All DP primary elements restrict the flow in some way. A restriction in a pipe results in an increase in the fluid velocity, according to Bernoulli’s law of conservation of energy. The ensuing conversion to kinetic energy reduces the static pressure. This pressure drop, the measured DP, is proportional to the square root of the flow rate and thus provides a means to measure flow.
The elements have no moving parts and can be fabricated in a wide selection of materials. Their purchase price is relatively low, even for large pipe sizes. Accuracy is moderate, ranging from 1% to 5%; compensation techniques can improve these values to better than 1%. Extensive research over decades has led to optimization of the flow elements and development of standards. DP meters are generally easy to select for a specific application.
Disadvantages revolve around rangeability, installation costs, density and flow profiles. The square-root relationship limits the range of flow rates that realistically can be measured in a particular application; the typical rangeability is 4:1 or slightly higher. Installation requires a DP transmitter, manifold, valving and impulse lines. The impulse lines leading to the DP transmitter can become plugged unless remote seals and filled capillaries transfer the pressures to the transmitter. The measured value varies with fluid density for both volumetric and mass flow. Additionally, flow elements tend to be sensitive to flow profiles within the pipe, requiring long upstream pipe runs or flow-straightening devices.
While the orifice is the most common restriction associated with DP measurement, several others — venturi tubes, flow nozzles, wedges and flow tubes — have found a solid place in process applications. Users should look at these designs, too, to come up with the optimum DP meter for the particular operating conditions and requirements. An important consideration is the pressure drop, which as a rule of thumb should be as small as possible. Technologies such as venturi tubes, flow nozzles, wedges and flow tubes feature very small pressure drops, leading to reduced energy loss and pumping requirements. So, in this article, we’ll look more closely at these elements.
Figure 1. Classic Venturi Tube -- The outlet cone
The classic venturi tube (Figure 1) is a robust flow element that’s useful for applications requiring low loss of line pressure. A venturi tube essentially is a section of pipe with a converging conical entrance (about 20°), a straight cylindrical throat section and a diverging conical exit with a smooth, gradually increasing (about 7°) diameter. Unlike the orifice, the interior surfaces always remain in contact with the fluid. Additionally, because dirt won’t build up as it passes through the contoured sections (as it does in the front of an orifice), this differential producer can serve in dirty flow applications. For the same differential pressure, the classic venturi can pass about 60% more flow than an orifice plate. Years of test data have documented and validated the flow coefficients for various sizes and fluids.
Four or more high pressure taps in an annular chamber leading to the straight throat section average the lower pressure reading. Initially designed for large line size (>6 in.) water and wastewater applications, the venturi today ranges in line sizes from 2 in. to 48 in.; installation possibilities include flanged, welded and threaded-end fittings. Manufacturers generally machine the smaller sizes from solid rods, while fabricating the larger sizes from rolled plate. The lengths of the elements typically run five pipe diameters.
Because most of the pressure recovers, the venturi is a good choice for large flows where the velocity is higher and Reynolds number, Re, is in the turbulent flow regime. And while it has a relatively long length, the venturi tube requires minimal upstream flow profiling, as its interior shape helps to condition the flow. Rangeability, while better than that of orifice plates, is less than 6:1, with typical accuracies of ±1% to 2% of full scale.
Variations of the classic venturi are available. Shorter versions increase the angle of the outlet cone with some sacrifice in pressure recovery. Eccentric inlet and outlet cones can handle mixed phases or build-up of heavy materials. Forms with a rectangular cross-section often serve in ductwork for gaseous flows.
Like the orifice and venturi tube, these are standard DP elements with extensive generic testing and documentation. Because of their rigidity, flow nozzles are dimensionally more stable at higher temperatures and velocities than an orifice plate. They typically measure fast flows that otherwise might damage an orifice plate from cavitation or erosion. Applications include high velocity steam or fluids with entrained solids.
Nozzles aren’t recommended for slurries or dirty fluids that might foul pressure taps. In contrast to an orifice plate, flow nozzles have no sharp edges that might wear over time and degrade performance — so they maintain long term accuracy and offer reduced possibility of distortion. Throat Re should exceed 10,000, although data are available down to about 6,000.
