Advances in piston and packing ring materials for oil-free compressors

ROBIN S WILSON
Technical Director

Presented to the 4th Workshop Piston Compressors 25 & 26 October 2000 Ktter Consulting Engineers Rheine Germany

-1-

Advances in piston and packing ring materials for oil-free compressors

ROBIN S WILSON
Technical Director

INTRODUCTION

Reciprocating gas compressors provide a uniquely efficient means of compressing special gases used in the oil, petrochemical, gases and general process industries. Another growing requirement for this type of compressor is in the PET bottle-blowing industry, where oil-free reciprocating air compressors are used to blow PET (polyethylene terephthalate) bottles used as drink containers. Reliability of process machinery is very high on the agenda of all plant operators, and much attention has been focussed on the need for the critical sealing components used in reciprocating compressors to deliver consistent and long running lives. To this end, a long-term, dedicated program of research and development of improved self-lubricating materials has been conducted. This paper outlines some key aspects of this program, and illustrates with actual field examples some important advances in materials which have helped to restore confidence in the reciprocating compressor as a first choice for oil-free gas compression. In short, the future of oil-free reciprocating compressors is secure. An important, and sometimes overlooked spin-off of this work has been to enable the conversion of lubricated compressors to operate oil-free, with the many attendant and obvious advantages which this can give, including: Removal of costly lubricant (major saving if synthetic oil is used) Elimination of downstream contamination Extended catalyst life in chemical reactors Reduced maintenance attention (e.g. lubricator top-up)

Fig. 1 shows an LNG boil-off compressor successfully converted from lubricated to oil-free operation.

-2-

Fig. 1 Natural gas compressor converted from lube to oil-free 2 THE API 618 CHALLENGE

This internationally recognised standard is regarded as the bible by most compressor contractors and users. In the latest edition of API 618 dated June 1995, an important ideal for the compressor designer is laid down in section 2.1.1: The equipment (including auxiliaries) covered by this standard shall be designed and constructed for a minimum service life of 20 years and an expected uninterrupted operation of at least 3 years Only 10 to 15 years ago, this seemed an impossible challenge for selflubricating piston and rod sealing rings of an oil free reciprocating compressor. In some instances, lives of only a few months, weeks or even days were possible with the best materials then available. For these and other maintenance-related reasons, other types of compressor (centrifugal, screw, etc) started to gain in popularity. Nowadays, the advances made in self-lubricating materials have put the reciprocating compressor firmly back into no.1 position - for reliability, -3-

efficiency and flexibility of application. Moreover, the lives which can now be achieved make the choice of oil-free gas compression a simple one. It also lowers the capital cost (due to the absence of oil separators) and lowers the running cost (due to the exceptional efficiency of a truly positive-displacement machine). Finally, reciprocating compressors are untouchable in their sheer range of application, with commercial uses up to pressures of 3500 bar (lubricated) or 350 bar (oil-free).

3 OIL-FREE GAS COMPRESSOR APPLICATIONS

Some common examples of gases which are compressed in the process industries using oil-free compressors are listed below:

Gas
AIR AMMONIA ARGON BUTANE CARBON DIOXIDE CARBON MONOXIDE ETHYLENE HELIUM HYDROGEN ISOBUTANE METHANE NATURAL GAS NITROGEN OXYGEN PROPANE PROPYLENE

Application examples
PET bottle blowing, air separation, pneumatic instruments refrigerant, chemical processing agent welding, lamps fuel gas, chemical manufacturing carbonation of drinks, fire extinguishing, cooling chemical processing, ore reduction, fuel gases plastics manufacture, antifreeze welding, lamps, cryogenics, balloons refining, food manufacture, ammonia synthesis plastics & chemicals manufacture fuel gas, chemical manufacture fuel gas inert gas purging, ammonia synthesis steel & chemical manufacturing, breathing systems fuel gas, refining plastics & chemical manufacturing

