A conventional grinding process applied to ceramic materials often results in surface fracture damage, nullifying the benefits of advanced ceramic processing methods [7]. These defects are sensitive to grinding parameters and can significantly reduce the strength and reliability of the finished components. It is, therefore, important to reduce the depth of grain penetration to minimum values so that the grain force is below the critical level for structural damage. The critical value of grain penetration depth for a hard ceramic is typically less than 0.2 μm. This small value of grain penetration depth is made possible using the ELID grinding technique with fine grain wheels.
Although ELID grinding is good for workpiece accuracy, it is not necessarily beneficial to workpiece strength as the following discussion of the effects of removal rates demonstrates [8]. Stock removal rate increases with increasing number of passes, higher stock removal rates being obtained with a stiffer machine tool in the first few passes. For grinding wheels of a similar bond type, a larger stock removal rate is obtained for the wheels of larger grit sizes. Cast iron–bonded wheels used during the ELID grinding allow a larger stock removal rate, yet a lower grinding force than a vitrified bond grinding wheel used in a conventional grinding process. Machine stiffness has little effect on residual strength of ground silicon under multipass grinding conditions, which can be attributed to the effect of the actual wheel depth of cut on workpiece strength [9]. As the number of passes increases, the actual depth of cut approaches the set depth of cut, which means that regardless of machine tool stiffness, grinding force does not necessarily alter workpiece strength in a stable grinding process. Also, more compressive residual stress can be induced with a dull grinding wheel, with a grinding wheel of a larger grit size, or with a wheel with a stiff and strong bond material. However, a larger grinding-wheel grit size causes a greater depth of damage in the surface of the ground workpiece. As the number of passes increases, the normal grinding force also increases. This increase of force is steep initially and slows down as the number of passes increases, a phenomenon more evident for a high stiffness machine tool. Due to machine tool deflection, the normal grinding force is initially smaller with lower machine stiffness [9]. Eventually, the normal force approaches a limit value, regardless of the machine stiffness characteristics [9]. To avoid damage to the workpiece, it is necessary to limit the grain penetration depth, which is more directly dependent on the removal rate than on the grinding force.
A controversial and little studied aspect of the ceramic grinding process is the pulverization phenomenon that takes place in the surface layer of a ceramic workpiece during grinding [10]. Surface pulverization makes ceramic grains in the surface much smaller than those in the bulk and gives the ground surface a smoother appearance.
7.7. Material removal mechanisms in grinding of ceramics and glasses
In general, two approaches in investigating abrasive-workpiece interactions in ceramics grinding [7]:
1.
The indentation-fracture mechanics approach models abrasive-workpiece interactions by the idealized flaw system and deformation produced by an indenter.
2.
The machining approach involves measurement of forces coupled with scanning electron microscope (SEM) and atomic force microscope (AFM) observation of surface topography and of grinding debris.
The stock removal during grinding of ceramics is a combination of microbrittle fracture and quasi-plastic cutting. The quasi-plastic cutting mechanism, typically referred to as ductile-mode grinding, depicted inFigure 7.8, results in grooves on the surface that are relatively smooth in appearance. Through careful choice of values for the grinding parameters and control of the process, ceramics can be ground predominantly in this so-called ductile mode. On the other hand, brittle-mode grinding shown in Figure 7.9 results in surface fracture and surface fragmentation. Ductile-mode grinding is preferred because negligible grinding flaws are introduced and structural strength is maintained.
Figure 7.9.
An abrasive grain depicted removing material from a brittle workpiece: (a) in ductile-mode grinding and (b) in brittle-mode grinding[7]
Figure options
As shown in Figure 7.9, a plastically deformed zone is positioned directly under the grit. In brittle-mode grinding, two principal crack systems are generated: median (radial) cracks and lateral cracks. Brittle-mode removal of material is due to the formation and propagation of these lateral cracks.
The specific depth at which a brittle-ductile transition occurs is a function of the intrinsic material properties of plasticity and fracture. According to Bandyopadhyay, the critical depth is [7]:
equation(7.1)
Turn MathJaxon
where d is the critical depth of cut.