The initial cost of a flow nozzle is substantially higher than that of an orifice plate but lower than that of a venturi. However, the permanent pressure loss is significantly greater than that of a venturi. A flow nozzle will pass about 60% more flow than an orifice plate of the same diameter and DP. Because of the nozzle’s streamlined interior, unrecoverable pressure loss is slightly less than that of an orifice but still can range to 40% or more of the DP.
The standards for nozzles include ones for the so-called 1932 ISA nozzle, which is uncommon in the U.S., as well as the long radius nozzle.
Figure 2. Flow Nozzles -- Low (right drawing) and high
The long-radius flow nozzle predominates in the U.S. It has a converging section that is a quarter ellipse followed by a cylindrical throat section, and comes in two design variations — one with a low beta ratio (throat/inlet-diameter ratios between 0.20 and 0.50) and the other with a high beta ratio (throat/inlet-diameter ratios between 0.45 and 0.80). The difference in geometry is a flattening of the elliptically shaped inlet in the high-beta-ratio version (Figure 2). The American Society of Mechanical Engineers (ASME) has developed standards for the nozzle geometries based on the beta ratio desired for the application.
Users either may weld the nozzles into the pipeline or mount them with a holding ring between flanges. Where inspections are required, the flange mounting provides accessibility. In the U.S., the DP taps are commonly found one pipe diameter upstream and one-half pipe diameter downstream from the inlet. Flow nozzles may be installed in any position. Vertical downward flows better suit wet steam or gases and liquids with suspended solids. Upstream and downstream piping requirements for flow conditioning are similar to those for orifices.
Vendors can make nozzles from any machinable material, such as aluminum, fiber glass, stainless steel and chrome-alloy steel. The bevel on the discharge side of the nozzle is a critical point of manufacture. The throat should be perfectly round with no taper. The standard surface finish is 16 roughage measurement system (RMS). The standard nozzle coefficient is 0.9962, which users can adjust for actual beta ratio and throat Re.
A special variation known as a venturi nozzle combines the 1932 ISA nozzle inlet profile with the divergent cone of a venturi tube. Taps in the throat transmit the lower DP pressure. This nozzle can serve as a secondary flow-rate standard when experiencing “choked” (sonic velocity) flow. It’s also commonly used to test steam turbines.
The wedge element consists of a V-shaped restriction welded into the top of the meter body (Figure 3). In profile the wedge looks like a segmental orifice plate to the incoming fluid; particulate matter or entrained gases easily pass through. This basic meter has been on the market for more than 40 years and has proven its ability to handle tough, dirty fluids. The slanted faces of the wedge provide self-scouring action and minimize damage from impact with secondary phases. The fixed-body design provides a constant discharge coefficient over a wide range, 8:1, which is relatively high for a DP element. Accuracies to ± 0.5% of full scale are possible.
Figure 3. Wedge -- The shape of the restriction allows
The wedge meter is popular for oil and gas applications, especially in production fields. For difficult fluids it can be equipped with a pair of remote seals that effectively isolate the metered fluid from the DP transmitter without affecting accuracy while keeping the flowing fluid contained within the pipe.
Wedge meters are characterized by an H/D ratio (analogous to the orifice’s beta ratio), where H is the height of the restricted opening and D is the unrestricted inside diameter of the wedge element (Figure 3). The ability of the wedge to produce varying DP ranges depends on the H/D ratio selected. Elements come with ratios in fixed steps (0.2, 0.3, 0.4, etc.) that allow a wide range of element sizing for a given pipe size. The meter coefficient is established at the time of factory calibration (in water).
Some wedge meters can handle flows with Re as low as 500, which is very helpful in metering slurries and liquids with high viscosities, and also allow measuring flows bidirectionally with the same degree of accuracy.
Wedge elements can be supplied with remote seals having large diameter diaphragms, affording more sensitive response to DP changes while eliminating plugging of impulse lines. These wafer-type seal connections are raised off the meter body and may be preferable for more aggressive and erosive applications. A second design permits use of isolating seals that can seat with diaphragms flush to the meter body. This seal arrangement keeps process fluid contained within the wedge element, with no dead zones under the seals where sludge and waste can build up. If the element is sized to have enough fluid velocity, a natural washing action will occur over the seal diaphragms and restriction, keeping the meter clean and sustaining maximum performance.
