Individual, Actual Leakage Rates? 1

[SamChem]

Is there an instrument for quantifying in the field the actual rate loss of all individual components? I have this possibility, but this analyzer is very expensive:

http://www.heathus.com/specs/hiflowsampler.pdf

Anyone know of a better way to put objective, money-lost numbers to leakage rates? I want to sell the idea that it will be cost-effective to fix, but I need solid numbers to demonstrate what amount of money is going up in vapor loss right now.

 

[mbeychok]

SamChem:

Are you familiar with the US EPA's publication AP-42, "Compilation of Air Pollution Emission factors"?

Go to www.epa.gov/ttn/chief/ap42/index.html. Scroll down to "AP-42, Volume 1, Fifth Edition".

If you work in a petroleum refinery , click on Chapter 5 and go to Section 5.1 "Fugitive Emissions" where you will find data on fugitive vapor losses from valves, drains, flanges, pump seals, compressor seals and other such leak points. You will also find data on fugitive vapor losses from other systems (such as the oil-water separator and the cooling tower). These data are all based on actual sampling done at a number of refineries.

If you work in a chemical plant, use the search function at the top of the index page to search for "fugitive emissions" and you will find a list of chemical-specific publications on loss sources of the specific chemical.

Hope this helps you.

Milton Beychok
(Visit me at www.air-dispersion.com)
.

 

[SamChem]

MBeychok:

Thank you for that reference, but the site I am dealing with is in a foreign country. I'm being asked to provide empirical leak rates because this particular facility's equipment is (unfortunately) in quite a bit worse shape. I fear these figures would not adequately represent the actual losses or make my financial case as well as a data collected from the field will.

I do really need to nail down some actual loss rates.

However, I also plan for monitoring at US sites, though, this will be very useful for those if not for the one I'm wrestling at the moment.

Sam

 

[mbeychok]

Sam Chem:

If you study that AP-42 web site carefully, you will find a Protocol for running tests such as you propose. As I recall the original survey program, they actually bagged most of the potential leak sources in a number of operating US refineries, collected samples over a measured period of time, and had them analyzed in a laboratory. I seriously doubt that you will find any automatic instruments that can do what they did. The entire program took a number of years to complete.

Even using the published emission factors in the EPA's AP-42 publication and going through all of a refinery's piping flow sheets to calculate the fugitive emissions from each valve, each flange, each drain, each relief valve, each sample point, each pump seal, and each compressor seal, etc. can easily take a few months to complete properly.

Some of our US state environmental regulatory agencies (notably in California) believe that a routine Inspection and Maintenance (I&M) program will mitigate the fugitive emissions and they allow refineries to take credit for implementing such programs. How to implement such programs is spelled out in the regulations promulgated by those state and local agencies.



Milton Beychok
(Visit me at www.air-dispersion.com)
.

 

[SamChem]

I'm coming around to your solution. Even though it will require months of number-crunching, doing that rather than individuall bagging and trying to measure leak rates in the field sounds both safer and less labor-intensive. Plus, as you've taken the time to point out, the bag and measure correlation has already been performed in this study - redoing bags ourselves doesn't seem necessary. I'll tell the boss as much.

I am curious, though - I've been doing some reading on the use of the Hi-Flow to detect leaks; it looks like the major area of impact is around compressors in natural gas. I notice there isn't a direct correlation listed in the EPA Table 2-10 under 2.3.3

What would be the most reasonable approach to targeting leaks from a compressor, in your view?

 

[mbeychok]

SamChem:

I notice there isn't a direct correlation listed in the EPA Table 2-10 under 2.3.3

I assume that in the above quote from your latest response, you are referring to the Protocol Document on how the emission factors were obtained. You are correct, Table 2-10 doesn't include a correlation for converting compressor seal Screening Values (SV) in ppmv to kg/hr.

However, Tables 2.2, 2.2, and 2.3 have the fugitive emission factors for compressor seals in kg/hr/source seal for chemical plants (Table 2.1), refineries (Table 2.2) and for marketing terminals (Table 2.3). I should think those would be adequate for what you want to do.

As for the Hi-Flow device, I am unfamiliar with it. I know that the I&M program in the regulations of California's South Coast Air Quality Management District (SQAMD) called for the use of hydrocarbon sniffers to detect leaks ... or at least they did some 15-20 years ago before I retired.

I hope all this is of some use to you.



Milton Beychok
(Visit me at www.air-dispersion.com)
.