36

The Current State of Disc Cutter Design and Development Directions

Joe RobyThe Robbins Company, Kent, Washington

Tyler SandellThe Robbins Company, Kent, Washington

John KocabThe Robbins Company, Solon, Ohio

Leif LindberghThe Robbins Company, Kent, Washington

ABSTRACT: Rolling disc cutters are the business end of hard rock and mixed face tunnel boring machines(TBMs). Since their first successful employment on a TBM over fifty years ago, disc cutter technology hasconstantly evolved, allowing modern TBMs to excavate very hard and abrasive rock efficiently. Disc cuttershave also been employed successfully on earth pressure balance (EPB) and slurry machines, cutting rockunder water and ground pressure. The range of materials excavated by machines today is broader, excavationrates higher and cutter costs lower than ever before, proving the value of investment in cutter development.

This paper describes recent improvements in disc cutter components including cutter ring materials andprocessing, as well as improvements in lubricants, bearings, seals and cutter condition monitoring. Thesedevelopments have resulted in more reliable cutters capable of operating in a wide range of geological condi-tions. Ultimately these improvements result in a better value for both contractors and project owners.

INTRODUCTION

Disc cutters are used on a wide range of tunnelingequipment, from pipe jacked slurry micro-TBMs lessthan a meter in diameter to 15 m diameter, hard rockboring TBMs. The geological conditions underwhich they are employed range from sands and grav-els with several bar of water pressure to extremelyhard, massive rock with UCS up to 420 MPa.

Regardless of the type of machine or geology,one thing remains constant: changing worn out cut-ters is costly. When cutters must be changed in themiddle of a tunnel, the contractor incurs the cost ofdowntime as well as the cost of refurbishing orreplacing the cutters. In the case of catastrophic cut-ter failures, the project can be stopped completelyfor long periods, and the costs mount up rapidlywhile tunnel production is at a stand still. Cata-strophic cutter failures include, for example, theoccasion when hard rock disc cutters are failed ingroups (called a “wipeout” phenomenon) and theoperator fails to stop the machine, which results insevere damage to the cutterhead. The cause of thiscan be either an undetected failed cutter propagat-ing damage to surrounding cutters, or operator errorin steering. Regardless of the cause, the resultingdamage can take days or even weeks to repair.

When cutters fail prematurely on EPB or Slurrymachines in certain geological conditions, it isimpossible to evacuate the chamber and thereforeimpossible to get into the chamber to change thecutters. The solution is frequently an unplannedintervention shaft which must be sunk at great cost.The importance of being able to predict cutter lifeaccurately, and in all geological conditions, cannotbe overemphasized.

For these reasons, cutters have been developedfor specific machine types and sizes, as well as forspecific geological operating conditions. Clearly,different size cutters are required for different sizemachines. For example, while 19 and 20-inch (483and 508 mm) cutters are used on large diameterTBMs, it is impossible to employ cutters so large ona small micro-TBM. Also, the cutter must bedesigned for the specific geological conditions underwhich it will be operating. The disc cutter ringsrequired to bore extremely hard rock are the mostexpensive of cutter rings; however, they provide lit-tle advantage in weaker rock formations where farless expensive rings will do the job as well. It isimportant to choose the correct cutter for themachine and the geological conditions in order toget the best balance of cost and risk.

37

LARGE DIAMETER, HARD ROCK DISC CUTTER DEVELOPMENT

In the early days of hard rock tunnel boring, TBMswere being employed primarily in weak to moderatestrength sedimentary formations. The first success-ful use of disc cutters was on the Robbins mainbeam TBM 910-101 in 1952 on the Oahe Damproject in South Dakota where the machine exca-vated faulted, jointed shale at only 1 to 3 MPa. Thecutters were small and looked little like modern hardrock disc cutters.

Because the rock could be excavated with lowcutter loads, the cutter bearings didn’t need to bevery large and the cutters were kept small whichmade them very easy to handle and change. How-ever, soon TBMs were forced into ever harder rocks,which resulted in an unacceptable rate of cutter wearon the small cutters, along with a rising number ofcutter failures due to catastrophic bearing failure. Asa result, over the succeeding years, cutter size andbearing capacity increased (see Table 1).

