Static pressure losses in nonmetallic flexible duct.
Prior measurements of static pressure loss for flexible ducts only considered fully stretched and naturally relaxed flat configurations that naturally contracted to about 4% with respect to the fully stretched case. Pressure loss calculation methods exist within the Air Conditioning Contractors of America (ACCA) Manual D (ACCA 1995). Chapter 35 of the ASHRAE Handbook--Fundamentals (ASHRAE 2005) also contains pressure loss data that have linear correction factors based on the percent of compression extending to 30%. Existing research by Abushakra et al. (2002a, 2002b, 2004) has shown the data included in the ACCA and ASHRAE Handbook references contains errors of 70%. The measurements presented in this paper extend the measurements previously taken to include fully stretched and compression values from 4% to 45%. In addition, the development of an "as-built" test protocol improves the applicability of the pressure loss data to actual installations. This protocol includes duct installations that are board-supported (flat), joist-supported with natural sag, and joist-supported with full or long-term sag. This provides a range of pressure losses that can be expected in field installations, depending upon the extent of the sag.
The data acquisition (DAQ) setup sequences and captures the measurements needed to determine the static pressure loss. Figure 1 shows a diagram of the test setup. ANSI/ASHRAE Standard 120-1999, Method of Testing to Determine Flow Resistance of HVAC Ducts and Fittings, requirements were used to design the system and process the data after acquisition. A computer-controlled variable-frequency drive (VFD) adjusts the airflow. The VFD allows for varying the fan rpm to provide a range of 50 to 600 cfm (24.0 to 283.1 L/s).
[FIGURE 1 OMITTED]
The pressure loss through the corresponding duct or fitting is then measured by an array of pressure transducers with an accuracy of [+ or -]0.25% for pressure losses of up to 0.25 in. [H.sub.2]O (62.2 Pa) and [+ or -]0.5% for pressure losses from 0.25 to 2.00 in. [H.sub.2]O (62.2 to 497.6 Pa). These 4-20 mA transducers produce a current proportional to the amount of applied pressure. A 249.0 [OMEGA] precision resistor ([+ or -]0.25%) converts the current loop outputs from the sensors to voltage inputs to the DAQ. The DAQ processes the voltages, and the program in the computer performs the requisite calculations and display functions.
The static pressure measurements in the test duct are performed through pressure taps set up in a piezometer ring. The ring functions as a static pressure averaging device. Each ring consists of four equally spaced and parallel connected taps. The piezometer rings meet ASHRAE Standard 120 requirements, with individual readings from each tap measuring within [+ or -]2% of one another.
The temperature measurement throughout the test run uses two silicon-junction transistor type devices and a third 1000 [OMEGA] combination platinum temperature-humidity unit. Sensor locations are (1) at the nozzle chamber, (2) at the beginning of the test section, and (3) at the end of the test section.
Figure 1 shows the nozzle chamber upstream of the duct, in accordance with ASHRAE Standard 120. This cylindrical nozzle chamber, used for duct sizes from 4 to 10 in. (10.2 to 25.4 cm), contains a 2.5 in. (6.4 cm) and a 5.0 in. (12.7 cm) flow nozzle. Pressure taps provide the pressure loss through the nozzles. The pressure transducers measure the pressure loss across the nozzles and produce the current value to the PC via the 4-20 mA loop connected to the DAQ card. The pressure loss in the nozzle is converted to mass flow rate using Equation 16 from ASHRAE Standard 120, Section 188.8.131.52.
The automated measurement control system acquires 5,000 readings for each point reported. This occurs by taking 100 readings each second and calculating the average. Next, this process repeats 50 times with each of the 50 point values stored on disk. An average of these 50 values produces a final average value. ASHRAE-validated spreadsheets were used to verify all pressure loss calculations.
AS-BUILT TEST PROTOCOL
Actual duct installations occur over joists and in hung configurations. To better approximate actual installation conditions, an "as-built" test protocol using two installation configurations was created. The first, termed "boardsupported," positioned a duct on top of a continuous flat horizontal board over the entire test length. The second, termed "joist-supported," replicates the duct installation over 1.5 in. (3.8 cm) wide supports on 24 in. (61 cm) centers. In this configuration, the duct sags between the joists when compressed and creates a test condition similar to actual installations. The natural sag test configuration occurred by removing the supporting board and allowing the flexible duct to sag under its own weight. The long-term sag, or "maximum sag" condition, required increasing the depth of the sag to represent a maximum or "worst-case" condition and then allowing the duct to retract. Minimal difference between the natural sag and the maximum sag occurred at 15% compression and below.
The tests used nonmetallic flexible duct with a singlehelix core, an R-6 insulation layer, and a foil facing outer layer (vapor barrier). The duct testing used numerous compression ratios to provide a spectrum of data for comparison. These ratios included 0% (a maximum stretch), 4%, 15%, 30%, and 45% compression. The compression ratio equals the difference between the compressed length and the maximum stretched length divided by the maximum stretched length. Setting up the compressed duct involved marking the duct in one-foot sections when fully extended and then axially compressing it to the desired ratio evenly over the length. Nonuniformities in compression increase the total pressure loss with respect to ducts with uniform compression. This approach ensures uniform longitudinal compression over the entire length of the duct under test.
