Fan Testing: Test Codes
Fan Testing: Test Codes
Fan Testing: Test Codes
Fan Testing
Test Codes
Various engineering societies and industry organizations throughout the
world have published fan test codes.
The Air Movement and Control Association, Inc. (AMCA), publishes a
number of standards dealing with fan testing. The American Society of
Heating, Ventilating and Air Conditioning Engineers, Inc. is a joint sponsor of
the basic laboratory standard, ANSI/AMCA 210-ANSI/ASHRAE 511 entitled
Laboratory Methods of Testing Fans for Rating. This standard provides the
rules for establishing the aerodynamic performance of a fan in the laboratory.
Other AMCA standards for aerodynamic testing are: ANSI/AMCA 230,
Laboratory Methods of Testing Air Circulator Fans for Rating, ANSI/AMCA
240, Laboratory Methods of Testing Positive Pressure Ventilator (PPV) Fans
for Rating, and AMCA 250, Laboratory Methods of Testing Jet Tunnel Fans
for Aerodynamic Performance Rating.
1
Most test codes bear a number signifying the year of last revision or reaffirmation. These
designators will generally be omitted in the text of this handbook. References to specific
items in a code will cite the year of publication.
The low velocities associated with some ratings cannot always be meas-
ured accurately by pitot-tube tests. Better accuracy can usually be obtained
by using flow nozzles, if substantial exit velocities can be generated. The
pressure needed to produce these velocities can be supplied by the test fan or
by an auxiliary fan if the test fan is incapable. If the fan does develop enough
pressure, the nozzles can be placed at the end of the test duct. Suitable
approach conditions must also be provided. When an auxiliary fan is used, a
series of nozzles can be permanently set up in a chamber as shown in Figure
13.1C. Variations in point of rating are produced by plugging and unplugging
various combinations of nozzles. Alternatively, a single nozzle or a nozzle
Venturi can be inserted into the duct between the auxiliary fan and the cham-
ber, as shown in Figure 13.1D.
Figures 13.1A to 13.1D give the chief dimensions for various laboratory-
test setups. Other combinations are possible; for example, either the pitot
traverse or the multiple nozzles can be used with an inlet chamber. Specific
numbers are assigned to various sections or planes of reference as shown.
If a fan is to be furnished with bearings, it should be tested on its actual
shaft and bearings after a suitable "run in" period. The inlet and outlet should
be unobstructed except for the bearings and supports, and any other appurte-
nances, such as screens or dampers, that are specified. Inlet bells and dis-
charge cones, if contemplated, should also be in place.
Suitable provisions should be made for driving the fan and measuring the
input power by using dynamometers, torsion elements, or calibrated electric
motors. If the driver is an integral part of the fan and helps to determine the
airflow passages, it should be either in place or adequately modeled.
The fan may have to be rotated to provide a suitable discharge direction
on the test block. But, the relative direction of inlet-box entry to the discharge
direction should not be changed, since this may affect performance.
The room in which the test is conducted should be free from any air
currents that might affect fan performance. Whenever it is necessary to
discharge the air into another room, provisions should be made for makeup
air. An adequate, whirl-free supply of uniform-density air should be available
to the fan.
The acoustic properties of the room will determine the type of sound tests
that can be performed.
For small fan units, it may be possible to modify the ductwork in order to
simulate laboratory-test-block conditions. However, for fans of any
appreciable size, this is not feasible. Similarly, for small fan units, the driving
motor can be removed and replaced by a calibrated driver. But, for large
units, this is not feasible.
Whenever the field installation differs from an ideal laboratory setup, the
effects of the differences should be taken into account. If this accounting is
done at the specification stage, the ratings should be based on the fan as it is
to be installed. This may require that the vendor derate the laboratory
performance or that special laboratory tests be conducted on models to
determine the effects of the differences in setup. However, if the accounting
is done after the fan has been installed, a tolerance may have to be applied to
the vendor's predicted performance. AMCA 201 includes system-effect-
factor information that is useful for this.
Field noise tests should be very carefully evaluated to take into account
other noise sources and the difficulties of measurement.
Precautions are necessary to ensure that the mass flow rate at the
measuring station is the same as that passing through the fan. Appropriate
tests for leakage between the measuring station and the fan should be
conducted. The area at the measuring station should be determined by
internal inspection. Internal bracing and accumulations of dust or other
materials should not be allowed.
The average specific kinetic energy eK and the average velocity pressure
pV can be obtained from Equations 2.110 and 2.111, respectively. These
include the distributional effects noted above. The average total pressure pT
is simply the sum of the average static pressure and the average velocity
pressure as indicated by Equation 2.112.
Based on the above definitions and averages, the fan specific energy y F
can be determined from
pS 2 − pS 1
yF = + eK 2 − eK 1 . (13.1)
ρm
pFT = pT 2 − pT 1 . (13.2)
Referring to Equations 2.23 and 2.26 will show that the potential energy terms
have been considered negligible, which is usual for fans.
