KCF Technologies Blog

Midlife Crises Now Under Way for 30-plus-year-old Generators with Hydrogen-Inner-Cooled Stators

Authored by professional engineers from Alstom Power and Portland General Electric, an article with the optimistically promising title "A permanent solution to generator vibration problems" appeared in the April 2006 issue of Power Magazine, the regularly published online journal of the Electric Power Conference and Exhibition.

It concerned upgrading the high-pressure turbine at PGE Boardman, a large coal-fired power plant producing 15 percent of the electricity in Oregon, in advance of a major refit to boost its output and run Boardman as a baseload unit.  Like many other coal-fired plants of the last 40 years, Boardman "has generators whose stators are directly inner-cooled by hydrogen.  With three decades of experience operating these generators now under its belt, the industry has identified several fleetwide chronic maintenance problems that can result in extended, unplanned outages and significant lost-generation and repair costs."

In 2004, PGE realized their planned upgrade also gave them the opportunity of "permanently solving...four known generator problems.  The solution would be costly, bu the payback would be significant--long-term, reliable operation of the plant at a much higher output."  Those problems included cracking of turbine retaining rings and rotor top teeth, loosening of stator core laminations, and high stator endwinding vibration levels that required significant annual downtime for inspections.  Even at 40 percent output, "the unit experienced relatively high endwinding vibrations, on the order of 10 mils [hundredths of an inch]."

During the overhaul, Alstom Power Inc. rectified each of these.  To address the vibration issue, "Alstom designed a free-floating support system...[with] multiple pressing plates equally spaced around the outside of the winding head," which, "creates a very cohesive endwinding basket structure that is resistant to vibration."  As a result, tests after the turbine overhaul had less than 25 percent of the vibration of pre-overhaul levels.  Significantly, this was verified by an "endwinding vibration monitoring system," enabling technicians to keep an eye on that aspect of turbine wear in the furture, too.

Disco was King--and Condition-Based Vibration Monitoring had Barely Begun

Archived online in PDF format, "Machine-Condition Monitoring Using Vibration Analysis--A Case Study from Alma Paper Mill, Quebec, Canada" is an "application note" about an "Abitibi Price" paper mill from European multinational firm Brüel & Kjær Sound and Vibration Measurement.  The report, dating between 1974 (when Abitibi Paper acquired Price Brothers & Co.) and 1979 (when the company name acquired a hyphen as "Abitibi-Price"), offers a glimpse of the origins of vibration monitoring in condition-based maintenance in the paper industry nearly 40 years ago.

If you download the report, be sure to note the photo at the top of page 11.  The caption notes that the firm's "Type 9613 Vibration Monitoring Set, comprising Vibration Meter, Stroboscope and Headphones, provides the tools for a fast, easy check of machine condition...using the ear as an effective frequency analyzer."  One wonders if the bespectacled worker with the notebook and scattered tools, squatting at one end of the massive multi-ton paper press, regarded his task as a "fast, easy check."

Heat Wave Came with Cascade of Chiller Failures at the University of Wisconsin

In July 2011, Wisconsin suffered a four-day heat wave.  Heat indexes ran as high as 117 degrees.  Among those feeling the heat from the first day on was the University of Wisconsin at Madison.  Two of its large chillers already were under repair when Madison Gas and Electric "experienced a failure with one of its chillers at the West Campus Co-generation Facility."

The result was a campus-wide shortage of chilled water used for air conditioning.  What little remained went to critical operations at the hospital, servers, animal labs, and experimental facilities, while maintenance staff sweated night and day all week to make the needed repairs.  The MG&E chiller was restored within 24 hours, and one day later cool water began to trickle to to 20 of the 47 buildings that had been without it.  But it would be two more days--and the end of the heat wave--before the need for chilled water was again fully met.  By then, many university classes and events had been canceled.

Murphy's Law pays no attention to the clock or the calendar.  Periods of peak need are precisely when your critical systems frequently fail.  Predictive maintenance, which helps you schedule your repairs and minimize the downtime for HVAC, and other key infrastructure, can be a powerful ally in the maintenance specialist's eternal struggle to keep "anything that can go wrong" from wrong at the worst possible moment.

(Sources: "UW only cooling essential buildings; chillers under repair," Wisconsin State Journal, July 18, 2011; and "Air conditioning reported at normal levels," University of Wisconsin - Madison News, July 22, 2011.)

SmartDiagnostics® Feature Higlight: Alarms and Warnings

SmartDiagnostics® VMS viewer showing warnings and alarms.
A key value of the SmartDiagnostics® Vibration Monitoring System (VMS) is that it facilitates exception-based monitoring of machine condition.  If you are implementing a facility-wide condition-based maintenance program, you may have a large number of machines to monitor on a regular basis.  Configure and apply monitoring bands with warning and alarm thresholds and the system can do much of that daily basic monitoring for you.  When VMS detects a machine's vibration level exceeding one of the thresholds, it automatically tells you of the event so that you can take action.

Navigation tree close-up.
VMS indicates alert conditions in several ways to allow you to quickly, and easily, navigate to the machine that is having a problem and to see when and why the alert condition was raised.  In addition to it's system-wide alert log, VMS presents a dynamic alert icons on the navigation tree (see image at left) an alert bar on the trend chart (see above image), monitoring bands alert thresholds on the vibration data chart, and an alarm log for every monitoring location.

