When postforming machining operations are required on high-strength structural components, tool life becomes a costly issue, often requiring external softening via techniques such as laser assistance for press-hardened steel components. Electrically assisted manufacturing (EAM) uses electricity during material removal processes to reduce cutting loads through thermal softening. This paper evaluates the effect of electric current on a drilling process, termed electroplastic drilling, through the metrics of axial force, and workpiece temperature when machining mild low carbon steel (1008CR steel) and an advanced high strength press hardened steel. A design of experiment (DoE) is conducted on 1008CR steel to determine primary process parameter effects; it is found that electricity can reduce cutting loads at the cost of an increased workpiece temperature. The knowledge generated from the DoE is applied to the advanced high strength steel to evaluate cutting force reduction, process time savings, and tool life improvement at elevated feedrates. It is found that force can be reduced by 50% in high feedrates without observing catastrophic tool failure for up to ten cuts, while tool failure occurs in only a single cut for the no-current condition. Finally, the limitations of the developed model in electroplastic drilling are discussed along with future suggestions for industrialization of the method.

## Introduction

The effect of electricity on deformation was first shown in 1959 when Machlin passed electricity through salt crystals and observed improvement in the crystals' ductility [1]. This work laid the foundation of the field which is known as electrically assisted manufacturing (EAM). The effect of electricity on deformation was later termed the electroplastic effect. The electroplastic effect has been shown to increase formability through increasing ductility, especially in brittle metals and reducing forming loads and springback in sheet forming [211]. In an effort to review the current state of the art, Nguyen-Tran et al. [12] summarized different EAM processes and suggested different areas of potential improvement in modeling. Ruszkiewicz et al. reviewed the existing theories of electroplastic effect and different modeling approaches in their work and suggested that the electroplastic effect was based on microscale thermal softening [13].

Most research on the electroplastic effect is focused on low strain rate operations, typically not higher than 10 s−1, except to the best of our knowledge for one article which showed an absence of the electroplastic effect at a strain rate of 1000 s−1 for both 304 stainless and Ti–6Al–4V [14]. This work was done at high strain rates in tension; therefore, the resultant test time was short. If the electroplastic effect is a thermal phenomenon, then, the short test times may not result in enough Joule heating to activate substantial thermal softening. However, drilling offers the possibility of studying the electroplastic effect at a high strain rate but at an elongated testing time, up to 12 s in this work.

Only a few papers exist on the topic of electrically assisted machining. The first work was a theoretical study of an electroplastic drilling process [15]. It was proposed that an electroplastic drilling process could reduce the friction force by upward of 30%, a number which is conservative compared to existing experimental papers. Jones et al. [16] showed that cutting forces could be reduced by up to 60% in orthogonal cutting of A2 steel with a continuous current ranging between 800 and 900 A. Similar results were found by Ulutan et al. [17] in electroplastic turning of grade 5 titanium and Inconel 718 where up to 70% force reduction was achieved. However, they showed that if enough electric current was applied to the workpiece, the cutting force would increase compared to the no electrical assistance condition. Egea et al. [18] were the first researchers to test a pulsed current (a technique typically used in tensile deformation to avoid overheating a necking region) in electrically assisted machining. A 90A square wave with a pulse duration ranging from 50 to 200 ms and a frequency of 100–300 Hz was applied to SAE 1020, 1045, and 4140 steels. It was found that the specific cutting energy and hardness decrease accompany an increase in surface quality.

The objective of this work is to offer the first in-depth experimental investigation of electroplastic drilling through the use of a generalized factorial design of experiments (DoE) study conducted on low-carbon 1008CR steel. The knowledge from this DoE is applied to high feedrate drilling of 1500 MPa press hardened steel (PHS1500) often used in the automotive industry, where extreme strength is required (e.g., B-pillar inner layer material). A simple force and temperature prediction model based on Merchant's machining model [19] for electroplastic drilling are presented, and its performance evaluated in different drilling conditions to show its limitations and areas for future improvement.

This paper has three major sections with the following experimental setup:

1. (1)

design of experiment to understand process parameter effect on an electroplastic drilling model of 1008CR steel;

2. (2)

electroplastic drilling model formulation and evaluation; and

3. (3)

elevated feedrate and current electroplastic drilling of PHS1500 and tool life evaluation.

## Experimental Setup

All testing in this work was conducted using 6.35 mm twist drill bits. Black oxide steel bits with 135 deg point angle were used for drilling 1008CR steel with workpiece dimension of 31.75 × 63.5 × 1.4 mm. Tungsten carbide-tipped steel bits with a point angle of 117 deg were used for drilling PHS1500 steel (since the black oxide is unable to cut the PHS1500). The top surface of each workpiece was painted black prior to cutting to replicate a black body for temperature acquisition via a FLIR A40 thermal camera, sampling at 12–25 Hz. A knee mill was used with a servomotor to control feedrate during the drilling process. Electricity was applied via a Darrah 4 kA power supply controlled through LabVIEW software.

