systematic error of measurement, expressed in units of volume;
VS is the selected test volume at the POVA under test.
or in percent using Formula (9):

Where:
ηs is the relative systematic error of measurement, expressed in percent.
In the case of fixed volume POVA the selected test volume Vs is the nominal
volume.
The systematic error of measurement in the ISO 8655 series is based on historic
convention within the pipetting industry and is reversed in sign compared to the
definition described in ISO/IEC guide 99:2007,2.17, for more information see ISO
8655-1:2022, 6.2.
9.5 Random error of measurement
Calculate the random error of measurement of the POVA as repeatability or
standard deviation sr using Formula (10):
Where:
sr is the standard deviation, expressed in units of volume.
This random error of measurement may also be expressed as a percentage by the
coefficient of variation, CV, using Formula (11):

Where: CV is the coefficient of variation.

9.6 Uncertainty of measurement
For calibrations (ISO 8655-1:2022, 6.4), the expanded measurement uncertainty of
the mean delivered volume for each selected volume shall be estimated and
reported. It shall include contributions from both the gravimetric measurement
procedure of this document and the device under test. For further information
see ISO/TR 20461[10]. When performing testing (ISO 8655-1:2022, 6.4) it is
optional to estimate and report the expanded measurement uncertainty of the
mean delivered volume for each selected volume.

An excel sheet was generated to perform the calculations that been prior
discussed but the general measurement steps should be followed as discussed in
each section.
References: