These are excerpts from a paper titled, ‘Revamp/re-rate design considerations’ presented at the 43rd Turbomachinery & 30th Pump Users Symposia (Pump & Turbo 2014) held from September 23 to 25, 2014 at Houston, TX.
As compressor users look to maximize the production of their facilities or to reduce their operating costs, they often face choices whether to purchase new equipment or to upgrade their existing turbomachinery. Given the cost to build an entire new compressor train and its related support systems, it can, at times, be far more cost effective to refurbish existing equipment. However, there are limitations to what can be achieved via the so-called “revamp” or “re-rate.”
Revamping, re-rating or upgrading centrifugal compressors have been common practices in the refining, chemical, petrochemical, gas pipeline, and oil & gas production industries. Given the current economic climate, there is increasing interest in revamping or upgrading existing equipment simply to reduce expenses. Further, the required payback period (the number of years needed for a project to turn profitable) has become increasingly short as competitive pressures increase and concerns arise over political uncertainties at some plant locations. Thus, the need to maximize the capability of existing rotating equipment is important.
Unfortunately, there are limits to what can be accomplished within a given compressor. That is, all compressors are not created equal. Numerous factors will determine the level of performance modifications that can be made. For example, if the goal of a revamp is to increase the operating capacity, one must determine whether the existing flange size on the case can accommodate the higher flow rates without introducing unacceptably high losses. Further, if the inlet flanges can pass the flow, will it be possible to install an impeller that is large enough to accept the incoming flow or will the compressor rotordynamics be acceptable with the new impeller line-up? Moreover, will the Mach numbers, flow range, efficiency, impeller stress levels, etc. be acceptable to the end user?
The terminology used by OEMs varies somewhat when it comes to revamping, re-rating or upgrading compression equipment. For clarity, in this tutorial, these terms will be defined as follows: Revamp / Re-rate –These terms will be used interchangeably. That is, there is no functional difference between revamping and re-rating compressor. The process involves changing the performance characteristics (aerodynamic and/or mechanical) of a compressor. The changes might involve increasing or decreasing the flow rate, increasing or decreasing the operating speed, increasing or decreasing the inlet and/or discharge pressure, etc. These will be accomplished by changing at least some, but possibly not all, of the compressor internal components. At the end of the revamp or re-rate, new performance curves are typically supplied and new operating requirements are specified. It might be necessary to have a new operator/service manual.
Upgrade – The term “upgrade” will refer to those instances when the operating requirements of the compressor do not change but changes are made to the components within the machine to improve their mechanical integrity, reliability or longevity. Examples could include replacing old riveted impellers with welded wheels, changing from aluminum labyrinths to polymer seals, changing bearings, or seals or the like. The primary objective of such upgrades is to improve a machine while not necessarily changing its performance characteristics.
Upgrades are not to be confused with normal preventative maintenance programs during which labyrinth seals, o-rings, bearing pads, etc. are replaced in kind with the same components as originally installed in the compressor. The choice between a revamp/re-rate or upgrade depends entirely on the end user’s needs. For example, the end user might have no interest in changing the performance characteristics of the compressor but might want to extend the operating life of the machine by switching to welded impellers. Similarly, changing from sleeve bearings to a tilt-pad bearing or from a tilt-pad bearing to more advanced damper bearings could improve the rotordynamic characteristics of the machine, with no impact on the aerodynamic performance.
The primary motivation for conducting re-rates or revamps is to re-purpose the compressor or to modify the compressor’s performance characteristics to better match new operating requirements. This might involve increased or decreased flow requirements, increased or decreased head or discharge pressure requirements, or some combination of both while simultaneously decreasing (or at least minimizing the increase in) the horsepower required to drive the compressor or compressor train. Simply put, the existing compressor or compressor train is incapable of meeting the new operating requirements or incapable of meeting the new operating requirements efficiently. For whatever reason, the end user has decided against purchasing new equipment and has, instead, decided to revamp the equipment to meet the new operating requirements.