The axial force was recorded using a 1 kN interface loadcell. The electricity was started and stopped using a load trigger. When the force from the load cell was greater than 66 N for 1008 steel or 333 N for PHS1500, electricity was applied, and when the force fell below these values, the electricity was stopped. A higher force was used for the PHS1500 to ensure sufficient tool contact area to avoid arcing for the steeper tipped tool and stronger material. Temperature and force data were filtered using a ten-point moving average prior to plotting and analysis. The experimental setup is shown in Fig. 1.

Fig. 1
Fig. 1
Close modal

### 1008cr STEEL Design of Experiment.

A four-factor generalized DoE was used to determine the impact of the input process parameters on electrically assisted drilling of 1008CR steel. The inputs used were: spindle speed (RPM), feedrate (mm/min), current (A), and the number of holes made. Both current and number of holes have three levels for their given factors, while both spindle speed and feedrate have two levels, shown in Table 1. The DoE resulted in 36 individual parameter sets, each replicated three times, for a total of 108 tests. The outputs studied were maximum part temperature (°C) and maximum axial force (N).

Table 1

Inputs and outputs for DoE for 1008CR steel

FactorsLevel 1Level 2Level 3
Feedrate (mm/min)12.525.4
Current (A)0150300
Spindle RPM350560
Number of holes123
Experimental outputs
Maximum temperature (°C)
Maximum axial drilling force (N)
Average flank wear (mm)
FactorsLevel 1Level 2Level 3
Feedrate (mm/min)12.525.4
Current (A)0150300
Spindle RPM350560
Number of holes123
Experimental outputs
Maximum temperature (°C)
Maximum axial drilling force (N)
Average flank wear (mm)

#### Axial Force.

The primary effects plot (see Fig. 2), shows that electric current level has the strongest impact on the axial force observed during the drilling process of 1008CR Steel. However, the effect of current magnitude shows a nonlinear behavior. At a current of 150 A, a force reduction is resultant from electrical softening of the base material, while at a current of 300 A, a large axial force increase of 41% is observed compared to the no-current condition. This is caused by arcing at the initial application of electricity, suggesting that the 66 N preload was not sufficient for the 300 A first cut test. The arcing dulls the tool by removing the chisel edge (top cutting edge) as well as damaging the helix cutting edge, which can be seen in the tool wear images shown in Table 2. The underlying reason for arcing can be explained by considering the voltage of the (current driving) power supply, which is determined based on the requested current. Therefore, for each of the 300 A tests, a higher voltage was supplied than the 150 A tests, which gives a greater potential to overcome air and contact resistance and therefore induce arcing.

Fig. 2
Fig. 2
Close modal
Table 2

Tool wear images for electrically assisted drilling of 1008CR steel at 350 RPM and 12.7 mm/min

Electric current
Cut number0150300
1
2
3
Electric current
Cut number0150300
1
2
3

Increasing feedrate will increase the cutting force, while increasing the rotational surface speed will decrease the cutting force. This can be explained by examining the shear area (discussed in the modeling section of this paper), where increasing the RPM decreases the chip thickness, resulting in a smaller shear area. Increasing feedrate will increase the chip thickness and higher force will be observed as more material is removed per rotation of the tool. Increasing the number of cuts per tool increases the forces as the cutting edges start to wear out during each cut. An example is shown in Fig. 3 for a 300 A test. The temperature is the highest in the first cut since the tool is still sharp, leading to a smaller contact area and higher current density, coupled with arcing at the initial application of current.

Fig. 3
Fig. 3
Close modal

#### Maximum Temperature.

Similar to force, current is the dominant input parameter affecting maximum temperature, as shown in Fig. 4. The effect is nonlinear, as current is increased from 0 A to 150 A, an average temperature increase of 100 °C is observed. However, when the current is raised to 300 A, the temperature rises to above 500 °C. A large portion of the nonlinearity is likely due to arcing in the 300 A tests. The difference in temperatures between parameter sets is shown for the first cut comparisons in Fig. 5. Without electricity, feedrate appears as the dominant heating parameter (0 A test). However, with current, higher temperatures is observed from lower feedrates due to smaller shear area resulting in a higher current density. It is worth mentioning that when the current is present, all other parameters have a negligible effect on the process temperature.

Fig. 4
Fig. 4
Close modal
Fig. 5
Fig. 5
Close modal

### Electoplastic Drilling Model Formulation and Evaluation.

Merchant's model [19] is used to compute axial cutting forces for comparison with experiment. The shear force is computed using the Johnson–Cook plasticity model, which is strain rate and temperature dependent. Strain rate is computed through the material models found in the literature [20], and temperature is computed via a two-dimensional (2D) finite volume heat transfer model. The formulation and integration of these models are described as follows.

