November 2017 by Energy-Tech Magazine

Centrifugal compressor 16 • Situational awareness 21 • Tube failures 26

ENERGY-TECH A WoodwardBizMedia Publication

NOVEMBER 2017 www.energy-tech.com

Dedicated to the Engineering, Operations & Maintenance of Electric Power Plants In Association with the ASME Power Division

Ensuring flexible operation of combined-cycle steam turbines

RECIPROCATING AIR/GAS COMPRESSORS National Compressor Services Your OEM Alternative! REPAIR & REPLACEMENT PARTS & SERVICES - Oil Pump Repair - Boring/Honing - Packing Case Repair - Lubrication Systems - Main Bearings - Crankshaft Repair - Seals - Shim - Pressure/Wiper Packing - Rider & Piston Rings - Regulators - Babbitt Bearing Repair - Crosshead & Wristpin Repair - Preventative Maintenance Programs - Motor Repair or Replacement - Lubricants - Component Repairs/Replacement - Controls (New & Retrofit) - Surplus and Reconditioned Air Compressors and Parts

- Piston/Rod Repair - Casing Repair - Crankpin Bearings - Oil /Air Filters - Gasket Kits - Valve Reconditioning - Welding & Fabricating - Machining - Metallizing - Compressors & Dryers - Cylinder Boring & Sleeving - Water Jacket De-scaling - Complete Overhauls - Portable Onsite Field Machining - Parts (New) - Metal Stitching - Intercooler & Aftercooler - Rentals

OEM EQUIPMENT WE SERVICE

www.national-compressor.com

Come check us out at the TurboMachinery & Pump Symposia Houston, TX December 12-14, 2017

BOOTH #2916!!

ENERGYT ECH P.O. Box 388 • Dubuque, IA 52004-0388 Phone: 563.588.3857 • Fax: 563.588.3848 Email: sales@WoodwardBizMedia.com

FEATURES

5 16

inaccuracies, errors or advertising content. Entire

contents © 2017 WoodwardBizMedia. All rights reserved; reproduction in whole or in part without permission is prohibited. Printed in the U.S.A. Group Publisher Karen Ruden – kruden@WoodwardBizMedia.com

Editorial views expressed within do not necessarily reflect those of Energy-Tech magazine or WoodwardBizMedia.

sue.babin@WoodwardBizMedia.com

keith.neighbour@WoodwardBizMedia.com or call 773-275-4020 Michelle Stappert michelle.stappert@WoodwardBizMedia.com or call 563-588-3856 Graphic Artist Eric Faramus – eric.faramus@WoodwardBizMedia.com Address Correction Postmaster: Send address correction to: Energy-Tech, P.O. Box 388, Dubuque, IA 52004-0388 Subscription Information Energy-Tech is mailed free to all qualified requesters.

NewTech Applied Tech

Tube failure

The science behind eddy current and remote field testing: for condenser and heat exchanger tubing By Christopher Van Name, Gary Fischer and James Kocher, Conco Services Corporation, Verona, Pa., USA

Sue Babin

Keith Neighbour

EPRI: Situational awareness of operators using modern distributed control systems

ASME FEATURE

Advertising Sales

or call 773-275-4020

Maintenance Matters

By Jack Odlum, Systems Engineer at HRST, Inc.

Randy Rodgers – randy.rodgers@WoodwardBizMedia.com Kathy Regan – editorial@WoodwardBizMedia.com

Centrifugal compressor field performance testing

By Dwayne Coffey, Electric Power Research Institute

General Manager Editor

Machine Doctor By Patrick J Smith, Engineer at Air Products & Chemicals

in print and digital format by WoodwardBizMedia, WoodwardBizMedia assumes no responsibility for

By Stephen R. Reid, Principal Engineer, TG Advisers Inc.

COLUMNS

Energy-Tech (ISSN# 2330-0191) is published quarterly a division of Woodward Communications, Inc.

Ensuring flexible operation of combined-cycle steam turbines

INDUSTRY NOTES

4 31

Editor’s Note and Calendar Advertiser’s Index

ON THE WEB Have you seen Energy-Tech’s weekly newsletter yet? If not, sign up at www.energy-tech.com for the latest technical articles, business news and products in the industry.

To subscribe, go to www.energy-tech.com or E-mail circulation@WoodwardBizMedia.com Media Information For media kits, contact Energy-Tech at www.energy-tech.com or sales@WoodwardBizMedia.com Editorial Submission Send press releases to: Editorial Dept., Energy-Tech, P.O. Box 388, Dubuque, IA 52004-0388 Ph 563.588.3857 • Fax 563.588.3848 email: editorial@WoodwardBizMedia.com Advertising Submission Send advertising submissions to: Energy-Tech, 801 Bluff Street, Dubuque, Iowa 52001 E-mail: ETart@WoodwardBizMedia.com

A division of Woodward Communications, Inc.

November 2017 ENERGY-TECH.com

EDITOR’S NOTE

What are your plans for 2018? 2017 is coming to a close and looking back at the Energy-Tech year, I’m happy with the promises that were made and the results that we delivered on. 2017 was a great learning year for me as I completed my first full year as editor for Energy-Tech magazine. I was able to attend both the Electric Power Conference and the ASME Power & Energy Conference in 2017 which really opened my eyes to the wealth of knowledge and expertise in the industry. I plan to tap into that knowledge and expertise to bring you even better technical articles, webinars and online training opportunities as we progress through 2018. In 2017, our authors were able to deliver excellent technical articles for each print issue and also our weekly e-newsletters. The lead story in the February issue on transient torsional vibration problems due to electrical noise by our Machine Doctor author, Patrick Smith, was a huge hit as well as Patrick’s other articles and the monthly columns provided by T.G. Advisers, ASME and EPRI just to name a few. In addition, the content for our website is growing with an increase in technical articles, industry news and new product introductions pertinent to the electric power industry. I expect this to continue in 2018. Beyond just reading our magazine, I hope that you’ve had a chance to attend one of our online opportunities this past year whether it was a technical online training session on gas and steam turbine vibrations or a FREE webinar sponsored by one of our advertisers. If you missed these educational opportunities, there’s still time to attend a webinar that was important to you by visiting the Energy-Tech website at www.energy-tech.com. Look for the webinars tab where all past webinars are still available conveniently for your download. Email me or call me at 563-588-3857 for a special 20% off promo code if you’d like to take advantage of a training session for PDH credit before year-end. As we move on to 2018, I’m excited for what we have to offer. Our technical content is filling up fast and I’m working on a full schedule of training opportunities for you as well. We’ll be partnering with Environment One Corporation (E-One) again to co-host the biennial Generator Auxiliary Systems Symposium to be held at the end of July in Saratoga Springs, NY. Make sure you get this in your budget and commit early. There is limited space available and this one will fill up fast! Watch for an email on this or call me for the details needed for your budget – 563-588-3857.

CALENDAR December 5-7, 2017 Power-Gen International Las Vegas Convention Center Las Vegas, Nev. www.power-gen.com December 12-14, 2017 Advanced Turbine Troubleshooting & Failure Prevention Online Training with Steve Reid & Tom Reid of TG Advisers www.energy-tech.com/turbine December 12-14, 2017 Turbomachinery & Pump Symposium George R. Brown Convention Center Houston, TX tps.tamu.edu March 19-22, 2018 Electric Power Conference & Exhibition Gaylord Opryland Convention Center Nashville, Tenn. 2018.electricpowerexpo.com June 24-28, 2018 ASME 2018 Power & Energy Conference & Exhibition Disney’s Contemporary Resort Lake Buena Vista, Fla. www.asme.org/events/power-energy July 30 – August 1, 2018 Generator Auxiliary Systems Symposium Hosted by Environment One Corporation (E/One) and Energy-Tech Magazine Saratoga Springs, N.Y. www.Energy-Tech.com/Gen-Sym Submit your events by emailing editorial@WoodwardBizMedia.com

Whatever the new year brings, look to Energy-Tech for the technical expertise needed in the power industry with our weekly enewsletter and our quarterly print publications. I’d love to hear from you. E-mail me at editorial@woodwardbizmedia.com with your ideas for topics that you’d like to see covered in articles or webinars. Thanks for reading,

Kathy Regan

4 ENERGY-TECH.com

November 2017

FEATURES

Ensuring flexible operation of combined-cycle steam turbines By Stephen R. Reid, Principal Engineer, TG Advisers Inc.

Dispatching requirements for many combined cycle plants (CCP) have pushed design limits for both load and on-off cycling operations. Gas turbine technology incorporated in CCP arrangements typically has the design capability to cycle multiple times a day. This is especially the case for smaller facilities wherein aero-derivative gas turbines have maintenance intervals biased more towards operating hours than on-off cycling. But where does this leave the steam turbine? If the facility is in the design phase, ensuring that the steam turbine chosen can respond rapidly and operate reliably under cycling conditions requires detailed turbine specifications and acceptance criteria along with field due diligence assessments. The due diligence must dig deep into cycling design parameters and damage mechanisms which most likely will not be evident until well after the manufacturerâ&#x20AC;&#x2122;s warranty has expired. The same is the case for CCPs that repower existing

steam turbines. In many of these cases, the steam turbine may have spent most of its operating life in a base loaded regime. Cycling will expose new weaknesses requiring repairs and upgrades as damage accumulates. This article will highlight the failure mechanism and critical design criteria essential for a new facility as well as address some of the pitfalls and issues associated with repowering applications.

