Veris Verabar Brochure

Verabar
®
Velocity Averaging
Flow Sensors
True Performance in
Flow Measurement
Superior Signal Stability and Greater Resistance to Clogging
Clogging can occur in low pressure ports located in or near the partial vacuum at the rear of the sensor. The Verabar design locates the low pressure ports
on the sides of the sensor, forward of the fluid
separation point and turbulent wake area. This virtually eliminates clogging
and produces an extremely
stable signal.
Rough Surface Smooth Surface
± 1/2 Percent
3%
2%
1%
0%
-1%
-2%
-3% 0% 5% 10% 15% 20% 25%
Percent Blockage
Percent Deviation from the Published K
Air Tests Water Tests
0.5 Percent Error Band
Lower Drag and Extended Turndown
Golf balls fly farther because they have a dimpled surface
that lowers aerodynamic drag.
The grooves and roughness on the Verabar’s frontal surface
apply the same principle. This simple design feature relieves the
partial vacuum at the rear of the sensor, reducing the pressure drag.
This extends the accuracy and rangeability to very low velocities.
The Most Accurate and Reliable Technology for Measuring Gas, Liquid and Steam
Developed from aerospace technology, the Verabar averaging pitot flow sensor provides unsurpassed accuracy and reliability. With its solid one-piece construction and bullet shape, the Verabar makes flow measurement clog-free and precise.
The unique sensor shape reduces drag and flow induced vibration. And the location of the low pressure ports elimi­nates the potential for clogging and improves signal stability.
Test Data Summary
From Veris Research…True Performance in DP Flow Measurement
In God We Trust… All Others Bring Data
The unique and exclusive break­through in improved accuracy derived from the development of a verified theoretical model predicts the Verabar flow coefficients. This elimi-
nates the need for calibration tests to characterize the flow coefficients.
Without such a model,
the uncertainty of the flow
coefficients is dramatically
increased and expensive
calibration may be required.
Empirical test data from indepen-
dent laboratories verified the theore-
tical model and flow coefficients as a
constant, independent of Reynolds number
and within ±0.5% of the predicted value. The
derivation of the theoretical model and test data
is published in the Verabar Flow Test Report (ED-100).
Unique Valve Head
Verabar offers a new concept… built-in valves in the head of the instrument.
This superior design:
Simplifies installation and
maintenance.
Lowers the
connecting
hardware
cost by reducing the number of fittings.
Transmount
A Transmount flow system is the first choice for all liquids; and for gas and steam applications, with slight variations in pressure and temperature.
Mass Transmount
A Mass Transmount flow system should be selected on steam and gas applications with variable temperature and pressure.
Field Flow Systems
Ready to install, the Verabar can be ordered with a manifold, transmitter or local indicating meter.
The Proof of Verabar Accuracy
Accurate Flow Coefficients
The true test of a flow measurement device is its ability
to repeat its published flow coefficient within its accuracy
band.
Verabar
has been
thoroughly
tested at
independent
flow laboratories
(all sensor sizes, in
multiple pipe sizes,
in gas and liquids).
Spring-Lock
Complete Installation in Less than an Hour
Partial Insert
Designed specifically for
high velocity cooling water applications, large diameter pipes, large vertical stacks and buried water lines
Extends 1/3 into pipe to
reduce procurement and installation costs— specifically useful when a hot tap is required
Partial Insert hot
tap sensors can be inserted/retracted with no reduction in flow rate
Spring-Lock… Offers a Superior Mounting Method
This advanced, patented design ensures the sensor remains sealed, locked and pre-loaded to the opposite wall regardless of changes in pipe diameter due to pres­sure, temperature or mechanical force.
This design has important advantages:
Fugitive emission and leak prevention…
The Spring-Lock continually compen­sates for the differential in packing and body growth rates due to increased temperature.
Increases sensor strength,
thereby eliminating the
need for an opposite wall
support. A locked, pre­loaded sensor is four times stronger than a non-preloaded, cantilevered sensor.