To predict temperature a 2D explicit finite volume heat transfer model was constructed with a time-step of 0.01 s, less than the required 0.0375 s step time for stability calculated using the Fourier number. A nodal spacing of 1.5 mm is used in both the x and y directions to mesh a workpiece with dimensions of 63.5 × 31.8 mm, as shown in Fig. 6. Convection conditions are placed on the top face and outside width edges of the workpiece, and a convection coefficient of 25 W/m2 K is applied, while the bottom and length edges of the workpiece are insulated due to contact with a Delrin plastic fixture. An energy balance is used to calculate the temperature of the workpiece, with a point heat source composed of three heating terms applied at the center of the workpiece.

Fig. 6
Fig. 6
Close modal
The three sources of heat are shear deformation, friction, and Joule heating. The Joule heat energy (Qelec) added to the system is given by Eq. (1), where I is the electric current, t is the time, T is the temperature, and R is the temperature-dependent electrical resistance given in Eq. (2), where $ρe$ is the electrical resistivity, L is the conduction length which is a length that electricity flows through for a given cross-sectional area assumed equal to the shear area, Ashear
$Qelec=I2R(T)t$
(1)
$R(T)=ρE(T)LAshear(t)$
(2)
The frictional heating energy (Qfriction) model is shown in Eq. (3), where μ is the coefficient of friction, τFSS is the fracture shear stress, r is radius of the tool in contact with the workpiece, λ is the percentage of heat that goes into the workpiece (assumed 50%), and ω is the angular velocity.
$Qfriction=λμτFSS(T)ωrAsheart$
(3)
The shear deformation energy (Qshear) model is shown in Eq. (4), where Fs is the shear force, and Vs is the shear velocity. The shear velocity is a function of average velocity (v) of the shear face found through the midpoint radius multiplied by the angular velocity and the shear plane angle (ϕ), as in Eq. (5). The shear plane angle is a function of friction angle (β) and the rake angle (α) assumed to be the helix angle for the drill bit, as shown Eq. (6). The friction angle (β) is found using Eq. (7). The shear force (Fs) is found using the fracture shear stress (τFSS) and the shear area (As), Eq. (8). The fracture shear stress (τFSS) is assumed equal to the half of the calculated fracture tensile stress (σFTS) using the Johnson–Cook plasticity model using Eq. (9). Where A is the yield stress, B is the strength coefficient, $εfracture$ is the fracture strain, n is the strain-hardening exponent, C is the strain rate sensitivity coefficient, $ε˙$ is the strain rate, $ε0˙$ is the reference strain rate, T* is the homologous temperature, and m is the temperature sensitivity exponent. The Johnson–Cook parameters for mild steel from the literature are given in Table 3 [21]
$Qshear=λFsVst$
(4)
$Vs=vcosϕ$
(5)
$ϕ=90+α−β2$
(6)
$β=arctan(μ)$
(7)
$Fst=τFSSAshear(t)$
(8)
$σUTS=A+Bεfracturen1+Clnε˙ε0˙1−T*m$
(9)
Table 3

Johnson–Cook plasticity model parameters for mild steel [21]

ParameterValue (units)
A217 MPa
B233 MPa
C0.0756
n0.6428
m1
ParameterValue (units)
A217 MPa
B233 MPa
C0.0756
n0.6428
m1
The shear strain rate ($γ˙$) was calculated using Eq. (10), where ah is the hyperbola curvature constant found using Eq. (11), with Csr as a material constant [20]
$γ˙=v(4ahsin2ϕtanα+cotϕ1.5)$
(10)
$ah=t12(16Csr2sin4ϕtanα+cotϕ)$
(11)
The shear face area is found using Eq. (12), which represents the area of contact between the bit's cutting edges and the workpiece [22,23], where a is the uncut chip thickness, D is the tool diameter in contact with the workpiece at a given time, d0 is the chisel edge diameter, and φ is half of the bit point angle. The chip thickness is found using Eq. (13), where f is the feed (mm/rev)
$Ashear(t)=aDt−d0cos90−φsinϕ$
(12)
$a=fsinφ2$
(13)

This paper examines the drilling of thin sheets (1.5 mm) as such the contact between the bit and workpiece is continually changing and can be broken into four segments for the given bit and workpiece thickness combination, which is outlined in Fig. 7. The first segment is when the tapered portion of the bit is in contact with the workpiece but the shoulder has not yet touched, the second segment has shoulder contact but the tip of the bit has not yet left the workpiece, the third segment is when the tip of the bit leaves the bottom of the workpiece but the shoulder continues cutting, and the fourth segment is the end of the cut when the bit shoulder penetrates through the bottom of the sheet. If the tapered portion of the bit is longer than the thickness of the sheet, then, the second segment would be skipped.