Like the nozzle, the wedge doesn’t rely on a restriction with sharp edges or machined bores. As a result, it will perform for long periods of service without the need for maintenance and repairs. Particulates and solid debris easily pass under the V-shaped wedge and the inherent ruggedness of the restriction resists damage to its measuring edge, sustaining the initial calibrated accuracy.
As an example of the wedge meter’s ability to maintain calibration in tough applications, consider two stainless steel meters — one 3-in. and one 4-in. with wafer-type seal connections — that ABB examined after 12 years of service. The meters measured steam-cracked tar, a byproduct of ethylene production. The cracked tar is kept at elevated temperature to prevent solidification of abrasive coke fines and other particles in the process stream. The meters endured temperatures in excess of 355°F and pressures up to 310 psi. Fluid viscosities of 22 cP produced Re of 1,870 and 2,850, respectively, at maximum flow rates. With more than 5 million pounds a month of the abrasive tar being produced, repeatable and reliable measurement was a prime concern.
Calibration before and after the 12 years of service demonstrated that neither unit exhibited a major change in meter coefficient. The 3-in. meter showed a deviation of 0.24% from its original testing while the 4-in. meter shifted 1.3% (based on averaged meter factor). Given typical calibration uncertainties, it’s safe to say that the meter factors remained virtually constant over the 12 years of operation.
Wedge meters can be manufactured in virtually any alloy for service temperatures up to 720°F and pressures exceeding 6,000 psi.
The ASME defines flow tubes as any DP element whose design differs from the classic venturi (a definition that includes short-form venturis, nozzles and wedges). In practice, flow tubes come in several proprietary shapes; all tend to be more compact than the classic and short-form venturis. Laying lengths typically run from two to four pipe diameters. Being proprietary, flow tubes vary in configuration, tap locations, differential pressure and pressure loss for a given flow. The user must depend on the manufacturer of the particular flow tube for sizing and calibration. The ASME recommends calibration with a piping section that replicates actual use over the full range of expected flows, which may be difficult and expensive for the larger sizes.
Flow tubes can have either static or corner pressure taps. Static taps, like those of a venturi tube, sense pressure where the fluid velocity doesn’t change direction and parallels the pipe wall. Otherwise the taps are called corner taps. Three types of flow tubes are available:
Figure 4. Flow Tube -- This Type 1 version comes close
• Type l (Figure 4) has static pressure taps at both the inlet and outlet;
• Type 2 has a corner tap in the inlet and a static tap in the throat; and
• Type 3 has a corner tap at both the inlet and outlet.
Laying lengths tend to decrease with type number. Flow coefficients range from 0.9797 for Type 1 to 0.75 for Type 3.
Type 1 flow tubes more closely approach the characteristics of the classic venturi. The inlet cone converges in two angles that condition the fluid as it enters the throat. Flow coefficents are relatively stable for a variety of flow conditions. For large pipe sizes, the shorter Type 3 flow tubes (such as the one shown in Figure 5) may be useful but also may require more upstream flow conditioning for good performance. Type 3 meter coefficients may change with variations in Re, line size and beta ratio.
Flow tube sizes range from 4 in. to 48 in. Justification becomes easier for the larger pipe sizes, where installed cost may be less than that of the venturi. However, accuracy depends on the manufacturer’s calibration data. Extrapolation of meter flow coefficients for large sizes from tests on smaller sizes may be problematic.
Flow tubes can be fabricated from a variety of materials. In some cases they’re available as inserts of fiberglass-reinforced plastic or metal.
Greg Livelli and Steve Pagano are senior product managers for ABB Instrumentation, Warminster, Pa. E-mail them at email@example.com and firstname.lastname@example.org.
1. Lipták, B., “Process Measurement and Analysis,” 4th ed., CRC Press, Boca Raton, Fla. (2003).
2. Livelli, G., “Matching the Flowmeter to the Application,” Flow Control, p. 14 (August 2007).
3. “Differential Pressure Flow Elements,” ABB Inc., Warminster, Pa. (2005). Downloadable via http://library.abb.com/global/scot/scot203.nsf/veritydisplay/ee6036f29cf9e8c68025708c0047f235/$File/SS_DP_3.pdf.