4 Methods of Test [7]
4.1 General
The following tests should be conducted for each type of
nozzle. Before testing, precise drawings of parts and the assembly
should be submitted together with the appropriate specifications (using
SI units). Tests should be carried out at an ambient temperature of
(20,±5)°C, unless other temperatures are indicated.
4.2 Visual examination [7.2]
Before testing, nozzles should be examined visually with
respect to the following points:
4.3 Body strength test (see 3.6) [7.3]
4.3.1 The design load should be measured on ten
automatic nozzles by securely installing each nozzle, at room temperature,
in a tensile/compression test machine and applying a force equivalent
to the application of the rated working pressure.
4.3.2 An indicator capable of reading deflection
to an accuracy of 0.01 mm should be used to measure any change in
length of the nozzle between its load bearing points. Movement of
the nozzle shank thread in the threaded bushing of the test machine
should be avoided or taken into account.
4.3.3 The hydraulic pressure and load is then
released and the heat responsive element is then removed by a suitable
method. When the nozzle is at room temperature, a second measurement
is to be made using the indicator.
4.3.4 An increasing mechanical load to the nozzle
is then applied at a rate not exceeding 500 N/minute, until the indicator
reading at the load bearing point initially measured returns to the
initial value achieved under hydrostatic load. The mechanical load
necessary to achieve this should be recorded as the service load.
Calculate the average service load.
4.3.5 The applied load is then progressively increased
at a rate not exceeding 500 N/minute on each of the five specimens
until twice the average service load has been applied. Maintain this
load for 15 ± 5 s.
4.3.6 The load is then removed and any permanent
elongation as defined in 3.6 is recorded.
4.4 Leak resistance and hydrostatic strength tests
(see 3.8) [7.4]
4.4.1 Twenty nozzles should be subjected to a
water pressure of twice their rated working pressure, but not less
than 34.5 bar. The pressure is increased from 0 bar to the test pressure,
maintained at twice rated working pressure for a period of 3 min and
then decreased to 0 bar. After the pressure has returned to 0 bar,
it is increased to the minimum operating pressure specified by the
manufacturer in not more than 5 s. This pressure is to be maintained
for 15 s and then increased to rated working pressure and maintained
for 15 s.
4.4.2 Following the test of 4.4.1, the twenty
nozzles should be subjected to an internal hydrostatic pressure of
four times the rated working pressure. The pressure is increased from
0 bar to four times the rated working pressure and held there for
a period of 1 minute. The nozzle under test should not rupture, operate
or release any of its operating parts during the pressure increase
nor while being maintained at four times the rated working pressure
for 1 minute.
4.5 Functional test (see 3.5) [7.5]
4.5.1 Nozzles having nominal release temperatures
less than 78°C, should be heated to activation in an oven. While
being heated, they should be subjected to each of the water pressures
specified in 4.5.3 applied to their inlet. The temperature of the
oven should be increased to 400 ± 20°C in 3 min measured
in close proximity to the nozzle. Nozzles having nominal release temperatures
exceeding 78°C should be heated using a suitable heat source.
Heating should continue until the nozzle has activated.
4.5.2 Eight nozzles should be tested in each normal
mounting position and at pressures equivalent to the minimum operating
pressure, the rated working pressure and at the average operating
pressure. The flowing pressure should be at least 75% of the initial
operating pressure.
4.5.3 If lodgement occurs in the release mechanism
at any operating pressure and mounting position, 24 more nozzles should
be tested in that mounting position and at that pressure. The total
number of nozzles for which lodgement occurs should not exceed 1 in
the 32 tested at that pressure and mounting position.
4.5.4 Lodgement is considered to have occurred
when one or more of the released parts lodge in the discharge assembly
in such a way as to cause the water distribution to be altered after
the period of time specified in 3.5.1.
4.5.5 In order to check the strength of the deflector/orifice
assembly, three nozzles should be submitted to the functional test
in each normal mounting position at 125 per cent of the rated working
pressure. The water should be allowed to flow at 125 per cent of the
rated working pressure for a period of 15 min.
4.6 Heat responsive element operating characteristics
4.6.1 Operating temperature test (see 3.3) [7.6]
4.6.1.1 Ten nozzles should be heated from room
temperature to 20 to 22°C below their nominal release temperature.
The rate of increase of temperature should not exceed 20°C/min
and the temperature should be maintained for 10 min. The temperature
should then be increased at a rate between 0.4°C/min to 0.7°C/min
until the nozzle operates.