Development of the 19-inch Cutter and Its Application

The story of the development of a successful 19-inchcutter and its application is one of incrementalimprovement in component design. As one compo-nent was improved, another component became theweak link and required further development. How-ever, it was not just the cutter assembly that requiredcontinual design improvement, but the application;cutterhead design, cutter management and cutterlubricants all received investigation and improve-ment over the years.

During development of the 19-inch cutters,Robbins engineers collected massive amounts ofdata from job site cutter shops, examining andcounting failed 17-inch cutter components. WhenRobbins introduced the first 19-inch cutter in 1989,many things were new in the design, and muchimproved compared to the existing 17-inch cutterdesign:

• The cutter ring wear volume of 19 inch cutterwas increased by 38% (see Figure 1).

• Ratio of Cutter Load Rating to Cutter BearingLoad Capacity was reduced. Robbins 19-inchcutters full load rating is only 84% of the bear-ing’s rated capacity (i.e., 32 t/38 t) whereas the17-inch cutter’s full load rating is 93% of thebearing’s rated capacity (i.e., 27 t/29 t) (seeFigure 2).

• The new Wedge-Lock cutter mounting systemproved a vast improvement in reliability versusthe previous V-block mounting system.

• The face seal torics were much larger in cross-sectional diameter, allowing greater misalign-ment. Seal torics, the silicone or nitrile o-ringwhich support the metal face seal, in the17-inch V-block cutter had a small cross-sec-tion diameter (6.35 mm) which allowed onlyslight misalignment of the shaft to the hub.Finite element analysis revealed that, when thecutter is severely impact loaded, deflection ofthe shaft can cause misalignment exceedingthat allowed by the small cross-section toricthus allowing ingress of foreign materials. Theoriginal toric material also tended to becomenon-elastic quickly if the cutter temperaturewas too high. The larger, silicone torics used onthe 19-inch Wedge-Lock cutters eliminatedboth of these problems.

• Cutter hub life was increased dramaticallythrough improved materials and processing togive a very wear resistant hub surface at thehub-to-ring interface.

• The new Wedge-Lock cutter mounting systemproved a vast improvement in reliability versus

Table 1. Cutter diameters and load rating vs. year

Diameter (inch)

Load (kN) Year Introduced

11 85 196112 125 196913 145 198014 165 1976

15.5 200 197316.25 200 1987

17 215 198319 312 198920 312 2006

Figure 1. Cutter ring wear volume: 17-inch vs. 19-inch

38

the previous V-block mounting system. Cutterhousing life was also dramatically improved.

Initially, cutter ring life on the 19-inch cutterwas not as long as had been expected. Too manyrings failed prematurely due to spalling rather thanslow wear, and some were fracturing. While thelarger bearings could provide good support at thehigher loads, clearly the existing cutter ring materi-als were not capable of withstanding the increasedHerzian contact stress. Increasing the cutter ring tipwidth would reduce the contact stress, but alsoresulted in reduced penetration into the rock for agiven cutter load, moving the problem from the cut-ter ring to the bearing. Increasing ring tip width wasthe short term solution; however continuing metal-lurgical research eventually resulted in rings madefrom tool steel and later, from proprietary modifiedtool steels. With proper heat treatment, these ringshave much higher hardness as well as increased frac-ture toughness compared to the previous materials.In addition, these steels retain their strength at theelevated temperatures incurred when boring veryhard rock. Today’s 19-inch cutter rings can be usedin most geology at the same tip widths as 17-inchcutters, allowing 19-inch cutters to penetrate at thesame rate with only a very slightly increased load.

Another problem with the early deployment ofthe 19-inch cutter was a tendency to be susceptibleto multiple cutter, wipeout failures. In effect, if onecutter failed catastrophically (i.e., broken ring, failedbearing) there was a tendency for the cutters in thepaths next to that cutter to fail also, and the failure

pattern might repeat until 5 to 10 cutters were failedin a group. This was eventually identified to be aresult of the cutter spacing and was corrected overthe following years. As cutterhead design evolved,19-inch cutter data was collected from the field.With improvements made over the years in cutterrings, 19-inch ring life became very good and theweak link in the cutter seemed to be the bearing,once the strongest part of the design.