Fig. 7
Fig. 7
Close modal
The diameter in contact with the workpiece for the first segment is shown in Eq. (14), where fr is the feedrate. The second segment uses the nominal bit diameter since the shoulder and cutting edges are both fully in contact. The third segment subtracts the protruding diameter from the nominal diameter. The protruding diameter is calculated using Eq. (15), where thick is the thickness of the workpiece. Finally, the fourth segment has no diameter contact as the cut is complete
$Dt=2rt=2frt*tanφ$
(14)
$Dprotrude=2rprotrude=2frt−thicktanφ$
(15)
The axial force relation, shown in Eq. (16), is calculated using angular relations of the bit and shear zone from Merchant's model [19] with the shear force calculated plugged in from Eq. (8)
$Faxial=Fshearcos(β−α+ϕ)$
(16)

The assumptions for solving the model are listed below:

• Constant friction coefficient, not a function of temperature, pressure, or velocity.

• Conduction length of electricity remains constant.

• Electricity flows uniformly through the shear area.

• 50% of heat generated by mechanical shearing and friction goes to the workpiece and the other 50% to the tool, and steel on steel contact.

• 100% of electricity goes to Joule heating.

• Fracture shear stress can be approximated as 50% of the fracture tensile stress for 1008CR steel.

• Hyperbola shape constant, Csr = 6.

• Rake angle = helix angle = 30 deg.

• Point angle = 135 deg, based on manufacturer specification.

• No tool deformation and no tool wear.

The model has the following tuneable parameters:

1. (1)

Friction coefficient found to be 0.6 by fitting model to 0A tests.

2. (2)

Fracture strain varied between 1 and 9 for Johnson–Cook model, fit to 0 A tests, found to be within 1–10 range given by Ref. [24].

3. (3)

Conduction length of electricity at tool–workpiece interface assumed to be 0.5 mm through the shear area, derived based on fitting model to 150 A tests since the 300 A experienced arcing resulting in a damaged tool early in the drilling process.

#### Electroplastic Drilling Model Evaluation, Limitations, and Suggestions.

The formulated model is used to predict temperature and axial force during the first cut of an electrically assisted drilling process. Comparisons between model and experiment are shown for each parameter set on the first cut for both axial force and temperature as listed in Tables 4 and 5, respectively. The temperature and axial force models' prediction is within 20% of experimental results of the 0 A and 150 A tests. However, the model deviates significantly for the 300 A tests, due to arcing. Arcing occurs at the onset of electric current and drastically changes the contact area between the workpiece and tool, as shown in column 3 in Table 2. The model assumes an undeformed tool, but the 300 A cases are severely deformed tools, even for the first cut. This deviation leads to a greater contact area between tool and workpiece in the experiment than predicted in the model. This resulted in a current density in the model higher than experiment, leading to a higher predicted temperature, correlated to a low axial force through a low material strength from the Johnson–Cook model. A large deviation between the shape of the model and experimental curves are shown in Fig. 8. This is from the electrical augmentation of the knee mill, which requires an electrical contact carrier to be held against the spindle using springs. The fixture slides along four guide rails but has a large copper cable connected to one side of the fixture resulting in eccentric rotation which transfers to the bit potentially causing the abnormal behavior. A comparison between the contour temperature plot results from the temperature prediction model and thermal camera data collection is shown in Fig. 9. The given image is shown at the end of drilling process, and it is found that the contours and magnitudes from the model are reasonable and resemble the experiment.

Fig. 8
Fig. 8
Close modal
Fig. 9
Fig. 9
Close modal
Table 4

Model accuracy results for axial force during drilling of 1008CR steel across all parameter sets, underlined tests experienced arcing

Current (A)RPMFeedrate (mm/min)Average max force (N)Predicted max force (N)Percent error (%)
035012.72452354
035025.43123043
15035012.719021715
15035025.426029514
30035012.73317677
30035025.434811068
056012.71741599
056025.42642438
15056012.717414716
15056025.42422363
30056012.74004289
30056025.428912059
Current (A)RPMFeedrate (mm/min)Average max force (N)Predicted max force (N)Percent error (%)
035012.72452354
035025.43123043
15035012.719021715
15035025.426029514
30035012.73317677
30035025.434811068
056012.71741599
056025.42642438
15056012.717414716
15056025.42422363
30056012.74004289
30056025.428912059
Table 5

Model accuracy results for maximum temperature during drilling of 1008CR steel across all parameter sets, underlined tests experienced arcing

Current (A)RPMFeedrate (mm/min)Average max temp (°C)Predicted max temp (° C)Percent error (%)
035012.7961037
035025.41181171
15035012.72031983
15035025.41661538
30035012.7578104481
30035025.4463940103
056012.79311018
056025.41351468
15056012.72272098
15056025.41801832
30056012.75801259117
30056025.44871064119
Current (A)RPMFeedrate (mm/min)Average max temp (°C)Predicted max temp (° C)Percent error (%)
035012.7961037
035025.41181171
15035012.72031983
15035025.41661538
30035012.7578104481
30035025.4463940103
056012.79311018
056025.41351468
15056012.72272098
15056025.41801832
30056012.75801259117
30056025.44871064119

