THE VALUE OF A CHECK METER AND SURVEILLANCE SYSTEM

Roland Rollins
Vastar Resources
Lafayette, LA.

17th Annual Measurement Short Course
Acadiana Flow Measurement Society
1996
Lafayette, LA

When thinking of a company's cash register, many of these noteworthy questions concern management.

• How much loss in revenue are we experiencing? Or perhaps more usefully, "How much loss in revenues can be recouped or avoided by using check meters and a surveillance system?"
• Can the expected pay out for a check meter be determined, given the flow rate?
• Do certain transporters/buyers have more loss than others?
• What types of errors are expected on a system and which ones are the most costly?
• Do stations with EFM (flow computers) have the same type and amount of losses as traditional chart drive stations?
• Can the expected pay out be determined for installations of secondary devices on the sales taps?
• Do the errors shrink after years of close scrutiny?

The following discussion highlights statistics gleaned from one surveillance system. It can direct firm answers to achieve an informed decision about check meters.

OVERVIEW

The data for the graphs was derived from the Gulf Coast Onshore Unit of Vastar Resources Inc. an explorer, producer, and marketer of natural gas. This Unit includes gas-condensate fields in southern Louisiana and south-eastern Texas, utilizing nine transporters with seventeen stations including a mix of large and medium sized pipe lines. A range of 29 to 93 BCF per year deliveries occurred during the study period of 1990 to 1996. When presented graphically, the answers to the questions posed become very clear.

DISCUSSION OF GRAPHS

1 - YEARLY ADJUSTMENTS BY PIPELINE 1990-1996

This graph depicts the revenue adjustments and losses avoided by each pipeline. The adjustments are from primary and secondary devices, while the losses avoided are from the primary device (meter tube) or systematic BTU problems.

Some method had to be arrived at to determine the revenue saved from systematic errors. A percentage change comparing before and after correction was used to derive the revenue saved. In practice, this percentage of change savings continues indefinitely, but was confined to six months for this study. A company with no surveillance or no measurement staff would indeed lose this revenue indefinitely.

2 - ADJUSTMENTS BY TYPE

This chart depicts the revenue received from primary and secondary devices. There is a direct inverse correlation to the number of adjustments for each type versus the revenue received. The secondary devices involved 4-15 yearly adjustments, while primary devices involved 0-3 incidents yearly.

3 - % ADJUSTMENTS BY TYPE

Primary, EFM, and Chart adjustments as a percentage of the total revenue savings is depicted. Of the 17 sales points, 12 switched from charts to EFM from 1990 to 1996.

4 - % ADJUSTMENTS Vs DELIVERIES ANNUALLY BY PIPELINE

This chart shows a percentage of volume returned versus total gas transported per pipeline. This chart and chart 5 are a much fairer look at pipeline performance of measurement errors, by factoring out large variances in volumes delivered to different pipelines.

5 - % AVERAGE ADJUSTMENTS

This pie chart depicts the average percentage of volume adjustments versus total deliveries for each pipeline.

6 - TOTAL ADJUSTMENTS BY PIPELINE

Chart 6 depicts total adjustments received and losses avoided by each pipeline.

ANSWERS TO THE QUESTIONS

· How much loss in revenue can be recouped or avoided by using check meters and a surveillance system?
Chart 4's data reveals an average approaching 0.42% of deliveries can be returned or saved annually.

· Can the expected pay out of a check meter be determined, given the flow rate?
Yes, based on chart 4. The minimum volume for a 6 month pay out to install a 3" check meter would be 2,000 mcf/day with 1000 BTU gas at $2/mcf sales price, based on current prices for meter tubes and flow computers.

· Do certain transporters/buyers have more loss than others?
Based on charts 4,5, and 6 - yes. This system shows 3 pipelines with average adjustments of 2.75%, 0.75%, and 0.61% compared to the 0.42% adjustment average. Conversely, 3 pipelines have average adjustments of only 0.05%, 0.08%, and 0.18%. The adjustments included both primary and secondary devices. Seven of the 9 pipelines had adjustments in 1991 to 1996 higher than the total averages. Since many of the large adjustments are primary device type, the remaining pipelines may experience these problems.

· What types of errors are expected on a system and which ones are the most costly?
Considering charts 2 and 3, by far the most costly errors are systematic, usually fouled or dirty tubes and BTU discrepancies. The single high value for most years on chart 1 involved a systematic error. On average, one tube fouled every 21 months during the period reviewed. Much more common, though less costly are the one-time secondary device errors.

· Do stations with EFM have the same type and amount of losses as traditional chart drive stations?
Charts 2 and 3 register the same amount of losses on EFM sites as chart drive sites, based on the number of each type in the system. The types of errors are similar if not the same; for instance, power failure instead of stopped clock, transmitter failure instead of bellows, range spring, or static element failure.

· Can an expected pay out be determined for installations of secondary devices on the sales taps?
Utilizing charts 2 and 3, one can determine a pay out. Primary device errors cannot be detected with a "poor boy" check (secondary device on the sales tube). The secondary errors would be the only ones discernible to seek adjustments. With 25% of the revenue coming from this type, a .10 - .12% return could then be expected. A rate of 4,000 mcf/day would be needed to expect a 6 month payout. The pay out is quicker to also buy the meter tube than to use only a secondary device.

· Do the errors shrink after years of close scrutiny?
The data for chart 1 does not infer this. The year 1994 proved as high a revenue adjustment year as the others. As can be seen with chart 4, the percentage of revenue adjustments for 1994 was the highest yet. At least two probable factors for this recent "high" year exist. For one, the pipe line industry has endured the same large cuts as producers in personnel, with a large experience base retired. Also, saddled with requirements to have current volumes available to interested parties via electronic bulletin boards, the pipe lines have far less time to internally audit their volumes with a smaller staff.

The deliveries per station have a direct effect on the number of adjustments. This is due to economics. The local Vastar Measurement Plan dictates measurement staff to review all errors, the larger of a set percent or dollars. Since small stations can have low volume, high percentage errors, these are not pursued. Rebooking volumes costs more than the value gained. Adjustments are requested below this threshold whenever it is likely the adjustment can be procured, at least orally, before the allocations are run.

AFTERTHOUGHTS

With constant use of a surveillance system, several issues arise that cannot be expressed with numbers or graphs.

After 20+ months of surveillance, Meter A has never had a daily variance over 0.5%. Should one continue to check volumes daily, monthly, or not at all? A reaffirmed lesson says "all tubes eventually foul". Performing monthly variances would appear a prudent option. Do we ignore the one-time secondary device failures? On a site with no failures yet, this might seem like overkill. After all, if the error is too great, it will show up on the monthly tally. Not necessarily! Meter A had a 5421 mcf adjustment (-33.6% variance for one day), but the month's variance was +0.8%. Is $10,000+ worth maintaining a daily variance spreadsheet? Now the question is easier to answer. Lesson two then is "all stations are subject to problems".

With all the documented revenue adjustments and losses avoided, not even accounting for the revenue saved from an absence of pipeline imbalance penalties in a post-FERC 636 world, one could say - MEASUREMENT PAYS$$$$$!

Web Administrator Kirk Harrison
E-mail us your thoughts or suggestions.
Last Update 2-11-04