When cutters are worn and removed from theTBM, they are subject to one of two treatmentsbefore being returned to the TBM:

• Re-ring: Remove and replace the cutter ring,and change the lubricant

• Rebuild: Completely disassemble and replacecutter rings, bearings, seals, other small parts,and/or lubricant.

Obviously, rebuilding is far more expensivethan re-ringing. Routine monitoring of the cutterre-ring-to-rebuild ratio is necessary to maintain thelowest consumable cost on projects. A high cutterrering-to-rebuild ratio is indicative of a high-quality cutter body assembly, and always results inlower total cutter cost for the project. For the 19-inchcutter, Robbins wanted to increase the cutter re-ring-to-rebuild ratio. The solution was two-fold: improvedlubricants and precise record keeping. The lubricantsrecommended today are quite expensive, but theirreturn on investment is substantial. Precise recordkeeping allows the cutter manager to be constantlyaware of each cutter’s time in service, providing the

Figure 2. Bearing load rating and assembly weight: 17-inch vs. 19-inch cutter

39

cutter manager the information necessary to predictthe probable end of life of the cutter assembly andrebuild it in advance of failure.

Ten years after its introduction, the 19-inch cut-ter proved itself a clear choice for excavating hardand mixed face rock. TBMs employing the 19-inchcutter were excavating the hardest rock worldwideand setting impressive production records whiledoing so. See Table 2 for a comparison of cutter ringlife at the introduction of the 19-inch cutter with thatfrom more recent projects.

Having increased the cutter re-ring-to-rebuildratio, the next step was for engineers to once againextend the life of the cutter ring.

Development of the 20-inch Cutter

There are two methods by which one can increasedisc cutter ring life:

• Increase cutter ring strength and abrasionresistance.

• Increase the wear volume available on the cut-ter ring.

Research of new disc metallurgy and processing toincrease cutter ring life in hard and abrasive rock hasyielded incremental improvements, at high cost forthe research. Currently 19-inch cutter rings are beingmanufactured using three different cutter ring metal-lurgies and heat treatment processes (Standard,Heavy Duty, and Extra Heavy Duty). Some researchtest rings have shown marked improvement in ringlife, but only at an unacceptable cost for the rings.While metallurgical research continued, in the shortterm there was no economical way to improve cutterring life through increases in material strength and/or abrasion resistance. It became clear that anincrease in wear volume was the most cost effectiveway to improve cutter ring life for the 19-inch cutter.

To increase the wear volume one can increasethe cutter ring tip width, the cutter ring diameter orboth. Obviously, increases in cutter ring tip widthhave an adverse effect on cutter ring penetration intothe rock. To achieve the same penetration, a cutterring with an increased tip width would require anincrease in thrust, which would have an adverseaffect on the cutter bearing. However, an increase indiameter would have only a negligible effect on cut-ter load to achieve the same penetration. Essentially,one can increase the “tip length” to provide morewear volume in the ring. The risk of a longer cutterring tip is the potential for fracturing of the long tip.

This story is somewhat similar to the advent ofthe 17-inch cutter, which was simply a change in thecutter ring diameter mounted on the original15.5-inch cutter assembly. And, the same logic wasused when the extended tip 17-inch cutter wasemployed successfully on many sedimentary rockjobs. The cutter provided deeper penetration capa-bility and larger wear volume compared to a stan-dard 17-inch cutter ring. This experience providedempirical field data to support the concept that anextended tip 19-inch cutter might prove successful.

Using a conservative approach, new 20-inch cut-ters were developed for use on TBMs which coulduse either 19 or 20-inch disc rings. The 19-inch cutterhub/bearing/shaft assembly is used for both 19- and20-inch cutter rings, so either cutter ring can beemployed. Cutter housings and cutterheads were ini-tially designed to accept either the 19- or 20-inch ringcutters for the following three TBMs:

• 14.5 m, Hard Rock Open, High PerformanceTBM (HP-TBM) for the Niagara, Canadahydropower expansion.