Crack initiation factors Every start, stop and load change decreases the overall low cycle fatigue life of a steam turbine. Mitigating fatigue related problems means reducing the potential for crack initiation and limiting subsequent crack growth. The operating profile specified for a CCP must take advantage of the fast start time of the combustion turbine with some minimal delay period

Double flow LP turbine November 2017 ENERGY-TECH.com

FEATURES rotor failure. The following factors highlight the design parameters that should be carefully evaluated.

Blade root stress modeling

for the heat recovery steam generator (HRSG) to produce superheated steam and pre warm the HPIP turbines. For those facilities with reheat steam turbines, the warm up process must address preheating requirements of both the HP and IP sections of the turbine. TG Advisers has found a number of repowered facilities that have HP steam admission on a combined HPIP turbine but no direct IP admission until the unit reaches full speed and some minimal load level. This is a dangerous process especially taking into consideration turbines built decades ago which have inherent flaws and embrittled rotor material conditions. Lack of preheat and cold bore temperatures put a HPIP turbine at great risk for brittle

Thermal stress management Lowering the temperature differential between metal turbine components and the steam flow is an important design factor to be considered. Thermal-related stress is the main variable that must be managed to mitigate cycling damage. The principal means by which to limit these stresses and cycling damage include: (1) extend ramp-up time, (2) limit the mass flow rate for steam during startups, (3) extend soak times; and (4) modify the ramp-down during shutdowns. The cycle analysis considers the fully reversed thermal and mechanical stress cycle which simply means both the startup stress and shutdown stress is used to count 1 damage cycle. With the need for rapid starting of CCPs, the shutdown cycle may be the best option for mitigating the total damage cycle. For one application evaluated by the TGA team, the shutdown preferred cycle consisted of rapidly going from steady-state operation to the no-load condition. The goal is to minimize the time period in which the turbine operates with both a high mass-flow rate and low steam temperatures. Gradually reducing load before shutdownâ&#x20AC;&#x201D;known as forced coolingâ&#x20AC;&#x201D;can decrease turbine life significantly. In fact, rapid shutdown from the full-load or steady-state condition can double the turbineâ&#x20AC;&#x2122;s expected life compared to forced cooling options. Centrifugal stress management In general, the combination of thermal and centrifugal stresses plays a major role in LCF damage to the high pressure (HP) and intermediate-pressure (IP) turbines. In contrast, centrifugal stress acting alone is a primary cause of LCF damage in low-pressure (LP) turbine and generator rotors. Choosing components designed for minimal centrifugal stress is a key aspect in any optimization effort. By evaluating the results of finite element analyses, where peak centrifugal stresses existed for various turbine rotor component designs, an optimized design can be developed. During a different project, finite-element analysis revealed that the filleted contours of one conventional LP rotor disk would exhibit high peak stress concentrations in the planned operating mode which were virtually eliminated by incorporating a rotor design with smoother, more gradual contours.

6 ENERGY-TECH.com

November 2017

FEATURES Extreme care is a must when comparing peak-stress analyses from different turbine suppliers. First, recognize that stresses greater than a given material’s yield strength must be corrected to reflect the true plastic strain. Most LP later stage bladeattachments feature peak stresses higher than the material’s yield strength. Thus, well-established methods of correcting for plastic strain, such as Neuber’s rule, should be understood and applied. The explanation of these methods goes beyond the scope of this article. But ignoring these methods may yield gross errors in predicted LCF damage and component life. Other factors that must be studied carefully when comparing peak stress analyses supplied for different designs include: (1) the effect of blade-root-togroove tolerance on peak stresses, (2) the type of finite element applied, (3) the effect of blade and root-groove boundary conditions, (4) mesh refinement; and (5) the number of nodes per element. Overall, you must determine the effects of any differences between finite-element analysis methods and assumptions used by different manufacturers.

Crack propagation analysis The presence of a small crack does not necessarily mean a turbine has reached the end of its useful life. Therefore, the factors causing a crack to grow to a critical size must also be studied. Every turbine component has flaws inherent to the forging and fabrication process. In many cases, the flaws are too small for accurate evaluation by existing NDE methods. Thus, linear-elastic fracture mechanics (LEFM) criteria

Provided to our customers For Over 135 years!

Creep In addition to cycling stress, longterm creep also plays a major role in the initiation of cracks in HP and IP turbines. Creep and fatigue do interact and can significantly decrease predicted life if they are not evaluated as interacting. Pure creep rupture characteristics for a given rotor material can be analyzed by studying the material’s Larson-Miller stress-rupture curves. The Larson-Miller parameter (LMP) correlates allowable rotor stress to exposure time, temperature, and material. Given a known steady-state value for maximum stress, the fractional life reduction for critical components can be found by using the LMP chart to identify a material’s LMP for creep rupture. Assuming that the resultant exposure time for creep rupture is greater than the plant’s design life, the fractional damage can be estimated by the ratio of the plant’s design life to creeprupture time. In some cases, the turbine manufacturer may include creep data as part of the LCF design charts.

Balancing Machines for all Industries Moment Weighing Solutions Spin Test Systems High Speed Facilities Vibration Analysis Equipment Seminars & Training

The Difference is Quality. The Difference is Schenck.

schenck-usa.com

1-800-873-2352 sales@schenck-usa.com

November 2017 ENERGY-TECH.com

FEATURES should be incorporated into the turbine specifications. These requirements dictate manufacturers must prove that any indication detected in the turbine will not grow to critical crack size within a given duty cycle.

Material optimization Regardless of the turbine supplier, power producers can ensure conservatism in the rotor design by specifying materials optimized for LCF and creep resistance. Example: Metals with high ductility will have larger critical crack sizes compared to less ductile materials. This translates into longer component

NEW

life. A material’s impact energy is one measure of ductility. Fracture toughness of a material, which can determine critical crack size, can be approximated directly if its impact energy is known. The critical crack size can then be estimated because it is directly related to the square of the fracture toughness. Another materials-related property that should be considered is the Fracture Appearance Transition Temperature (FATT), which is the temperature at which a material exhibits halfbrittle and half-ductile characteristics during impact testing. Specified materials with the lowest FATT possible will ensure the best possible toughness characteristics during startup with minimal prewarming requirements. ■

Deluxe Kits!

Everything you’ll ever need

VIDEO BORESCOPES all in one indestructible carrying case!

UP TO

2,000

IN SAVINGS! 4mm & 6mm Deluxe Kits now starting at $9995!

Stephen R. Reid, P. E. is president and principal engineer of TG Advisers Inc. (TGA) Steve's company has provided troubleshooting and condition assessments on more than 350 STG units including assessments on all nuclear units in the U.S., Spain and Belgium. TGA also provides expert witness services, failure analysis and maintenance optimization services on both steam and gas turbine units. He has numerous patent disclosures/awards, and has published more than 50 papers. He received the ASME George Westinghouse Silver Medal, is past chairman of the ASME Power Operations Committee and is a Registered P.E. He is a short-course instructor for all major conferences, EPRI and a number of OEMs. You may contact him by emailing editorial@WoodwardBizMedia.com.

Videoscopes now starting at $7995! All Hawkeye® Deluxe Kits Contain: • 4-way Articulating V2 Videoscope • All Standard Accessories

PLUS: • 90° Prism Tip • Close-Focus Tip • V2 Stand • V2 Rigidizer

gradientlens.com 8 ENERGY-TECH.com

800.536.0790

Made in USA November 2017

Keeping the World Flowing

Improving efﬁciency & minimising downtime It’s in our DNA

For sixty years our customers have relied on Rotork for innovative and reliable ﬂow control solutions. Rotork products and services help companies in the oil & gas, water & waste water, power, marine, mining, chemical, pharmaceutical, and food industries around the world.

T 1 585 247 2304

Rotork Innovation A Client Support Programme that helps you to: Ÿ Protect your investment Ÿ Increase plant availability Ÿ Maximise productivity Ÿ Reduce cost of ownership Ÿ Protect the environment

E info@rotork.com

W rotork.com

ASME FEATURE

The science behind eddy current and remote field testing: for condenser and heat exchanger tubing By Christopher Van Name, Gary Fischer and James Kocher, Conco Services Corporation, Verona, Pa., USA

Abstract With the increasing demand on the world’s power grids, now more then ever it is important to keep power plant condensers, feedwater heaters and balance of plant heat exchangers running at peak efficiency. While it is well known that keeping these units clean is important for maximizing power output, so too is monitoring each unit’s tube integrity and taking corrective action to prevent tube failure. The best way to monitor a unit’s tube integrity, detect patterns of tube wear and damage, and determine the specific wear and damage to a particular tube is through Non-Destructive Testing. Depending on the tube material, the best NonDestructive Testing method to employ would be either eddy current testing, remote field testing or other variations of these electromagnetic techniques.

the probe. When the probe is inserted into a metal tube, a circular flow of electrons will begin to move through the metal, generating its own magnetic field. This circular flow of electrons is the eddy current. As the probe moves through the tube, the magnetic field generated by the eddy current will interact with the coil’s magnetic field. Defects in the tube wall, such as pitting or cracking, and changes in wall thickness will interrupt or alter the amplitude and pattern of the eddy current, changing its magnetic field. This change in the magnetic field then affects the coil by varying its electrical impedance, which is monitored by the test instrument. By plotting the changes in the impedance amplitude and phase angle on a monitor, a trained operator can compare the pattern displayed on the monitor to patterns of known test samples to determine the condition of the tube being inspected.