Of these, many are processed in very pure and dry conditions. As recently as the 1980s, users of oil-free compressors handling such gases were experiencing unacceptably short piston and rod sealing ring lives, leading to frequent, often unplanned production stoppages, causing huge losses of revenue and high maintenance overheads. At the other end of the spectrum, atmospheric air, which contains water vapour is never constant in quality: some locations can be 100% humid while others are almost dry. As a consequence the level of water vapour (or liquid) which appears -4-

in the compressor cylinders can vary to a wide degree, significantly affecting the wear behaviour of the self-lubricating sealing rings. To address these many and varied problems, an extensive and long-term program of research and development was started in the 1980s, and continues today. This has focussed upon the wide range of available polymers, fillers and processing techniques which may be used to fabricate useful compounds, whose selflubricating capabilities can be matched to each and every potential application.

RESEARCH & DEVELOPMENT

The program has included the following areas of study: literature search moulding techniques commercially available polymers and fillers field experience in oil-free compressors failure analysis of used components wear testing microscope analysis physical and mechanical testing influence of component design field evaluation of new materials

Of paramount importance to the success of this program has been the intimate and continuous liaison with users of oil-free compressors without whose cooperation and active contribution - including a willingness to evaluate new materials minimal progress would have been made. Ultimately, those users have helped to make significant gains for their own companies, and in so doing have unselfishly aided other users throughout the process industries. Some key elements of this work are outlined below.

EARLY FIELD EXPERIENCE

The following are some examples of user experience in oil-free compressors around the early 1980s:

Bone-dry nitrogen
Variable ring and packing lives, some as short as 1000 hours (while there were some proprietary materials available capable of lasting for several years in this application, the results were inconsistent, and the relatively brittle properties of this family of thermosets led to occasional breakages, either in service or during fitting). The widespread use of dry nitrogen as an inert purging agent in almost all process industries, along with the bulk compression of this gas in the air separation industry has resulted in a high population of dry nitrogen oil-free reciprocating compressors. -5-

This was a well defined problem waiting to be solved. Now it has been well and truly solved, using a novel polymer alloy which has demonstrated lives running to tens of years.

Isobutane
This gas, usually in dry condition, and combined with other hydrocarbon gases, is recycled as part of the manufacture of certain grades of polyethylene. Oil-free reciprocating compressors used for this process are commonly of large bore sizes, and the unreliability of the piston and rider rings has been an important cause of plant stoppages due to short ring lives, which used to be as short as 1000 hours. Here again, the development of new polymer alloy ring materials has transformed the availability of these important compressors, and allowed a return to the luxury of planned shutdowns. With such recycling of gas, it is fairly normal for abrasive polymer and catalyst dust to be returned to the compressor, and this can of course compromise the result, but in clean gas conditions ring lives in the range 2 to 3 years are now available.

Pure hydrogen
The demand for compressed hydrogen for various processes is on the increase. Back in the early 1980s the piston and rod sealing rings used on dry hydrogen compressors were cause for much concern, and depending on the working pressure, typical lives of these components were as short as 1000 hours (6 weeks). With the advent of the new breed of self-lubricating materials, this life-expectancy has been raised by a factor of up to 40 times. Examples of such gas compressors with totally acceptable ring lives are given later on.

Atmospheric air
As mentioned in the introduction, oil-free air compression has become a growth business, based on the escalating demand for PET bottles, and the trend away from glass towards PET containers for many kinds of beverage. PET bottles are blown in a variety of blow-moulding machines, but a typical process air pressure is 40 to 45 bar. Traditional non-lubricated air compressors delivered at around 7 to 10 bar, and being mostly of standardised design, the piston and rod sealing materials used were able to provide adequate lives (about 1 to 2 years) using basic filled-PTFE grades. The increased delivery pressure has however exceed the capability of most standard materials, and improved grades have had to be developed to restore acceptable life to the higher pressure stage(s). -6-

While this has been very successful, further improvements will continue to be made.

LNG boil-off
Natural gas is liquified for storage at below 160 C. Since 100% insulation is not possible, a continuous evaporation of stored gas occurs, and this is usually either refrigerated back to liquid for re-storage, or exported. Either way, oil-free compressors are used to compress this extremely dry methane-rich gas mixture. For high volume boil-off, rotary compressors are sometimes used, but the reciprocating compressor is the ideal candidate, especially since reliable sealing ring lives can now be assured - typically 1 to 3 years, sometimes longer. Before the development of special self-lubricating materials, the rings on these boil-off compressors rarely lasted more than 2000 hours.

