Quality control during reservoir fluid sampling

Quality control is an area in which there have been significant improvements in recent years, with significantly more details of quality-control tests being reported by laboratories, yet only limited information has been published on this aspect of fluid sampling. This section highlights the principal controls that should be performed and gives guidelines for selecting which samples are most likely to be representative.

Selecting samples for study

The primary objectives must be selection of a fluid that is most representative of the reservoir fluid and identification of any serious quality problems related to the samples or the sampling data; these problems must be communicated to the client before proceeding with the fluid study.

Poor sample quality can arise from such sources as sampling nonrepresentative fluid, human error during sampling or field transfers, contaminated sample containers, and leaks during shipment.

For separator-gas samples, the quality checks that should be made when the sample bottles have been heated to, or slightly above, separator temperature are

• Determination of opening pressure
• Compositional analysis, including air content
• Determination of residual liquids, possibly from carry-over.

Separator liquids transported with a gas cap must be homogenized by pressurization and agitation. In this instance, the controls that must be performed are:

1. Determination of initial opening pressure
2. Determination of bubblepoint pressure at ambient or (preferably) separator temperature
3. A check for presence of sediments or an aqueous phase; and, when feasible
4. Flash separation to give gas/oil ratio (GOR), shrinkage, gas gravity, or composition.

Downhole samples should be checked in the same way, except that bubblepoint pressure can be measured at either ambient or reservoir temperature. Measurement at reservoir temperature takes longer but is preferable for comparisons with downhole static or flowing pressures. Ambient bubblepoint-pressure estimates are often available from field transfers, but they should be used only as a guide because thorough mixing of the sample may not be achieved during recompression, and temperatures may be unstable. In downhole samples of a highly volatile oil or a gas condensate, no "break point" will be seen on the recompression curve, and a saturation pressure must be determined in a windowed Pressure/Volume/Temperature PVT cell.

Selecting best sample from duplicates

The following are parameters that should be used, in order of preference, when a sample is selected on the basis of sample quality alone (i.e., when samples are essentially duplicates collected at the same time and under the same conditions):

• An adequate sample volume or pressure
• A downhole sample bubblepoint pressure lower than downhole pressure during sampling
• Contamination levels lower than, or similar to, duplicate samples
• bottle opening pressures that agree with sampling data (i.e., leaks are unlikely)
• Surface sample bubblepoint pressures that agree with separator data
• A close correlation between laboratory measurements on duplicate samples
• One sample that represents "average" properties of duplicates.

Separator samples of oil and gas

For gas and oil samples collected from a separator, if at all possible, production test reports or other documentation should be studied in addition to the sampling sheets that normally accompany samples because a high proportion of sampling sheets contain inconsistencies. Data should be studied with the following objectives:

1. To identify what well or plant conditioning has been performed
2. To look for the stability of gas and liquid rates when the surface samples were taken and to calculate averages at the time the samples were taken
3. To ensure that the GOR is based on oil flow rate at separator conditions
4. To determine which gas gravity and nonideality ( Z ) factors were used, as well as the reference pressure and temperature
5. To verify reservoir temperature and static pressure.

Selecting samples based on field data

If all samples meet the quality criteria, the choice can be made on the basis of field data alone, although the selection tends to be a compromise in some cases. Both operator and laboratory personnel must be involved in these choices. Primary emphasis should be given to

1. Samples collected after proper well conditioning
2. Surface oil and gas samples taken simultaneously or close together
3. A downhole sample that was collected above its bubblepoint that compares well with the bubblepoint pressure for duplicate samples
4. A good downhole sample in preference to a recombined surface sample
5. A recombined surface sample if doubt exists about the quality of downhole samples

Creation of a recombined surface sample

In cases in which downhole samples have been backed up by surface samples (an excellent practice), creation of a recombined surface sample from the best surface samples might be worthwhile, especially if there are only one or two downhole samples that appear to be valid. This allows comparison of the recombined surface sample with the downhole samples. In fact, in important wells, complete analyses of the two types of reservoir-fluid samples might be useful. Such an approach can give a high level of confidence in the data and could provide a crosscheck of separator GORs.

All validation data and analyses do not need to be included in the report when laboratory measurements are reported, but it is good practice to use a minimum of one page to explain sample selection and to detail any quality or field-data problems. This information can be of major value to engineers using the measured data.