Most models are created with the potential promise of being used as predictors such that less experimentation is necessary to determine process parameters for a given process, in this case electroplastic drilling. However, this proposed model cannot be used for prediction of process parameters and was unable to predict force and temperature for a parameter set outside of the tested parameter range used in this paper, even without arcing. There are three main reasons conjectured, attributed to limitations in knowledge and technology that require further advancement before the proposed model could be used as a process output predictor.

1. (1)

Friction modeling: The current model took a simplified approach to friction and assumes a constant friction without dependency on temperature and pressure. While some temperature and pressure friction models exist, they are empirical in nature. The interaction between electricity and friction has not yet been studied, making it difficult to use existing empirical models with any degree of certainty. In addition, a slight change to the friction coefficient has a significant effect on the model output since the shear plane angle which is used to calculate shear area, and axial force is a function of friction angle which is derived from the friction coefficient, making the model highly sensitive to friction modeling.

2. (2)

Temperature measurement and model fitting (once again an issue with friction): The resolution of the thermal camera used in this work plays an important role in the accuracy of the temperature reading (see Fig. 9) compared to what the true maximum temperature is. The thermal camera works by dividing the given picture into a series of small regions, each captured by a pixel with average temperature of each as measurement results. However, even a high resolution thermal camera cannot clearly resolve the temperature at the tool–material interface, which essentially introduces a temperature reading error in the model. This error subsequently affects curve-fitting parameters for the drilling model. It is worth noting that the temperature results can be used to compare different tests but finding the true maximum temperature in the drilling process is prone to error.

3. (3)

Electrical conduction length, path, and current density: In this work, electricity flows through the tool, then into the workpiece, and grounds at an aluminum ring underneath the workpiece. A constant conduction length value was assumed for simplifying the model; however, it will continually change. Determining the conduction length and resultant current density with good degree of accuracy would require a finite element simulation. However, this too presents challenges. To the best of our knowledge, the commercialized finite element explicit solvers do not support thermal–electrical–structural elements. Explicit solvers are traditionally used for machining simulation or high strain rate processes.

#### Conclusions From 1008CR Design of Experiment and Model Formulation

• Electric current applied to a drilling process may result in reduction of cutting forces (150 A) though increasing in process temperature was observed. Due to the low initial strength of 1008CR steel, the force reduction was only in the range of 20%.

• The increase in cutting force observed at 300 A is likely caused by two phenomena. First, is the arcing observed at the initial application of current. Second, is the thermal softening of both the tool and the workpiece leading to smearing of the cutting edges. A tool material with a higher working temperature than the workpiece should result in greater electroplastic drilling benefits.

• An electroplastic drilling model, the first of its kind to attempt to predict cutting forces and temperatures in the presence of electricity was formulated and evaluated. The model formulation highlights the need for advancement in the fields of friction modeling, electrical conduction length modeling, and in situ temperature measurement to reach its predictive potential. For these reasons and the lack of a published material model for PHS1500 high strength steel, the model will not be applied to the remainder of this paper.

### Elevated Feedrate Drilling of Press-Hardened Steel (PHS1500).

This section examines the feasibility of augmented drilling in aggressive cutting conditions to determine the effect of electricity on tool life. This is done to evaluate the potential of electric augmentation to overcome limitations of traditional drilling through a process time and feedrate study in drilling high strength steel. PHS1500 steel was used and was prepared using the setup as explained in the Experimental Setup section.

A DoE study similar to what was used for 1008CR steel is conducted with input parameters shown in Table 6. Spindle speed is held constant at 560 RPM, and current is varied between 0 A and 600 A. Feedrates of 50.8 and 101.6 mm/min are used, and three holes are made per bit with two replications per parameter set. Fewer repetitions than the 1008CR steel study are used due to a limited supply of PHS1500. Following the DoE, each parameter set was tested until bit failure to determine potential tool life increase in the presence of electric current.

Table 6

Design of experiment for PHS1500 steel

FactorLevel 1Level 2Level 3
Feedrate (mm/min)50.8101.6
Current (A)0300600
Spindle (RPM)560
Number of holes123
Experimental outputs
Maximum temperature (°C)
Maximum axial drilling force (N)
Average flank wear (mm)
FactorLevel 1Level 2Level 3
Feedrate (mm/min)50.8101.6
Current (A)0300600
Spindle (RPM)560
Number of holes123
Experimental outputs
Maximum temperature (°C)
Maximum axial drilling force (N)
Average flank wear (mm)

In the 0 A test in both 50.8 and 101.6 mm/min feedrates, catastrophic failure of the bit was observed within the three cuts, resulting in unreliable force data (following tool failure, the limit of the loadcell was reached). As such, the output of the DoE is set to two current values (300 and 600 A) to allow for proper evaluation of the effect of electricity.