4.6.1.2 The nominal operating temperature should
be ascertained with equipment having an accuracy of ±0.35%
of the nominal temperature rating or ±0.25°C, whichever
is greater.
4.6.1.3 The test should be conducted in a water
bath for nozzles or separate glass bulbs having nominal release temperatures
less than or equal to 80°C. A suitable oil should be used for
higher-rated release elements. The liquid bath should be constructed
in such a way that the temperature deviation within the test zone
does not exceed 0.5%, or 0.5°C, whichever is greater.
4.6.2 Dynamic heating test (3.14)
4.6.2.1 Plunge test
4.6.2.1.1 Tests should be conducted to determine
the standard and worst case orientations as defined in 1.4 and 1.5.
Ten additional plunge tests should be performed at both of the identified
orientations. The worst case orientation should be as defined in 3.14.1.
The RTI is calculated as described in 4.6.2.3 and 4.6.2.4 for each
orientation, respectively. The plunge tests are to be conducted using
a brass nozzle mount designed such that the mount or water temperature
rise does not exceed 2°C for the duration of an individual plunge
test up to a response time of 55 s. (The temperature should be measured
by a thermocouple heatsinked and embedded in the mount not more than
8 mm radially outward from the root diameter of the internal thread
or by a thermocouple located in the water at the centre of the nozzle
inlet.) If the response time is greater than 55 s, then the mount
or water temperature in degrees Celsius should not increase more than
0.036 times the response time in seconds for the duration of an individual
plunge test.
4.6.2.1.2 The nozzle under test should have 1
to 1.5 wraps of PTFE sealant tape applied to the nozzle threads. It
should be screwed into a mount to a torque of 15 ±3 Nm. Each
nozzle is to be mounted on a tunnel test section cover and maintained
in a conditioning chamber to allow the nozzle and cover to reach ambient
temperature for a period of not less than 30 min.
4.6.2.1.3 At least 25 ml of water, conditioned
to ambient temperature, should be introduced into the nozzle inlet
prior to testing. A timer accurate to ±0.01 s with suitable
measuring devices to sense the time between when the nozzle is plunged
into the tunnel and the time it operates should be utilized to obtain
the response time.
4.6.2.1.4 A tunnel should be utilized with air
flow and temperature conditionsfootnote at
the test section (nozzle location) selected from the appropriate range
of conditions shown in table 2.
To minimize radiation exchange between the sensing element and the
boundaries confining the flow, the test section of the apparatus should
be designed to limit radiation effects to within ± 3% of calculated
RTI valuesfootnote.
4.6.2.1.5 The range of permissible tunnel operating
conditions is shown in table 2.
The selected operating condition should be maintained for the duration
of the test with the tolerances as specified by footnotes 4 and 5
in table 2.
4.6.2.2 Determination of conductivity factor (C)
[7.6.2.2]
The conductivity factor (C) should be determined using the
prolonged plunge test (see 4.6.2.2.1) or the prolonged exposure ramp
test (see 4.6.2.2.2).
4.6.2.2.1 Prolonged plunge test [7.6.2.2.1]
.1 the prolonged plunge test is an iterative process
to determine C and may require up to twenty nozzle samples. A new
nozzle sample must be used for each test in this section even if the
sample does not operate during the prolonged plunge test;
.2 the nozzle under test should have 1 to 1.5
wraps of PTFE sealant tape applied to the nozzle threads. It should
be screwed into a mount to a torque of 15 + 3 Nm. Each nozzle is to
be mounted on a tunnel test section cover and maintained in a conditioning
chamber to allow the nozzle and cover to reach ambient temperature
for a period of not less than 30 min. At least 25 ml of water, conditioned
to ambient temperature, should be introduced into the nozzle inlet
prior to testing;
.3 a timer accurate to ± 0.01 s with suitable
measuring devices to sense the time between when the nozzle is plunged
into the tunnel and the time it operates should be utilized to obtain
the response time;
.4 the mount temperature should be maintained
at 20 ± 0.5°C for the duration of each test. The air velocity
in the tunnel test section at the nozzle location should be maintained
with ± 2% of the selected velocity. Air temperature should
be selected and maintained during the test as specified in table 3;
.5 the range of permissible tunnel operating conditions
is shown in table 3. The
selected operating condition should be maintained for the duration
of the test with the tolerances as specified in table 3; and
.6 to determine C, the nozzle is immersed in the
test stream at various air velocities for a maximum of 15 min.footnote Velocities are chosen such that actuation
is bracketed between two successive test velocities. That is, two
velocities must be established such that at the lower velocity (uj)
actuation does not occur in the 15 min test interval. At the next
higher velocity (uh), actuation must occur within the 15
min time limit. If the nozzle does not operate at the highest velocity,
select an air temperature from table
3 for the next higher temperature rating.