2 π LFN
Ps = . (13.3)
CW
The conversion constant CW has a value of 33000 for power in hp, distance in
ft, force in lb and speed in rpm. For SI units of W, m, N, and rps, CW = 10
. .
Table 13.1 lists some convenient lengths and the corresponding U.S.
customary values of the dynamometer constant 2πL CW .
The restraining force on the arm can be measured with a yardarm type
balance, a dial-type scale, or a suitable force gauge. If the weight of the arm
is not perfectly counterbalanced (by means of an exact counterweight on the
stator), the indicated force Fi will have to be corrected. This correction,
called tare Ft can be determined by running the dynamometer, without load,
at the proper speed and noting the scale reading. The corrected force Fc may
be smaller or greater than the measured force, depending on whether there is
positive or negative tare, so
Fc = Fi + Ft . (13.4)
The arm should be in exactly the same position for each reading. When
dial scales are used, the deflection may become so great that an adjustment
will be needed. Such an adjustment is never required with a yardarm-type
balance, which always returns the arm to the same position.
Calibration tests are not usually needed for electric dynamometers. By
measuring tare at the proper speed, the effects of bearing and windage losses
may be cancelled out. Electrical connections must be completely flexible, and
the bearings should be in good condition and well lubricated.
Fan input power can also be found by measuring the electrical input to a
calibrated motor. Calibration tests are made with some form of absorption
dynamometer. If the electrical input is measured in terms of current I in
amps and potential E in volts and if the corresponding efficiency η is
determined from the calibration, then the input power of the fan, if driven by a
direct-current motor, can be calculated from
EIη
Ps = . (13.5)
K
When alternating-current motors are used, the actual power will differ
from the apparent power indicated by the amps and volts. Power factor cosθ
is the ratio of actual to apparent power for single-phase or three-phase AC
motors.
For single-phase AC motors,
EI cosθ
Ps = . (13.6)
K
3EI cosθ
Ps = . (13.7)
K
The conversion constant K has a value of 746 for power in hp and 1000 for
power in kW. A wattmeter can be used to find the actual power input without
having to measure amps, volts, and power factor separately. For three-phase
current, either the two-wattmeter method of measuring power or a polyphase
wattmeter should be used.
If a torsion meter is used, the torque T is obtained from a calibration of
the change in resistance for a strain-gage element bonded to the transmission
shaft. A separate prime mover is needed. Fan input power can be determined
from the torque and speed by using
2 π TN
Ps = . (13.8)
CW
The conversion constant CW has a value of 33000 for power in hp, torque in
ft-lb, and speed in rpm. For SI units of W, m-N, and rps, CW = 1.0.
In field testing a fan driven by an uncalibrated prime mover, reasonable
values of motor or engine efficiency based on the manufacturer's tests of
similar models should be used. The uncertainty in the fan input power will, of
course, depend on the uncertainty in the efficiency values used.
120 f 60n
NF = − . (13.9)
np t
This equation will yield fan speeds in rpm for frequencies in Hertz and time
intervals in seconds.
Measuring Thrust
Thrust determinations for jet fans require that the fan either be suspended
or supported in such a way that it is held level while allowing sufficient
horizontal movement to measure the force needed to restrain it. Various
means of measurement are detailed in the test standards including load cells
and spring balances.
Calibrations
The instruments used in a fan test should be calibrated and the calibration
corrections applied to individual measurements. Measurements at a single
point should be temporally averaged before applying the calibration correc-
tion. However, calibration corrections must be applied to these temporal
averages before spatial averaging is performed. Refer to the measurement-
Test Results
The results of a fan test should be expressed in terms of either the fan
mass flow rate or the fan volume flow rate, either the fan specific energy or
the fan pressure, and the fan input power. These three quantities represent the
basic performance of the fan. However, performance also depends on the
operating conditions, so fluid density and fan speed, too, should be included
in the results. It may be desirable to calculate the fan efficiency and list it
with the results. The uncertainties in each of the results should be calculated
from an analysis of the uncertainties in each of the measurements leading to
that result. Refer to the section on propagation of uncertainties into a result,
which appears in the chapter on engineering statistics.
If the operating conditions differ from the specifications, use the fan laws
and appropriate compressibility corrections to convert actual results into
results corresponding to specifications. Refer to the chapter on fan laws for
discussions and examples of the use of the fan laws and compressibility
coefficients. Note that additional uncertainties will be introduced if conver-
sion calculations are performed. Both AMCA 210 and ASME-PTC 11
address this matter.
When tests are performed at various points of rating, curves can be fitted
using procedures outlined in the chapter on engineering statistics. Flow rate
should be plotted as abscissa, and both specific output and input power should
be plotted as ordinates.