Minimizing cost is another key motivator in revamping or upgrading existing equipment. In many situations, it is possible to re-use the majority of the compressor components including the case, heads, rotor, bearings, seals, and many of the internals. This can result in significant cost savings as the casing is quite often one of the more expensive items. If the casing is re-used, then the foundation, baseplate and existing piping arrangement can also be left in place, again resulting in substantial savings. In many situations, it is also possible to re-use many of the compressor internal components, such as the impellers, return channels, inlet, discharge volute, and inlet guides. Care must be taken to ensure the components are still structurally sound and that the flow passages are aerodynamically acceptable (i.e., not fouled, eroded or otherwise damaged).
The OEM or third-party must thoroughly inspect these components to: (a) ensure they are fit for continued service and (b) acquire the necessary geometry to conduct the necessary analyses (aerodynamic and mechanical) of the parts. It might also be possible to decrease the long-term operating cost of the compressor or compressor train by installing state-of-the-art aerodynamic components in the older machine. For example, older impellers with simple, less than optimal blading could be replaced by more modern impellers with sophisticated 3-D, inducer-style blades, resulting in higher efficiency and lower operating costs.
Likewise, more modern, machined stationary components with their inherently smooth surface finishes and precise dimensions could replace old, rough cast components; again increasing performance and lowering energy costs. In some cases, replacing a component can increase the overall efficiency of the compressor. In this instance, the evaluating engineer must make an economic analysis based on either dollar per horsepower (HP) of electricity or dollar per pound of steam.
Another advantage of the revamp or re-rate when compared to purchasing new equipment is that one can typically minimize the cycle time for the turn-around or, stating this another way, one can minimize the production down time. Clearly, if there is no need to replace piping, build a new foundation or lift a new casing in place, time (and money) are saved. Some end users also see revamps or re-rates as a way of minimizing risk. They might have good operating experience with the existing compressor and want to maintain the current or similar mechanical components in hopes of avoiding any potential mechanical and rotordynamic issues with a new machine.
Similarly, if the aerodynamic components applied in the revamp are from the same OEM product line, the revamp components are perceived as “proven” or “grandfathered” technology. Again, the perceived risk is minimized. Finally, as suggested above, revamping a compressor does allow the end user to apply new technology to an old machine. Assuming the new components can be made to fit, the end user can have the “latest and greatest” impellers, diffusers, return channels, bearings, seals, materials, etc. and the improved aerodynamic and mechanical characteristics that come with each.
Limitations of re-rate: When an OEM is consulted regarding the rerate of an existing compressor, there are certain physical and mechanical constraints that can be checked quickly to determine whether or not a revamp is possible. These constraints limit the maximum performance capability of the machine. The physical constraints include the casing internal dimensions (i.e., length and diameter) and the nozzle sizes. The mechanical constraints include casing pressure and temperature ratings as well as any operating speed limits. If any of these constraints are exceeded by the re-rate requirements, the manufacturer will recommend replacing the compressor with a more suitable model.
In many cases, the end user plans to use the available driver. This is acceptable as long as the driver can produce sufficient power to drive the revamped compressor. An assessment must be made on the driver capabilities relative to the new compressor operating conditions. For example, if the compressor speed is dropping considerably, it is possible the driver will not be able to provide the necessary horsepower at the reduced speed. Further, the rotordynamic characteristics of the driver must be assessed at the new required operating speed.
There are a number of criteria that will limit the design speed for a given model of compressor. These include rotational stresses, Mach number, critical speeds, and (potentially) the driver.
Rotational Stress Limits
Impellers for the API industry are designed with tip speeds of up to 1,000 fps. As a general rule, riveted impeller construction is limited in tip speed to 800 fps; welded impeller construction allows tip speeds up to 1,000 fps. It is good design practice to limit rotational impeller stresses to 70% of the impeller material yield strength. This is done so that at over-speed test conditions, no material yielding will occur. Rotational stresses are especially important in sour gas service in which the impeller material yield strength is generally limited to 90 ksi with a reduced hardness requirement of Rockwell C22. Since the compressor head is proportional to the impeller tip speed, the head requirement for the rerate is limited to using the maximum impeller diameter allowable for the casing and an impeller designed for a tip speed of approximately 1,000 fps.
James M. Sorokes is Principal Engineer, Dresser-Rand, Olean, NY, USA.
Steven T. Kaulius is Manager Engineered Solutions, Dresser-Rand, Bethlehem, PA, USA.
Edmund A. Memmott is Principal Rotordynamic Engineer, Dresser-Rand, Olean, NY, USA.