Improving progressing cavity pumps to increase MTBR

When initially setting up a vibration monitoring program for progressing cavity (PC) pumps at my facility, we used the overall vibration analysis parameter alongside ISO severity charts, which included higher vibration limits due to the field-expedient installation methods commonly employed in the oil and gas industry. However, - the overall vibration analysis parameter quickly proved to be ineffective for monitoring the pumps, prompting us to seek a better indicator for assessing their operational health.

The first part of this article presented a novel approach to assessing the health of a progressing cavity pump through vibration analysis, highlighting the limitations of traditional overall vibration measurements. This next part reviews the improvements we have made to our progressing cavity pumps to increase the mean time between repairs (MTBR), reduce repair costs, minimize lost production, and enhance safety.

Equipment setup

My plant uses progressing cavity (PC) pumps for pipeline applications involving both water and oil, in both tanked and tankless systems. In tanked systems, the suction pressure ranges from 1 to 2 psi, while tankless systems typically operate around 100 psi. The pump's discharge pressure ranges from 100 to 165 psi. The pumps are driven by a variable frequency drive (VFD), with speed ranges of 55-440 rpm or 63-339 rpm, depending on the system. The VFD speed is adjusted based on the suction vessel's level.
The most common failure mode in the PC pumps at this plant is loss of flow and pressure, typically caused by stator failure. Selecting the right stator material is essential, as it must be compatible with the type of fluid being pumped and its chemical properties and the operating pressures, temperatures. Choosing the wrong stator material can drastically shorten its lifespan.

Another common failure mode of the PC pumps at this plant is a failure within the connecting rod assembly. The pump uses a connecting rod with a pin-to-joint connection to link the rotor to the drive shaft. These pin-to-joint connections were failing, caused by the multiple starts resulting in rapid and excessive wear on both connections. (Part 1 of this article describes how my team came to understand these two major failure modes, and why vibration data collection was crucial in determining how to use this data to predict the equipment's health.)

Pump design improvement 1: stator material

PC pumps are the main source for hydraulically moving oil from our facility to the customer, so each pump failure constitutes a big loss in production. Analysis of the failure data for the past two years indicates that most of the premature pump failures due to stator damage. Multiple pumps at different tankless facilities were disassembled to study various parts of the pump and to eliminate any site-specific bias. The stator was pinpointed as the most frequent/highest damaged part that resulted in a complete PC pump failure. Figure 1 shows hysteresis damage to the stator.

Hysteresis damage results from excessive temperature rise of the stator elastomer, which causes the material to swell and reduces the clearance between the stator and rotor, causing rubbing. Several factors contribute to the extent of temperature rise and swelling of the elastomers. These include pump rotation frequency, discharge pressure, chamber size (clearance between stator and rotor), wall thickness of the rubber, oil temperature, chemical attacks, and stator materials. 

The oil temperature did not exceed 150°F across the field, all the pumps were spinning at recommended speed by the manufacturer, and no chemical attack was known to exist in the past, which suggested that the problem was not common. Since there was not a satisfactory answer to the second “Why” question to narrow down the root cause for stator premature failure, we decided to take a step back and review the material specifications of the existing stator design (see Figure 2).

The default stator design for Oxy has always been 61T hydrogenated acrylonitrile butadiene rubber (HNBR). This material provides a cost-effective solution with a maximum temperature rating of 140°C/284°F,  well above the required temperature for operating these pumps. However, since the failures keep recurring very frequently at the worst possible times (during highest production flow), we decided to review different stator materials to see if a more durable material would fit our needs. Table 1 shows other material ratings considered before we landed on a high fluoro FKM (HFKM) and AFLAS material for their temperature and abrasion rating.

Both AFLAS and HFKM show high temperature ratings and chemical resistance. The two materials were tested in the lab with produced water at 95°C/203°F, 4-5% sand, heavy crude oil, and some gaseous H2S present. The physical results are captured by the manufacturer in Figure 4. HFKM has a 69.5% fluorine content compared to 54% in AFLAS. The fluorine content is completely bound so there are no dangers with handling, release, or disposal. 

Due to its lower fluorine content, AFLAS is more suitable for field use with a high concentration of wet H₂S, which will prevent cracking and hardening of the elastomer. Wet H₂S can be absorbed into standard Viton rubber (a brand of FKM) under high temperature and pressure. After cooling, the H₂S releases from the elastomer, causing problems. However, the concentration of H₂S is minimal at this plant, so there is no observable advantage of one material over the other. HFKM was considered more durable for our oil field due to its sturdier construction and has a shorter lead time.