• Two 10.0 m, Hard Rock Double Shield, HP-TBMs for the AMR water transfer project inIndia.

The 20-inch cutter rings produced offer a 58%increase in wear volume compared to the 19-inchcutter ring (see Figure 3).

Table 2. Improvements in 19-inch cutter ring life

Project Location

Svartisen, Norway

1990Atlanta, Georgia

2005Rock Types Micaschist,

granite, chalkstone

Very fine-grained medium grade metamorphic

rocksUCS Range (MPa)

49 to 196 MPa Average UCS of 255 MPa and

localized maximum of

530 MPaCutter Life (m3/ring)

146 m3 187 m3

Figure 3. Wear volume: 19-inch vs. 20-inch ring

40

Initial reports from the Niagara project aregood, with the 20-inch cutter rings in continual usein the sedimentary rock, giving good penetration andgood life to date. The AMR project requires boringthrough hard rock from 160 to 190 MPa UCS. Thiswill be the first hard rock test of the new 20-inch cut-ter ring. The machines will start boring in mid-2008and cutter ring performance will be reported afterdata has been collected from the project.

CUTTER BEARING DEVELOPMENT

Cutter Bearings

Most rolling disc cutters use two tapered roller bear-ings that are arranged in what the bearing industrycalls “indirect mounting.” These bearings are askedto cope with extremely dynamic loading and vibra-tion in a very harsh environment. As cutter ringshave grown, so have the bearings and the loads theymust handle. In this section we will discuss bearinglife and failure modes.

Catastrophic Failures—Post Mortem Inspection

When a cutter assembly fails catastrophically (seeFigure 4), it is never easy to determine how the fail-ure was initiated. A post mortem inspection is theonly way to diagnose and remedy the problems.There is always some amount of doubt as to why thecutter assembly is full of tunnel muck, which resultsin some classic questions. Did a bearing(s) fail,allowing misalignment and letting muck pass theseals? Or did the seals fail and allow contaminationof the bearing? Fortunately most cutter/bearing fail-ures are not this severe and are usually detected andrepaired/replaced before they get to this point.

Seal Inspection

Seal failure is a death sentence for a mechanicalassembly in an underground environment andnowhere is this more certain than with TBM cutters.Post mortem inspections of catastrophically failedcutters generally start with the seals. If for no otherreason they are the first part to be removed duringdisassembly. In some instances post mortem sealinspection will immediately reveal the “smokinggun” but quite often the mode of failure isn’t soobvious. Complicating things further, it is notalways easy to determine the cause of failure oncethe failure mode is isolated.

Modes of seal failure:

• Damaged torics• Fused faces• Abrasive wear

Causes of seal failure:

• Assembly error – Too much drag– Face pressure too low

• Rust from long periods of no use• Packing—Seal filled with clay or mud that is

allowed to harden while cutter is not turning • Use past service life

Bearing Inspection

In most hard rock tunneling conditions the cutterbearings will remain serviceable through multiplecutter rings if maintained properly. Proper mainte-nance requires frequent inspections and theseinspections often reveal impending problems. If fail-ing bearings are not detected, adjacent cutters willbecome overloaded and prematurely fail and lead toa chain reaction of multiple cutter failures. The TBMindustry calls this phenomenon a “wipeout,” whichcan cost hundreds of thousands of dollars in equip-ment and downtime (see Figure 5). Much like sealfailures, this mode of failure does not always leave aclear path to the original cause of the failure.

Modes of bearing failure:

• Spalled or grooved raceways (see Figure 6)• Worn or broken roller cage• Damaged rollers• Brinelling/false brinelling

Causes of bearing failure:

• Extreme dynamic loading• Overloading

Figure 4. The cutter on the left quit turning and wore flat. The cutter on the right experienced a massive bearing failure that resulted in a damaged seal retainer and shaft.

