Union Gas' Bright Compressor Station has two RB211 gas turbines that were installed 25 years ago with tenkay style pulse inlet systems. As part of their maintenance excellence program, and in order to optimize their soak wash schedule, Bright's operations staff began searching for potential filtration upgrades. These upgrades essentially allow for continuous operation during winter, therefore eliminating the need to shut down the turbines for maintenance. To determine whether they could improve operations, Bright received a free consultation with Camfil to better understand the latest media developments.
The Site
Blandford-Blenhein is a very clean, low dust rural area located in Ontario, Canada. The PM2.5 dust concentration averages 6 µg/m3. In comparison, the average USA PM2.5 concentration is 9 µg/m3 1. With 6 µg/m3, the area is cleaner than 90% of all sites. Despite the clean conditions, the turbines were fouling. As a result, staff would have to soak wash the engine offline every 1000 hrs and would experience 2.5 MW of power degradation (6% of the total output) over that period.
Performance Monitoring
In order to understand filter performance and impact on the turbines, a side-by-side performance comparison test was conducted between Bright‘s existing filters (M6 blended filters) and Camfil‘s GTC F92 (ISO ePM1 80) synthetic filters. To measure the efficiency of both filters, a particle counter measured the concentration of particles before and after the filters. The compressor cleanliness was then evaluated with the particle concentration in the soak wash fluid. Finally, fuel consumption, as well as the impact on compressor performance was tracked for over a year.
Test Results
As compared to the blended filters, the GTC filters showed increased efficiency without an increase in airflow restriction. After 1000hrs, the efficiency on 0.4um particle penetration was 75% as compared to the blended media efficiency of 5%. This resulted in a cleaner engine, a higher stable compressor efficiency, and lower fuel consumption.
Despite the difference in efficiency, both the blended and GTC filters have the same ASHRAE MERV 163 rating. The ASHRAE standard tests filters from initial efficiency, which may be electrostatically charged; they are also loaded quickly to their maximum capacity. Both the electrostatic charge and the dust loading have a positive impact on filter efficiency and as such can exaggerate filter performance, resulting in a high classification. In real life, any charge would wear off rapidly, causing efficiency to plummet. When the filter is loaded in the lab, or if installed in a very high dust area, efficiency would increase due to dust build-up on the filter – that dust cake acts as an additional barrier, allowing it to capture smaller particles, therefore increasing average efficiency.
When a site is extremely clean, a filter may stay at its minimal discharged efficiency for years since the electrostatic charge wears off and there is not enough dust accumulation for the efficiency to increase. This may cause fouling, corrosion, or plug the cooling holes of the turbine, which can cause an increase in temperature, shortening the life of the turbine blades.
Contrary to ASHRAE, the EN779: 2012 standard requires both an average efficiency and a discharged minimal efficiency, starting from the F7 filter class. The blended filters at Bright do not meet the minimum discharged efficiency requirement and therefore are classified as an M6. The GTC filters have a minimum discharged efficiency of 70%, meeting the F9 requirements.
New Zealand