Both eddy current testing (ECT) and remote field testing (RFT), also known as remote field eddy current testing, use the principles of electromagnetic induction to detect defects in condenser and heat exchanger tubes. In both ECT and RFT probes, an alternating current flows through a wire coil or coils, generating an alternating magnetic field around 10 ENERGY-TECH.com

Figure 1. Probe coils in an absolute arrangement used for surface scanning.

(Photo Courtesy of Conco Services Corp.)

Introduction The modern eddy current testing industry owes its existence in a very real sense to Michael Faraday, (1791 to 1867). This brilliant scientist’s discovery of and experiments into electromagnetic induction laid the foundation for the many Electromagnetic Testing techniques in use today. Though there has been much advancement in test instrument technology, computer software and test coil design, the basis of the electromagnetic techniques still rely heavily on the experiments performed by Faraday in the mid 1800’s.

(Photo Courtesy of Conco Services Corp.)

This paper will discuss the science behind eddy current and remote field testing, how they differ and which one to select depending on the situation. It will look at the construction of the probes and how they work. It will explain the difference between use of a single frequency or multiple frequencies and the advantages of multi-frequency testing. The paper will also identify the necessary procedure for a successful NonDestructive Test, including the types of tubes that can be tested and tube preparation.

Figure 2. Probe coils mounted on a fixture to spin the coil inside of a tube to provide a focused surface scan. ASME Power Division Special Section | November 2017

ASME FEATURE (Photo Courtesy of Conco Services Corp.)

the tube. This type of coil is the most widely used and is considered the “workhorse” of the tube testing industry. 3. Encircling coil (Fig. 5) – Also known as feedthrough coils, this coil type allows the inspection of round objects such as tubes, wires and rods from the outside diameter (OD) surface. Much like the bobbin coil, the encircling coil also interrogates the entire circumference of the material as it is passed over the material or the material is fed through the coil. This type of coil is used mostly in production monitoring activities.

(Photo Courtesy of Conco Services Corp.)

Figure 3. Standard issue barnacle scraper style bobbin coil in a self comparison differential coil arrangement. This can be configured to run in the external reference mode simultaneously with the self comparison mode.

Test coil arrangements How a coil is electrically configured to operate with the test instrument is an essential variable to optimal performance. The three basic coil arrangements are: [1] 1. Absolute (Fig. 1 and Fig. 2) – An arrangement where the coil works independently, making no reference to any other coil, and is affected by all changes in the material. This coil is usually limited to use by conductivity testers, coating thickness gauges and small surface riding pancake coils for surface scanning.

Figure 4. Flexible bobbin coil design that can negotiate U-bend tubes.

Probe construction All ECT and RFT probes consist of a coil or coils wrapped around a structure to form the coil. It is the manner in which these elements are designed and their interaction with each other that will determine how the eddy currents are induced and how flaws are detected in the test material. Most coils are built on a non conductive body (air core) but many can be constructed using ferrite cores and conductive shielding to help shape the eddy current field for special applications. Three basic coil types According to James Cox, author of “Nondestructive Testing, Eddy Current: Classroom Training Handbook”, there are three basic coil types: [1] 1. Probe coil (Fig. 1) – Also referred to as a pancake coil, it is designed to test the surface of materials and can be applied to plates, welds or even tubing when fixed to a special device that spins the coil (“spinning probe technique”) inside a bolt hole or a tube. When a probe coil is fixed to this type of spinning device it is commonly referred to as a motorized rotating pancake coil (MRPC) (Fig. 2) in the heat exchanger industry. This probe type can provide some very detailed information but is time consuming and expensive to operate. 2. Bobbin Coil (Fig. 3 and Fig. 4) – This coil type allows for the inspection of installed heat exchanger tubing from the inside diameter (ID) surface. The bobbin coil interrogates the entire circumference of the tube as it is drawn through November 2017 | ASME Power Division Special Section

2. Differential – An arrangement where two or more coils are electrically connected in some fashion to oppose each other and look for an imbalance or “difference” between the coil impedance when a flaw is encountered. Differential coil arrangements are sub categorized into two types. a. Self Comparison (Fig. 3, Fig. 4 and Fig. 5) – In this differential arrangement, at least two coils are electrically connected, placed in close proximity to each other and wound in opposition. If both coils are affected by the same condition, the output or “difference” is zero. This arrangement is very sensitive to small volume flaws such as pits, cracks and any abrupt changes in wall thickness such as those caused by tube-to-baffle wear, while minimizing noise due to probe motion (wobble) as the probe traverses the tube, temperature variations and deposits in the tube. While effective in detecting abrupt changes in wall thickness, the self comparison differential cannot detect gradual wall loss associated with steam erosion or tube-to-tube wear. b. External Reference – In this differential arrangement, at least two coils are electrically connected to each other but not in close proximity. The coils are either separated on the same test part by a distance that does not allow any direct coupling between the two coils or one coil is on the test part while the other coil sits in a fixed location on a reference sample that represents nominal material conditions. This arrangement is sensitive to all measurable changes (much like an absolute coil but with better detection) including abrupt changes, gradual wall loss, temperature variations, probe wobble, and any other gradual condition that can produce noise. The data can be erratic and is typically reserved for defect confirmation against the self comparison differential channels and for detection of specific damage like erosion and tube-to-tube wear. ENERGY-TECH.com

ASME FEATURE

Figure 5. Encircling probe (coil exposed) designed to test tubing from the OD. It is a similar configuration as the bobbin coil.

To perform an adequate examination on any condenser or heat exchanger, it is imperative to utilize both modes of differential operation. The industry typically refers to the self comparison mode as differential and to the external reference mode as absolute because of the data display. Sensitivity to all measurable changes in the material under test is very similar to an absolute coil response. 3. Hybrid â&#x20AC;&#x201C; Hybrid coils, also called driver pickup or reflection coils, have the widest range of configurations. The possibilities are as limited as the engineersâ&#x20AC;&#x2122; imagination. The basic configuration utilizes a separate excitation coil and an independent sensing (pick-up) coil or set of sensing coils. The excitation and sensing coils can be incorporated into each other or separated by certain distance. The coils can even be on opposite sides of thin foil or plate (through transmission) passing the eddy current field through the test part and measuring the change in field on the opposite side. Hybrid coils have endless configurations to meet special needs of the inspection industry, but are not necessarily the most affordable breeds.

ECT vs RFT probe Eddy current probes used for inspection of heat exchangers in the majority are bobbin probes operating in the differential modes, self comparison and external reference. There are some minor variations in design such as a narrow groove bobbin which would have an enhanced sensitivity to small pits and cracks and magnetic bias probes to overcome slight permeability issues, but for the most part function the same way over a host of manufacturers. The biggest concern comes into the probe selection being compatible (impedance matching) with the test system being used.

Eddy current testing using absolute coils is not completely out of the question. Some MRPC probes and profilometry probes (8 x 1 pancake coil arrangement) use absolute, surface riding pancake coils to scan tubes from the inside and provide detailed defect information. Other specialty eddy current tube probes include a myriad of magnetic bias probes which incorporate strong rare earth magnets to overcome mild permeability variations in certain metal alloys. Remote field probes (Fig. 7) used for inspection of ferromagnetic tube materials all fall into the hybrid coil design. The basic function is the same for all, a large excitation coil generating the magnetic field to penetrate the tube wall and a pickup coil located two to three tube diameters from the exciter coil detecting the changes in field strength as the energy enters back into the tube wall. One can find RFT probes in many variations of exciter/pickup coil configurations, each having their own benefits for certain applications. A variant to the RFT probe is a NFT probe. This probe is also a driver/pickup mode of operation but the sensing coil is placed in the near zone (close to the exciter) rather than using the far (remote) field energy for detection. This probe is helpful in applications where there is OD copper or aluminum fins on carbon steel tubes where RFT testing is ineffective.

Fill factor For all bobbin probe applications for tube testing, probe size is an essential variable. Ideally the probe would occupy as much of the diameter of the specimen to be tested as possible [2]. The probe size is a compromise between accessibility to the tube and the best possible energy intercepting the tube to provide for strong eddy currents. The stronger the eddy current field, the better the results. For most eddy current (Photo Courtesy of Conco Services Corp.)

(Photo Courtesy of Conco Services Corp.)

array probe (Fig. 6) designs incorporate a self comparison differential bobbin coil in the array probe to provide conventional eddy current data right along with the specific array data. Of course there is an added cost for complex design configurations, but overall, advanced array probe technology has reached a price level that is more affordable for the common condenser and heat exchanger inspection application.

A smaller grouping of eddy current probes falls into the hybrid category for specialized configurations. Some multi sensing element probes, such as array probes, operate in a driver pickup mode of operation. Care needs to be exercised in coil selection for hybrid designs; as sensitivity is gained for one specific damage mechanism, other mechanisms may be missed. Probe cost is also a consideration for hybrid probes as they can be very complex and very expensive. More advanced Figure 6. Advanced array probe which includes a conventional bobbin coil.