41

• Loss of lubrication• Overheating• Assembly error—Too much or too little pre-load• Contamination—ingress of tunnel muck—

usually a seal failure

Bearing Life—Theoretical vs. Actual

There is a wealth of information and many studieshave been done on calculating and estimating bearinglife in mechanical assemblies. Bearing manufacturersand 3rd party associations such as ISO, SAE and theASME have done comprehensive scientific researchand have developed mathematical models that pre-dict serviceable duration under controlled conditions.ISO 281 is generally accepted as the industry stan-dard that provides an estimated number of cycles orhours that the bearing can be expected to perform.Unfortunately, experience tells us that these modelsare not able to accurately predict bearing life in disccutters. Problems with predicting bearing life in disccutters are two-fold. First and foremost, the industryhas not been able to accurately define and quantifythe extreme dynamic loads to which these bearingsare subject and secondly geology dictates that condi-tions are always far from controlled or consistent.

Theoretical LifeThe following definition is from the ISO’s website(ISO 281, 2007). “[ISO 281:2007] specifies methodsof calculating the basic rating life, which is the lifeassociated with 90% reliability, with commonlyused high quality material, good manufacturingquality and with conventional operating conditions.In addition, it specifies methods of calculating themodified rating life, in which various reliabilities,lubrication condition, contaminated lubricant andfatigue load of the bearing are taken into account.ISO 281:2007 does not cover the influence of wear,corrosion and electrical erosion on bearing life.”Table 3 provides examples from two long term tun-nel projects. One of the jobs used 17" cutters for aduration of 7,950 total operating hours. The otherjob utilized 19" cutters for 3,117 hours. In both casesthe rock was generally considered to be hard toextremely hard (100 MPa UCS and higher). The cal-culated bearing set life per ISO 281 should havebeen 43 hours for the 17" cutters and a very impres-sive sounding 2,165 hours for the 19" cutters.

Actual LifeIn practice, actual service life varies drastically fromtheoretical life estimate. In addition to extremedynamic loads on bearings, cutters can also be dam-aged due to service factors such as ingress of debris

Figure 5. Example of a cutter wipeout phenomenon

Figure 6. Close up of spalling in the load zone of the inner race

Table 3. Comparison of calculated vs. actual bearing life

17" Cutters 19" CuttersCalculated Bearing Life (Hours)

43 2165

Actual Field Data: Manapouri, NZ

Cobb County, Atlanta

TBM Working Time (Hours)

7950 3117

Actual Bearing Sets Used 1612 191Average Bearing Set Life (Hours)

5 16

Ratio of Actual/Standard L10

11.47% 0.75%

42

and water (seal failure). The damaged cutters typi-cally show bearings with severe spalling in the loadzone, damaged rollers, broken cages and or over-heating. Sometimes the normal inspection or servic-ing of cutters may show early impending bearingdamage, and in these cases bearing components arereplaced much more frequently than their theoreticallifespan. Table 3 shows that actual bearing life in theaforementioned projects was orders of magnitudeshorter than the ISO 281 calculation predicted.

CUTTER SEAL DEVELOPMENT

One of the most significant developments in disccutter technology came with the application of theCaterpillar metal face seal in the early 1970s (Hand-book of Mining and Tunnelling Machinery, 1982).This seal type uses two metal rings that are loadedaxially such that they ride against each other with afilm of lubricant between them, creating a dynamicface-seal interface. Each ring, one on the rotatinghub, and one on the stationary shaft, is sealed to themounting gland by a rubber ring (toric, or o-ring),which also acts to allow a relatively generous hub-to-shaft misalignment tolerance, and as a spring toload one seal ring against the other (see Figure 7).Although proven effective over 30 years of hard rockdisc cutter application, there are problems associatedwith the seals in some specific applications.

One of the problems encountered with this sealis when a slurry of certain rock types pushes throughthe labyrinth created by the hub and seal retainersand then, given time, dries, effectively cementingone metal face seal ring to the other. When the cutterstarts rolling again, the cemented metal rings maymomentarily spin with each other and stretch therubber toric holding the metal seal rings. This canrip the toric, allowing leakage of lubricant out of thecutter, as well as ingress of dirt and abrasives intothe bearing cavity. To alleviate this problem, thetoric should minimize the amount of metal ringexposed to the slurry, hopefully allowing the rings tobreak free and rotate against each other as intended.