Table 2 Plunge oven
test conditions
|
|
Air temperature ranges*
|
Velocity ranges**
|
Normal Temperature, °C
|
Standard Response, °C
|
Special Response, °C
|
Fast Response, m/s
|
Standard Response, m/s
|
Special Response, m/s
|
Fast Response Nozzle, m/s
|
| 57 to
77
|
191 to
203
|
129 to
141
|
129 to
141
|
2.4 to
2.6
|
2.4 to
2.6
|
1.65 to
1.85
|
| 79 to
107
|
282 to
300
|
191 to
203
|
191 to
203
|
2.4 to
2.6
|
2.4 to
2.6
|
1.65 to
1.85
|
| 121 to
149
|
382 to
432
|
282 to
300
|
282 to
300
|
2.4 to
2.6
|
2.4 to
2.6
|
1.65 to
1.85
|
| 163 to
191
|
382 to
432
|
382 to
432
|
382 to
432
|
3.4 to
3.6
|
2.4 to
2.6
|
1.65 to
1.85
|
Note
* The selected air
temperature should be known and maintained constant within
the test section throughout the test to an accuracy of
±1°C for the air temperature range of 129 to 141°C within
the test section and within ±2°C for all other air
temperatures.
Note
** The selected air velocity
should be known and maintained constant throughout the
test to an accuracy of ±0.03 m/s for velocities of 1.65 to
1.85 and 2.4 to 2.6 m/s and ±0.04 m/s for velocities of
3.4 to 3.6 m/s.
|
Table 3 Plunge oven
test conditions for conductivity determination
Nominal nozzle temperature, °C
|
Oven temperature, °C
|
Maximum variation of air temperature during
test, °C
|
| 57
|
85 to
91
|
±
1.0
|
| 58 to
77
|
124 to
130
|
±
1.5
|
| 78 to
107
|
193 to
201
|
±
3.0
|
| 121 to
149
|
287 to
295
|
±
4.5
|
| 163 to
191
|
402 to
412
|
±
6.0
|
Test velocity selection should ensure that:
The test value of C is the average of the values
calculated at the two velocities using the following equation:
where:
Δ TgActual gas (air) temperature minus the mount temperature (Tm) in °C.
Δ TeaMean liquid bath operating temperature minus the mount temperature (Tm) in °C.
u Actual air velocity in the test section in m/s.
The nozzle C value is determined by repeating the bracketing
procedure three times and calculating the numerical average of the
three C values. This nozzle C value is used to calculate all standard
orientation RTI values for determining compliance with 3.14.1.
4.6.2.2.2 Prolonged exposure ramp test [7.6.2.2.2]
.1 the prolonged exposure ramp test for the determination
of the parameter C should be carried out in the test section of a
wind tunnel and with the requirements for the temperature in the nozzle
mount as described for the dynamic heating test. A preconditioning
of the nozzle is not necessary;
.2 ten samples should be tested of each nozzle
type, all nozzles positioned in standard orientation. The nozzle should
be plunged into an air stream of a constant velocity of 1 m/s ±
10% and an air temperature at the nominal temperature of the nozzle
at the beginning of the test; and
.3 the air temperature should then be increased
at a rate of 1 ± 0.25°C/min until the nozzle operates.
The air temperature, velocity and mount temperature should be controlled
from the initiation of the rate of rise and should be measured and
recorded at nozzle operation. The C value is determined using the
same equation as in 4.6.2.2.1 as the average of the ten test values.
4.6.2.3 RTI value calculation [7.6.2.3]
The equation used to determine the RTI value is as follows:
where:
tr Response time of nozzles in seconds
u Actual air velocity in the test section of the tunnel in m/s from table 2
Δ Tea Mean liquid bath operating temperature of the nozzle minus the ambient temperature in °C
ΔTg Actual air temperature in the test section minus the ambient temperature in °C
C Conductivity factor as determined in 4.6.2.2
4.6.2.4 Determination of worst case orientation
RTI
The equation used to determine the RTI for the worst case
orientation is as follows:
where:
Tt-wc Response time of the nozzles in seconds for the worst case orientation
All variables are known at this time per the equation in
paragraph 4.6.2.3 except RTIwc (Response Time Index for
the worst case orientation) which can be solved iteratively per the
above equation.
