|
Field Survey of Trunk Line Heating Gas Leaks in Commercial Broiler Houses
Jeremiah D. Davis1,2*, Baylor D. Arnold2, Cody R. Smith2, Kelly G. Griggs2, Martha S. Rueda2, Jesse C. Campbell1,2, Carson M. Edge2, Joseph L. Purswell3
Published in Applied Engineering in Agriculture 39(3): 279-283 (doi: 10.13031/aea.15470). 2023 American Society of Agricultural and Biological Engineers.
1Biosystems Engineering, Auburn University, Auburn, Alabama, USA.
2National Poultry Technology Center, Auburn University, Auburn, Alabama, USA.
3Poultry Research Unit, USDA Agricultural Research Service, Mississippi State, Mississippi, USA.
* Correspondence: j.davis@auburn.edu
The authors have paid for open access for this article. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License https://creative commons.org/licenses/by-nc-nd/4.0/
Submitted for review on 22 November 2022 as manuscript number PAFS 15470; approved for publication as a Research Brief by Associate Editor Dr. Yang Zhao and Community Editor Dr. Shafiqur Rahman of the Plant, Animal, & Facility Systems Community of ASABE on 24 March 2023.
Mention of company or trade names is for description only and does not imply endorsement by the USDA. The USDA is an equal opportunity provider and employer. Trade and brand names used in this publication are given for information purposes only. No guarantee, endorsement, or discrimination among comparable products is intended or implied by the National Poultry Technology Center. Any opinions, findings, conclusion, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture.
Highlights
Abstract. A field survey was conducted on 23 farms (101 broiler houses ranging in age from new to 33 years old) in central and south Alabama to evaluate the presence of gas leaks on the high-pressure trunk line. Gas trunk lines usually run on the outside of poultry houses exposing them to temperature fluctuations that cause expansion and contraction of the rigid steel piping. Each threaded connection has the potential to leak. A soap solution in a hand sprayer was used to spray each connection and evaluate the presence of a leak through bubble formation. Leaking fittings were evaluated as percentage of total fittings to account for variation across houses. Pearson correlation matrices were used to determine correlations of house age and floor area on total number of fittings and percentage of leaking fittings. Total number of fittings ranged from 36 to 170 per house and was strongly correlated to floor area [r(99) = 0.70, p < 0.0001]. Larger floor areas require more heaters and longer piping systems. Percentage leaking fittings ranged from minimum of 0% to maximum of 32% per house and was not correlated with either house age [r(99) = 0.03, p = 0.74] or floor area [r(99) = 0.05, p = 0.65] for the houses surveyed. Leaks seemed to be more about regulator quality and installation craftsmanship with variation across farms and houses within farms. Of houses surveyed, the overall mean leaking fittings was 6.9% ± 0.6%. Poultry producers should get written verification from the gas line contractor/installer that newly constructed houses are free of leaks before operating the farm. It would be useful for poultry producers to check for leaks annually to verify system integrity and to monitor any developing leaks.
In recent years, there have been many studies working to estimate fugitive emissions (gas leaks) from natural gas distribution systems at the city (Phillips et al., 2013; Jackson et al., 2014; Gallagher et al., 2015) and national levels (Weller et al., 2020). Jackson et al. (2014) found 5,893 gas leaks in the pipeline across 1,500 miles of roads in Washington, DC. Gallagher et al. (2015) identified 1050, 351, and 132 leaks in Manhattan, Cincinnati, and Durham, respectively, across 247, 750, and 595 road miles. In developing an improved national estimate of fugitive emissions, Weller et al. (2020) found an interaction between pipeline materials and system age with the number of leaks increasing with age. The authors were not able to find any similar studies performed on agricultural operations as it pertains to fugitive emissions from heating gas piping distribution systems.