As shown in tool flank face images in Table 7, the 0 A case bits are broken. However, in the 300 A test, drill bits showed near sharp condition after three cuts. In the 600 A test drill bits experienced arcing and softening of the chisel edge, but it could retain sharp cutting edges.

Table 7

Tooling images after three cuts in drilling PHS1500 steel for two different feedrates and three different currents

Electric current (A)
Feed (mm/min)0300600
50.8
101.6
Electric current (A)
Feed (mm/min)0300600
50.8
101.6

The main effects plot for maximum axial force during electroplastic drilling of PHS1500 is shown in Fig. 10. The feedrate and number of cuts show trends similar to the 1008CR steel, as in Fig. 2. However, 1008CR steel had 150 N difference between the mean results of all parameter sets, while PHS1500 has a 500 N difference, showing that the effect of electric current is much greater on the PHS1500. The 1008CR steel had a force rise at 300 A, while for both 300 A and 600 A tests the force decreases for PHS1500. This is due to the strength of the steels; PHS1500 is roughly five times stronger than 1008CR. Therefore, increasing the temperature through increased current will have a greater effect. In addition, the WC-tipped tool has a higher working temperature than the black oxide steel tool, allowing for greater temperatures without significant softening. However, excessive temperature can lead to thermal expansion of the steel portion of the bit, allowing the carbide tip to separate, resulting in tool failure.

Fig. 10
Fig. 10
Close modal

The experimental results at 50.8 and 101.6 mm/min tests at various electric current inputs for cuts 1, 2, and 3 are listed in Table 8 and shown for cuts 1 and 3 in Fig. 11. The large force drops for the 0 A cases indicate where the tool failed. The 101.6 mm/min feedrate at 0 A and 300 A reached the maximum of the loadcell, resulting in the plateau. The oscillation in the beginning of the cutting process in the 600 A test of Fig. 11 is caused by thermal softening and its resultant cutting force reduction. The loadcell reaches 333 N, triggering the electric current, which then causes the load to drop back below 333 N, turning the current off until the load exceeds 333 N again. The force reductions for the PHS1500 are much greater than 1008CR. At 600 A tests, the resultant axial force reduction compared to the 0 A case for 50.8 mm/min is 54%, 55%, 51% for the first to third cut, respectively. The force reduction for 101.6 mm/min is 47%, 44%, 51% for the first to third cut, respectively. While the 600 A case had the lowest cutting force, it also suffered more wear than the 300 A case due to arcing or excessive softening at the onset of electric current, which is shown in Table 7.

Fig. 11
Fig. 11
Close modal
Table 8

Maximum force and temperature results from high feedrate drilling of PHS1500

Axial force results (N)
Feedrate (mm/min)Current (A)Cut numberTrial 1Trial 2AverageSTD. dev.
50.8019931059102647
50.802Bit failure
50.803
101.60110591043105111
101.6021059105910590
101.603Bit failure
50.830018551001928103
50.8300285197891590
50.8300386399592994
101.630011059105910590
101.630021059105910590
101.630031059105910590
50.8600147145646311
50.8600245948647219
50.860034974944962
101.6600159464161833
101.6600264469466935
101.6600372766369545
Axial force results (N)
Feedrate (mm/min)Current (A)Cut numberTrial 1Trial 2AverageSTD. dev.
50.8019931059102647
50.802Bit failure
50.803
101.60110591043105111
101.6021059105910590
101.603Bit failure
50.830018551001928103
50.8300285197891590
50.8300386399592994
101.630011059105910590
101.630021059105910590
101.630031059105910590
50.8600147145646311
50.8600245948647219
50.860034974944962
101.6600159464161833
101.6600264469466935
101.6600372766369545
Temperature results (°C)
Feedrate (mm/min)Current (A)Cut #Trial 1Trial 2AverageSTD. dev.
50.80113715814815
50.802Bit failure
50.803
101.60126924425618
101.6022962852908
101.603Bit failure
50.8300128140434287
50.8300229741435583
50.8300330541636078
101.6300125937731884
101.6300227436732066
101.6300326837332174
50.860015805805800
50.860025805805800
50.860035805805800
101.660015805805800
101.660025785775771
101.660035805805800
Temperature results (°C)
Feedrate (mm/min)Current (A)Cut #Trial 1Trial 2AverageSTD. dev.
50.80113715814815
50.802Bit failure
50.803
101.60126924425618
101.6022962852908
101.603Bit failure
50.8300128140434287
50.8300229741435583
50.8300330541636078
101.6300125937731884
101.6300227436732066
101.6300326837332174
50.860015805805800
50.860025805805800
50.860035805805800
101.660015805805800
101.660025785775771
101.660035805805800