Let's look at dynamic alert icons.  Any time a vibration sample exceeds an alarm, or warning, threshold in one or more monitoring bands, an alert is generated and presented as an alarm icon, or warning icon on the monitoring band, monitoring location, machine and facility levels in the navigation tree.  If there are multiple alerts triggered at the same level of the navigation tree, the higher level will propagate up the tree to the next level.

The alert icons are dynamic and only show the alert level of the most recent vibration sample.  As such, if the machine temporarily exceeds a threshold vibration level and then returns to normal operation, the alert icon will show up briefly and then go away.  In this way, the alert icons on the navigation tree always give you a snap shot of the current state of your monitored equipment.

Predictive Maintenace Tip: Making a Condition-Based Case for New Equipment

One practical problem with introducing predictive maintenance is that sooner or later it requires the maintenance engineer, or manager, to ask the plant manager, company president, or board to approve costly repairs to, or replace, an expensive piece of critical equipment that shows only the first subtle signs of impending failure.  Decision-makers faced with smoldering wreckage don't hesitate to fund its replacement.  However, especially with today's tight budgets, it's a harder sell to procure the money to replace a big piece of costly gear that appears--for the moment at least--to be perfectly fine.

This predictive maintenance tip comes from Lindsay Audin.  An energy consultant, customer, and supplier with over 30 years' experience, Audin is the president of Croton, New York-based energy consulting firm EnergyWiz, and a contributing editor for Building Operating Management.  Confronting this very problem in an excellent, long article in FacilitiesNet®, Audin notes that, "trying to scare management into action...may not be effective."

"When asked how they have successfully made their cases, three basic themes emerged from interviews with facility executives, operating staff, and consultants."
  1. "You won't get to first base without hard information on the condition of a piece of equipment."
  2. "Never go into a boardroom with only one reason to replace; two is a minimum."
  3. Every supporting argument should relate back to revenue, cost, value, or a combination of those factors."
Here are some key criteria Audin says you can use to persuade skeptical decision-makers:
  • Energy efficiency: How much could be saved with a new unit?
  • Code compliance:  Is the unit causing a code violation (i.e., health, fire, building)?
  • Reliability: Does it conk out often enough to irritate staff and students?
  • Capacity: Have loads or output changed to a point where it's not doing the job?
  • Liability: Could a failure cause a lawsuit, insurance claim, or other damage?
  • Appearance: If the system is visible and ugly, would a new unit look better?
He concludes, "At least two of these points, each with supporting documentation, should be used when presenting a case," for major HVAC repair or replacement.

(Source: "Selling the CEO on HVAC Upgrades," Lindsay Audin, FacilitiesNet®, Milwaukee, WI, January 2009, ©Trade Press Media Group, Inc.)

Vibration Monitoring Gives Impressive ROI at Tampa WWTP

Career Vibration Analysts Tom LaRocque, Gary Kaiser, and Joe Spencer co-authored "Continuous Vibration Monitoring of Wastewater Pump Stations" in the August 2008 online monthly trade journal Pumps & Systems "the voice of the pump and rotating equipment industry," with "more than 42,500 readers across the globe."

"The City of  Tampa's Howard F. Curren Wastewater Plant uses vibration analysis hardware and process controller equipment to protect critical machinery against damage from mechanical failures or environmental changes, ensure survivability, and prevent unscheduled downtime and costs.  This system uses relays to trigger alarms, or shutdowns, and is integrated to the main plant's Supervisory Control and Data Acquisition (SCADA) system.

The plant is "a state-of-the-art facility that treats all wastewater discharged from approximately 100,000 accounts in the City of Tampa system.  The plant has a permitted capacity of 96 [million gallons a day], with an average daily flow of 60mgd."  The stakes are high; "Pump failures can often be damaging to the pumps and auxiliary equipment.  Moreover, the cost of a new pump motor can be as high as $450,000, and the cost to repair an existing unit can approach $175,000 after a catasrophic failure."

However, "A protection system that monitors the vibration levels and can be integrated to a shutdown circuit minimizes flow interruptions and the amount of damage to that equipment...The Wastewater Department installed on unit as a trial on a large (700 hp) pump and motor combination at a major pump station.  Vibrations were detected and repairs were made for less than $500 that saved damage to the expensive pump and motor.  Plans are in place to install monitors at all major pump stations over the next two years."

"Periodic monitoring might be sufficient to identify general, long-term machinery conditions, but to capture transient conditions that can cause catastrophic failures the team determined that continual monitoring was required.  Given the unmanned pump stations, an integrated system that could alert a technician at the plant of an issue with the pump station equipment."

The report gives serious consideration to how each of the hardware components and locations in Tampa's wastewater monitoring system were carefully selected.

"Comparing the cost of repairing a pump station with with an 800-hp motor ($175,000) with the price of a typical monitoring system (a two-channel system is approximately $2,500, or approximately $1,500 per measuring point) justified the project.  The initial approval to outfit one major lift station was decided in 2006, and a unit has been in service since then.  Another pump motor failure costing an estimated $160,000 further justified the project and renewed interest in a relatively low-cost 24-hour protective device."

The study concludes with a list of "factors...critical in convincing management of the benefits of vibration monitoring to the predictive maintenance program and the need to expland the program to other pump stations."  Leading the list is this: "Cost of the [monitoring] equipment is much less than the cost of repair or replacement of pump and motor."

Click for a Virtual tour of the Howard F. Curran Advanced Wastewater Treatment Plant.

KCF Making the Things You Work with Smarter