In the Southeastern U.S., propane and natural gas are the common sources of heat energy for broiler houses during brooding and the early portion of the growout phase. Poultry farms have approximately 75 to 150 m (250 to 500 ft) of rigid gas piping and connections. Gas is pumped from a tank or meter through a higher pressure (69 kPa;10 psi) trunk line made of black steel piping and is then stepped down using a regulator to a lower pressure line (3.5 to 34.5 kPa; ½ to 5 psi) to the heaters (fig. 1). High-pressure gas trunk lines usually run on the outside of poultry houses exposing them to temperature fluctuations that can cause expansion and contraction of the steel piping. Each threaded connection has the potential to leak costing the producer money and creating a potential fire safety hazard. These gas leaks can also increase a farm’s greenhouse gas footprint. During a few preliminary surveys, we found leaks to be more common and larger in size on the high-pressure trunk lines than in the low-pressure lines inside the houses that connect the heaters. Therefore, a field survey was conducted on commercial broiler houses in central and south Alabama to quantify the presence of gas leaks on the high-pressure trunk line.
 |
| Figure 1. Simplified schematic of a broiler house heating system layout illustrating the high-pressure trunk line between the fuel source and the low-pressure heater branches. Stars indicate representative areas where piping was rigidly fastened to the structure or fixed at the ground and does not allow for complete expansion and contraction. |
Materials and Methods
Leak Detection Methods
Before conducting the test, circuit breakers to the heaters were turned off to prevent accidental ignition while heaters are in the stowed position. Gas was turned on at the tank or at the natural gas meter to pressurize the trunk line. Section 8.1.5 of NFPA 54 (2021) states that leaks should be located with an approved gas detector, a noncorrosive leak detection fluid, or other approved method. Soaps containing ammonia can react with brass fittings when present and should be avoided. A soap solution was created using 4 L (~1 gal) of water mixed with 180 mL (¾ cup) of liquid dish soap (Dawn, Proctor & Gamble, Cincinnati, Ohio) and 60 mL (¼ cup) of glycerin (Equate, Walmart, Bentonville, Ark.). Industry professionals have found that glycerin increased viscosity and bubble stability outdoors in the wind. The soap solution was mixed in a 7.6 L (2 gal) hand pump sprayer and used to evaluate each connection on the trunk line. Beginning at the propane tank or meter, all threaded fittings (regulators, threaded couplings, unions, 90° elbows, tees, and caps) were sprayed with the soap solution. After a few seconds, the fittings were evaluated for the presence of a leak. Observed bubble formation was considered to be evidence of a leak. When a leak was found, the connection was flagged with marking tape, or marking paint was sprayed on the ground. If both sides of a coupling had leaks, then each side was flagged. The number of leaks for each house was totaled with a hand tally counter. A complete list of steps and a worksheet used to identify the gas leaks are detailed in Arnold et al. (2021).
Statistics
Our team worked with integrators to find approximately 100 participant farms that had availability without birds during our weekly testing windows. The number of leaks for each house were evaluated as a proportion of total fitting connections to account for the varying number of fittings used on each farm dependent of house floor area and type and size of heaters utilized. Pearson correlation matrices were calculated to determine possible correlations between broiler house age and floor area to the total number of fittings as well as the percentage of leaking fittings.
Results and Discussion
During the summer and fall of 2021, a total of 101 houses on 23 broiler farms, ranging in age from new to 33 years old, were evaluated for gas leaks on the high-pressure trunk line (table 1). The number of houses on each farm ranged from two to eight. The survey included 90 houses using propane gas and 11 houses using natural gas. Trunk lines were constructed of 9 to 13 cm (3/8 to ½ in.) copper tubing between the propane tank regulator and the house and then 13 to 19 mm (½ to ¾ in.) black rigid steel piping fastened to the roof eave along the house length. It took approximately 0.5 personnel-hours to evaluate each house with two people; one spraying each connection while the second followed and marked leaks. To reduce time and the need for a ladder, the team used marking spray paint to mark the ground beneath each leak. A four-house farm could be inspected in approximately 2 h.