During peak production, a trial was set up with a thermocouple inserted into the stator to measure the actual stator temperature when in use. Two PC pumps on pads were set up with new HFKM stators, while the four existing pumps were using HNBR stators. Within the first two months of high oil production, one of the existing HNBR stator failed and was replaced twice within a two-month period of high oil flow. None of the HFKM stators failed. 
Eventually, all the stators on pads were swapped out to HFKM and the material performed flawlessly through the period of high oil flow. The first pump stator was disassembled afterward for inspection and only minimal damage was found. The preliminary results are very positive for fieldwide implementation of this new HFKM stator material. More trials were run to understand the effect of temperature on the new stator material with different rotor designs, aiming to extend the lifespan of the PC pump. The results are outlined in the section below.

Pump design improvement 2: stator temperature and rotor design

Another factor contributing to early stator failure is the operating temperature of the stator assembly. Failures tend to occur soon after a facility experiences an increase in production, which lead to a higher daily runtime. It was suspected that the stator could potentially be exposed to temperatures above the rotor undercut temperature of 140°F (60°C). The rotor undercut temperature defines the amount of clearance needed between the rotor and stator to ensure that they do not contact each other as the stator swells a certain amount at due to temperature. This suspicion was based on the manufacturer’s root cause failure analysis showing the stator had experienced “high point” failures (see Figure 4) that are caused by swelling of the stator. 

To test the theory of higher run times and temperatures, a series of controlled tests was performed to understand the relationship between pump speed (Hz) and stator temperature. A resistance temperature detector (RTD) was installed in a drillout within the stator to measure the stator temperature. Typically, the pump is allowed to operate at speeds of up to 60 Hz, per the manufacturer’s recommendation. Testing began at a low speed of 35 Hz with plans to run the pump continuously for a period of time before increasing the pump speed incrementally. 

Trend 1 (Figure 5) shows the temperature trend while running at 35 Hz. The interesting finding is that, after the pump initially ramped up in speed, the temperature eventually increased and stabilized at 144°F. This result was surprising due to the fact that the stator temperature was already running at the limits of the rotor undercut. The temperature did fluctuate +/-10°F due to ambient conditions.

Increasing the speed to 40 Hz resulted in an average temperature of 152°F (see Trend 2, Figure 6). At this point, the stator was expected to swell and rub against the rotor, causing the elevated temperatures. 

The pump speed was increased to 45 Hz. This time, the average stator temperature rose to 164°F, which is well above the acceptable range (see Trend 3, Figure 7). 

For the final test, the pump speed was increased to 50 Hz, resulting in an average stator temperature of 194°F (see Trend 4, Figure 8). At this point, the trial was stopped to avoid damage to the pump.

Based on the data gathered, it is clear that the current stator and rotor design was unable to operate within an acceptable temperature range with the current rotor undercut (140°F). As a result, an additional trial was completed with a 176°F (80°C) undercut to understand if this would reduce swelling in the stator.  

With the 176°F (80°C) undercut rotor installed, the trial was started at 40 Hz and stepped up in 5 Hz increments to 60 Hz. Results showed that the stator temperature averaged 115°F while running at 40 Hz speed, which was within the acceptable temperature range (see Trend 5, Figure 9). 

As the speed was increased on the pump, a significant temperature increase was not observed. In fact, temperatures averaged 115°F even at 60 Hz (see Trend 6, Figure 10). The pump was left to run continuously at 60 Hz for a period of 6 days, after which the stator was inspected. The inspection revealed the stator had very little to no damage. 

The increased clearance between the stator and rotor did reduce failure frequency of the stator material in facilities with high production. However, the increased clearance from the larger undercut allowed more oil slippage, resulting in a reduced flow rate efficiency on the pump. At 60 Hz, the pump with the 176°F (80°C) undercut rotor was found to have a 9% lower throughput (5,500 BOPD) compared to the 140°F (60°C) undercut rotor (6,000 BOPD). 

Pump design improvement 3: pump pressure rating and PRV setpoint

Besides investigating the stator and rotor design, the team also looked at other failure mechanisms to improve pump design. Another source of catastrophic failure has been exceeding the pump differential pressure rating of 295 PSIG. In these scenarios, the pump experienced a discharge pressure of higher than 700 PSIG, far above the 295 PSIG differential pressure rating (see Trend 7, Figure 11). The pump would immediately fail by being unable to build any pressure even when the motor was spinning.  

During outage events for midstream, the sales skid would slam shut and the pressure would spike. Although the pump had an automated shutdown at 450 PSIG, it was discovered that the pump was still exposed to 700+ PSIG of pressure, at which point the pressure relief valve (PRV) would relieve pressure. Although the PRV was installed to protect the piping at 750 PSIG MAOP, the PRV was also serving as a stopgap for protecting the pump.