Flow measurement gets high-pressure treatment

December 4, 2013

Scotland’s NEL center will launch a new multiphase flow measurement test facility by year’s end. A second, able to operate at up to 150bar, is under development. Elaine Maslin paid a visit to learn more.

NEL’s laboratories were originally set up in the 1940s, as the National Engineering Laboratory (NEL).

It was one of several large governmentfunded research laboratories, staffed by scientists and engineers, with a remit to research subjects from early wind turbines to control systems.

While the scope of its once-wide range of activities has become more focused, since it was privatized in 1995, and bought by Germany’s TÜV SÜD Group, the research has not stopped. Flow measurement and fluid mechanics are now the main foci for NEL, which holds the UK’s National Standards for Flow Measurement and is a United Kingdom Accreditation Service (UKAS) accredited laboratory.

Image Caption: Inside NEL’s flow measurement facility.

Multiphase metering has been one of its areas of research since the 1980s, and multiphase testing since the 1990s, when the world’s first traceable calibration facility was developed.

By the end of this year, the center at East Kilbride, near Glasgow, will launch a new multiphase flow measurement test facility, able to operate at up to 60bar g. Next year, NEL will start construction of a second multiphase facility, able to operate at guage pressures up to 150 bar. The upgrades are to meet future demand for higher pressure testing and meters verification, because production is moving into deeper waters, says Phil Mark, sales and marketing director, NEL.

Use of multiphase flow meters is increasing, driven by the need to monitor individual flow streams where fiscal monitoring is required on fields with multiple ownership, and where test separation facilities are impractical or uneconomical. Deepwater and remote facilities are also strong candidates for multiphase metering, for the same reasons.

Developing accurate multiphase flow meters for simultaneous measurement of commingled oil, gas and water streams has long been a key issue for the oil and gas industry, however. Metering inaccuracy, even when marginal, canresult in significant errors when billions of barrels are involved.

Image Caption: The planned high pressure multiphase test facility.

A key issue is measurement uncertainty. A manufacturer can claim singlefigure uncertainties for a multiphase flow meter under certain conditions, but independent experts, including NEL and Det Norske Veritas (now part of DNV GL), suggest 10% would be more optimistic as a practical figure, especially considering the potential for fluid phase properties to change by the second.

Where a test separator is unavailable to verify flow meter results, operators have limited options to verify a flow meter’s results. Uncertainty of a multiphase flowmeter can be claimed to be as low as 2.5%. “If you could get 10% in service you would be doing well,” says Richard Harvey, lead multiphase flow engineer, at NEL. “They will have sweet spots, but 10-15% would be a good result.”

“Measurement uncertainty is a big issue,” Phil Mark says. “You can have constantly varying mixtures of gas, water, oil, fluids and other material going through a pipe at any one time, from all gas to all liquid, and anywhere in between. Take that to the bottom of the ocean, and there are additional complexities. So accurate multiphase metering technology is taking a long time to develop, but it is getting better.”

Image Caption: Phil Mark, sales and marketing director, NEL

Multiphase upgrade

To create its new high-pressure multiphase facility, NEL converted an existing two-phase (gas/liquid) separator into a gravity-based three phase separator. It operates with nitrogen, water, and oil (Exxsol D80, a kerosene substitute), at up to 60bar g, and typically 20°C, with a gas volume fraction up to 100% and water cut of 0-100%.

Its flow range rate is up to 1800cu m/hr for dry gas, and up to 80cu m/hr for both water and kerosene. The test section, which can be orientated horizontally or vertically, is 18m long and 8in (200mm) in diameter.

The second upgrade will be on a multiphase flow test facility, built around a full-scale three phase test separator, which also provides storage for the oil and water phases. Each phase is pumped as a separated single-phase stream and measured separately before being recombined into a multiphase flow and transported through the test loop. It can be constructed in vertical, horizontal or inclined piping configurations.