| Table 1. Summary statistics for leaking fittings in the 101 broiler houses tested. | ||||||||||
| Farm ID | Houses Per Farm (No.) | Farm Age (years) | House Width (ft) | House Length (ft) | Floor Area (ft2) | Fuel (type) | Pipe Fittings Range (No.) | Leaking Fittings | ||
| Range (No.) | Range (%) | Mean ± SEM (%) | ||||||||
| A | 4 | 0 | 60 | 500 | 30,000 | propane | 102 | 0-1 | 0.0-1.0 | 0.7 ± 2.1 |
| B | 4 | 0 | 60 | 500 | 30,000 | propane | 126 | 0-1 | 0.0-0.8 | 0.6 ± 2.1 |
| C | 4 | 2 | 60 | 500 | 30,000 | propane | 86-87 | 13-26 | 15.1-29.9 | 19.1 ± 2.1 |
| D | 3 | 4 | 66 | 600 | 39,600 | natural gas | 170 | 2-10 | 1.2-5.9 | 2.9 ± 2.4 |
| E | 4 | 5 | 60 | 500 | 30,000 | propane | 122 | 0-1 | 0.0-0.8 | 0.2 ± 2.1 |
| F | 2 | 6 | 60 | 500 | 30,000 | propane | 117 | 2-9 | 1.7-7.7 | 4.7 ± 3.0 |
| G | 2 | 7 | 55 | 550 | 30,250 | propane | 88 | 1-2 | 1.1-2.3 | 1.7 ± 3.0 |
| H | 4 | 7 | 60 | 500 | 30,000 | propane | 107 | 6-15 | 5.6-14.0 | 10.5 ± 2.1 |
| I | 3 | 8 | 60 | 500 | 30,000 | propane | 106 | 0-8 | 0.0-7.5 | 3.1 ± 2.4 |
| J | 4 | 8 | 60 | 500 | 30,000 | propane | 115 | 4-12 | 3.5-10.4 | 6.1 ± 2.1 |
| K | 8 | 7-13 | 43 | 510 | 21,930 | natural gas | 36-57 | 4-16 | 8.5-32.0 | 18.1 ± 3.1 |
| L | 2 | 13 | 66 | 600 | 39,600 | propane | 115-117 | 9-11 | 7.8-9.4 | 8.6 ± 3.0 |
| M | 6 | 14 | 42 | 500 | 21,000 | propane | 75-83 | 2-9 | 2.4-12.0 | 6.3 ± 1.7 |
| N | 2 | 14 | 66 | 600 | 39,600 | propane | 115-126 | 7-12 | 5.6-10.4 | 8.0 ± 3.0 |
| O | 3 | 14 | 66 | 600 | 39,600 | propane | 113-117 | 15-17 | 13.3-14.8 | 13.9 ± 2.4 |
| P | 4 | 15 | 42 | 500 | 21,000 | propane | 67-76 | 1-4 | 1.5-5.9 | 4.2 ± 2.1 |
| Q | 6 | 16 | 40 | 500 | 20,000 | propane | 78-83 | 1-10 | 1.3-12.7 | 4.8 ± 1.7 |
| R | 6 | 17 | 40 | 500 | 20,000 | propane | 69-83 | 1-5 | 1.2-7.2 | 3.1 ± 1.7 |
| S | 8 | 18 | 40 | 500 | 20,000 | propane | 73-110 | 1-4 | 1.0-5.5 | 3.0 ± 1.5 |
| T | 2 | 19 | 66 | 600 | 39,600 | propane | 53-54 | 3-9 | 5.6-17.0 | 11.3 ± 3.0 |
| U | 8 | 23 | 40 | 500 | 20,000 | propane | 90-98 | 5-16 | 5.5-17.0 | 10.3 ± 1.5 |
| V | 6 | 23-24 | 40 | 500 | 20,000 | propane | 61-69 | 1-7 | 1.4-10.8 | 4.9 ± 1.7 |
| W | 6 | 33 | 40 | 330 | 13,200 | propane | 49-61 | 1-5 | 1.8-9.6 | 5.7 ± 1.7 |
| Overall | 101 | 0-33 | 36-170 | 0-26 | 0.0-32.0 | 6.9 ± 0.6 | ||||
 |
| Figure 2. A selection of locations on the trunk line with leaks illustrated with foaming soap solution. Leaks at threaded connections include tees (a and h), couplings (b), unions (c and d), regulators (e, f, and g), and at the fuel tank (i). Regulator gaskets were a common source of large leaks (j-n). |
Figure 2 illustrates the variety of gas leaks found with the soap solution across the houses surveyed. Figures 2a-2h shows leaking threads in pipe connections along the trunk line. Visual inspections of these leaks varied from large leaks demonstrated by quick formation of large bubbles (fig. 2a) to relatively small leaks demonstrated by slow formation of small bubbles (figs. 2d, h). Leaks were observed at locations across the length of the trunk line, however, leaks were more concentrated near transitions with piping rigidly fastened to the structure or fixed at the ground that does not allow for complete expansion and contraction (starred locations in fig. 1, between the tank or meter entrance and the trunk line, and at the heater branch connections entering the house). The most common component of the trunk line that demonstrated large leaks was regulators at the propane tank that step the tank pressure down to line pressure (figs. 2j-2n). The four regulators with large leaks shown in figures 2k-2m were from each of the four broiler houses on Farm P.
