Unexplained Centrifugal Pump Failures And Solutions

What is this Twilight?

I'm talking about the distorted period between the time the pump is shipped from the factory and the subsequent start-up; I'm talking about the time and place between pump disassembly for servicing and reinstallation; I'm talking about the unexplained changes that occur when pump "A" is removed from the system and pump "B" is installed in place.

This is the transitional dimension of what I previously called the Land of Magic and mystery. Recently, I have referred to this dimension as "Twilight."

1. Absence of impeller

Problem

At a steel mill in the central Midwest, a six-level boiler feed pump was stopped for routine maintenance to restore the original operating clearance. Note that the pump is driven by a multistage steam turbine. After more than eight years of continuous use, the pump worked well, but it was time for an overhaul.
The OEM service center repaired the pump and returned it to the customer. When first started, the pump ran smoothly, but the hydraulic performance decreased significantly. Both flow and head decreased compared to the performance parameters published by the manufacturer. The distraught customer claimed that the repair center missed several stages of the impeller that were not installed during the repair process. The pump was originally designed as an eight-stage pump, but was designed as a six-stage pump for this customer. It is common practice for the impeller to use a no-stage design when a lower outlet pressure is required. The customer believes that the current pump should be a four-stage pump, rather than a six-stage pump.
The manufacturer's engineers were called to the site to fix the problem. He arrived on site and performed all the normal performance checks, including vibration, valve configuration, rotor setup, and verification of differential pressure and flow. From the appearance, it seems that there is a real possibility that the two-stage impeller is not installed, but this is unlikely, and the OEM repair center strongly insists that the six-stage impeller is installed.
From the performance indicators, it seems to be the problem of low speed. The customer was informed of several key points in the troubleshooting process, including that they did not use the steam turbine during the downtime and that the first thing they checked when the problem arose was the operating speed. The turbine/pump speed has been verified using a calibrated stroboscopic tachometer (stroboscope) at an accurate speed of 3,600 rpm. Luckily, the young pump engineer had a lot of experience with steam turbines and stroboscopes.

Solution: The impeller is not missing

In the first case, the pump plant engineer has extensive experience in steam turbines, steam governors and stroboscopes. He knew how easy it was to get a false reading from a strobe if you weren't properly trained. The steel mill strobe operator focused his attention on a coupling bolt to measure the rotor's speed, which he thought was 3,600 rpm because he had no reason to be suspicious. The turbine actually runs at 1,800 rpm, and what the strobe operator sees is multiple images (two coupling bolts are shown as one).
For strobe work, it is important to always capture a unique object, such as a key or keyway, from an axial circumference Angle. If a coupling bolt must be used, then one should be marked in some way to make it different from the other bolts (in these cases, luminous paint/reflective paste works well).
The customer had not previously mentioned that the turbine speed-regulating valve (W-TG-10) had been modified during the pump overhaul. The plant engineer who discovered the root cause had the honor of adjusting the turbine speedegator valve Settings to bring the equipment to the correct speed of 3,600 rpm and return its performance to its original state.

2. Overheating of the transformer problem

Scene: A power plant on the East Coast. In one 20-year-old power plant, a generator booster transformer (GSU) was severely overheating at full capacity from the day it was installed. As a result, the unit derated and operated at 85% full capacity. Since then, derated transformers have limited the output of the entire power station.
Fast forward 20 years, and heat-resistant transformers are planned to use new dielectric oils and brand new washers. In addition, all six transformer oil circulating pumps will be retrofitted (note: Power transformers use a circulating transformer oil system as an insulating dielectric and a heat transfer cooling medium).
These pumps are expedited to the plant repair shop for modification and returned to the site. The customer checked the rotation direction of the pump before installation and found that the rotation direction of the pump was wrong. The customer asked the factory to provide warranty compensation to solve the problem. Plant engineers said that unless the customer had reversed two of the three phases on the power line in the field, the pumps could not have operated in the wrong direction. The customer said that no wiring was changed on site.

Solution: Cooling solution

Plant engineers are confident that transformer pump motors are wired correctly because they have internal quality assurance/quality control (QA/QC) procedures. After arriving at the work site, to the surprise of the engineers, it was confirmed that the direction of rotation of the pump was indeed wrong. In fact, the pump has been rotating in the wrong direction for 20 years.
Note: Centrifugal pumps operating in the wrong direction still produce flow in the right direction, and the customer's basic mirror flow indicator cannot distinguish between 600 gpm and 1200 gpm flow. After the rotation direction of all pumps is corrected, the power plant is started and the transformer is energized. For the first time in 20 years, the transformer did not overheat at full load.