Another problem arises when the cutters areused in high pressure applications, such as on EPBand slurry machines. The standard seal is rated, perthe manufacturers’ recommendations, for about 3 barof pressure differential between the atmosphere andthe bearing cavity. Three bar of pressure is reached at30 meters below the water table and many tunnels arefar deeper. The problem that arises with higher pres-sure is the fluidized slurry pushes against the toric,forcing it down the seal gland ramps. This forces themetal rings tighter against each other, increasing facecontact pressure of the seal surfaces, causing the sealrings to act as a brake. This has the effect of ‘lockingup’ the cutter, preventing it from rolling in softermaterial. In extreme cases, with the smaller diametertorics used in many older cutter designs of the 1970sand 1980s, a portion of the toric on the non-rotatingside of the seal can push past the back of the seal,causing a major leak (see Figure 8). This problem islessened considerably by use of larger diametertorics, such as those used today.

Various methods have been tried over the yearsto try to compensate the internal pressure in cuttersto match the exterior pressure, in order to overcomethe pressure differential problems noted in the para-graph above. The following is a list of some of themethods by which pressure compensation has beenattempted:

1. A pressure compensating piston in the center ofthe shaft, which moves with the increase inexterior pressure, causing the interior pressureto rise to an equal level. The problem with thismethod is plugging of the exterior side of thepressure compensator with muck, preventingthe piston from traveling in the opposite direc-tion, which can over-pressurize the interior,destroying the seals.

2. A diaphragm, in the shape of an annular ringmounted in the seal retainer, which moves as theatmosphere side pressure rises, causing the inter-nal pressure to rise. This, too, has problems with

Figure 7. Cross-sectional diagram showing cutter seal Figure 8. Seal toric displaced by fluidized

slurry at high pressures

43

the breather holes in the diaphragm’s protectivecover becoming easily clogged.

3. A spring-loaded piston, which can supplybetween 2 and 3 bar of pre-load pressure inter-nally, causing a maximum differential of 3 barwhen the exterior pressure is up to 6 bar. Thismethod was left open to atmosphere (through abreather) on one early version; however, it wasfound that closing the atmosphere side andallowing a partial vacuum to develop stillallowed the devise to work satisfactorily.

Methods 1 and 2 have been successful in appli-cations where the amount of fluid greatly exceedsthe amount of muck, thus having a flushing affect onthe exterior side of the pressure compensator. Wherethe muck is in more of a paste form, plugging of theexterior side of the devices is a major problem forthese two methods.

All of these types of cutters generally use greaseas the lubricant, and in most cases the cutters have apressure relief valve. It has been shown that without apressure relief valve, heat from normal operation cancause internal pressure to increase sufficiently topush the torics hard toward each other, locking thecutter rotation as the seal rings act as a brake (seeFigure 9).

An obvious, but as yet untried method of pres-sure compensation would be to have all of the cuttersplumbed together to a common lubricant supply.Pressure in the cutterhead chamber would be sensedand the cutter hydraulic system would automaticallyapply the proper internal pressure to the cuttersthrough the plumbing lines. The drawbacks to thismethod would be isolation of the lubricant duringcutter change and the danger of a leak anywhere in acommon lube system.

The overall goal is to develop either an effectivepressure compensation devise, or a modified/newseal design that uses the robust characteristics of themetal-to-metal seal currently in widespread use, butmake modifications that prevent pressure differentialfrom affecting seal operation.

To provide rapid development of an improvedEPB cutter, a test fixture was built that allows mount-ing either a single or twin disc cutter inside a pres-sure vessel (see Figure 10). Both the atmosphericside and cutter bearing cavity can have their pressureadjusted and monitored independently. The cuttercan be rotated during testing at any chosen rpm. Therolling torque of the cutter can be measured at vari-ous pressure differentials or temperatures. The tem-perature of the lubricant adjacent to the seal can bemonitored. The pressure (atmosphere and/or bearingcavity) can be raised to as much as 60 bar.

