Micro Motion Guide: Guidelines for the Selection and Operation of Provers with ELITE Coriolis Flow Meters Manuals & Guides

Reference Guide
MC-001597 Rev E
1/2018
ELITE® Coriolis Flow Meters
Guidelines for the Selection and Operation of Provers with Micro Motion ELITE
®
Coriolis
Flow Meters
Micro Motion® ELITE® ow meters are high-
precision Coriolis ow meters that are often used in the oil and gas industry in conjunction with volume provers. These guidelines are designed to aid in the selection of a prover size that will result in consistent proving repeatability, while taking into consideration the balance between:
• Maximum proving efciency
• Minimum prover size and cost
• Minimum prover wear and maintenance
Substantial experience from laboratory testing and eld proving forms the basis of the prover size recommendations and the Total Prove Time (TPT) predictions in these guidelines. However, results may vary if unstable process conditions exist during proving.
It should be noted that these guidelines are based on conservative estimates from the available data. If repeatability requirements are already consis­tently being met, there is no need to change the process or the prover size.
Proving Methods and Proving Data Evaluation
The American Petroleum Industry (API) Manual of Petroleum Measurement Standards (MPMS) Chapter 4.8, Second Edition, Operation of Proving Systems, Annex A, Evaluating Meter Proving Data explains the relationship between the number
of proving runs, the observed repeatability, and the random uncertainty of the resulting meter factor. One important principal is that a lower meter factor uncertainty will always result as more runs are collected and averaged.
Prover Sizing and Selection for Fixed­Volume Provers (Does Not Apply to Master Meter Provers)
Important Note: The prover size should never result in a pass time of less than 0.5 seconds or a pre-run time of less than 0.25 seconds.
Total Prove Time (TPT) is dened as the total
accumulated amount of time during which the prover displacer was travelling between the detector switches. The minimum TPT that is needed to achieve the target meter factor random uncertainty can be used to size the prover.
Estimated minimum TPT values that may be expected to pass repeatability requirements for
different meter sizes and ow rates are shown in Table 1. The velocity of the uid as it travels through the meter ow tubes is also shown in Table 1 in
units of feet per second (fps). Velocities above 60 fps are not recommended when proving with
a xed-volume prover and may result in excessive
TPT to pass repeatability. For applications above 60 fps, master meter proving is recommended instead of xed-volume provers.
Equation 1 describes TPT:
www.Emerson.com/MicroMotion
Base Prover Volume(BPV)
𝑇𝑃𝑇 
Flow rate
×(# 𝑜 𝑟𝑢𝑛𝑠󰇜 × (# 𝑜 𝑝𝑎𝑠𝑠𝑒𝑠 𝑝𝑒𝑟 𝑟𝑢𝑛󰇜
Reference Guide
MC-001597 Rev E
1/2018
ELITE® Coriolis Flow Meters
To use Table 1 to size a prover, there are two methods to select from:
Method 1 (Determine the BPV): Find the minimum
TPT value from Table 1. Multiply the ow rate by
0.0117 to convert from BPH to gallons per second. Then enter the TPT, ow rate, the number of runs,
and the passes per run (if averaging multiple passes
per run) into Equation 2 to nd the minimum BPV
needed in gallons.
Equation 2:
𝐵𝑃 
TPT x Flow Rate
(# 𝑜 R𝑢𝑛𝑠󰇜×(Passes pe Run󰇜
Example: CMFHC4 meter at 6500 BPH
• From Table 1: Velocity ≈ 30 fps. TPT =
20 seconds.
• Convert: 6500 BPH X 0.0117 = 76 gallons per second
• If 5 single-pass runs are required, BPV = (20 seconds X 76 gallons per second) ÷ (5 runs X 1 pass per run) = 304 gallons or 7.2 BBLS.
• If 10 runs and 3 passes per run are acceptable, BPV = (20 seconds X 76 gallons per second) ÷ (10 runs X 3 pass per run) = 50 gallons or
1.2 BBLS.
Method 2 (Determine the Number of Passes Needed): Find the minimum TPT value from
Table 1. Multiply the ow rate by 0.0117 to convert from BPH to gallons per second. Then, insert the ow rate and a BPV in gallons into Equation
3 to estimate the total number of passes that
will be needed for a prover size with that BPV. If
averaging multiple passes per run, divide the total number of passes needed by the number of runs and round up to determine the minimum number
of passes per run needed. For single-pass runs,
the number of runs needed will equal the total number of passes needed.
Equation 3:
𝑇𝑜𝑡𝑎 # 𝑜 𝑃𝑎𝑠𝑠𝑒𝑠  𝑇𝑃𝑇×
Flow rate
Base Prover Volume(BPV)
Example: CMF400 meter at 2300 BPH
• From Table 1: Velocity ≈ 40 fps. TPT ≈
30 seconds.
• Convert: 2300 BPH X 0.0117 = 27 gallons per second
• If the BPV is 170 gallons (4 BBLS), Total # of passes = 30 seconds X 27 gallons per second ÷ 170 gallons = 5 total passes
(5 total passes ÷ 5 runs = 1 pass per run
for 5 runs).
Page 2
• If the BPV is 65 gallons (1.55 BBLS), Total # of passes = 30 seconds X 27 gallons per second ÷ 65 gallons = 12 total passes
(12 total passes ÷ 5 runs = 3 passes per run for 5 runs).
Reference Guide
MC-001597 Rev E
1/2018
ELITE® Coriolis Flow Meters
Table 1. Estimated Minimum Total Prove Time (TPT)
Values shown are the velocity inside the meter in feet per second (fps)
followed by the Total Prove Time (TPT) in seconds.