12 ENERGY-TECH.com

ASME Power Division Special Section | November 2017

ASME FEATURE applications, 85% fill factor is a good target percentage. For remote field testing, fill factor can be reduced to 70% and still produce viable results. When calculating fill factor, the following the formula is important. FF = d2/D2 In the above formula FF is the fill factor, d is the outside diameter of the coil and D is the inside diameter of the tube. Many times the calculation is run without squaring the diameters resulting in over inflated fill factor percentage which provides for a poor eddy current energy. Poor fill factor can also result in increased baseline noise in the eddy current data. When considering fill factor it is important to clean the tubes immediately prior to testing. This will eliminate blockages and allow for the use of maximum fill factor for best test results.

Eddy current vs remote field testing Though both eddy current and remote field techniques rely on electromagnetic induction as a function of the inspection process, they are very different in operation and application. Eddy current testing relies on direct coupling between the inspection coil and the test material and works very well for non-ferromagnetic materials. Materials which are magnetic have a major impact on the penetration of the eddy current field. Also, the permeability varies throughout the material and causes erratic signals and increased noise. ECT can be used on mildly permeable tubes by use of rare earth magnets (mag bias) placed near the inspection coil to â&#x20AC;&#x153;zeroâ&#x20AC;? out the permeability effects and let the eddy current alternating magnetic fields run free to do their job.

(Photo Courtesy of http://www.et-ndt.org/ wp-content/uploads/2015/07/RFT.png)

Remote field testing is designed to overcome the permeability effects in ferromagnetic tubing such as carbon steel and ferritic stainless steels. Before the development of RFT, test methods for inspecting carbon steel were very

limited. As the name implies, remote field testing does not work in the direct coupled zone. The remote field zone is the region in which direct coupling between the exciter coil and the receiver coil(s) is negligible. Coupling takes place indirectly through the generation of eddy currents and their resulting magnetic field. The remote field zone starts to occur at approximately two tube diameters away from the exciter coil [3]. RFT does theoretically work on non permeable materials but it is not as accurate or effective as conventional ECT.

Applicability Eddy current testing for tubing applications is limited to non permeable materials like copper, brass, copper-nickel, austenitic stainless steels and similar alloys. ECT can be used for mildly permeable tubes like Monel and ferritic stainless steels with the use of magnetic bias probes, provided they are strong enough to saturate the material. A limitation may be that with strong magnets traversing the probe can be difficult as the tube supports attract the probe as it passes intensifying the physical aspects of the inspection process. For materials which are permeable and highly permeable like ferritic stainless steel, carbon steel and similar alloys, remote field eddy current testing is a viable option. Going beyond the limits of eddy current testing using a magnetic bias probe, RFT can inspect the toughest carbon steel material with no restricted mobility. Since the energy used to penetrate the material is an alternating current field and is not a DC magnetic field, the probe is not attracted to the tube wall or tube supports as it passes through the tube. RFT is not as definitive as ECT, but for applications where tubes are not testable by ECT, RFT provides valuable data to provide for a reasonable condition assessment.

Calibration For all techniques of electromagnetic testing, it is crucial to perform a valid test system calibration to ensure functionality and sensitivity are adequate for the intended inspection. The ASME Boiler and Pressure Vessel Code, Section V identifies the basic requirements for nondestructive testing [4]. More specifically, article 8 and article 17 specify the requirements for eddy current testing and remote field testing, respectively. The articles define all essential elements of the inspection system including test probes, calibration standards, test system requirements, frequency selection, calibration settings and documentation. ASME section V also defines certification requirements for inspection personnel. The inspection company is required to develop a written practice to define how they qualify and certify inspection personnel in accordance with recommended practices. From the applicable codes, inspection companies need to develop inspection procedures to provide instruction to the inspectors and maintain consistency and repeatability from one inspection to another and from one inspection performed over time on the same heat exchanger.

Figure 7. RFT Probe â&#x20AC;&#x201C; Basic operation is Driver/Pickup (receiver) and falls into the hybrid coil arrangement category. November 2017 | ASME Power Division Special Section

ENERGY-TECH.com

ASME FEATURE Single frequency vs multi frequency testing More is better, correct? Well, at least for eddy current testing that is true. The ASME code only defines the prime frequency needed to obtain a desired response. It allows for the use of additional alternate frequencies, but does not require them. Subsequent frequencies each have their strengths for detection and allows for signal mixing to eliminate unwanted interference like tube support plate signals. The same holds true for the mode of operation where ASME defines differential and absolute modes but calls for either or both as an option. Running both is highly recommended. The variety of a number of eddy current channels and mode combinations allows an extensive analysis of defect (flaw) depth and characterization [5] . Unfortunately, remote field testing does not easily lend itself to the variety of frequencies and signal mixing that ECT does. Due to the characteristic low frequency operation, one or two test frequencies are typical for a RFT inspection. Adding too many low frequencies has an impact on production by reducing the sample rate and in turn forces slower scanning speeds.

Conclusion Electromagnetic testing techniques have been proven effective for many years and continue to provide viable inspection data for heat exchanger tubing condition assessment. Armed with the detailed information on a unit’s tube integrity provided by eddy current testing and remote field testing, plant managers can take proactive steps to either repair, replace or plug damages tubes before they fail, preventing a forced outage. Understanding how each testing technique works and the capabilities and limits of each will provide the plant with the ability to choose which direction you should take, ECT with multiple frequencies for non-ferromagnetic tubes or RFT with one or two frequencies for ferromagnetic tubes. One size does not fit all, as each will provide information that is vital to your equipment assessment. Don’t go it alone. As discussed there are a wide range of variables to consider when selecting the best inspection technique for your specific application. Consulting with eddy current equipment manufacturers and service providers can help to navigate the many options available and select what will work the best for your circumstances. With continued technological improvements, options for tubing applications are always expanding. However, we can revert back to the basics when selecting the appropriate technique for a specific application.■

Applications: • Steam pressure vessels • Hot water heaters • Demineralizers • Steam humidifiers • Water purifiers • Refrigeration units • Liquid treatment vessels

• Compressed air tanks • Filtering units • Dryer cans in paper mills • Water towers • Water softeners • Deaerators • Make-up tanks

The steam boiler market makes extreme demands on gaskets; Topog-E® Series 2000 gaskets are specifically designed and fabricated to provide the superior performance users of Topog-E® gaskets have come to expect – but at higher operating pressures and temperatures.

1224 North Utica | Tulsa, OK 74110 T: 918-587-6649 | E: info@topog-e.com W: www.topog-e.com

14 ENERGY-TECH.com

References [1] Cox, J., 1997, “Classroom Training Handbook, Nondestructive Testing, Eddy Current,” PH Diversified, Inc., Harrisburg, NC Chap. 3. [2] Droesch, D., Maestas R., Saxon Jr., G., 2006, “The Advanatges of Using Multiple Frequencies for Eddy Current Examination of Condenser Tubing,” ASME Paper No. POWER2006-88149, Proceedings ASME Power, Atlanta, GA May 2006. [3] NDT Education Resource Center (2001-2014) “Introduction to Eddy Current Testing,” Education Resources, NDT Course Material, EC Testing, NDE-ED. org. [4] ASME Boiler and Pressure Vessel Code, Section V, 2007 edition, Article 8 and Article 17. [5] Innospection (ND) “Multiple Frequency Eddy Current Technique,” retrieved from http://www.innospection.com/ pdfs/Multiple%20Frequency%20Eddy%20Current.pdf on 12-3-2015. Editor’s Note: This paper, Power Energy PowerEnergy2016-59273, was printed with permission from ASME and was edited from its original format. To purchase this paper in its original format or find more information, visit the ASME Digital Store at www.asme.org.

ASME Power Division Special Section | November 2017

HARNESS THE POWER OF CUSTOM ENGINEERING

When RENTECH custom engineers a boiler to your exact specs, you get peak operation and lower emissions. Great custom engineering starts with expert engineers—and our people are the most resourceful and experienced in the business. For durability, energy efficiency and clean operation, count on RENTECH’s custom-engineered solutions. We master every detail to deliver elemental power for clients worldwide. HARNESS THE POWER WITH RENTECH.

HEAT RECOVERY STEAM GENERATORS WASTE HEAT BOILERS FIRED PACKAGED WATERTUBE BOILERS SPECIALTY BOILERS

WWW.RENTECHBOILERS.COM

MACHINE DOCTOR

Centrifugal compressor field performance testing By Patrick J Smith, Engineer at Air Products & Chemicals

Condition monitoring is a key part of an overall reliability program. Sometimes the signs of progressive wear and/or damage are very subtle and difficult to detect. In these cases, a machine may appear to suddenly degrade or fail without warning. The purpose of condition monitoring is to detect thermodynamic and/or mechanical degradation at an early stage. This allows for action to be taken before the condition worsens to a point that becomes critical or causes a machine failure. Not having a condition monitoring program puts much more reliance on the compressor protection system and operators to detect a problem. This often results in a delay in detecting a problem.

Introduction This case study pertains to a 4-stage, dual service integrally geared centrifugal compressor driven by a 1787 RPM, 5000 HP induction motor. The gearbox consists of a bullgear and two rotors. The low speed rotor comprises the first two stages of a supplemental main air compressor (SMAC) service. The HS rotor comprises the 3rd stage of the SMAC service and a single stage of compressed dry air (CDA) service. The CDA flow comes from multiple sources and is a higher delivered flow than the SMAC service. All the impellers are a semi-open type. In this design the front side of the impeller is open and the vanes run against a close clearance, noncontacting stationary shroud. The compressor configuration is shown in Figure 1.