Table 2 summarizes the Pearson correlation matrix for house age and floor area on the total number of fittings and the percentage of leaking fittings. There was a significant (p<0.0001) but moderate (r = -0.66) negative correlation between house age and the floor area of broiler houses house. There has been a trend in the industry to build new houses with larger floor areas to reduce the overall cost of construction.
Floor area was significantly (p<0.0001) and strongly correlated (r = 0.70) to the total fittings on the gas trunk line (table 2). The total number of connections between the fuel source and the low-pressure step-down regulator ranged from 36 to 170 per house and depended upon the house length and the number of branches required by the heater layouts. Total fittings increased approximately four times per house between the oldest and newest farms surveyed.
The percentage of total fittings that were leaking did not correlate to either house age (r = 0.03, p = 0.74) or floor area (r = 0.05, p = 0.65) for the houses surveyed (table 2). The number of gas leaks per house ranged between zero and 26 leaks (table 1). As a percentage of total fittings, leaks ranged from 0% to 32% per house. The overall mean rate of leaking fittings was 6.9% ± 0.6%.
| Table 2. Pearson correlation matrix with coefficients (r), probability values (p), and number of observations (n), for house age, floor area, total number of fittings and proportion of leaking fittings on tested broiler houses. | ||||
| Floor Area | Total Fittings | Percent Leaking Fittings | ||
| Value | (m2) | (No.) | (%) | |
| House Age (years) | r | -0.66 | -0.57 | 0.03 |
| p | <0.0001 | <0.0001 | 0.74 | |
| n | 101 | 101 | 101 | |
| Floor Area (m2) | r | 1.00 | 0.70 | 0.05 |
| p | <0.0001 | 0.65 | ||
| n | 101 | 101 | ||
Figure 3 illustrates the variation in mean percentage leaking fittings across the 23 farms with farms sorted from newest to oldest. The overall mean (6.9% ± 0.6%) is shown for comparison. The eight newly constructed houses on Farms A and B had minimal leaks, 0.7% ± 2.1% and 0.6% ± 2.1%, respectively. Two houses had zero leaks and the remaining six had a single leak. It would be expected to have zero leaks for new houses as these systems should have been pressure tested according to Sections 8.1.4 of the NFPA 54 (2021) by the plumbing contractor/installer to verify the piping could hold a sustained gas pressure over a period of time.
Farm C was only three years old and had a farm leaking fittings rate of 19.1% ± 3.6% with one house having 30% of fittings leaking. This farm was almost three times the overall mean. Farm K was an eight house farm that was built in three stages. Mean leaking fittings was 18.1% ± 3.1% with houses ranging from 8.5% to 32%. The house with 32% leaking fittings had little visible pipe thread compound on each fitting. This producer stated that his fuel consumption had increased but did not know there were that many leaks.
 |
| Figure 3. Box and whisker plot of the variation in percent leaking fittings for houses on each farm. Farms were sorted from newest to oldest. The farm mean is shown by (+) and the farm median is shown by (-). The overall mean for the percent leaking fittings was included for reference. |
Most producers in this survey had never checked for gas leaks unless there was a catastrophic event that either made significant noise, smelled, or had emptied a propane tank. Based on this survey, it is recommended to conduct an annual gas leak check using the steps detailed in Arnold et al. (2021) and recording any leaks using the accompanying “identifying gas leaks worksheet.” It is important for producers to monitor gas piping over time as this survey showed that the number of leaks can be variable across farms. This survey also showed that one or two houses on many farms can have much higher leaks and should be monitored and addressed first.
Commercial broiler houses in the U.S. use rigid black steel piping to plumb the heating systems. After installation, plumbing contractors typically use a protocol based on Section 8.1.4 of NFPA 54 (2021) to pressure test the system for a period of time to ensure it is free of leaks. If any leaks are present, the contractor should repair the leaks and confirm the system will hold pressure. Poultry producers should get written verification from contractors/installers that every newly constructed house on their farm is free of leaks before operating the farm. There are no required follow ups or routine inspections made by contractors after initial installation to identify gas leaks during the life of the system, this responsibility falls to the producer.