Other mounting methods
do not pre-load the sensor or the packing seal and are subject to increased sensor vibration, metal fatigue, breakage and leakage.
0.74
.073
0.72
0.71
K 0.70
0.69
0.68
0.67
0.66 0 5 10 15 20 25
Velocity (fps)
Test Location: Utah water Research Laboratory
Test Data: November 13, 1991 Fluid: Water at 60 deg F
Meter Size: 6 inch, Schd 40, -10 Serial No: V1000.05.1
±1.0%
Visual load indicator
Spring travel
Compensates for changes in pipe diameter
Actual Flow Test
The Verabar offers the widest application range of any flow sensor. It accurately measures gas, liquid and steam.
High Pressure Design
2500# ANSI Class
6000PSI and 1000ºF
The Verabar’s versatile design lends itself to a wide range of applications. Contact Veris application engineering for your special requirements.
Square and Rectangular Ducts
Large Stacks and Ducts
Up to 21 feet (6 meters)
Specialized Mounting
PVC Concrete
FRP Cast Iron Pipe
Regular Models — (Threaded Components)
Model Number Type of Mounting
Tube Fitting V100 (Single Support)
V110 (Double Support)
Spring-Lock V150 (No opposite
support required)
Hot Tap Models — (Threaded Components)
Screw Drive V200
Low Pressure Hand Insertion V250
Hot Tap Models — (Flanged Components)
Model Number Type of Mounting
Screw Drive V400
Low Pressure Hand Insertion V450
Flanged Models — (Flanged Components)
Flanged V500 (Single Support) V510 (Double Support)
Flanged Spring-Lock V550 (No opposite
support required)
V100
Verabar Model Selector
Verabar Applications
V110 V150
V500 V510 V550
V200 V250
V400 V450
Gas Liquid Steam
Natural Gas Cooling/Chilled water Saturated Compressed Air Boiler Feed Water Superheated Combustion Air De-Mineralized Water Main Header Hydrocarbon Gas Hydrocarbon Liquids Custody Transfer Hot Air Cryogenic Distribution Blast Furnace Gas Thermal Transfer Fluids Energy Studies
Extended Range Applications
Fast and Easy Model Selection
The easy-to-operate Veracalc computer program features:
Flow Calculations: DP from flow rate, or flow rate from DP.
Model Selection: Complete model selection from drop down menus.
Structural Analysis: Verifies sensor strength at flowing conditions.
Temperature and Pressure Limits: Error warnings if limits are exceeded.
The Veracalc PC program is available from your local representative, the factory or it can be downloaded from our website at www.veris-inc.com.
Verabar Lowers Installed Costs
Verabar can save you more than 60% in installation costs over an orifice plate in a 10” pipe.
The graph shows the total installed cost by pipe size of the orifice plate, the Verabar, and the resultant Verabar savings. The most significant portion of the savings is the reduction in the linear inches of weld.
Savings in Weld Time Installed Cost Savings
Verabar Has the Lowest Operating Costs
Verabar pays for itself in less than a year.
The graph shows the yearly operating cost savings and equivalent horsepower savings of the non-constricting, low permanent pressure loss Verabar compared to the extremely constricting, high permanent pressure loss orifice plate. Savings are shown for gases, liquids and steam— at typical design velocities, by pipe size.
Verabar vs. Orifice Operating Cost Savings
Through Accuracy of Measurement, Low Installed and Operating Costs,
Verabar Proves Its Performance, Efficiency and Value.
Verabar Maintains Its
Accuracy
Orifice plates show long term deterioration of accuracy.
The initial accuracy of the orifice plate is ±1%. However, long term accuracy deteriorates unless the plate is periodically inspected. Senior, dual chamber fittings are available to check the plate without requiring system shutdown, but such fittings are very expensive.