Thermodynamic performance monitoring is only part of an overall condition monitoring system. Mechanical (vibration) monitoring, cooler performance monitoring, function testing of key valves, lube oil analysis, water washing, etc. are other parts. Thermodynamic performance monitoring includes both efficiency and overall machine capacity. Operating pressures, temperatures, flow and power need to be measured and recorded so that calculations and trends can be performed to access if there has been any degradation. The August 2009 Energy-Tech article entitled â&#x20AC;&#x153;Centrifugal compressor thermodynamic performance monitoring and analysisâ&#x20AC;? describes one technique for evaluating overall centrifugal compressor efficiency. Although this analysis can be done by manually, having an automated system is Figure 1. Compressor configuration much more effective. Once a compressor thermodynamic performance problem has been identified, it is usually advantageous to determine the cause so that a more effective plan can be developed to address the problem. In many cases this involves analyzing the compressor stage performance to determine if a stage or component has degraded. The purpose of this article is to present an example of how centrifugal compressor stage performance testing was useful in identifying and correcting a performance problem.

16 ENERGY-TECH.com

Compressor control system The basic piping and instrumentation diagram (P&ID) for this compressor is shown in Figure 2. The SMAC control system includes a 1st stage IGV, a discharge flow meter, a discharge pressure transmitter, and a discharge vent valve. In this application, the SMAC capacity is controlled with the IGV and the machine discharge pressure is a function of the system back pressure. A DCS (Digital Control System) is used to operate and control the compressor. The control system also includes a surge avoidance system which opens the discharge vent valve if the machine operates within

November 2017

MACHINE DOCTOR a predefined margin to surge. The CDA control system is similar except that rather than a discharge vent valve, a recycle valve is used. The recycle valve is used to recycle cooled discharge gas back to the suction so that the process gas is not lost when the valve is open.

History There are two identical SMAC/CDA compressors in this plant. Both machines were commissioned and were operated in continuous service for approximately 16 years before a performance problem was identified with the SMAC “A” compressor during the summer months. The “A” compressor was not able to maintain flow and pressure compared with the “B” machine. At the same discharge pressure, the “A” compressor flow was lower and was approaching the surge controller limits. The operators noticed that small movements and swings in plant flow would cause the “A” compressor to back out of the header and the vent valve would open. At the time the problem was identified, automated thermodynamic performance monitoring of these compressors had not yet been implemented. Condition monitoring was managed locally.

Basic thermodynamics The basis for this discussion is that the process gas can be treated as an ideal gas. Centrifugal compressor performance can be impacted by many operating conditions, including inlet/discharge pressures, compressor inlet and interstage suction temperatures, gas properties, flow, IGV opening and other factors. The interstage temperatures are generally a function of cooling water temperature. A centrifugal compressor is designed for a certain volume flow and pressure rise. Mass flow is the product of volume flow and gas density. So, in the summer, when the air is less dense, the maximum capacity of an atmospheric air compressor is lower compared when the air is cooler and more dense. Centrifugal compressors also impart energy into the process gas, which results in an increase of the Enthalpy. The increase in Enthalpy is typically referred to as head.

A compressor process can be modelled as an isothermal, adiabatic or polytropic process. The basic equations used to model these processes are derived from the first and second laws of thermodynamics. Then, certain criteria, based on the process that is used for the model, further refines the equations. For example, if a compression process is modelled as adiabatic, there is no heat transfer. The head equation for an adiabatic compression process is: Ha = Z*R/MW *T*k/(k-1)*(rp((k-1)/k) - 1) where: Ha = adiabatic head, ft-lbf/lbm Z = compressibility factor; 1.0 for an ideal gas R = ideal gas constant; 1545.32 ft-lbf/(lbmole * ºR) MW = Molecular Weight T = Inlet Temperature; ºR k = ratio of specific heats r p = pressure ratio = discharge pressure / outlet pressure; pressures are psia As seen in this equation, if the inlet temperature is higher, then the pressure ratio is lower for the same head. The term Z is used to correct for deviations from the ideal gas law. For moderate deviations from the ideal gas law, the Z at the inlet and outlet is sometimes averaged as an approximation. The efficiency equation for an adiabatic process is: ηa = Tin * (r p((k-1)/k)-1) / (Tout – Tin) where: ηa = adiabatic efficiency Tin = stage inlet temperature, ºR Tout = stage outlet temperature, ºR Stage power is the product of mass flow times head divided by the efficiency. For an adiabatic process, the stage power equation is: Wa = Ha * w / ηa where: Wa = stage adiabatic power, BHP w = mass flow, lbs/hour The stage power equation only accounts for the thermodynamic power. In addition to this, there are mechanical losses, such as bearing are gear losses, which add to the overall stage power. These basic equations form the basis for the stage performance analysis.

Figure 2. PEID November 2017 ENERGY-TECH.com

MACHINE DOCTOR Basic aerodynamics The amount of head a centrifugal compressor stage can impart is a function of the impeller geometry, number of impeller blades, impeller speed, and stationary member design. Stationary members include the diffuser, casing, diaphragm, etc. The ideal head a compressor stage can impart is a function of the impeller tip velocity (at the OD) and the impeller vane backward lean, which is the curvature of the impeller vanes at the OD. Due to something referred to as slip, the actual head is somewhat lower and varies with flow. Actual head can typically be represented by the following equation:

The compressor supplier stage curves were based on an isentropic process. An isentropic process is a reversible adiabatic process. Stage curves included Q/N verses μ and Q/N verses ηa where: Q = volume flow in ACFM (Actual cubic feet per minute) N = impeller speed in RPM The shop performance test stage curves were then converted to ACFM verses head using equation (4), and ACFM verses efficiency.

Field test The compressor testing was limited to the SMAC service 2 since there was not an issue with the CDA service. To H = (1/g)*μ*u2 perform the test, the following data was collected: where: • IGV opening (plant instrumentation) g = gravitational constant • Inlet temperature (ambient temperature, locally measured) μ = head coefficient • Relative humidity (used information from local weather u2 = impeller tip velocity service) • Cooling water temperature (plant instrumentation) Assessment • SMAC discharge flow (plant instrumentation) The reason for the “A” compressor performance • Motor power (plant instrumentation) degradation was not obvious with the available operating • Inlet pressure each stage (locally measured) data. Reviewing past performance, it appeared the “A” • Discharge pressure each stage (locally measured) compressor capacity had degraded about 5% when compared • Inlet temperature (locally measured) with past summer performance. It was decided to perform • Discharge temperature (locally measured) a stage by stage performance test on both compressors to compare the individual stage performance of both machines Shop test data was not available at different IGV angles. So, relative to one another and to compare with the original to compare with field performance with shop test data, it was “A” compressor shop test curves. Hopefully this would necessary to test with the SMAC IGV’s at 100% open. validate the “A” compressor degradation and determine the cause and possible corrective actions. Note that only the “A” A couple of things to note. First, the compressor stage compressor was shop performance tested. performance curves are based on total pressures, which includes the static head and the velocity head. The stage pressures that were measured were static pressures and did not include the velocity head. The velocity head is a small component of the total pressure in these type of compressors, but it does introduce some errors when comparing field data with factory test data. Secondly, calculated compressor stage efficiency is very sensitive to the temperature measurements. And, with field instruments, there can be a moderate amount of error. All the measurements are also based on single instruments, not the average of multiple instruments as described in many industry performance test codes. And, finally, the location of the instruments used to measure the stage data is not necessarily the same as done in the shop test or described in industry codes. All these factors add a small amount of discrepancy when comparing field performance Failure to properly with shop test performance.

Are Shaft Currents Destroying Your Machinery?

ground rotating equipment can result in expensive bearing, seal, & gear damage.

SOHRE TURBOMACHINERY® INC.

128 Main Street, P.O. Box 1099 Monson Massachusetts, USA 01057-1099 PH: 413.267.0590 • 800.207.2195 • FX: 413.267.0592 info@sohreturbo • www.sohreturbo.com 18 ENERGY-TECH.com

November 2017

MACHINE DOCTOR

Figure 5. 2nd stage

“A” and “B” compressors. The 1st stage deviation for the “B” compressor was not unexpected since the IGV’s were only partly open. The largest discrepancy was the “A” compressor 2nd stage. This is also shown in Figures 3 and 4 below which compare the field measured head and efficiency with the shop tested stage curves. The “B” compressor 2nd stage head fell right on the compressor shop tested curves. The efficiency tested a little above the “A” compressor shop test curve. This probably had to do with the accuracy of the field data.

Figure 3. 2nd stage head performance

Based on this testing, it was decided to inspect the SMAC “A” compressor 2nd stage.

Inspection The 2nd stage was inspected. As shown in Figures 5 and 6, there was some light fouling and some minor erosion of the leading edge of the diffuser vanes. The diffuser vane damage was reviewed with the compressor supplier, who advised that this erosion could account for the measured degradation in stage performance. The impeller was cleaned up and the diffuser was replaced. The diffuser was procured prior to the inspection based on the suspicion that there could have been diffuser damage.

