Some 20-years ago, I lived and traveled the Caribbean islands with my pump rebuild facility based in Puerto Rico. An old customer, a petrochemical plant, contacted me recently and asked me to return to look at some sick pumps.
I thought I’d see some of my old drinking buddies in the maintenance shop. As I signed-in at the guard shack, I imagined ending the day sipping rum with coconut water in the bar around the corner. I was dismayed when I arrived at the engineering office – everyone had changed. The people I knew had retired or left for some reason. The new engineers all had smooth, clean faces. I don’t like drinking alone.
However, the pumps had not changed, although they had been painted a different color. The cooling tower was different, and they had installed a desalination plant for industrial water from the ocean.
I recognized many of the pumps and seals I had installed 12 and 18 years ago. The plant had 870 total pumps within the perimeter fence, of which about 200 pumps were from the same manufacturer in three different models (small, medium and large). With 200 mostly identical pumps, the shop mechanics were familiar with the rebuild procedures, including the specs on bearing and seal installation. The seals were “easy-to-install” cartridge models in three sizes for the 13⁄8”, 17⁄8”, and 25⁄8” pump shafts.
Twenty-two of these pumps were experiencing elevated maintenance. Two pumps suffered runaway maintenance, with constant seal and bearing failures, which led to my visit. We started the day in the company conference room.
The plant engineer said these “Bad Actors” were critical to production. Thus, the pumps were installed side-by-side into the same suction and discharge piping system. We put the vibration statistics on the wall with a projector. The vibrations rose and fell over time in an erratic pattern. The technician’s report labeled most of the vibrations as “undetermined.”
We looked at the maintenance history on the pumps. There were periods when the pumps would eat a couple of mechanical seals each week. These periods paralleled the high vibrations. Mysteriously, there were also times when the mechanical seals would run for months between failures.
As we walked to the pumps, I asked, “Do you alternate these pumps?” The engineer said, “Yes. One pump of the pair always operates. When it fails, we send the power end to the shop and operate the other pump. We rotate again to the first pump when the second pump leaks.”
I asked, “Do you ever run these two pumps together?” The engineer said, “Yeah, when we get a big order from a customer, and when we ship to a sister plant.” I asked, “Are these alternating pumps or parallel pumps?” The engineer asked me, “What’s the difference?”
People, there is a big difference! A century ago, the philosopher Santayana wrote, “Those who cannot learn from history are doomed to repeat it.” Our ancestors could walk into the forest to gather food and medicines. Our ancestors could distinguish edible plants and leaves from poisonous vegetation. Today, TV commercials and maybe the produce manager at the supermarket tell us what we can eat. We’d starve if we had to gather our food in the forest because we’ve lost our history.
Our ancestors built homes in harmony with the weather, the seasons, the sun, and the wind. Today’s houses are out of sync with the climate. We use electrical gadgets to combat the cold and the heat because we didn’t learn from history.
We learned nothing from the energy crisis of the early 1970s, when the price of gasoline doubled to 50-cents per gallon. President Carter installed solar collecting panels on the roof of the White House and urged us to conserve energy.
President Reagan related conservation to deprivation saying, “It’s un-capitalistic for Americans to suffer.” Reagan promptly removed the panels. Now in 2011, we’re still using gasoline powered cars as the price approaches $6-USD a gallon. We didn’t learn the lessons of history.
I began my career in maintenance at a steel mill in Alabama. I distinctly remember my mentors and teachers. I also had mentors when I served in the Navy. I had mentors at the university, and when I began my formal work career.
About 25-years ago, many of the world’s great industrial companies did away with “mentoring” as a teaching tool. The older workers were fired or pushed into early retirement. The young workers lost their mentors and teachers.
Today, a young process engineer sees two pumps mounted side-by-side in a piping system, and he can’t determine if the pumps are “parallel” or “alternate.” He guesses wrong. With no mentor, the young engineer assumes he inherited a couple of “Bad Actor” pumps.
Two (or more) pumps mounted side-by-side with common suction and discharge piping may be “Parallel Pumps,” or they may be “Alternate Pumps.” They appear similar, but they are not.
Think of it this way: A jug of sweet tea appears similar to a jug of gasoline. However, sweet tea and gasoline are two different products for two different purposes. The two should not be confused. Likewise, a piping system for parallel pumps appears similar to a piping system for alternating pumps. The two should not be confused.
Parallel pumps and alternate pumps are two different devices or tools. Each has its application in the chemical process plant. The process engineer must know when, and how, to use each tool. The two tools should not be confused.
In simple terms, parallel pumps can operate together at the same time, or with operator training, they can operate independently in the piping system. Alternating pumps can only operate independently in the system, but not together at the same time. When alternating pumps run together at the same time in the system, the reliability engineer will perceive he has two “Bad Actor Pumps.”
The difference is in the way the system curve mates with the pump curves. This is the lost or forgotten information.
Forty years ago, an engineer would have developed his system curve first, and mated it to the pump curve. This technique would show the engineer what the pump is doing. With no mentor or foundation, today’s reliability engineer starts with a vibration meter.
The Pump Guy advises, “Start with the system curve (Figure 2) to analyze a pump problem.” The system curve is the graphical picture of the energy (resistance) contained in a piping system. It is the resistance the pump must overcome. The next step is to impose the side-by-side pumps onto the system curve.
In a parallel pump system (Figure 3), the system curve mates with the pump curves at best efficiency for both (or all) pumps running together.
In an alternating pump system (Figure 4), the system curve mates with either pump A or pump B operating alone at best efficiency.
Both (or all) pumps will be stressed (Figure 5), when pumps “A” and “B” operate together in an alternating system. The system throttles both (or all) pumps. The pumps operate at the combined “X” flow and head, the intersection of the system curve with the pump curves.
Figure 5 shows that neither pump operates at the preferred design or best efficiency point. The system drags both pumps away from their preferred operating zones. Both pumps suffer mysterious vibrations, internal recirculation, shaft deflection and stress. It is akin to a traffic jam forming as five lanes of cars try to squeeze onto a two-lane highway.
These vibrations and stress produce the runaway bearing and seal failures. The reliability engineer, with no mentor or history mistakenly perceives these pumps are “Bad Actors.”
We returned to the plant conference room and the air conditioning. The high vibrations paralleled the periods of elevated production when both pumps ran together at the same time.
The engineer said, “Larry, you solved the problem. Those are not parallel pumps. We’ve been operating alternate pumps together.” The engineer called his secretary and told her to contact me for a return visit.
We said goodbye, and I gathered my stuff to leave The engineer said, “We get off at four. There’s a bar around the corner. Have you ever sipped rum-with-coconut-water?”
System curves are an important, though mostly forgotten reliability tool with industrial pumps. In the Flow Control Pump Guy Seminar Series, we dedicate a lot of time to system curves and their contribution to extended uptime. Come if you can. We’ll bring you up to date. For our upcoming training schedule, visit FlowControlNetwork.com/events.asp.
Larry Bachus, founder of pump services firm Bachus Company Inc., is a regular contributor to Flow Control magazine. He is a pump consultant, lecturer, and inventor based in Nashville, Tenn. Mr. Bachus is a retired member of ASME and lectures in both English and Spanish. He can be reached at firstname.lastname@example.org.