In the case of fast response nozzles, if a solution for
the worse case orientation RTI is unattainable, plunge testing in
the worst case orientation should be repeated using the plunge test
conditions under Special Response shown in table 2.
4.7 Heat exposure test [7.7]
4.7.1 Glass bulb nozzles (see 3.9.1):
.1 glass bulb nozzles having nominal release temperatures
less than or equal to 80°C should be heated in a water bath from
a temperature of (20 ± 5)°C to (20 ± 2) °C below
their nominal release temperature. The rate of increase of temperature
should not exceed 20°C/min. High temperature oil, such as silicone
oil should be used for higher temperature rated release elements;
and
.2 this temperature should then be increased at
a rate of 1°C/min to the temperature at which the gas bubble dissolves,
or to a temperature 5°C lower than the nominal operating temperature,
whichever is lower. Remove the nozzle from the liquid bath and allow
it to cool in air until the gas bubble has formed again. During the
cooling period, the pointed end of the glass bulb (seal end) should
be pointing downwards. This test should be performed four times on
each of four nozzles.
4.7.2 All uncoated nozzles (see 3.9.2) [7.7.2]
Twelve uncoated nozzles should be exposed for a period of
90 days to a high ambient temperature that is 11°C below the nominal
rating or at the temperature shown in table 4, whichever is lower, but not less than 49°C. If
the service load is dependent on the service pressure, nozzles should
be tested under the rated working pressure. After exposure, four of
the nozzles should be subjected to the tests specified in 4.4.1, four
nozzles to the test of 4.5.1, two at the minimum operating pressure
and two at the rated working pressure, and four nozzles to the requirements
of 3.3. If a nozzle fails the applicable requirements of a test, eight
additional nozzles should be tested as described above and subjected
to the test in which the failure was recorded. All eight nozzles should
comply with the test requirements.
4.7.3 Coated nozzles (see 3.9.3) [7.7.3]:
.1 in addition to the exposure test of 4.7.2 in
an uncoated version, twelve coated nozzles should be exposed to the
test of 4.7.2 using the temperatures shown in table 4 for coated nozzles; and
.2 the test should be conducted for 90 days. During
this period, the sample should be removed from the oven at intervals
of approximately 7 days and allowed to cool for 2 h to 4 h. During
this cooling period, the sample should be examined. After exposure,
four of the nozzles should be subjected to the tests specified in
4.4.1, four nozzles to the test of 4.5.1; two at the minimum operating
pressure and two at the rated working pressure, and four nozzles to
the requirements of 3.3.
Table 4 Test temperatures
for coated and uncoated nozzles
|
Values in degrees Celsius
|
|
Nominal release Temperature
|
Uncoated nozzle test temperature
|
Coated nozzle test temperature
|
| 57-60
|
49
|
49
|
| 61-77
|
52
|
49
|
| 78-107
|
79
|
66
|
| 108-149
|
121
|
107
|
| 150-191
|
149
|
149
|
| 192-246
|
191
|
191
|
| 247-302
|
246
|
246
|
| 303-343
|
302
|
302
|
4.8 Thermal shock test for glass bulb nozzles (see
3.10) [7.8]
4.8.1 Before starting the test, condition at least
24 nozzles at room temperature of 20 to 25°C for at least 30 min.
4.8.2 The nozzle should be immersed in a bath
of liquid, the temperature of which should be 10 ± 2°C
below the nominal release temperature of the nozzles. After 5 min.,
the nozzles are to be removed from the bath and immersed immediately
in another bath of liquid, with the bulb seal downwards, at a temperature
of 10 ± 2°C. Then test the nozzles in accordance with 4.5.1.
4.9 Strength test for release elements [7.9]
4.9.1 Glass bulbs (see 3.7.1) [7.9.1]
4.9.1.1 At least 15 sample bulbs in the lowest
temperature rating of each bulb type should be positioned individually
in a text fixture using the sprinkler seating parts. Each bulb should
then be subjected to a uniformly increasing force at a rate not exceeding
250 N/s in the test machine until the bulb fails.
4.9.1.2 Each test should be conducted with the
bulb mounted in new seating parts. The mounting device may be reinforced
externally to prevent its collapse, but in a manner which does not
interfere with bulb failure.