Rigid gas plumbing systems are typically not designed for easy field adjustment, tightening of fittings, or repairing of leaks that develop after the initial installation. Many of the leaks that were identified in this survey could not easily be tightened or repaired. In many cases, it would require a total replumbing of the entire system, approximately 75 to 150 m (250 to 500 ft), to fix a few leaks. It will likely be cost prohibitive to repair a few small leaks in the middle of the system. Though we did find a few leaking unions on some of the houses, it would be helpful for periodic unions to be installed along the system that would allow for quick field repairs to reduce the cost of repair, reduce fuel loss, improve safety, and save operating costs.
Although, this study illustrates broiler houses having a large proportion of fittings leaking on the high-pressure gas trunk line above ground, the piping below ground and inside the houses was not evaluated for leaks and could be an additional source of fuel loss. Although Section 8.1.4 of NFPA 54 (2021) provides a method to determine that leaks are present, there is not a good method for producers to estimate how much fuel is wasted through these leaks to determine the cost benefit of making repairs. Further research should be performed to quantify the volume of gas lost through these trunk line leaks to provide a cost estimate to assist poultry producers when making decisions to repair.
Conclusions
- Poultry producers should get written verification from gas line contractors/installers that newly constructed houses are free of leaks before operating the farm.
- The percentage of leaking fittings on broiler houses surveyed were not correlated with either house age or floor area. However, leaks seemed to be more about regulator quality and installation craftsmanship with variation across farms and houses within farms. Of houses surveyed, the overall mean leaking fittings was 6.9% ± 0.6%.
- A producer can spend less than 30 min per house to evaluate the high-pressure gas piping trunk lines and should check for leaks annually or when they notice an increase in fuel usage to verify system integrity.
- Rigid gas piping systems should include periodic union fittings to allow for easy repair of leaks.
- Further research should be performed to quantify the volume of gas lost through these trunk line leaks to provide a cost estimate to assist poultry producers when making decisions to repair.
Acknowledgments
This work was completed as part of the National Poultry Technology Center, Research and Extension Experience for Undergraduates (NPTC REEU). Funding was provided in part by USDA project [58-6064-0-010], by the Alabama Agricultural Experiment Station, Auburn University, the Alabama Cooperative Extension System, and the Hatch program of the National Institute of Food and Agriculture, U.S. Department of Agriculture. We would like to thank Wayne-Sanderson Farms, Koch Foods, and Tyson Foods and their producers for participating in this effort.
References
Arnold, B. D., Davis, J. D., Campbell, J. C., Griggs, K. G., Smith, C. R., Rueda Lastres, M. S.,... Purswell, J. L. (2021). NPTC Tools of the Trade: checking for gas leaks. ANR-2835. Alabama Cooperative Extension System. Retrieved from https://www.aces.edu/blog/topics/farming/nptc-tools-of-the-trade-identifying-gas-leaks/
Gallagher, M. E., Down, A., Ackley, R. C., Zhao, K., Phillips, N., & Jackson, R. B. (2015). Natural gas pipeline replacement programs reduce methane leaks and improve consumer safety. Environ. Sci. Technol. Lett., 2(10), 286-291. https://doi.org/10.1021/acs.estlett.5b00213
Jackson, R. B., Down, A., Phillips, N. G., Ackley, R. C., Cook, C. W., Plata, D. L., & Zhao, K. (2014). Natural gas pipeline leaks across Washington, DC. Environ. Sci. Technol., 48(3), 2051-2058. https://doi.org/10.1021/es404474x
National Fire Protection Agency. (2021). NFPA 54: National fuel gas code. Quincy, MA: NFPA.
Phillips, N. G., Ackley, R., Crosson, E. R., Down, A., Hutyra, L. R., Brondfield, M.,... Jackson, R. B. (2013). Mapping urban pipeline leaks: Methane leaks across Boston. Environ. Pollut., 173, 1-4. https://doi.org/10.1016/j.envpol.2012.11.003
Weller, Z. D., Hamburg, S. P., & von Fischer, J. C. (2020). A national estimate of methane leakage from pipeline mains in natural gas local distribution systems. Environ. Sci. Technol., 54(14), 8958-8967. https://doi.org/10.1021/acs.est.0c00437