Although it is not always easy to observe microcracks produced by grinding, the depth of a median crack can be estimated using the following formula [11]:
equation(7.2)
Turn MathJaxon
where ψ is the indenter angle, F is the indentation load, E is the modulus of elasticity, and Kc is the fracture toughness of the material. Therefore, the depth of the median crack depends on the material properties, force, and grinding grit shape.
Indentation load F is estimated by dividing the grinding force by the number of active cutting edges in the contact area between the grinding wheel and the workpiece. This relationship applies above a threshold force F*. The critical force F* that will initiate a crack can be estimated by:
equation(7.3)
Turn MathJaxon
where α is a coefficient that depends on the indenter geometry, and H is the hardness.
In conclusion, crack size can be estimated theoretically, from equation (7.2) when the force exceeds a certain critical value, determined by equation (7.3). Typical values of critical force to propagate subsurface damage are presented in Table 7.1[7]. In order to cut in the presence of plastic deformation, the grain load should be less than 0.2 N and 0.7 N for SiC and Si3N4, respectively, for a typical grain shape.
Table 7.1.
Critical Force Required to Propagate Subsurface Damage [7]
Materials H (GPa) E (GPa) Kc (kN/m3/2) F* (N)
SiC 24.5 392 3400 0.2
Si3N4 14 294 3100 0.73
Table options
Scanning electron microscope (SEM) and atomic force microscope (AFM) techniques can be utilized to evaluate surface and subsurface fracture damage. Typical micrographs are shown in Figure 7.10 and Figure 7.11. From SEM and AFM micrographs, one can assess the difference between brittle mode and ductile mode material removal.
Figure 7.10.
SEM micrographs [7]: (a) #325 and (b) #8000
Figure options
Figure 7.11.
AFM micrographs [7]: (a) #325 and (b) #8000
Figure options
7.8. Comparison between ELID and other grinding techniques
Significant reduction in grinding force has been reported with the application of ELID for workpieces ground both in the longitudinal and transverse directions.
ELID Grinding
Protruding grains abrade the workpiece. As a result, the grains and the oxide layer wear down. The wear of the oxide layer increases the electrical conductivity of the wheel. As a result, the electrical current in the circuit increases, leading to an intensification of the electrolysis. Consequently, protrusion of the abrasive grains increases and, during a short period, the thickness of the oxide layer recovers. The described electrical behavior is nonlinear, due to the formation of this insulating oxide layer. The oxide layer has a beneficial lubricating role in the grinding process. The process of wear and recovery of the oxide layer follows in a rather stable manner during the entire ELID grinding operation.
Other In-Process Dressing Technologies
Nakagawa and Suzuki investigated various techniques of in-process dressing [12]. The effects of in-process dressing using a dressing stick were studied. The wheel is dressed at the beginning of each stroke. Higher material removal rates were reported. An application of this procedure to side-grinding is difficult. However, use of a dressing stick accelerates wear of the superabrasive grains.
The technique of electrochemical dressing was introduced by McGeough in 1974 [13]. An electroconductive metal-bond wheel forms the anode and a fixed graphite stick forms the cathode. The dressing process takes place by electrolysis. Welch and others employed sodium chloride electrolyte [14]. However, sodium chloride is corrosive and is therefore harmful to the machine tool.
Another dressing technique is based on the principle of electrical discharge machining (EDM). The conductive grinding wheel is energized with a pulsed current. The flow of ions creates hydrogen bubbles in the coolant, creating an increasing electric potential. When the potential becomes critical, a spark is generated that melts and erodes the material that clogs the wheel. This procedure does not continuously provide protruding abrasive grains, being considered unsuitable for ultrafine grinding of materials, especially with micrograin-size grinding wheels.
Other nonconventional machining processes based on electrochemical metal removal include electrochemical machining, electrochemical grinding, and electrochemical polishing.
7.9. Applications of ELID grinding
ELID grinding has been investigated for various materials, including ceramics, hard steels, ceramic glass, and ceramic coatings, having a variety of shapes (plane, cylindrical external and internal, spherical and aspherical lens, etc.) and of greatly varying dimensions. Lately, new applications of ELID principle were tested for biomedical materials [15] and [16].
Recent investigations reported include:
•
ELID-side grinding [4], [8] and [9]
•
ELID double-side grinding [17]
•
ELID-lap grinding [18] and [19]
•
ELID grinding o