Figure 4. 2nd stage efficiency

Results It was not possible to take the compressors offline to take data over a complete range of flows. And, it was not possible to operate the “B” compressor with the IGV’s at 100% open due to process limitations. The results of the field test data for the “A” and “B” compressors are shown in the table below. As shown, there was a moderate amount of deviation from the shop tested curves for the 1st and 3rd stages for both the

The machine was re-tested after the repair and as shown in Figures 7 and 8, the 2nd stage head and efficiency tested higher than the shop tested performance curve. Although there is some instrument error, the data did show a significant improvement in the 2nd stage performance which fully resolved the overall performance problem with this machine.

Conclusions In this case, a compressor performance problem was identified in the early stages of degradation. Through stage testing, the problem was confirmed and the problem area

November 2017 ENERGY-TECH.com

MACHINE DOCTOR

Figure 6. 2nd stage diffuser vane Figure 7. 2nd stage efficiency - after repair

was identified making the physical inspection and corrective action much easier. Had the type of degradation described in this article gone un-noticed, it is possible that the compressor could have been driven into unstable operating conditions, possibly leading to a failure. One such example is the impeller failure described in the March 2012 Energy Tech article, “Destructive forces in centrifugal compressors.” In this example, the 1st stage of a centrifugal compressor stage was driven into stall when the 2nd stage degraded due to fouling. In another case, a compressor suffered from a significant drop in capacity. Based on stage testing, there was no stage degradation, but the intercooler pressure drops were much higher than design. Machine capacity was restored once the intercooler bundles were replaced. As discussed in the introduction and in the examples discussed, early identification of machinery degradation through a robust condition monitoring program is a key factor in a successful reliability program. ■

Figure 7a. 2nd stage head performace - after repair

References 1. Brown, Royce, “Compressors Selection & Sizing,” Houston, Texas, Gulf Publishing Company, 3rd Edition, 2005. 2. Smith, Patrick J., “Centrifugal compressor thermodynamic performance monitoring and analysis,” Energy-Tech Magazine, August 2009 3. Smith, Patrick J., “Destructive forces in centrifugal compressors,” Energy-Tech Magazine, March 2012 Patrick J Smith is lead machinery engineer at Air Products & Chemicals in Allentown, Pa., where he provides technical machinery support to the company’s operating air separation, hydrogen processing and cogeneration plants. You may contact him by emailing editorial@woodwardbizmedia.com.

REACH YOUR

H2IGHEST

POTEN2TIAL STREAMLINE OPERATIONS WITH ON-SITE HYDROGEN GENERATION • SAFE • RELIABLE • COMPACT ProtonOnSite.com | 203.678.2000 20 ENERGY-TECH.com Info@ProtonOnSite.com • COST-EFFICIENT ™

November 2017

MAINTENANCE MATTERS

EPRI: Situational awareness of operators using modern distributed control systems By Dwayne Coffey, Electric Power Research Institute

For some years, the Electric Power Research Institute (EPRI) has sponsored ongoing research on situational awareness of control room operators using modern Distributed Control Systems (DCSs). This research has tried to understand how power plant monitoring and control systems, in particular control room graphics, provide information that operators can use to make the best decisions in a timely fashion. Research findings have helped to illuminate the differences between data and information, how humans perceive and act upon visual cues, and how computer graphics can be improved to provide more actionable information.

Situational awareness The term “situational awareness” describes the recognition or consciousness of humans to their surroundings. For researchers of control room operations, the term means three things: (1) being aware of what is happening around you, (2) understanding what that information means to you now, and (3) understanding what that information means to you in the future.

Design engineers responsible for control systems often start with a mental model of the decision-making process that differs from how operators actually reach decisions. Without understanding the end user’s goals on situational awareness, the information provided by the control system will have little or no meaning. For situational awareness to work, the cues provided by control displays need to lead to three levels of understanding (Figure 1). First, the cues must lead to perception of clear information. Second, the cues should lead to comprehension, or an understanding of the meaning of the information. Third, the cues should provide a projection of actions to be taken. When problems arise with operators gaining these three levels of understanding, studies have shown that 78% of the time, operators don’t get the information that is needed (Level 1); 17% of the time, they don’t correctly understand the information that they do get (Level 2); and 5% of the time they don’t project what will happen in the future (Level 3).

Figure 1. For situational awareness to work, the cues in display screens need to lead to perception, comprehension, and projection.

November 2017 ENERGY-TECH.com

MAINTENANCE MATTERS Control rooms and alarm management Power plant control rooms have changed dramatically over the years. Prior to 1990, all operations could be monitored and indicated with a single glance at one place on a benchboard. An immediate visual imagery was available of the entire plant. Also, at that time, alarms had to be generated by a physical device—a pressure switch, a relay, or Figure 2. Digital data versus analog indicator. another instrument. If a pump tripped, the circuit breaker that fed that pump went on trip status. As a result, a practical limit existed to the number of alarming devices. The total number of possible alarms in a plant was typically in the hundreds. In today’s control room, everything is monitored and indicated on 50-60 different computer graphic screens. Extensively more visual imagery is possible but not always interpretable. As regards alarms, with the invention of digital alarming and the continued improvement in automated controls in power plants, the number of alarmable power plant items seen by a control room operator increased exponentially from a few hundred alarms to tens or hundreds of thousands alarms. In this environment, the alarming capabilities of the plant can exceed the information-handling capacity of the operator. Industry studies have shown that operators can effectively manage up to 150 alarms per day, and are likely not able to manage systems that generate more than 300 alarms per day. An overabundance of alarms causes the control room operator to perform an internal mental “alarm shedding” to determine what is important and can lead to incorrect event diagnosis. Due to the continuous high alarm counts found in many current alarm systems, the operator is actually desensitized to the alarms, increasing the likelihood of equipment damage.

Figure 3. Comparisons of poor and improved displays of information.

Data is not information Many DCSs provide displays of raw numeric data. Such displays raise a number of questions: What numbers would indicate optimal operation? How long does it take you to scan and interpret the numbers? Do the numbers mean anything to you? How much training would you require before you could interpret the numbers? Instead of columns of raw data, the displays should provide data that supports comprehension. Abnormal values should be seen at a glance. In this sense, the definition of “information” is data in context made useful. As a result, control room operators need the digital numeric data in an analog display format. Merriam-Webster’s dictionary

In 2014, EPRI published the Alarm Management Philosophy Document (3002004840), which aims to help utilities meet these challenges by providing a “how-to” guide to implementing an alarm management program. The guide describes best practices based on the experiences of utilities and the knowledge and experience of industry subject-matter experts and EPRI managers. Figure 4. Vessel level indicators with improved information.

22 ENERGY-TECH.com

November 2017

MAINTENANCE MATTERS defines “analog” as “signals or information represented by a continuously variable physical quantity such as a spatial position.” The analog information that control room operators need is a pointer to a scale, or a clear indication of the normal working range, which shows upper and lower limits (Figure 2). The human brain interprets an analog display quicker than a number. Such a display allows us to easily see where the value is, as well as what it is, and to easily see rates of change.

Designing DCS screens Control room operators interact with the control system through a set of graphics that constitute the human machine interface (HMI). These graphic HMIs for most power plants were generally developed in the 1980s and 1990s as older technology control systems were replaced with computerbased DCSs. At that time, no guidelines existed on how to create effective HMIs. Display design and style have a significant impact on the speed and accuracy of operators’ interactions with a display. Common misuse of color and animation are impediments to easy and early recognition of a significant event. Improper design slows response times, and contributes to errors in perception and comprehension. Attributes of DCSs that can hinder situational awareness include: • High-contrasting colors cause eye strain and fatigue. • Movement of graphics, which draws your attention to them without obvious meaning. • Complex graphics and 3D objects, which make it difficult to form a mental model. Poorly designed HMIs can be responsible for: • Encouraging poor operating techniques, such as “running by alarms” • Actively impeding proper situational awareness • Contributing to higher numbers of avoidable upsets • Increasing the likelihood of suboptimum response to abnormal situations As a result, poor HMIs have often been cited as contributing factors to major industrial accidents. Proper HMI design increases the ability of operators to distinguish different conditions, recognize important information, and respond to abnormal plant conditions. Characteristics of properly designed HMIs include: • Gray backgrounds are used to minimize glare. • Low-contrast depictions are used for improved comprehension. • Color is used in a very limited way and is used consistently. • Animation is limited and used only to highlight abnormal situations. • Information is only shown that supports comprehension of a process or plant.

Figure 5. Overview DCS display screen.

• Key performance data is represented as trends. • Attention is attracted to an area only if there is an issue. • Information is provided that helps operators retain data in their short-term memory. • Information is grouped together so it can be processed as one data point. • Layout is consistent with an operator’s mental model of the process.

Thermal Performance Nuclear Fossil Combined Cycle

Heat Balance Analysis and Plant Evaluation Thermal Performance Training Courses Plant DVR (Data Validation and Reconciliation) ASME PTC Plant Testing (Performance Test Codes) TP-Plus

TSM (Thermal Monitoring Software)

TP-Plus

CIM Software (Cycle Isolation Monitoring)

TNC has worked with over 100+ sites worldwide and has the knowledge, tools, and expertise to solve your thermal performance problems, no matter how big or small

www.tnorthconsulting.com

1-970-964-2573

November 2017 ENERGY-TECH.com

MAINTENANCE MATTERS • A hierarchical structure supports progressive exposure of detailed information. • Process values are depicted in the context of information, not as raw data. • Alarms are prominently depicted and important. • Alarm colors are used only to display alarms. Figures 3 and 4 show poorly and properly designed HMIs. Figure 5 shows an overview display with the broadest available view of the facilities under a single operator’s control.