Other gases
Across a wide range of process gases, new ring materials have enabled the achievement of vastly increased ring lives. The selection of ring material is deeply influenced by the chemistry of the gas itself, which can affect the transfer process by which self-lubricating materials are able to stabilise and to resist wear. Accordingly, an extensive field experience has resulted in material selection guidelines, so that costly mistakes can be avoided. For example, materials expressly developed for use in bone-dry nitrogen compressors will be found completely unsuitable in, for example dry ethylene or crude argon. Materials developed expressly for dry oxygen compressors will not perform well in dry hydrogen. And so on. In other words there is no universal self-lubricating material which can be selected for oil-free gas compressors which will suit the full range of gases. A number of gases exhibit unique chemical behaviour, especially when subjected to heat, either due to gas compression or frictional rubbing heat from the rings. Examples of such gases, in which great care is needed for ring design and material selection are:

Ethylene : decomposes to form a soft polymer dust usually of limited

detriment to ring life

Carbon monoxide : decomposes into carbon dioxide and a highly abrasive

carbon dust - lives of conventional rings can be dramatically reduced

-7-

WEAR MECHANISM

Self-lubricating materials operate by what is termed an adhesive wear mechanism. They deposit a thin film onto the sliding counter-surface. This film, once stabilised becomes the lubricant against which the component (piston or packing ring) can continue to slide with reduced friction and an acceptably low wear rate. A graph depicting an ideal wear profile is shown in fig.2 . The initial part of the profile corresponds with the deposition of the transfer film, and the time for this transfer to stabilise may be tens or hundreds of hours, depending on the working conditions of the compressor.

Stabilised Wear Transfer Process

Transfer Process

Time
Fig. 2 Ideal Wear Profile
A satisfactory transfer film is illustrated in fig.3. This film was deposited by special polymer alloy packing rings running against an induction hardened stainless steel piston rod. The gas compressed in this instance is dry natural gas, used to fuel a gas turbine for power generation.

-8-

Transfer Film

Fig. 3 Satisfactory transfer film

The transfer mechanism or film formation is extremely complex, and is certainly not fully understood to this day. However, certain guiding principles have become clear as a result of many years of testing and observation. Perhaps the most important single conclusion is that chemistry plays a crucial role. The transfer process is therefore not just a mechanical transference of material to a semi-rough surface; it is strongly affected by chemistry. That is, the chemistry of the ring material, of the countersurface (or more correctly its surface oxides), and of the surrounding gas. An understanding of these fundamental principles has helped to speed the progress towards better self-lubricating materials.

Factors influencing the wear mechanism

Apart from the proper selection of self-lubricating material, there are a number of important operating factors which influence the actual running life which can be obtained in a given compressor system. These include: Differential pressure Gas temperature Counter-surface material Counter-surface finish Gas type Gas dryness -9-

Gas cleanliness (e.g. absence of solids or liquids) Cylinder cooling Packing cooling Speed & stroke

Expanding on just some of these:

Counter-surface materials and finishes

Cylinder liners: For the majority of applications grey (flake graphite) cast iron has proved to be an ideal choice. The preferred hardness should be at least 200 Hb. In conditions where severe corrosion is possible, then ni-resist or austenitic (flake graphite) cast iron A436 type 2b is advised, with the same hardness recommendations as above. Some special liner treatments and hard coatings have been used with success, but also some have proved disastrous in terms of accelerated ring wear, and caution is advised, with the reminder that chemistry (not just surface hardness and texture) is important. Optimum liner surface finish is in the range 0.4 to 0.6 microns Ra. Liners which have become polished by abrasive wear should be re-roughened to restore optimum ring life. Piston rods: Alloy (AISI 4140) or stainless steels (AISI 420) have proved successful, but these must be hardened to 50 / 55 Rc to give adequate stability for the transfer film, and sufficient long-term wear resistance. Nitrided piston rods offer harder conditions (60 / 65 Rc) and are also acceptable. Hard coatings such as tungsten carbide, applied by the detonation gun or HVOF processes are now widely used, and appear generally acceptable. These coatings are very hard (70 Rc) and therefore provide good wear resistance. However, in certain combinations of gas type and packing ring material, results have been less than successful. Therefore, as with liner coatings, caution is advised. If in doubt, the best guide is existing field experience. Piston rod surface finishes should be in the range 0.2 to 0.3 microns Ra, for hardened surfaces and 0.1 / 0.2 microns Ra for the harder coatings. Gas type The chemistry of the gas can literally have a dramatic impact on the ring wear behaviour. An important example is oxygen, whose influence has been found to be crucial to the wear mechanism. To verify this, wear tests were carried out on materials selected for dry inert gas, with the environment altered from pure nitrogen to dry air, then reverting to pure nitrogen. The wear patterns show the - 10 -

immediate and severe impact of the oxygen component during this phase of testing. The chemistry of all gas types is to a greater or lesser extent influential upon wear behaviour. So, when a new application is considered, it is actual field experience in a closely matching gas type which is the best guide to the proper selection of ring (and countersurface) material. Gas cleanliness Compressors are fed via pipelines and other upstream systems. For reasons of economy, these are often fabricated in carbon steel, which is subject to corrosion. It is uncommon for compressors to be equipped with upstream filters (also for economic reasons) and very common for rust and pipescale to build up, and to be sucked into the compressor. The abrasive effect of such particles can be highly destructive to a set of selflubricating rings and to the transfer film which they have so carefully laid down. The same results can occur with other abrasive particles, such as polymer powder. The only possible ways to address this risk, are by careful cleaning of pipework, and ideally by the use of upstream filters (8 micron typically), preferably of the duplex type, allowing on-line element changes. Liquids can have an equally destructive effect, in that they can disrupt the transfer film, leading to accelerated ring wear. Liquids may be removed by upstream and interstage separators. Otherwise, it is helpful to inhibit condensation of liquids in the cylinders, by raising the temperature of the cooling jackets to 5 above the gas inlet temperature. C

Speed & Stroke In order to achieve maximum capacity for a given physical size of compressor, there has been a trend towards the use of higher speed, shorter stroke units. For example, a medium speed compressor having a speed of 400 RPM and a stroke of 260 mm will have roughly the same average piston speed as a high speed compressor running at 833 RPM with a stroke of 125 mm. However, the higher speed unit is likely to be less reliable in terms of ring life, due to the fact that the frictional heat generated by both the piston rings and the packing rings will have a less effective path for heat dissipation clearly the heated areas of the counter-surface will be less exposed to their surrounding heat sink.

- 11 -

WEAR TESTING

It is obviously impossible to ask compressor users to field-evaluate hundreds of new materials, and therefore a special material wear test rig was purpose designed and built (fig. 4). This unique rig allows comparative wear tests to be conducted in a chosen gas environment - typically bone dry nitrogen, bone dry air, atmospheric air, etc. The special geometry of the wear test pieces, and of the sliding countersurface (sleeve) (fig. 5), ensures that wear behaviour can be observed over a relatively short test period (typically 24 hours). This procedure is known as accelerated wear testing, and has become a valuable means of rapidly screening new materials for their suitability. The wear profile of each grade tested is plotted on a wear / time graph to establish its behaviour and especially its wear stabilization, which is an indication of the success or otherwise of the transfer process. The test rig counter-surface materials, hardnesses and finishes are chosen to closely match those of the compressor cylinder liner and / or piston rod. The sliding speed of 4 m/s is close to the upper level of average piston speed normally encountered in an oil-free reciprocating compressor. Those experimental grades which show outstanding promise during the screening test are subjected to further wear testing, along with physical and mechanical testing to assess their all-round suitability for the harsh conditions they will find in a real compressor application.