4.9.1.3 Record the failure load for each bulb.
Calculate the lower tolerance limit (TLI) for bulb strength. Using
the values of service load recorded in 4.3.1, calculate the upper
tolerance limit (TL2) for the bulb design load. Verify compliance
with 3.7.1.
4.9.2 Fusible elements (see 3.7.2)
4.10 Water flow test (see 3.4.1) [7.10]
The nozzle and a pressure gauge should be mounted on a supply
pipe. The water flow should be measured at pressures ranging from
the minimum operating pressure to the rated working pressure at intervals
of approximately 10% of the service pressure range on two sample nozzles.
In one series of tests, the pressure should be increased from zero
to each value and, in the next series, the pressure shall be decreased
from the rated pressure to each value. The flow constant, K, should
be averaged from each series of readings, i.e., increasing pressure
and decreasing pressure. During the test, pressures should be corrected
for differences in height between the gauge and the outlet orifice
of the nozzle.
4.11 Corrosion tests [7.12]
4.11.1 Stress corrosion test for brass nozzle parts
(see 3.11.1)
4.11.1.1 Five nozzles should be subjected to the
following aqueous ammonia test. The inlet of each nozzle should be
sealed with a nonreactive cap, e.g., plastic.
4.11.1.2 The samples are degreased and exposed
for 10 days to a moist ammonia-air mixture in a glass container of
volume 0.02 ± 0.01 m3.
4.11.1.3 An aqueous ammonia solution, having a
density of 0.94 g/cm3, should be maintained in the bottom
of the container, approximately 40 mm below the bottom of the samples.
A volume of aqueous ammonia solution corresponding to 0.01 ml per
cubic centimetre of the volume of the container will give approximately
the following atmospheric concentrations: 35% ammonia, 5% water vapour,
and 60% air. The inlet of each sample should be sealed with a nonreactive
cap, e.g., plastic.
4.11.1.4 The moist ammonia-air mixture should
be maintained as closely as possible at atmospheric pressure, with
the temperature maintained at 34 ± 2°C. Provision should
be made for venting the chamber via a capillary tube to avoid the
build-up of pressure. Specimens should be shielded from condensate
drippage.
4.11.1.5 After exposure, rinse and dry the nozzles,
and conduct a detailed examination. If a crack, delamination or failure
of any operating part is observed, the nozzle(s) should be subjected
to a leak resistance test at the rated pressure for 1 min and to the
functional test at the minimum flowing pressure (see 3.5.1).
4.11.1.6 Nozzles showing cracking, delamination
or failure of any non-operating part should not show evidence of separation
of permanently attached parts when subjected to flowing water at the
rated working pressure for 30 min.
4.11.2 Stress-Corrosion Cracking of Stainless Steel
Nozzle Parts (see 3.11.1)
4.11.2.1 Five samples are to be degreased prior
to being exposed to the magnesium chloride solution.
4.11.2.2 Parts used in nozzles are to be placed
in a 500-millilitre flask that is fitted with a thermometer and a
wet condenser approximately 760 mm long. The flask is to be filled
approximately one-half full with a 42% by weight magnesium chloride
solution, placed on a thermostatically-controlled electrically heated
mantel, and maintained at a boiling temperature of 150 ± 1°C.
The parts are to be unassembled, that is, not contained in a nozzle
assembly. The exposure is to last for 500 hours.
4.11.2.3 After the exposure period, the test samples
are to be removed from the boiling magnesium chloride solution and
rinsed in deionised water.
4.11.2.4 The test samples are then to be examined
using a microscope having a magnification of 25X for any cracking,
delamination, or other degradation as a result of the test exposure.
Test samples exhibiting degradation are to be tested as described
in 4.11.1.5 or 4.11.2.6, as applicable. Test samples not exhibiting
degradation are considered acceptable without further test.
4.11.2.5 Operating parts exhibiting degradation
are to be further tested as follows. Five new sets of parts are to
be assembled in nozzle frames made of materials that do not alter
the corrosive effects of the magnesium chloride solution on the stainless
steel parts. These test samples are to be degreased and subjected
to the magnesium chloride solution exposure specified in paragraph
4.11.2.2. Following the exposure, the test samples should withstand,
without leakage, a hydrostatic test pressure equal to the rated working
pressure for 1 minute and then be subjected to the functional test
at the minimum operating pressure in accordance with 4.5.1.