It is a “big picture,” at-a-glance view of the process unit. It provides a clear indication of the current performance of the process by tracking the key performance indications. The overview display is frequently placed on a larger monitor mounted from the overhead or a wall in the operator’s line of vision for ease of use. In many facilities, operating areas are closely coupled through feed and product streams, heat integration, and dependent utilities, so the overview of any given area may be useful to adjacent operators. The display screen in Figure 5 has several important features. The two rows of indicators across the top and middle of the screen are “moving analog indicators.” These indicators display not only the process value (number) but also its proximity to a normal or expected range, and a high or low alarm range. At a single glance, an entire bank of such indicators can be evaluated.Values outside of normal range are easily detected, and values in alarm stand out strongly.

PERPETUAL MOTION

The graphs on the bottom third of the screen are “embedded trends elements.” The depiction of trends is fundamental to high-performance graphics. Even in trends, it is important to indicate the boundaries of normal and abnormal operation.Various DCSs have differing abilities in this regard.

What every plant manager wants… To achieve high productivity and financial success, today’s complex process operations require unyielding power that won’t quit. That’s why thousands of smart process operators have chosen custom-built Skinner Steam Turbine Generator packages to produce power & savings in such diverse applications as refineries, petrochemical plants, food processing and other facilities. We configure the optimum system for your requirements up to 2.0 MW including a steam turbine generator along with other key accessories. Contact us for a reliable smaller-scale package that does the job you want it to do with a price to match.

“A half-century of outstanding steam turbine experience” Skinner Power Systems A Division of Time Machine, Inc. 8214 Edinboro Road Erie, Pennsylvania 16509, U.S.A. Toll-free: (877) 868-8577 www.skinnerpowersystems.net

24 ENERGY-TECH.com

The design of the display screen in Figure 5 also has several useful characteristics: • Normal operational values are shown in white. • High and low alarm ranges are shown in dark gray. • Desirable operational ranges are shown in light blue. • Alarm indicator is shown with priority level and color. • Different shape is used for alarm priority. EPRI research in graphic design for DCS displays is summarized in the report Operator Human Machine Interface Case Study: The Evaluation of Existing “Traditional” Operator Graphics Versus High-Performance Graphics in a CoalFired Power Plant Simulator (1017637).■ Dwayne Coffey (dcoffey@epri.com) is Principal Program Manager of EPRI’s Operations Management and Technology Program. You may contact him by emailing editorial@WoodwardBizMedia.com

November 2017

Running on MindSphere

at if you could spendWhat if y re time on the more tim en, while we help green, w p you out of the red?keep you Running on MindSphere

What if you could spend more time on the green, while we help ng your assets is just one part of the service plan. Maintaining your ass keepmodeling, you out ofBythe red?advanced ng advanced analytics and predictive applying

elp you optimize everything from process, we can help you opti nce, output, and maintenance. With improved performance, output and availability, you can pursue new market reliability and availab Maintaining your assets is just one part of the service plan. By applying advanced analytics and predictive modeling, ities, and most importantly collect more profitsâ&#x20AC;Ś opportunities, and m we can help you optimize everything from process, u putt. while you putt. performance, output, and maintenance. With improved reliability and availability, you can pursue new market opportunities, and most importantly collect more profitsâ&#x20AC;Ś while you putt.

Digital Services for Energy-servicing more than your Siemens Digital Servi rvicing your profitability. assets, your Siemens Digital Services for Energy-servicing more servicing than your assets, servicing your profitability.

Visit siemens.com/power-generation-services

NEW TECH APPLIED TECH

Tube failures By Jack Odlum, Systems Engineer at HRST, Inc.

Introduction Tube leaks in HRSGs are a common cause of outage expenses and downtime.There are numerous causes of tube failures, with some more common or difficult to detect than others.This article details four HRSG tube failure mechanisms: corrosion fatigue, short-term overheat, pitting and flow accelerated corrosion. Each mechanism has a failure mechanism, indications of impending failure, and methods to prevent it from occurring at your plant. Knowing the signs can help identify risk areas and take proactive measures to prevent failure from occurring.

Figure 1. Typical failure location. Failed tube marked with X, water can be seen coming from the inner radius of the return bend.

Corrosion fatigue Corrosion fatigue is the degradation of a material under the combined actions of corrosion and cyclic loading. In the case of HRSG economizers, it is the result of cyclic fracturing and reforming of the protective oxide layer on the internal tube surface. In order to materialize, corrosion fatigue requires a corrosion component and a stress component.

Figure 2. Detail of corrosion fatigue crack Figure 3. Corrosion fatigue failure at economizer tube-to header joint. on economizer return bend

Cause of failure: Corrosion fatigue results in tube failure through the combination of a stress and corrosion component.Areas damaged by corrosion act as stress concentrators when the material is stressed. Small cracks initiate from the stress concentrator.These cracks expose new base metal which is subjected to further corrosion.The crack becomes partially filled with corrosion products.When the material is under compression, the corrosion products act as a wedge further propagating the crack.When the material is under tensile stress, the corrosion products are exfoliated exposing new base metal and the process begins again. The combined effect of corrosion and stress is greater than the sum of stress and corrosion acting separately.

Are Shaft Currents Destroying Your Machinery?

Failure to properly ground rotating equipment can result in expensive bearing, seal, & gear damage.

Sohre Turbomachinery® Inc. designs and manufactures shaft grounding brushes and related systems. Thousands of Sohre shaft grounding brushes have been installed and are being used in many kinds of rotating equipment (steam turbines, gas turbines, compressors, pumps, propeller shafts, reduction gears etc.). Over ten thousand Sohre Turbomachinery® brushes are installed, protecting machinery all over the world. These brushes can also be used as part of a shaft current monitoring program. have been installed to replace brushes initially supplied by the turbomachinery OEM. In many cases, Sohre Turbomachinery® brushes are specified as standard equipment on new machinery.

SOHRE TURBOMACHINERY® INC.

128 Main Street, P.O. Box 1099 Monson Massachusetts, USA 01057-1099 PH: 413-267-0590 • FX: 413-267-0592 • info@sohreturbo.com • www.sohreturbo.com 26 ENERGY-TECH.com

November 2017

NEW TECH APPLIED TECH How to spot the issue before failure occurs: Corrosion fatigue can be difficult to spot during inspections. Economizers and feedwater preheaters are the most vulnerable areas within the HRSG for this damage mechanism. Problem areas include: â&#x20AC;˘ Tube to header welds, especially if the tube has a bend just prior to the header. â&#x20AC;˘ Return bends at the top of an economizer that that also serves as part of the support for the adjacent tubes. The cracks can initiate from the OD or ID of the tube, depending on the original design and fabrication quality. Often, visual inspection can reveal clues to motivate more detailed and focused inspection. Bowed tubes and return bends showing signs of compression (oval instead of round bends) indicated high stresses in those tubes. Unloaded return bends, those free to move side to side during an inspection, indicate high stress in neighboring loaded bends.The high stress and corrosion risk situations typically occur during cycling (high thermal stresses and waterside chemistry upsets), meaning HRSGs that cycle more often have a higher likelihood of failure.

Preventing failure: Preventing corrosion fatigue is a matter of mitigating the effects of the corrosion and stress.The failure mechanism, by definition, requires a combination of stress and corrosion to propagate cracks in the tube wall. Smart cycling procedures that prevent a large amount of cold water entering a hot economizer helps prevent high stresses in the return bends, while wet layup with a nitrogen blanket helps prevent oxygen ingress, which is the most likely source of the corrosion aspect. Controlling feedwater chemistry during start-ups is also critical to preventing the corrosion aspect of the mechanism.The best way to deal with corrosion fatigue is to minimize the stress and corrosion contributors, as the issue is difficult to inspect for or repair without equipment replacement. Sometimes, changes to the original design can reduce the stress. Examples include: modifying water flow circuitry at the economizer water inlet or changing the support strategy to sudden temperature changes during start-up do not cause excessive stress. Pitting Pitting is a failure mechanism in HRSG tubes resulting from an improper wet layup in standing units, or neglect for HRSG components prior to installation. If tube bundles are allowed to lay open for an extended period during shipping and construction, untreated water could pool on the tube surface, combining with ambient atmosphere to start corrosion that forms pits in the tube surface. For units in operation, extended periods of wet layup without an oxygen-displacing gas or chemical will cause pits in the same way.

Figure 1. Pitting found in HP Figure 2. Pitting found steam header due to oxygen ingress in tubes due to oxygen during wet layup. ingress during wet layup (same drum as Fig 1).