Fig. 4 Comparative wear test rig

- 12 -

Fig. 5 Wear test sleeves and test sample

THE NEW MATERIALS

The program of research and development has yielded many hundreds of new experimental compounds, some of which are referred to as polymer alloys. This term reflects the apparent alloying of different polymers within the matrix when such compounds are processed at elevated temperature. In fact it is clear that the physical and chemical properties of these unique materials are modified to an extent that is substantially different to those of the individual polymers which are used. The commercial advantage which has resulted from proven success in this important area, means that the actual compositions and processing methods used to manufacture these compounds obviously cannot be disclosed. Details of some of these new materials are summarised below, along with a representative selection of field results in section 9.

- 13 -

8.1

CPI 184
Dry hydrogen & hydrocarbons (+ other selected gases) in oil-free compressors Piston rings, rider rings, packing rings 11 5 70 90 x 10-6 1.9 65 - 70 125

Typical application (low to medium duty) Products Tensile strength (MPa) Elongation @ 20 (%) C Coeff. thermal exp. (/ C) Specific gravity Hardness (Shore D) Suggested mean temp. limit for oil-free compressors ( * C)

* mean temp. is defined as (T1 + T2) / 2 T1 = gas suction temp. T2 = gas discharge temp.

8.2

CPI 192
Dry hydrogen & hydrocarbons (+ other selected gases) in oil-free compressors Piston rings, rider rings, packing rings 35 1-2 40 x 10-6 1.5 80 - 85 175

Typical application (high duty) Products Tensile strength (MPa) Elongation @ 20 (%) C Coeff. thermal exp. (/ C) Specific gravity Hardness (Shore D) Suggested mean temp. limit for oil-free compressors ( C)

8.3

CPI 188
Pure, bone-dry nitrogen, argon & helium in oil-free compressors Piston rings, rider rings, packing rings 9 2-4 70 90 x 10-6 2.0 65 - 70 120 - 14 -

Typical application (low to medium duty) Products Tensile strength (MPa) Elongation @ 20 (%) C Coeff. thermal exp. (/ C) Specific gravity Hardness (Shore D) Suggested mean temp. limit for oil-free compressors ( C)

8.4

CPI 196
Pure, bone-dry nitrogen, argon & helium in oil-free compressors Piston rings, rider rings, packing rings 30 1-2 40 x 10-6 1.7 80 - 85 175

Typical application (high duty) Products Tensile strength (MPa) Elongation @ 20 (%) C Coeff. thermal exp. (/ C) Specific gravity Hardness (Shore D) Suggested mean temp. limit for oil-free compressors ( C)

8.5

CPI 124
Atmospheric air in oil-free compressors Piston rings, rider rings, packing rings 15 120 70 90 x 10-6 3.8 65 - 70 130

Typical application Products Tensile strength (MPa) Elongation @ 20 (%) C Coeff. thermal exp. (/ C) Specific gravity Hardness (Shore D) Suggested mean temp. limit for oil-free compressors ( C)

RECENT FIELD EXPERIENCES

To illustrate the advances achieved with the new materials, the following are a few examples of actual field experiences, which are representative of the results being achieved across a wide range of oil-free gas compressors:

9.1

DRY HYDROGEN

- 15 -

A)

Horizontal Compressors : Gas Processing Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bore (mm) Suction pressure (bar) Discharge pressure (bar) Materials Selected Piston and rider rings Packing rings Background & Results The compressors were supplied in 1993, and due to good previous experience with these materials, the user specified CPI 184 and CPI 192. All components are giving a repeatable running life of 16000 hours (2 years) or more. CPI 184 CPI 192 Horizontal, oil-free 2 Bone-dry hydrogen 327 320 4 1 350 25.3 41.5

B)

Vee Compressors : Gas Processing Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) Vee, oil-free 2 Bone-dry hydrogen 570 125 2 2 240 / 160 7 40

Materials selected Piston and rider rings Packing Rings CPI 184 CPI 184

- 16 -

Background & Results The compressors were originally supplied with carbon + ceramic filled PTFE piston rings, rider rings and packing rings. These gave a typical life of 1000 to 2000 hours only. These were replaced with CPI 184 piston rings, rider rings and packing rings in 1987. Lives of about 16000 hours are now regularly obtained.