The process for assembling the board-supported as-built test required uniformly compressing the duct supported by a board in a flat configuration and then performing all measurements. The process for creating the natural sag configuration required removing the board supports and letting the flexible duct sag over the 1.5 in. (3.8 cm) wide, 24 in. (61.0 cm) centered joists and then performing all measurements. Since the amount of sag can vary depending upon the installation, pressure loss measurements using two extremes of sag were measured. For natural sag, the midpoint sag distance ranges from 1 to 3 in. (2.54 to 10.6 cm) for duct compressions ranging from 4% to 45%.
Long-term sag was achieved by depressing the duct midpoint between the joists and then allowing each section between the joists to retract, emulating a longer-term sag condition. Table 1 shows the approximate sag at the midpoint between the supporting joists for the natural and the longterm sag condition, measured from duct centerline to sag centerline. At duct compression below 15%, natural sag and long-term sag are equal since insufficient duct material exists to maintain a deeper sag condition. Above 30% duct compression, long-term sag will exceed natural sag as shown in Table 1. Sag creates a dramatic increase in the pressure loss through flexible duct and needs to be taken into account in any pressure loss calculation.
Table 1. Flexible Duct Midpoint Sag Distances Sag, in. (cm) Compression 4% 15% 30% 45% 6 in. (15.2 cm) flex 0.5 2 4 7 natural sag (1.27) (5.1) (10.2) (17.8) 6 in. (15.2 cm) flex 0.5 2 6 11.5 long-term sag (1.27) (5.1) (15.2) (29.2) 8 in. (20.3 cm) flex 0.5 2 3 4 natural sag (1.27) (5.1) (10.6) (10.2) 8 in. (20.3 cm) flex 0.5 2 6 7 long-term sag (1.27) (5.1) (15.2) (17.8) 10 in. (25.4 cm) flex 0.5 1.5 2 3.5 natural sag (1.27) (3.8) (5.10) (8.90) 10 in. (25.4 cm) flex 0.5 1.5 4.5 6.5 long-term sag (1.27) (3.8) (11.4) (16.5)
The test procedure for joist and board-supported configurations exceeded the requirements in ASHRAE Standard 120-1999 (ASHRAE 1999) for all assembly, leak testing, and measurements. Measured air property variables include ambient dry-bulb temperature, barometric pressure, chamber drybulb temperature and relative humidity, and dry-bulb temperature at two points within the test duct. Measured pressure loss variables include nozzle plate static pressures, nozzle differential pressure, upstream and downstream static pressure, and test duct differential pressure.
The resulting data plots display static pressure loss as a function of volumetric flow rate for each of the three sizes of 6, 8, and 10 in. (15.2, 20.3, and 25.4 cm) duct. In each of the plots, the static pressure loss through rigid sheet metal duct of the same diameter is presented as a comparative baseline for the results. The compression configurations tested include rigid sheet metal duct, maximum stretch flexible duct, 4% compressed flexible duct, 15% compressed flexible duct, 30% compressed flexible duct, and 45% compressed flexible duct. Each compression configuration contains data for both board and joist-supported configurations.
Rigid Sheet Metal Duct
Rigid sheet metal duct was tested for each size for agreement with existing ASHRAE/ACCA numbers (ASHRAE 2005). The rigid duct was tested under the same volumetric flow rate range as the flexible duct. Resulting values for the rigid duct showed agreement to within [+ or -]5% of ASHRAE values in the 2005 Handbook.
Maximum Stretch Configurations
Results for the maximum stretch case and rigid duct showed agreement within 2% (Figure 2). For comparative purposes, rigid sheet metal duct was tested utilizing both 3 ft (0.914 m) and 5 ft (1.524 m) standard, commercially available section lengths in the 6 in. (15.24 cm) size. This comparative testing allows the individual contributions of transition and length to be ascertained. The resulting data showed that section length has less than a 5% effect on the static pressure loss over the measured flow range.
[FIGURE 2 OMITTED]
4% Compression Configurations
Four percent compression revealed substantial increases in static pressure loss as shown in Figure 3. A 4% compression rate results in 1 ft (30.48 cm) of compression for a 25 ft (7.62 m) length, resulting in 25 ft (7.62 m) of flexible duct running 24 ft (7.32 m). The duct weight caused the natural sag to occur when the supporting boards were removed at the completion of the board-supported tests (flat configuration). At 4% compression, very little sag occurred. The data from the ASHRAE Handbook generally agree with the data taken, with the condition that the Pressure Loss Correction Factor increases when the ducts sag. Some variations from experimental setups are expected due to the sensitivity to the pressure loss as a function of the evenness of the compression and the uniformity of sag.