Cause of failure: The root cause of pitting is untreated water staying in contact with the metal surface for extended periods of time, either during layup or construction. Oxygen in the water, often from the ambient air, will make its way throughout the tube section during a shutdown, unless an inert gas like nitrogen is used to displace it.Attention and care are required when storing the unit, especially in wet layup. Units that are required to be in stand-by for long periods but rarely run are the most susceptible to pitting failures. How to spot the issue before failure occurs: The best way to check for pitting in HRSGs is a combination of borescope inspection and tube sampling. Units with layup issues will have rust tubercles form on the tube and header IDs.This is especially true in the evaporators below the steam drum waterline. During a drum inspection, tubercles found below the waterline can be rubbed away to reveal pits. Pits in the drum are a good indicator that the issue is occurring in the tube bundle itself. Once pitting is observed in the drums, borescoping and sampling of tubes becomes a high priority. Even if no pitting is seen inside the bundle, tubercles indicate the need for an improved wet layup procedure. For units that experience pitting from improper storage prior to construction, pitting can be caught prior to start-up with a thorough borescope inspection after construction is complete. Preventing failure: When pitting is found, the course of action varies based on pit depth and amount. Shallow pits allow flow over the affected area during operation once the tubercles are removed. In many cases, the pits will re-passivate when the HRSG is running. It then becomes an issue of removing existing tubercles and avoiding lay-up periods with oxygen- laden water. Putting a nitrogen blanket on the HRSG water side when it is in wet layup may be required to prevent further pitting. Deep pits, however, do not allow for treated water to effectively re-passivate the surface.As there is no way to economically fill pits, this situation requires constant monitoring of the existing pits and their growth. Knowing how rapidly the problem is progressing could allow the plant to replace the affected components before failure occurs. Once replaced, preventing pits from re-forming again becomes an issue of wet layup and preventing oxygen intrusion. Short-term overheat Short-term overheat tube failures in HRSGs are typically caused by foreign objects within a tube bundle that block some or all of the flow through a tube.This results in a sudden increase in tube metal temperature, leading to yielding and eventually stress rupture.These failures are identified by excessive bowing and swelling of the tube.

Figure 3. Pitting caused by water resting Figure 4. Pitting failures manifest as small in panel prior to installation. Discovered pinholes on the tube. via borescope following numerous pitting leaks in nearby tubes.

November 2017 ENERGY-TECH.com

NEW TECH APPLIED TECH

Figure 1. Swollen tubes and failure resulting from short-term overheat

Figure 2. HP Evap tube bowed over 12” out of position. This was the first indicator of the problem

Objects known to cause these failures range from foreign objects, e.g. aerosol cans dropped down a downcomer, as well as broken steam or waterside components from attemperator spray probes and liners, valve internals and drum internals.

Cause of failure: The failure is caused by the tube metal at temperatures significantly higher than design.As a material increases in temperature the maximum allowable stress decreases, reaching a point where there is a significant drop in allowable stress per extra degree.Tube temperature is more closely controlled by the cooling side (water, steam) than the heating side, meaning during normal operation the working fluid keeps the tube metal temperature near its design value.When the flow is blocked, the temperature of the tube is closer to that of the gas, which can be several hundred degrees above the design temperature. How to spot the issue before failure occurs: The window of opportunity to catch the failure is short, hence the name short-term overheat. Failures of this kind are best avoided by minimizing the risk of foreign objects blocking flow. Periodic borescope inspection to check for foreign objects and damaged attemperator components are good steps.Tubes experiencing short-term overheat tend to exhibit a significant amount of bowing and swelling. Look for new bowed tubes or check for tube swelling in concern areas. Lone bowed tubes in a superheater or evaporator can be an indicator of foreign object blocking flow, especially if sections of bare tube are noted to have a swollen diameter. Preventing failure: Any object dropped into a downcomer or riser needs to be retrieved prior to start-up, especially one large enough to fully or partially block a tube.To prevent workers dropping items into the downcomer, sites could install an expanded metal cover over the down-

Figure 1. FAC tube failure at the top of Figure 2. Close-up view of tube failure an HRSG LP Evaporator tube. area in previous photo.

28 ENERGY-TECH.com

Figure 3. Borescope revealed tube inlet was 100% blocked

Figure 4. Culprit was found to be an aerosol can dropped in the downcomer by a contractor.

comer and vortex breaker. HRSG attemperators and their liners should be regularly inspected by pulling the probes and borescoping inside the pipe to check the liner conditions. Missing parts or tools in steam drums should be investigated thoroughly if they are at risk to blocking a tube opening.

Flow accelerated corrosion Flow accelerated corrosion (FAC) is a damage mechanism that causes metal loss and failure in water and two-phase piping, HRSG tubes, steam drums, and other components composed of carbon steel. It is one of the top two most prevalent causes of pressure part damage in HRSGs (per EPRI Heat Recovery Steam Generator Tube Failure Manual dated May 2002). Cause of failure: Flow accelerated corrosion (FAC) is a material wastage process where the ID surface of the metal continually forms as a magnetite layer that is then persistently eroded and corroded away. FAC typically affects carbon steel, the most prevalent material used in LP and IP systems.This does not make the use of carbon steel bad practice, as there are numerous factors outside of material that affect FAC: • Temperature: FACis most common when boiler water temperatures are 200°F and 400°F. • Pipe and tube geometry: Impingement zones such as elbows and tees are more suspect to FAC • Flow velocity: Higher water velocities can promote FAC • Chemistry: Low pH and dissolved oxygen are correlated to FAC. • Water: Liquid water is a requirement for FAC. Steam-only piping does not experience FAC in HRSGs.

Figure 3. Red arrows point to tube wall thinning from FAC. Yellow arrow points to typical “orange peel surface” indicative of FAC.

November 2017

Cogeneration biomass

waste to watts

Boiler Replacement Parts

toll free 1.877.774.8778 24hr emergency service

229.221.4690

November 2017 ENERGY-TECH.com

NEW TECH APPLIED TECH The above contributors all work in combination. For example, if the velocity (scrubbing action) is very high, then FAC can occur with almost perfect water chemistry. Conversely, if the chemistry exhibits pH below 9.2 or levels of dissolved oxygen below 5 ppb, then FAC can occur in areas with moderate velocity. Understanding the flow pattern and chemistry in various HRSG components can allow predictive lists to be generated on where FAC might strike.

How to spot the issue before failure occurs: Understanding the flow pattern and chemistry in various HRSG components can allow predictive lists to be generated on where FAC might strike.Without such a list, the best way to detect FAC is routine UT testing and borescope inspection on LP Evaporator and IP Evaporator components, especially the belly pans, upper tube bends, and riser bends. Borescoping can find the distinctive FAC indications visually and UT can identify thinning.The best practice is to use the borescope findings to identify areas for UT. FAC is not limited to LP and IP Evaporators, and any economizer piping that falls into the temperature range should be tested on elbows and tube bends. Preventing failure: Preventing FAC is a difficult undertaking, with many new ideas still being attempted and discovered within the industry.There is nothing to be done about the temperature profile, but it is used to guide effort within the HRSG.When damage is found and pressure part replacement is needed, upgrading the tubes, headers and piping from carbon steel to steel containing chrome, is an effective way to reduce or solve the problem. Striving to maintain water chemistry

30 ENERGY-TECH.com

pH in LP and IP systems above 9.6, with a dissolved oxygen range between 5-10 ppb reduces the risk of FAC. Diligent inspecting for FAC should still be undertaken even with good chemistry control, with an engineering assessment guiding the locations being monitored.

Conclusion There is no way to prevent 100% of the problems that an HRSG will experience. However, knowledge of failure mechanisms and the ways they manifest themselves can help a plant be prepared to tackle the issues in scheduled outages rather than lose operating time to a forced outage. Understanding why and where these failures occur is the first step in finding and fixing them before they become a problem. â&#x2013; Jack Odlum is a Systems Engineer at HRST, Inc. His responsibilities include on-site HRSG inspections, Roof Cause Analysis of material failures, analysis projects (FAC Risk Assessments and Performance Studies), Technical Advisory service, and training. He received his B.S. degree from University of California, Berkeley in Mechanical Engineering. Jack works out of the HRST Headquarters in Walnut Creek, CA. You may contact him by emailing editorial@WoodwardBizMedia.com

November 2017

Remember in the movies when the Cavalry came to your rescue, just in the nick of time? We got your back like that. • Staff augmentation - skilled, trained, and experienced • Strict compliance with site and operational safety • Precision in-place machining and code welding • OEM support ensures optimal equipment for your job • Project planning and management assistance • Engineering support and custom equipment mfg.

Rapid response you can depend on, and you’ll like our charges.

Give us a call...

(916) 288-6100 (888) 345-5015

Regional Service Offices across the U.S. • www.tritool.com

November • Advertisers’ Index A-T Controls, Inc. Gradient Lens Corporation

Back Cover 8

Hexeco 31 HRST, Inc.

Hurst Boiler & Welding Company

National Compressor Services

Proton Onsite

Rentech Boiler Systems, Inc.

Rotork Controls, Inc.

Schenck Balancing & Diagnostic Systems

Siemens Energy Inc.

Skinner Power Systems

Sohre Turbomachinery Inc.

18, 26

Topog-E Gasket Company

Tri Tool Inc.

True North Consulting

November 2017 ENERGY-TECH.com

The leader in Process Ball Valves and Automation now offers a complete range of Butterfly Valves for Power Generation and Geothermal industries!

AT Controls offers resilient seat, PFA-lined, high performance, metal seat, double flanged, fire safe, alloy and high temperature butterfly valves. We have the right valve for condensate, compressed air, flue gas, steam, fly ash and numerous water services in todayâ&#x20AC;&#x2122;s power generation industries. We feature substantial inventories, technical expertise and comprehensive TRIAC pneumatic and electric valve automation systems.

9955 International Boulevard Cincinnati, Ohio 45246 (513) 247-5465 FAX (513) 247-5462 sales@atcontrols.com www.a-tcontrols.com