The main effects plot for maximum temperature is shown in Fig. 12, with the maximum temperature of each test in Table 8. The trends are similar to 1008CR steel, except that the difference between the means with respect to current is lower for PHS1500 with a spread of 250 °C versus a difference of 400 °C for 1008CR steel. This is caused by the differences in process time. Since the 1008CR steel was processed at 12.7 and 25.4 mm/min feedrates, it resulted in more time for Joule heating and a higher resultant temperature. In contrast, a higher feedrate used in drilling PHS1500 requires a higher electric current to achieve the same temperature. It may be possible to predict this if a current density could be calculated, though as shown in the modeling section, this presents some problems based on friction assumptions.

Fig. 12
Fig. 12
Close modal

The temperature results for both feedrates for the first and third cuts are shown in Fig. 13. There is a temperature increase from the first to the third cut for both feedrates. However, the difference is greater in the 50.8 mm/min test.

Fig. 13
Fig. 13
Close modal

To evaluate the potential for tool life savings, the first bit for each parameter set is tested until failure. In the 0 A tests for both 50.8 and 101.6 mm/min, the drilling bit reached catastrophic failure mode after three initial cuts, and therefore, are not tested further. In the 101.6 mm/min test, the drill bit fails after the second cut in both replications, while in the 50.8 mm/min test, the drill bit fails after the first cut in both replications. This is likely due to the vibration of the fixture coupled with the slower penetration of the lower feedrate. The vibration of the spindle from the connected large electrical leads can cause the bit to grab the part more than it would without the excess vibration, therefore leading to high stresses on the bit and early failure. The vibration of the bit caused by the electrical lead attached to the spindle can cause the bit to initially contact the part with the cutting edges rather than the tip of the bit. This shock loading of the bit can cause significant tool damage and early fracture. This is accelerated when the bit contacts a stronger steel. Higher feedrates force the bit into the material faster, reducing the chance and time that the cutting edges rather than the tip of the bit are in contact, early in the drilling process. This is evident as the 50.8 mm/min test with 0 A failed on the first cut, while the 101.6 mm/min test failed on the second cut. The results for three sequential cuts are shown in Fig. 14. The force increases with each cut on the 0 A and 600 A cases. However, at the 300 A test, no significant change observed in force between three sequential cuts.

Fig. 14
Fig. 14
Close modal

The force results from the run-until-failure testing are shown in Fig. 15. For the 101.6 mm/min tests, bits failed after five and four cuts in 300 A and 600 A, respectively. Electric current was only able to extend the life of the tools for the 101.6 mm/min case by three and two cuts for 300 A and 600 A, respectively. However, this does show that electricity has the potential to allow for higher feedrates in the presence of electric current as the electrically applied bits outlasted the 0 A bits. A higher current magnitude and load trigger may lead to more desirable results. Tool images following the endurance testing are shown in Table 9.

Fig. 15
Fig. 15
Close modal
Table 9:

Tooling after failure in PHS1500 testing, all bits are failed except for 600A at 50.8 mm/min. 300 A at 101.6 mm/min has chips and a large crack through the carbide portion of the bit.

Electric current (A)
Feed (mm/min)300600
50.8
101.6
Electric current (A)
Feed (mm/min)300600
50.8
101.6

In the 50.8 mm/min tests a significant tool life improvement in the presence of electric current was observed. The drill bit in 0 A tests failed during the first cut for both feedrates. However, in 300 A, eight cuts could be completed before tool failure. The drill bit in the 600 A lasted for ten cuts and did not fail. Due to material restrictions, the testing was stopped at ten cuts. Comparing the tool wear image for this case at ten cuts, in Table 9, to the wear images after three-cut wear shown in Table 7, there appears to be insignificant difference with respect to the tool wear area. The chisel edge is more worn, which resulted in an extra 100 N of force throughout the testing but failure was not observed. This shows that electric current has the ability to drastically improve tool life and allows for drilling at a higher feedrate than what is possible without an electric current.

## Conclusions

The objective of the paper was to examine the electroplastic drilling of mild 1008CR steel and high strength PHS1500 steel through experimentation and modeling. The following conclusions were found:

• Electric current has a larger effect on axial force and temperature than feedrate or spindle RPM on drilling.

• Electric current can result in a reduction of cutting forces and tool wear. The higher the strength of the base material, the higher the force reduction (up to 50% for PHS1500).

• Electric current allows for improved tool life and the ability to cut at higher feedrates. At 600 A, the bit lasted ten cuts in comparison with the 0 A tests where the tool failed at the first cut.

• Tungsten carbide (WC) bits can handle electric current better than traditional black oxide steel bits and allow for higher current magnitudes with less damage.