TPT = prover pass time X passes per run X runs per proving* (*total number of runs required to reach ±0.027% meter factor uncertainty may vary.)
Page 3
Reference Guide
MC-001597 Rev E
1/2018
ELITE® Coriolis Flow Meters
• The liquid inside the piping connecting the meter to the prover should remain stable.
o
Minimize piping between the meter and
prover.
o
Avoid dead-end branches between
meter and prover that may act as a volume
“spring” with compressible uids.
• Sufcient back pressure must be maintained on
both the prover and the meter to avoid vapor
breakout and to maintain a stable ow rate during displacer launch and travel. Minimum recom­mended back pressure is shown by Equation 4 (from API MPMS Ch. 5.6).
Table 1 (cont.) Estimated Minimum Total Prove Time (TPT)
Increasing the BPV will result in meeting the mini-
mum TPT with fewer passes. Increasing the number of passes will allow reaching the minimum TPT with a smaller prover. A decision may be made either to
size the prover with a smaller BPV (lower capital
investment) with a longer overall proving time, or
with a larger BPV with reduced runs (less long-term
wear and tear) and shorter overall prove time.
Prover Conditions
It is important to prove at conditions that are as similar as possible to the expected operating con­ditions. There are many conditions and factors that
can inuence the success of proving systems.
• Prover equipment and all supporting reference
measurement devices must be well-maintained and veried to ensure measurement traceability,
reproducibility, and repeatability (API MPMS Ch. 4 and Ch. 21.2, paragraph 2.11).
• Stability of ow rate, density, temperature, and
pressure is critical during proving. System design, prover settings, and maintenance can all impact
ow rate stability during proving.
Equation 4: ρb ≥ 2 Δρ + 1.25 ρ
e
Where: ρb = Minimum back pressure (psig)
ρ = Pressure drop across meter at max.
Δ
ow rate
ρ
= Equilibrium vapor pressure at
e
operating temperature (psia)
• Accurate prover density measurement is crucial when mass proving with a volumetric prover. The following tolerances are advised when using a pycnometer (API MPMS Ch. 14.6).
o
Max. temperature difference = 0.2 °F
o
Max. pressure difference = 1 psi
o
Density Meter Factor (DMF) repeatability
should be 0.05% or better between consecutive
pycnometer tests
• Flow pulsation from PD pumps, including lack of back-pressure regulation, may inuence
repeatability and additional passes may be needed to meet random uncertainty requirements.
• Enabling compensation for the effect of pressure
on the meter (consult the Transmitter Congura-
tion and Use Manual) can improve repeatability in applications where line pressure varies by more than 30 psig during proving runs.
Page 4
Reference Guide
MC-001597 Rev E
1/2018
ELITE® Coriolis Flow Meters
API MPMS Chapter 5.6 Measurement of Liquid Hydrocarbons by Coriolis Meters denes a pulse
scaling factor (PSF) as the number of pulses output
by a Coriolis meter per unit of ow. In this way, a Coriolis meter PSF is like a mechanical meter’s K-factor, but adjustable.
Micro Motion transmitters typically provide a fre­quency output of up to 10,000 Hz (10,000 pulses per second). Many eld device pulse input specications
have a similar maximum pulse input rate of 10,000
Hz. However, some eld devices have a maximum pulse input rate that is less than 10,000 Hz. The PSF of the Coriolis owmeter can be adjusted to stay within the constraints of any eld device that is
counting the pulses from the meter per the equations shown here:
Meter Operation
When using smaller provers, it is important to ensure
that the meter is congured for optimum ltering and
speed of response.
1. Select the fastest speed of response available:
o
5700 transmitter: select “Low Filtering”
response mode
o
2700 transmitter: select “Special” for Update
Rate and either “Special” or “Low Filtering”
for Calculation Speed
2. Set ow damping to a value between 0.0 and
0.08 seconds.
3. Set density damping to 0.16 seconds.
• Proving Wizard software is available from Emerson to aid in preparing Micro Motion Coriolis meters for proving
Coriolis owmeters do not have a xed K-factor (number of pulses output per unit of ow). The number of pulses output per unit of ow (e.g. pulses
per barrel) from a Coriolis meter is an adjustable parameter that can be set to any desired value.
However, the frequency of pulses during the highest ow rate must not exceed the pulse input capacity
of the prover pulse counting device.
When the meter factor remains stable between proving events, this indicates that the meter zero setting value is good. A change in the meter factor may or may not be related to the meter zero, so it is
important to always perform a Zero Verication Test (consult Transmitter Conguration and Use manual) before making any zero adjustments. Only adjust the meter zero if advised to by the Zero Verication Test.
If a meter zero is adjusted, reprove the meter.
Page 5
Reference Guide
MC-001597 Rev E
1/2018
Master Meter Proving
Micro Motion Coriolis meters can be used as master
meters per API MPMS Ch. 4.5 for proving with the
following advantages:
• A Coriolis master meter can be used to prove in either volume and/or mass units.
• Pass duration can be lengthened to improve repeatability.
• Maintaining stable process conditions is much easier with no effects due to a displacer launch.
• Low maintenance and high reliability, with no seals or moving parts.
ELITE® Coriolis Flow Meters
Need More Information?
Emerson has extensive eld experience in mass
and volume proving of our Micro Motion Coriolis meters.
Contact us at 1-800-522-6277 or visit our website at
www.MicroMotion.com
Emerson Automation Solutions
7070 Winchester Circle Boulder, CO USA 80301 T +1 303 527 6277 F +1 303 530 8459
www.Emerson.com
©2018 Micro Motion, Inc. All rights reserved.
Loading...