[FIGURE 3 OMITTED]
15% Compression Configurations
Figure 4 shows the 15% compression data. These values were found to be quite sensitive to the uniformity of the compression, and variations from these values should be expected in field installations.
[FIGURE 4 OMITTED]
30% Compression Configurations
Figure 5 shows the 30% compression pressure loss. Again, these values were found to be quite sensitive to the uniformity of the compression.
[FIGURE 5 OMITTED]
45% Compression Configurations
Figure 6 shows the 45% compression pressure loss.
[FIGURE 6 OMITTED]
COMPARISON TO PREVIOUS WORK
A previous study (Abushakra et al. 2002a, 2002b, 2004) examined the effects of compression on the static pressure loss through flexible duct. The study tested flexible duct in a drawthrough configuration with nominal compression ratios of maximum stretch, 15%, and 30%. Tables 2 and 3 display the results of the current measurements and the Abushakra et al. measurements for nonsag straight ducts with 0% (maximum stretch), 15%, and 30% compression. It should be noted that the current testing used a blow-through configuration while Abushakra et al. used a draw through configuration, so the data cannot be directly compared. However, the two data sets do show similar results.
Table 2. Comparison to Previous Work (I-P Units) Flow, Max. Stretch, 15% Board 30% Board cfm in. Supported, in. Supported, in. [H.sub.2]O/100 [H.sub.2]O/100 [H.sub.2]O/100 ft ft ft Abushakra et al. 6 in. 100 0.109 0.458 0.984 8 in. 200 0.078 0.308 0.498 10 in. 300 0.062 0.221 0.344 Measured 6 in. 100 0.081 0.561 1.052 8 in. 200 0.073 0.382 0.718 10 in. 300 0.054 0.229 0.361 Table 3. Comparison to Previous Work (SI Units) Flow, L/s Max. Stretch, 15% Board 30% Board Pa/m Supported, Supported, Pa/m Pa/m Abushakra et al. 15.2 cm 47.2 0.893 3.740 8.035 20.3 cm 94.4 0.640 2.518 4.063 25.4 cm 141.6 0.507 1.802 2.805 Measured 15.2 cm 47.2 0.658 4.577 8.585 20.3 cm 94.4 0.598 3.116 5.859 25.4 cm 141.6 0.437 1.867 2.949
Nonmetallic flexible duct pressure losses, at maximum stretch, fall within [+ or -]2% of rigid sheet metal losses. At compression values over 4%, nonmetallic flexible duct exhibits 2 to over 10 times increased pressure losses over fully extended flex duct.
The experimental results also demonstrate that with compression ratios exceeding 4%, the duct performance varies considerably with slight variations in the installation. The results for the as-built test protocol need to be used as a range of values that can be encountered in field installations since nonuniform compression increases duct pressure loss above the values derived from the pressure loss equations for straight, natural sag, and maximum sag configurations.
This work was funded by the Air Distribution Institute, ASHRAE Research Project 1333, Texas Utilities, and Lennox Industries. We thank them for their support of this research.
REFERENCES AND BIBLIOGRPHY
Abushakra, B., D. Dickerhoff, I. Walker, and M. Sherman. 2002. Laboratory study of pressure losses in residential air distribution systems, LBNL Report LBNL-49293. Berkeley, CA: Lawrence Berkeley National Laboratory.
Abushakra, B., I.S. Walker, and M.H. Sherman. 2004. Compression effects on pressure loss in flexible HVAC ducts. HVAC&R Research 10(3):275-89.
Abushakra, B., I.S. Walker, and M.H. Sherman. 2002. Laboratory study of pressure losses in residential air distribution systems. Proc. ACEEE Summer Study 2002. Washington, DC: American Council for an Energy Efficient Economy.
ACCA. 1995. Residential Duct Systems, Manual D. Washington, DC: Air Conditioning Contractors of America.
Altshul, A.D., and P.G. Kiselev. 1975. Hydraulics and Aerodynamics. Moscow: Stroisdat Publishing House.
ASHRAE. 2005. 2005 ASHRAE Handbook--Fundamentals. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
ASHRAE. 1999. ANSI/ASHRAE Standard 120-1999, Methods of Testing to Determine Flow Resistance of HVAC Air Ducts and Fittings. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
Student Member ASHRAE
Charles Culp, PhD, PE
Kevin Weaver, now a senior engineer with BesTech, Dallas, Texas, conducted this research as part of his Master's thesis at Texas A&M University, College Station, Texas. Charles Culp is an associate professor in the Department of Architecture and associate director of the Energy Systems Laboratory, Texas A&M University.
|Printer friendly Cite/link Email Feedback|
|Author:||Weaver, Kevin; Culp, Charles|
|Article Type:||Technical report|
|Date:||Jul 1, 2007|
|Previous Article:||Simulation of the impact of commercial building envelope airtightness on building energy utilization.|
|Next Article:||Study on the energy conservation performance of the plate enthalpy exchanger.|