Correction techniques for wells drilled with oil based mud

Wells drilled with oil-based mud (OBM) are particularly problematic for formation-test-tool sample quality, and many research and development (R&D) centers worldwide are working on correction techniques. Figs. 1 and 2 demonstrate the sort of contamination that can occur; because the base oil is miscible with reservoir oil, it is impossible to remove this contamination from samples. It is essential to perform "fingerprint"-type gas chromatography (GC) analyses as a minimum quality control on formation-test samples to provide a qualitative indication of contamination, and even on production-test samples for which thorough cleanup may not have been achieved. Because of the wide range of fluids and sampling conditions, comparison of duplicate (or, more correctly, replicate) samples is generally the best method of evaluating whether the sample is representative.

• 

  Fig. 1 – Chromatogram showing broad OBM peak.

• 

  Fig. 2 – Chromatogram showing narrow OBM peak.

Special correction techniques are increasingly used within the oil industry, and because these techniques vary between organizations and laboratories, sample selection should be done only after considering which method to use. Many companies are forced to use oil-based drilling muds to manage drilling costs in water-sensitive formations, and the added expense of handling contaminated samples (and the risk associated with poorer-quality data) must be used to evaluate the overall economic balance.

Correction techniques for water sample

For water samples, comparisons of duplicates also give a good indication of quality. Where fluid concentration may be stabilizing (e.g., at the end of a cleanup), sequential samples should be used to look for compositional trends and thus to help decide if representative fluid has been sampled. For some sampling procedures involving trapping or precipitation of particular components, it is highly recommended to use blank "samples," which undergo exactly the same treatment and storage as the actual sample and provide a reference measurement to assist with the interpretation of laboratory measurements. More details are available in API RP 45.

Selecting fluid samples for storage

Decisions concerning sample storage involve the following constraints:

1. Discarding samples may prevent future measurements or checking of dubious results
2. Long-term storage of pressurized samples may incur very high rental costs
3. Sample-bottle purchase involves higher "short-term" cost (instead of long-term rental charges) but may sidestep the issue of deciding on a long-term storage policy
4. Long-term storage requires a safe storage area and a catalog to be maintained (subcontracting is an option to be considered by producers).

One policy could be to keep duplicate samples for a short length of time on a rental basis and then to transfer minimum sample quantities into bottles purchased specifically for long-term storage. In all cases, it is advisable to budget for long-term storage within the project costs.

A good approach is for the laboratory report to recommend which samples should be kept or discarded on the basis of the quality checks and the study itself. An initial selection can be made at the end of the study, or even at the quality-control stage, if useless samples are identified. Then, the "customer" can respond on the basis of this information and the other needs of the project.

Further measurements may be warranted for a number of reasons:

• Doubts about initial measurements
• Measurements required at different temperature or pressure conditions
• Advanced PVT measurements that are deemed necessary (interfacial tension, phase diagram, etc.)
• New analytical techniques that are developed before the reservoir is developed
• An asset purchase or joint venture that changes needs or requires independent measurements.

Because of the complexity of the heavy components in reservoir fluids, it is impossible to make an adequate synthetic mixture based on the liquid composition determined. It is thus good practice to store at least 1L of stock-tank liquid. This can be stored cheaply in a low-pressure closed container, but it also could be blended with a synthetic gas mixture (based on separator-gas and stock-tank gas analyses) to recreate a sample close to the original fluid composition, if further work was eventually required on the reservoir fluid itself.

Quality control for gas chromatography

Although this chapter concentrates on sampling rather than on analytical measurements, it is worth providing simple quality-control guidelines for GC here. This is because of the importance of GC analytical techniques both in the quality-control procedures described above and, increasingly, in on-site measurements and because simple guidelines of the sort given in Table 1 are not widely reported.

• 

  Table 4.9

A small volume (e.g., 100 cm³) of separator gas should be stored under pressure in case more-accurate or new analytical measurements are required on the gas later. For a downhole sample, a flash separation can be made at a convenient pressure to generate a suitable gas sample for storage.

Note that the quantities specified above are recommended in the case that no further laboratory measurements are anticipated. If further work appears likely, appropriate quantities of samples should be stored in addition to these minimum volumes.

This work recommends that a suitable sample-storage policy be identified and implemented in operating companies.

Noteworthy papers in OnePetro

Use this section to list papers in OnePetro that a reader who wants to learn more should definitely read

External links

Use this section to provide links to relevant material on websites other than PetroWiki and OnePetro

See also

Fluid sampling

Measurements affecting reservoir fluid sampling

Fluid sampling safety hazards

PEH:Fluid_Sampling