4.11.2.6 Non-operating parts exhibiting degradation
are to be further tested as follows. Five new sets of parts are to
be assembled in nozzle frames made of materials that do not alter
the corrosive effects of the magnesium chloride solution on the stainless
steel parts. These test samples are to be degreased and subjected
to the magnesium chloride solution exposure specified in paragraph
4.11.2.2. Following the exposure, the test samples should withstand
a flowing pressure equal to the rated working pressure for 30 minutes
without separation of permanently attached parts.
4.11.3 Sulphur dioxide corrosion test (see 3.11.2
and 3.14.2)
4.11.3.1 Ten nozzles should be subjected to the
following sulphur dioxide corrosion test. The inlet of each sample
should be sealed with a nonreactive cap, e.g., plastic.
4.11.3.2 The test equipment should consist of
a 5 litre vessel (instead of a 5 litre vessel, other volumes up to
15 litre may be used in which case the quantities of chemicals given
below shall be increased in proportion) made of heat-resistant glass,
with a corrosion-resistant lid of such a shape as to prevent condensate
dripping on the nozzles. The vessel should be electrically heated
through the base, and provided with a cooling coil around the side
walls. A temperature sensor placed centrally 160 mm ± 20 mm
above the bottom of the vessel should regulate the heating so that
the temperature inside the glass vessel is 45°C ± 3°C.
During the test, water should flow through the cooling coil at a sufficient
rate to keep the temperature of the discharge water below 30°C.
This combination of heating and cooling should encourage condensation
on the surfaces of the nozzles. The sample nozzles should be shielded
from condensate drippage.
4.11.3.3 The nozzles to be tested should be suspended
in their normal mounting position under the lid inside the vessel
and subjected to a corrosive sulphur dioxide atmosphere for 8 days.
The corrosive atmosphere should be obtained by introducing a solution
made up by dissolving 20 g of sodium thiosulphate (Na2S203H2O) crystals in 500 ml of water.
4.11.3.4 For at least six days of the 8-day exposure
period, 20 ml of dilute sulphuric acid consisting of 156 ml of normal
H2SO4 (0.5 mol/litre) diluted with 844 ml of
water should be added at a constant rate. After 8 days, the nozzles
should be removed from the container and allowed to dry for 4 to 7
days at a temperature not exceeding 35°C with a relative humidity
not greater than 70%.
4.11.3.5 After the drying period, five nozzles
should be subjected to a functional test at the minimum operating
pressure in accordance with 4.5.1 and five nozzles should be subjected
to the dynamic heating test in accordance with 3.14.2.
4.11.4 Salt spray corrosion test (see 3.11.3 and
3.14.2) [7.12.3]
4.11.4.1 Nozzles intended for normal atmospheres
4.11.4.1.1 Ten nozzles should be exposed to a
salt spray within a fog chamber. The inlet of each sample should be
sealed with a nonreactive cap, e.g., plastic.
4.11.4.1.2 During the corrosive exposure, the
inlet thread orifice is to be sealed by a plastic cap after the nozzles
have been filled with deionised water. The salt solution should be
a 20% by mass sodium chloride solution in distilled water. The pH
should be between 6.5 and 7.2 and the density between 1.126 g/ml and
1.157 g/ml when atomized at 35°C. Suitable means of controlling
the atmosphere in the chamber should be provided. The specimens should
be supported in their normal operating position and exposed to the
salt spray (fog) in a chamber having a volume of at least 0.43 m3 in
which the exposure zone shall be maintained at a temperature of 35
± 2°C. The temperature should be recorded at least once
per day, at least 7 hours apart (except weekends and holidays when
the chamber normally would not be opened). Salt solution should be
supplied from a recirculating reservoir through air-aspirating nozzles,
at a pressure between 0.7 bar (0.07 MPa) and 1.7 bar (0.17 MPa). Salt
solution runoff from exposed samples should be collected and should
not return to the reservoir for recirculation. The sample nozzles
should be shielded from condensate drippage.
4.11.4.1.3 Fog should be collected from at least
two points in the exposure zone to determine the rate of application
and salt concentration. The fog should be such that for each 80 cm2 of
collection area, 1 m1 to 2 ml of solution should be collected per
hour over a 16 hour period and the salt concentration shall be 20
± 1% by mass.
4.11.4.1.4 The nozzles should withstand exposure
to the salt spray for a period of 10 days. After this period, the
nozzles should be removed from the fog chamber and allowed to dry
for 4 to 7 days at a temperature of 20°C to 25°C in an atmosphere
having a relative humidity not greater than 70%. Following the drying
period, five nozzles should be submitted to the functional test at
the minimum operating pressure in accordance with 4.5.1 and five nozzles
should be subjected to the dynamic heating test in accordance with
3.14.2.