9.2

DRY ETHYLENE Horizontal Compressors : Ethylene Storage Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) Materials Selected Piston and rider rings Packing rings CPI 184 CPI 192 Horizontal, oil-free 2 Bone-dry ethylene 585 203.2 4 3 457.2 (2) / 381 / 241.3 0 21

Background & Results These compressors are used to compress ethylene gas from liquid storage. They have been operated for many years, originally equipped with carbon filled PTFE rings which lasted only around 1000 to 2000 hours. The rider rings were of solid, stretched-on design. These rings were replaced with CPI 184 piston and rider rings, initially of similar design, and CPI 192 packing rings, which resulted in a major improvement in running life, ranging up to 5 years or more. Subsequent improvements included a redesign to use split rider rings (for easier maintenance) and the use of grey cast iron liners to replace ni-resist liners where these were used. The current life projections range from 20000 hours to over 100000 hours.

- 17 -

9.3
A)

DRY NITROGEN Vee Compressors : Plastics Manufacturing Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) Vee, oil-free 2 Bone-dry nitrogen 660 100 3 (stage 2 & 3 tandem) 3 200 / 190 / 130 5.9 39.3

The compressor is illustrated in fig. 6. Materials Selected Piston and rider rings CPI 188

Fig. 6 3-stage bone-dry nitrogen compressor

- 18 -

Fig. 7 Stage 2 & 3 piston & rider ring inspection

Background & Results Originally supplied filled PTFE piston and rider rings were lasting only a few weeks. The stage 1 and 2 liners are grey cast iron, while the stage 3 liner is treated with a special chrome oxide coating. Comparative wear tests were conducted on CPIs test rig using dry nitrogen and exactly equivalent counter-surface materials. These tests demonstrated a dramatic improvement in wear behaviour when using polymer alloy CPI 188, compared with the originally supplied ring material. Accordingly one compressor was equipped with CPI 188 piston and rider rings on all 3 stages. Wear checks were made at regular intervals over a year (fig. 7 shows the stage 2 and 3 rings removed for one of these inspections), and the projected lives of all rings based on these measurements are at least 148000 hours (17 years!). B)

Horizontal Compressors : Air Separation Plant

Compressor type No. of compressors Gas Speed (rpm) Stroke (mm) Horizontal, oil-free 3 Bone-dry nitrogen 326 310 - 19 -

No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar)

2 2 355 / 255 17.8 72.2

Fig. 8 Shows one of these compressors (stage 2). Materials Selected Piston and rider rings Packing rings CPI 188 CPI 196

Fig. 8 Bone-dry nitrogen compressor

Background & Results Various ring materials have been used on these compressors, with the best life obtained being in the range 1 to 2 years. The user was seeking to obtain longer lives between overhauls, and decided to conduct an evaluation of - 20 -

CPI 188 piston and rider rings on one compressor. The packing rings were already CPI 196, and remain today in this grade. After a running period of 4000 hours, a wear check indicated the following projected lives: Stage 1: 132000 hours (15 years) Stage 2: 77000 hours (9 years) The pattern of results with CPI 188 rings has been repeated on a large number of bone-dry nitrogen compressors, and as a consequence many users are specifying CPI 188 material exclusively for all new dry nitrogen compressors which they purchase.

9.4

ISOBUTANE Horizontal Compressors : Polyethylene Manufacturing Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bore (mm) Suction pressure (bar) Discharge pressure (bar) Horizontal, oil-free 4 Bone-dry isobutane 422 254 2 2 710 / 385 1.1 19.4

Materials Selected Piston and rider rings Packing rings Background & Results Three of these compressors have been in operation since the 1970s and were originally fitted with filled PTFE piston rings, rider rings and packing rings. Life expectancy was variable and short, sometimes only a few days, and with a maximum of 6 months. In the 1980s CPI 184 piston and rider rings were evaluated on one compressor, and the first set of rings lasted 2 to 3 years. All three compressors were changed to the same grade, and continue to give long and totally acceptable average ring lives (variations can occur due to occasional recycling of polymer and catalyst from the reactor, and iron oxide - 21 CPI 184 CPI 192

from the pipework). Subsequently, the packing rings were upgraded to CPI 192 material, and water cooling removed from the packing cases. A fourth compressor was later installed, the user specifying CPI 184 rings and CPI 192 packings to be fitted by the manufacturer. CPI materials have now been installed in numerous similar compressors performing this duty at polyethylene plants all round the world.

