The primary goal for this project is to develop a liquid seal that demonstrates the potential for methane emissions reduction of at least 95% of 1% of the total compressor mass flow compared to the typical leakage rate of state-of-the-art dry seal packing systems. The work proposes to design, build, assemble, and operate a liquid seal. This includes the design, development, and fabrication of all components related to the seal such as the hydraulic support system. Component level commissioning will be performed in a staged order. Both static and dynamic testing of the system will be performed. A secondary goal is to develop and validate a leakage model for existing reciprocating compressor packing leakage.
Southwest Research Institute (SwRI), San Antonio, TX 78238
NextSeal AB, Stureplan 15, SE-11145 Stockholm, Sweden
Williams Gas Pipeline, Houston, TX 77056
Methane emissions from reciprocating compressors in the U.S. natural gas industry account for over 72.4 Bcf per year according to a 2006 statement by the United States Environmental Protection Agency . Methane has a global warming potential 50 times stronger than carbon dioxide, and reciprocating compressors are the machinery type with the highest contribution to methane emissions at natural gas transmission stations . The largest contributing factor to emissions is leakage from the sealing components in the packing systems around the piston rods.
Current technology uses a series of specifically-cut, dry-ring seals held in place with springs and cups. However, designing seals based on today’s technology inevitably leads to a trade-off between leakage reduction (with minimal gaps between the seals and the rod) and allowing sufficient gaps (such that the friction between the parts is sufficiently reduced allowing for movement). Once the piston moves, the pressure differential across the packing seals creates a twisting effect on the seal allowing substantial amounts of natural gas to leak into the casing. Ring twisting also causes increased friction and wear to the sealing rings and compressor rod. This gas is typically vented into the atmosphere, normally exceeding 11.5 standard cubic feet per hour for new, correctly-installed packing systems on well-aligned shafts .
This project will take the concept of liquid sealing and combine it with a novel, patented arrangement for pressure balancing across a seal arrangement (Patent No: US 7,757,599 B2 ) to allow for successful implementation in a dynamic environment with moving parts. The proposed seal design has been successfully implemented and tested at the bench-scale level. The seal will be designed, modeled, and fabricated for full-scale operation and tested in a reciprocating compressor system for various scenarios in a step-wise iterative method.
The primary benefit of this program is the development of a technology that will reduce and nearly eliminate methane emissions from reciprocating compressor packing. In this work, a liquid seal will be designed and tested at typical midstream pipeline operating scenarios to ensure the reduction of the environmental impact of process gas leakage from reciprocating compressor operation.
There are significant benefits to the development and implementation of a liquid packing seal, some of which are:
The novel liquid packing seal design has been designed, fabricated, and validated successfully in static and dynamic performance testing in low-pressure and high-pressure cylinders at speeds of 300-1,300 rpm and pressures up to 1,200 psia. No gas leakage was detected into the packing vent, distance piece, or oil stream by various flowmeters or gas bubble visualization while operating at these conditions. Additionally, no oil leakage into the cylinder was measured during testing. With these results, project work related to the novel packing seal has concluded and current project efforts are focused on development and validation of a leakage model for existing packing. Leakage data was obtained from low-pressure and high-pressure cylinders on a JGA/2 compressor, and a JGT/4 compressor is being prepared and commissioned for additional leakage data. In addition, both a literature review identifying several existing leakage models and guidelines from the industry have been obtained and compared with JGA/2 leakage data obtained over a range of conditions to identify focus areas for model improvement.