4.11.4.2 Nozzles intended for corrosive atmospheres
[7.12.3.2]
Five nozzles should be subjected to the tests specified
in 4.11.4.1 except that the duration of the salt spray exposure shall
be extended from 10 days to 30 days.
4.11.5 Moist air exposure test (see 3.11.4 and
3.14.2) [7.12.4]
Ten nozzles should be exposed to a high temperature-humidity
atmosphere consisting of a relative humidity of 98% ± 2% and
a temperature of 95°C ± 4°C. The nozzles are to be
installed on a pipe manifold containing de-ionized water. The entire
manifold is to be placed in the high temperature humidity enclosure
for 90 days. After this period, the nozzles should be removed from
the temperature-humidity enclosure and allowed to dry for 4 to 7 days
at a temperature of 25 ± 5°C in an atmosphere having a
relative humidity of not greater than 70%. Following the drying period,
five nozzles should be functionally tested at the minimum operating
pressure in accordance with 4.5.1 and five nozzles should be subjected
to the dynamic heating test in accordance with 3.14.2footnote.
4.12 Nozzle coating tests [7.13]
4.12.1 Evaporation test (see 3.12.1) [7.13.1]
A 50 cm3 sample of wax or bitumen should be placed
in a metal or glass cylindrical container, having a flat bottom, an
internal diameter of 55 mm and an internal height of 35 mm. The container,
without lid, should be placed in an automatically controlled electric,
constant ambient temperature oven with air circulation. The temperature
in the oven should be controlled at 16°C below the nominal release
temperature of the nozzle, but at not less than 50°C. The sample
should be weighed before and after 90 days exposure to determine any
loss of volatile matter; the sample should meet the requirements of
3.12.1.
4.12.2 Low-temperature test (see 3.12.2) [7.13.2]
Five nozzles, coated by normal production methods, whether
with wax, bitumen or a metallic coating, should be subjected to a
temperature of -10°C for a period of 24 hours. On removal from
the low-temperature cabinet, the nozzles should be exposed to normal
ambient temperature for at least 30 min before examination of the
coating to the requirements of 3.12.2.
4.13 Heat-resistance test (see 3.15) [7.14]
One nozzle body should be heated in an oven at 800°C
for a period of 15 min, with the nozzle in its normal installed position.
The nozzle body should then be removed, holding it by the threaded
inlet, and should be promptly immersed in a water bath at a temperature
of approximately 15°C. It should meet the requirements of 3.14.
4.14 Water-hammer test (see 3.13) [7.15]
4.14.1 Five nozzles should be connected, in their
normal operating position, to the test equipment. After purging the
air from the nozzles and the test equipment, 3,000 cycles of pressure
varying from 4 ± 2 bar ((0.4 ± 0.2)MPa) to twice the
rated working pressure should be generated. The pressure should be
raised from 4 bar to twice the rated pressure at a rate of 60 ±
10 bar/s. At least 30 cycles of pressure per minute should be generated.
The pressure should be measured with an electrical pressure transducer.
4.14.2 Visually examine each nozzle for leakage
during the test. After the test, each nozzle should meet the leakage
resistance requirement of 3.8.1 and the functional requirement of
3.5.1 at the minimum operating pressure.
4.15 Vibration test (see 3.16) [7.16]
4.15.1 Five nozzles should be fixed vertically
to a vibration table. They should be subjected at room temperature
to sinusoidal vibrations. The direction of vibration should be along
the axis of the connecting thread.
4.15.2 The nozzles should be vibrated continuously
from 5 Hz to 40 Hz at a maximum rate of 5 min/octave and an amplitude
of 1 mm (1/2 peak-to-peak value). If one or more resonant points are
detected, the nozzles after coming to 40 Hz, should be vibrated at
each of these resonant frequencies for 120 hours/number of resonances.
If no resonances are detected, the vibration from 5 Hz to 40 Hz should
be continued for 120 hours.
4.15.3 The nozzle should then be subjected to
the leakage test in accordance with 3.8.1 and the functional test
in accordance with 3.5.1 at the minimum operating pressure.