9.5

NATURAL GAS
The storage of LNG has become widespread in the natural gas industry, both for holding large stocks of gas onshore, or for its transportation worldwide. Fig. 9 shows a typical natural gas storage installation. Natural gas is liquified at 160 and any moisture is frozen out, leaving C, the gas very dry (less than one part per trillion of water vapour). Boil-off gas is compressed for re-liquifaction or for export, and traditionally used ring materials (filled PTFE or resin bonded composites) have exhibited irregular and generally short lives in this application, commonly less than 6 months. CPI 184 material has been successfully applied to LNG boil-off compressors, and two examples are given below:

Fig. 9 LNG storage plant

- 22 -

A)

Horizontal Compressors : LNG Storage Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) Materials Selected Piston and rider rings Packing rings Background & Results Originally fitted with filled PTFE piston and rider rings, the life achieved was inconsistent, ranging from only 4000 to 8000 hours. After installing CPI 184 piston and rider rings and CPI 192 packing rings, the life has increased to 25000 hours, allowing the operator to schedule planned overhauls every 2 years. CPI 184 CPI 192 Horizontal, oil-free 4 Bone-dry natural gas (LNG) 480 250 2 2 790 / 450 1.1 5.5

B)

W Compressors : LNG Storage Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) W, oil-free 5 Bone-dry natural gas (LNG) 990 85 3 2 215 (2) / 160 0 12

Materials Selected Piston and rider rings CPI 184 - 23 -

Packing rings Background & Results

CPI 184

These compressors were delivered with filled PTFE piston and rider rings and packing rings. The achieved running life between ring changes was no more than 2000 hours. All compressors have been changed over to CPI 184 piston and rider rings and packing rings, which have increased the running period to 20000 to 25000 hours between stoppages for ring renewals.

9.6

ATMOSPHERIC AIR Horizontal Compressor : Chemical Plant

Compressor Type No. of compressors Gas Speed (rpm) Stroke (mm) No. of cylinders No. of stages Bores (mm) Suction pressure (bar) Discharge pressure (bar) Horizontal, oil-free 1 Atmospheric air 425 267 4 3 820 (2) / 650 / 420 1 14.5

Materials Selected Piston and rider rings CPI 124

Background & Results This compressor is used in a chemical oxidation process, and has to operate continuously to sustain the process. Originally used carbon filled PTFE piston and rider rings on stage 2 gave short lives, typically about 4000 hours or 6 months. The operator wished to extend the life beyond 1 year, to allow annual plant stoppages.

- 24 -

A set of CPI 124 piston and rider rings was installed for evaluation, and were found in good condition after a running period of 12000 hours, three times the original ring life, and meeting the criterion for acceptance. The first and second stage cylinders have now been fitted with CPI 124 piston and rider rings. Owing to the outstanding performance of CPI 124 piston rings, rider rings and packing rings in atmospheric air compressors, several notable compressor manufacturers have chosen to adopt this grade for their full range of PET bottle-blowing compressors.

10

CONCLUSIONS

The development and introduction of a new family of self-lubricating materials has enabled the users of a wide range of oil-free process gas compressors to profit from a dramatic improvement in machine reliability and availability. This has resulted in major savings in operating and maintenance costs, and has provided oil, petrochemical, gases and general process industries with significant gains in productivity. Many existing compressors remain to be similarly upgraded, and users who need to procure new compressors can purchase oil-free units with full confidence in their performance, provided of course that they specify that the compressor manufacturer incorporates the best ring materials, proven for the respective application. Ongoing research and testing will ensure that further strides are made in the introduction and application of new materials. Not only will this allow yet further increases in life expectancy of wearing parts, but it will also enable the existing limits on operating parameters , such as pressure, sliding speed, etc., to be raised.

- 25 -