The current facility operates at pressures of 0-15bar g, but the new facility, currently being designed by NEL, will increase this to up to 150bar. Temperatures can be 5-40°C on 1-6in. line sizes with a 30m test section, or 10m when in a horizontal configuration.

The facility will operate with crude oil (API 30) at up to 145cu m/hr (22,000bbl/d), salt water at the same rate, and nitrogen gas at up to 1500cu m/ hr (1.3MMcf/d), at gas volume fraction (GVF) and water cuts of 0-100%. Both facilities are to be used for testing and development as well as performing certification, factory acceptance and calibration services.

Image Caption: Muir Porter, business manager, NEL

NEL determines the performance of a meter by measuring the single-phase flows to a very low uncertainty, and combining the understanding of different multiphase flow regimes at the given process conditions. Based on its UKAS accreditation, liquid phases are measured within 1.5% uncertainty for gas under most conditions, and within 1% for liquid, which ensures that we can assess multiphase flow very well and to a traceable standard, says Muir Porter, business manager, NEL.


The overall “uncertainty budget” consists of the summed uncertainty of measurement of various controlled elements of the process, including temperature, pressure, physical properties, and accuracy of secondary instrumentation.

“It is not possible to test flowmeters under the variable conditions they will see in the field, but we can measure them under very precise conditions through very careful control of the single phases,” Porter says. “We know how multiphase fluid flow regimes can change according to flow rate, gas volume fraction (GVF), pressure, temperature and even in-line disturbances and installation effects.”

Erosion and research

NEL also has a recently opened erosive flow test facility, and is involved in a number of joint industry projects (JIPs), from temperature and pressure effects on Coriolis flow meters to sampling needs for water-in-oil.

NEL’s High Viscosity Fluids JIP recently completed and resulted in the upgrade of an existing facility at NEL to allow it to test viscous fluids up to 1500 centistokes (cSt).

Image CaptionNEL’s laboratories at East Kilbride, near Glasgow.

To date, the most appropriate technologies for viscous flow measurement have not been defined, NEL says. The JIP saw established flowmetering technologies (Coriolis, Venturi and ultrasonic devices) evaluated across a Reynolds number (Re) range of 200–100,000 at kinematic viscosities of 20cSt, 100cSt, 175cSt, 300cSt and 500cSt.


A second stage of test work addressed gas entrainment, which has the potential to lead to substantial mis-measurement, at a kinematic viscosity of 500cSt with a GVF range of 0-5%.

The research found that viscosity effects are significant and must be taken account, as most meter technologies have a high dependence on Reynolds number, particularly at low Reynolds numbers. They must be calibrated at the viscosity/ Reynolds number range they will be operated at. Pressure drop is significantly higher than for conventional oils, and therefore means meters are more costly to operate.

“Alternatively, they are oversized (to reduce velocity and hence pressure drop). However, this means they are operating at the low end of their turndown, i.e. the place they are most inaccurate— it’s a balancing act,” NEL says.

Recently, the Temperature and Pressure Effects on Coriolis Flow Meters JIP was launched, with participation from Shell, BP, Nexen Petroleum, ConocoPhillips, Talisman-Sinopec, TAQA, and CNR. It is due to complete next year.

NEL says the standard practice for calibrating Coriolis flowmeters for the oil and gas industry has been to match the fluid viscosity and, if possible, the fluid temperature and pressure.

However, matching all parameters is seldom possible, due to the limitations set by the calibration facilities test fluids. Because of this, the parameter that is most often matched is the fluid viscosity.

A limitation of this approach is that temperature and pressure variations are known to influence properties, other than fluid viscosity, that may also be critical to measurement uncertainty.

Although there has been some research into the performance of Coriolis flow meters at high temperature and pressure, only a small amount of independent and traceable data exist on certain meter types and diameters. The JIP will look at the performance of Coriolis flow meters with onboard temperature and pressure compensation to provide traceable data. OE