We can reliably test your gas/gas condensate well from the surface!
By Rod Kelly |
Tue, 19 May 2009
Donít believe it?Weíll prove it at no cost to you!
Data Retrieval Corporation (DRC) pioneered the technology to test gas and gas condensate wells from the wellhead as a NO RISK and LOW COST alternative to running wire and pressure gauges downhole. If this population of wells is in critical flow and not slugging at the surface, they are a candidate for this technology. Since DRCís inception, in 1985, weíve invested significant time and money into R&D in both hardware (SPIDR gauge) and software (bottomhole modeling). These modeling improvements serve to make our technology viable over a wider range of wellbore (high rate, higher yields, sour gas, deviated or horizontal wellbores, etc.) conditions.
Most petroleum engineering textbooks consider that surface pressure conversion to bottomhole pressure (BHP) is acceptable under ìdry gasî conditions using the Cullendar & Smith single phase conversion model. Dry Gas is often defined in these text books as any well that produces less than 10 BBL / MMSCFD. Here at DRC we are more concerned about the wellbore pressure than the separator pressure and how that corresponds to the ìhydrocarbon phase envelopeî. We test many wells that produce large amounts of condensate at the separator but may still be in single phase gas in the wellbore.
DRC places the work; we perform for our customers, in 3 principal categories.
1. In larger fields and reservoirs, it is typically ongoing production / reservoir surveillance. The primary objective is to see how or if well performance has changed and is the reservoir pressure deviating from the expected trend. 2. In smaller or one well fields or reservoirs SPIDR technology is used to characterize or define the reservoir via limits testing and initial completion evaluations. Subsequent testing to characterize skin accretion is also commonly performed in higher permeability and gravel packed completions. 3. In tight gas applications the SPIDR is an economical choice for not only collecting pre-frac injection/fall-off or ìGî function data but also for capturing flowback rates and pressures after the well is brought online.
Today DRC performs over 800 well tests a year around the world for clients both large and small. Most customers immediately see the value of our service on those wells that are deep, hot, high pressure, offshore, unmanned, sour, remote, or other difficult conditions compared to traditional methods. There is a big advantage for the operator in proving up our system under these conditions. If you know that a service can provide you with similar results as the standard bearer but costs much less and would incur no risk in implementing itÖÖlogically there would have to be a strong reason that you would not consider adoption of the technology. In the case of DRCís surface testing methodology, I believe that reason is because many engineers do not believe in DRCís ability to ìaccuratelyî convert surface pressure data to BHP.
DRC has built a Modified Cullendar and Smith model that requires:
a. Model inputs ñ Information about the present conditions of your well such as flow rate, fluid yield, densities, completion diagram, and PVT (if applicable). b. Flow model - We determine the proper flow model to employ based on fluid volume and velocities at the wellhead (single phase, mist or annular mist flow). c. Thermal Decay ñ A proprietary model that allows us to adjust for wellbore cooling at the wellhead during a pressure build up. We model the rate of change in density due to cooling. One of the most important takeaways from this discussion surrounds our comparison of our conversion model to several popular commercial wellbore simulators in the marketplace that employ a WHP to BHP routine. In all cases, using the same input data, not only is DRCís model closer to actual downhole gauge pressure but even more importantly no routine in the industry has ANY solution for the effects of THERMAL DECAY. d. Phase effects - Flash Calculation to determine the amount of liquids that stay in solution with the gas
As we always explain to our customers, the process involves artificially or synthetically converting a surface pressure response to BHP conditions. This results in BHPís and reservoir properties (k, S, P*) that are equivalent to downhole gauge results. However, if our MODEL INPUT information (information supplied by the customer) is not correct or if it is not a reflection of the current well conditions, our conversion to BHP will not reflect true downhole conditions. We find the single biggest error contributor to our calculations is inaccurate flow rate measurement (both gas and water / condensate). It is not unusual for the operator to have a +/- 10% error in these rates. In addition, if it is a high condensate producer with old or outdated PVT information, we find that this can reduce the effectiveness of our models. The effect of uncertainty of these inputs on our BHP model and the analysis can typically be quantified and we can supply customers with an expected error range before the test if requested.
One of the ways we want to PROVE UP our technology to a new customer is to perform a simultaneous test with gauges downhole (permanent or wireline deployed). We will convert our surface data to downhole gauge depth and the operator will then make an objective judgment of our BLIND conversion and compare it to the recorded downhole gauge. Itís the single most effective way to demonstrate the effectiveness of our system under our customerís conditions. We also want our customer to be aware of our ability to improve upon or ìtuneî our conversion algorithms based on DHG data feedback received.
Another added benefit for our customers is the use of SPIDR technology as an inexpensive independent verifier of pressures recorded and for the Pressure Transient Analysis of those pressures. It is not unusual for us to encounter downhole gauge problems (calibration, drift, tandem or triple gauges that donít agree, etc.) during the comparison tests. Many of our customers are working with service providers as single source providers for multi-year contracts. If there is no independent method to check and confirm both the recording of the data and the analysis of that same data set, problems will occur. We have also discovered that both service and operating company engineers bring a wide range of experience to bear during the analysis process. Most of the commercial PTA software packages are straightforward to utilize, but as with any other software package, if you donít utilize it on a REGULAR basis you will not have an effective command of it. We continue to see analysis performed, on both downhole gauge data and converted SPIDR data, which differs dramatically from our analysis of both sets.