A variety of tests have already been run on bothstandard and modified seal sets. These tests haveshown the effect of various pressure compensationdevises. While new designs have not yet been final-ized, a better understanding of standard seal perfor-mance has been realized. Tests on slightly modifiedseals (material and surface finish changes) haveshown promising improvements are possible (atsome cost trade-off, naturally). While initial majorredesigns of seal configurations have been lesspromising so far, the test fixture has proven quitevaluable in efficiently reaching conclusions at aspeed which cannot be duplicated in the field.

CUTTER RING DEVELOPMENT

The Goal: Increase Advance Rates, Decrease Downtime

Cutter rings have frequently proven to be the limit-ing component for further improved cutter perfor-mance. Recent advancements in cutter technologyhave continually increased production while lower-ing overall cutter costs. Continual refinement in pro-prietary alloys and metallurgical processes hasimproved the hardness, abrasion resistance and most

Figure 9. Locking cutter rotation through internal pressure

Figure 10. Test fixture to examine cutter rolling torque at various pressure differentials and temperatures

44

importantly the toughness of today’s high perfor-mance Heavy Duty (HD) and Extra Heavy Duty(XHD) cutter rings. Cutter ring development hasimproved production in two ways:

• Higher penetration per revolution is possiblebecause the Heavy Duty cutter ring stayssharper longer than standard rings. Further-more, because the HD cutter ring is strongerthan standard material, narrower tip width ringscan be used without breaking.

• Longer cutter life means less down time forcutter changing and more time available forboring.

Hardness

The high contact stress seen by the highly loaded19-inch cutters has pushed metallurgical research tonew highs. Engineers quickly moved from commer-cial steels to proprietary steels in the search forhigher strength and toughness. In 1995, a MainBeam TBM was put on to the Midmar project inSouth Africa. This project included dolerites andsandstones with strengths to 350 MPa. The demandson rings were severe and resulted in another roundof materials and processing R&D. The result wasstate-of-the-art Heavy Duty rings, in use on morehard rock projects today than any other ring. Thisnew material improves the cutter life because it hashigh “hot strength.” The material has a reduced wearrate compared to the standard cutter when boringhard rock.

Toughness

It is intuitive to think that the key to creating a highperformance disc cutter is extreme hardness. This istrue to some degree but it is not the sole factor. Hard-ness is needed to retard deformation when the ring ispressed against rock but the hardness is useless if it istoo brittle, therefore toughness and hardness are themost important properties for the cutter designer tomanage. The standard toughness test for steels is theCharpy Impact test which consists of one blow from aswinging pendulum under defined, standard condi-tions. To increase toughness, sophisticated analysisand post processing methods have been used to moni-tor and control grain size, microstructure and chemi-cal composition. Continual refinement of the heattreating and tempering processes has proven to be ofequal importance to specifying the proper alloy if thegoal is to increase toughness and hardness simulta-neously. The toughness obtained in proprietary steelrings has allowed the manufacture of rings that areeven harder than was considered possible just a fewyears ago.

DEVELOPING TECHNOLOGY: REMOTE CUTTER STATUS MONITORING

The current focus in disc cutter design and applica-tion is finding the optimum cross point of disc cuttercost vs. boring system performance, measured inpenetration rate and availability. There are three basiccomponents to TBM downtime related to cutters:

1. Cutter inspection: Hard rock TBMs must beroutinely stopped to allow workers to inspectthe condition of all the cutters, to insure thatthere are no failed cutters or severely worn cut-ter rings, to insure that the TBM can continueto be operated until the regularly scheduled cut-ter change.

2. Cutter change: Routine maintenance encom-passing: Inspections for mechanical fitness andreplacement of cutters with worn rings or dam-aged bearing/seal assemblies, as well as movingcutters for even ring wear.

3. Cutterhead repairs: This is related to eitherextreme geological conditions (abrasive rock infault zones or blocky rock) or the failure of aseries of cutters, either of which can lead to cut-terhead damage. The work required may includerepair of cutter housings, cutterhead wear plat-ing, muck buckets and/or conveyor components.