4.16 Impact test (see 3.17) [7.17]
4.16.1 Five nozzles should be tested by dropping a mass
onto the nozzle along the axial centreline of waterway. The kinetic energy of the
dropped mass at the point of impact should be equivalent to a mass equal to that of
the test nozzle dropped from a height 1 m (see figure 2). The mass is to be prevented from impacting more
than once upon each sample.
4.16.2 Following the test a visual examination
of each nozzle shall show no signs of fracture, deformation, or other
deficiency. If none is detected, the nozzles should be subjected to
the leak resistance test, described in 4.4.1. Following the leakage
test, each sample should meet the functional test requirement of 4.5.1
at a pressure equal to the minimum flowing pressure.
4.17 Lateral discharge test (see 3.18) [7.19]
4.17.1 Water is to be discharged from a spray
nozzle at the minimum operating and rated working pressure. A second
automatic nozzle located at the minimum distance specified by the
manufacturer is mounted on a pipe parallel to the pipe discharging
water.
4.17.2 The nozzle orifices or distribution plates
(if used), are to be placed 550 mm, 356 mm and 152 mm below a flat
smooth ceiling for three separate tests, respectively at each test
pressure. The top of a square pan measuring 305 mm square and 102
mm deep is to be positioned 152 mm below the heat responsive element
for each test. The pan is filled with 0.47 litres of heptane. After
ignition, the automatic nozzle is to operate before the heptane is
consumed.
4.18 30-day leakage test (see 3.19) [7.20]
4.18.1 Five nozzles are to be installed on a water
filled test line maintained under a constant pressure of twice the
rated working pressure for 30 days at an ambient temperature of (20
± 5°C).
4.18.2 The nozzles should be inspected visually
at least weekly for leakage. Following completion of this 30-day test,
all samples should meet the leak resistance requirements specified
in 3.8 and should exhibit no evidence of distortion or other mechanical
damage.
4.19 Vacuum test (see 3.20) [7.21]
Three nozzles should be subjected to a vacuum of 460 mm
of mercury applied to a nozzle inlet for 1 min at an ambient temperature
of 20 ± 5°C. Following this test, each sample should be
examined to verify that no distortion or mechanical damage has occurred
and then should meet the leak resistance requirements specified in
4.4.1.
4.20 Clogging Test (see 3.22) [7.28]
4.20.1 The water flow rate of an open water-mist nozzle
with its strainer or filter should be measured at its rated working pressure. The
nozzle and strainer or filter should then be installed in test apparatus described in
Figure 3 and subjected to 30 minutes of continuous flow at
rated working pressure using contaminated water which has been prepared in accordance
with 4.20.3.
4.20.2 Immediately following the 30 minutes of
continuous flow with the contaminated water, the flow rate of the
nozzle and strainer or filter should be measured at rated working
pressure. No removal, cleaning or flushing of the nozzle, filter or
strainer is permitted during the test.
4.20.3 The water used during the 30 minutes of
continuous flow at rated working pressure specified in 4.20.1 should
consist of 60 litres of tap water into which has been mixed 1.58 kilograms
of contaminants which sieve as described in table 6. The solution
should be continuously agitated during the test.
4.20.4 Alternative supply arrangements to the apparatus
shown infigure 3 may be used where damage to the pump is possible.
Restrictions to piping defined by note 2 of table 5 should apply to such systems.
Table 5 Contaminant for the
contaminated water cycling test
|
SIEVE DESIGNATION*
|
NOMINAL SIEVE OPENING, MM
|
GRAMS OF CONTAMINANT (± 5%)**
|
|
PIPE SCALE
|
TOP SOIL
|
SAND
|
| No. 25
|
0.706
|
-
|
456
|
200
|
| No.
50
|
0.297
|
82
|
82
|
327
|
| No. 100
|
0.150
|
84
|
6
|
89
|
| No. 200
|
0.074
|
81
|
-
|
21
|
| No. 325
|
0.043
|
153
|
-
|
3
|
|
|
TOTAL
|
400
|
544
|
640
|
|
* Sieve designations correspond with
those specified in the standard for wire-cloth sieves for testing
purposes, ASTM E11-87, CENCO-MEINZEN sieve sizes 25 mesh, 50 mesh, 100
mesh, 200 mesh and 325 mesh, corresponding with the number designation
in the table, have been found to comply with ASTM E11-87.
** The amount of contaminant may be
reduced by 50 per cent for nozzles limited to use with copper or
stainless steel piping and by 90 per cent for nozzles having a rated
pressure of 50 bar or higher and limited to use with stainless steel
piping.
|
|