Condition % Deviation
Wear of knife edge:
0.010” -2.2
0.020” -4.5
0.050” -13.1
Dirt and grease
deposits in pipe -11.1
Valve lubrication upstream:
one side of plate -15.8 both sides -24.0
Leaks around plate -8.2
Plate warpage -9.6
Orifice Plate Test Results
Florida Gas Transmission Company conducted a test to quantify various conditions which can result in inaccu­rate measurement. A partial list of the results is shown:
Pipe Size (Inches)
10 20 30 40 48
$ 1,000
$ 500
$ 10,000
$ 5,000
$ 100,000
$ 50,000
Dollar Savings
Verabar Installed Cost
Orifice Installed Cost
Verabar Installed Cost Savings
Natural Gas
Steam
Air
Horsepower
Savings @
$ 0.05/KWH
300 200
100
50
10
5
1
Pipe Size (Inches)
10 20 30 40 48
$ 1,000
$ 500
$ 10,000
$ 5,000
$ 100,000
$ 50,000
Dollar Savings
Liquids
Valve lubr. (-24%)
Warpage (-9,6%)
63 inches of linear welding
Only 4 inches of linear welding
Wear (-13.1%)
Dirt & grease (-11.1%)
Leaks (-8.2%)
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
Aerodynamic-Shaped Sensors
Extreme aerodynamic shapes that permit the stream lines to reattach are subject to airfoil type lift forces. This occurs when the angle of attack varies due to sensor misalignment, or the direction of the fluid varies, as is common in industrial piping with upstream disturbances.
Result:
The lift forces can cause an unpredictable shift in the low pressure distribution, producing inaccurate measurement.
Quality Assurance
Veris manufactures its own leak-proof, solid one-piece sensor. Our primary goal is to provide the highest quality and most accurate sensor in the industry.
This can result in a significant undetec­table loss in accuracy.
Verabar is designed to meet or exceed
applicable ANSI and ASME codes. The Verabar is available to meet
B31.1,
B31.3, B31.8, NACE MR-01-75, etc.
Additional QA capabilities include code welding, hydrostatic and other non-destructive testing.
Why Average the Velocity Profile?
Verabar averages the velocity profile through multi-sensing ports which span the entire pipe diameter. Other types of non-averaging insert meters are SINGLE POINT INSERT METERS (turbine, vortex, magnetic, sonic, etc.). They assume a “textbook: turbulent velocity profile, and use a single “critical” point to infer an
average velocity. In actual industrial appli­cations, sensors are located downstream of disturbances, such as elbows or valves, which produce non-uniform velocity pro­files. This makes it virtually impossible to locate a single point that represents the average velocity.
Result: Inaccuracy ranging from ±10% to ±20%.
Round Sensors
Round sensors produce unpredictable accuracy. The original round sensors were designed for economical fluid balancing and did not meet industrial demands for accuracy. Round sensors have a variable fluid separation point that causes an unstable low pressure distri­bution around the sensor.
Result:
Inaccuracy in excess of ±5% and as high as ±10%.
ISO 9001 Certified
Problems with Other Sensor Shapes
Diamond and T-Shaped Sensors
These sensors produce pulsating, noisy signals. They improved accuracy by use of a sharp edge to fix the fluid’s separation point. However, this greatly amplified the vortex shedding forces.
Result:
The sharp edges generate extreme vortices, causing sensor vibration, pulsa­tions and a noisy signal to the point that transmitter dampening and signal averaging are recommended.
Other manufacturers use a three-piece sensor design that has no positive mechanical method of maintaining a seal between the tubes. Therefore, temperature, pressure, vibration and even manufacturing variations can
cause leakage between the chambers.
HL
HL
Verabar Round T-shaped
sensor sensor sensor
Textbook
profile
Actual profiles
Other
Verabar
Flow
Flow
Flow
Flow
Lift forces
Flow
Flow
Fixed separation point
Variable separation point
Control valve
Elbow
Low
Reynolds
Number
Location of average velocity
6315 Monarch Park Place •Niwot, CO 80503 USA •Phone: 303-652-8550 •Fax: 303-652-8552 VB-100 (2/09) E-Mail: contact@veris-inc.com •Website: www.veris-inc.com Printed in USA
Verabar…True Performance in Flow Measurement
H L H L H
L
L
Loading...