Clearly, a system which constantly monitors the sta-tus of every cutter on the cutterhead would provide atremendous advantage. If one could monitor all thecutters at all times, it would not be necessary tomake routine physical inspections (#1 above), whichwould reduce TBM downtime. Furthermore, if itwas possible to constantly know the status of everycutter, then the operator would know immediatelywhen a cutter failed, or was beginning to fail, andstop the machine thereby preventing a wipeout cut-ter failure before its occurrence and preventing dam-age to the cutterhead.

Historically, it was entirely up to the TBMoperator to detect a failure by noting an increase incutterhead torque requirements, which is very diffi-cult to detect given a single cutter failure on amachine which may be fitted with 20 to 80 cutters.Some TBM operators have reported being able todetect the aroma of overheating cutter lubricant inthe tunnel from a very hot cutter—not a desirable ordependable failure detection system. TBM usershave long sought a cutter monitoring systemwhereby the operator would get more precise infor-mation on the status of each cutter. In the past, theonly way to get this data from the rotating cuttersand cutterhead to the operator was via a hard-wiredsystem with multiple contact rotating union, whichproved wholly unreliable in the harsh environment.

45

New Developments in Remote Cutter Status Monitoring

Today, engineers are developing real-time cutter mon-itoring systems with the following specifications:

• Monitoring of every cutter on the TBM. • Indication of smooth rolling of all cutters on an

human-machine interface (HMI) display in theTBM.

• Cutter rolling alarm, with cutter number, whenany cutter stops rolling

• Indication of cutter wear, for every cutter, on anHMI display

• Indication of cutter temperature, for every cut-ter, on an HMI display

• Cutter temperature alarm, with cutter number,when any cutter temperature exceeds the pre-set limit

• Wireless transmission of all data

Rotational Speed of the Cutter

Cutters are not individually powered. They roll onlybecause the cutterhead in which they are mountedpresses the cutters against the rock face as the cutter-head rotates. The diameter of their circular, concen-tric path is determined by how far they are mountedfrom the rotating cutterhead center. The rotationspeed of the cutter is a function of the radius of theirmounting position on the cutterhead, the rotationalspeed of the cutterhead and the diameter of the cut-ter ring, which changes as the cutter wears. Knowingthe rotating speed of each cutter will provide twopieces of information:

1. Whether the cutter is rolling or not rolling. Thisinformation helps us to detect catastrophicallyfailed cutters before the failure propagates toother cutters or causes cutterhead damage.

2. The diameter of the cutter, which gives the cut-ter ring wear. This information eliminates theneed for downtime for routine cutter inspec-tions, and helps to plan in advance for cutterchanges.

Temperature of the Cutter

The temperature of the cutter is also an indicator ofmechanical condition of a cutter. Unusually hightemperature can indicate slipping seals, failing bear-ings, loss of lubricant or ingress of foreign materials.The temperature of the entire cutterhead increasesduring a boring stroke. Other TBM mechanical prob-lems might be indicated by the temperature risingabove the average normal increase during a boringstroke. Results obtained from the first machines fittedwith this system will be presented in a future paper.

CONCLUSIONS

Industry investment in cutter development hasresulted in cutters today which efficiently borethrough an ever wider range of materials at ever lowercost in real, inflation-adjusted dollars. Improvementsin pressure compensation devices will improve thesurvivability of cutters operating under pressure. Con-tinuous, real-time cutter condition monitoring willbecome a standard in a very short time and will resultin saving scores of times the system cost.

Substantial challenges, however, remain. Wemust continue to fund metallurgical and materialsprocessing research to find the next cutter ring mate-rial, allowing rings to be loaded ever higher, increas-ing TBM advance rates in hard rock. We must searchfor more abrasion resistant materials and improvedsealing systems to improve cutter life on EPB andslurry machines.

Improvements in cutter technology haveresulted in lower total project costs for contractorsand owners worldwide. However, we must allremain aware of the need to continually fundresearch and development in this area if we hope tocontinue the trend of improved systems reliabilityand performance with reduced cost.

REFERENCES

ISO 281:2007. Roller bearings—Dynamic load rat-ings and rating life. (2007). Retrieved January 8,2008 from http://www.iso.org.

Stack, B. Handbook of Mining and TunnellingMachinery. (1982). New York: John Wiley andSons Ltd.