• An electroplastic drilling model, the first of its kind to the best of our knowledge, to predict cutting forces and temperatures in the presence of electricity was formulated and evaluated. The model formulation itself is sound but it requires advancement in the fields of friction modeling, electrical conduction length modeling, and temperature acquisition in machining to reach its predictive potential.

## Future Work

In the future, the authors will examine the effect of electroplastic drilling on the base material. This will include hardness, microstructure, and strength evaluation of the PHS1500 specimens used in this research.

The authors will attempt to control cutting forces to create a constant cutting force application using electricity. Force oscillation was produced when the load exceeded the trigger and electricity was applied, which then reduced the load back under the trigger requirement, turning off the electricity. This was repeated to created force oscillation which will be mitigated in future tests through control techniques such as bang-off-bang control to avoid this chatter. Using this same concept, it is also possible to create constant force drilling, though thicker parts than used in this paper are required.

## Acknowledgment

Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors(s) and do not necessarily reflect the views of the National Science Foundation. The authors would also like to thank Mr. Gary Lee Mathis for fabricating the fixtures used in this paper.

## Funding Data

• National Science Foundation Graduate Research Fellowship (DGE-1744593).

## Nomenclature

a =

chip thickness

A,B,C,m,n =

material constants for Johnson–Cook model

Ah =

hyperbola constant

Ashear =

shear face area

Csr =

material constant

D,d =

diameters

f =

feed

fr =

feedrate

Fs =

shear force

I =

electric current

L =

conduction length

Q =

energy

r =

R =

electrical resistance

t =

time

T =

temperature

v =

average velocity

Vs =

sliding velocity

α =

rake angle

β =

friction angle

$γ˙$ =

shear strain rate

ε =

strain

λ =

heat fraction

μ =

friction coefficient

ρe =

electrical resistivity

σ =

flow stress

τFSS =

fracture shear stress

φ =

point angle

Φ =

shear plane angle

ω =

angular velocity

## References

1.
Machlin
,
E.
,
1959
, “
Applied Voltage and the Plastic Properties of ‘Brittle’ Rock Salt
,”
J. Appl. Phys.
,
30
(
7
), pp.
1109
1110
.
2.
Jones
,
J. J.
,
Mears
,
L.
, and
Roth
,
J. T.
,
2012
, “
Electrically-Assisted Forming of Magnesium AZ31: Effect of Current Magnitude and Deformation Rate on Forgeability
,”
ASME J. Manuf. Sci. Eng.
,
134
(
3
), p.
034504
.
3.
Jones
,
J. J.
, and
Mears
,
L.
,
2011
, “
Constant Current Density Compression Behavior of 304 Stainless Steel and Ti–6Al–4V During Electrically-Assisted Forming
,”
ASME
Paper No. MSEC2011-50287.
4.
Bunget
,
C.
,
Salandro
,
W.
,
Mears
,
L.
, and
Roth
,
J. T.
,
2010
, “
Energy-Based Modeling of an Electrically-Assisted Forging Process
,”
Trans. North Am. Manuf. Res. Inst. SME
,
38
, pp.
647
654
5.
Salandro
,
W. A.
,
Bunget
,
C.
, and
Mears
,
L.
,
2010
, “
Modeling and Quantification of the Electroplastic Effect When Bending Stainless Steel Sheet Metal
,”
ASME
Paper No. MSEC2010-34043.
6.
Ruszkiewicz
,
B. J.
,
Scriva
,
C.
,
Reese
,
Z. C.
,
Nikhare
,
C. P.
,
Roth
,
J. T.
, and
Ragai
,
I.
,
2015
, “
Direct Electric Current Spot Treatment's Effect on Springback of 90 Degree Bent 2024-T3 Aluminum
,”
ASME
Paper No. MSEC2015-9433.
7.
Ruszkiewicz
,
B. J.
,
Roth
,
J. T.
, and
Johnson
,
D. H.
,
2015
, “
Locally Applied Direct Electric Current's Effect on Springback of 2024-T3 Aluminum After Single Point Incremental Forming
,”
ASME
Paper No. MSEC2015-9429.
8.
Ross
,
C. D.
,
Kronenberger
,
T. J.
, and
Roth
,
J. T.
,
2009
, “
Effect of DC on the Formability of Ti–6Al–4V
,”
ASME J. Eng. Mater. Technol.
,
131
(
3
), p.
031004
.
9.
Kim
,
M.
,
Vinh
,
N. T.
,
Yu
,
H.
,
Hong
,
S.
,
Lee
,
H.
,
Kim
,
M.
,
Han
,
H. N.
, and
Roth
,
J. T.
,
2014
, “
Effect of Electric Current Density on the Mechanical Property of Advanced High Strength Steels Under Quasi-Static Tensile Loads
,”
Int. J. Precis. Eng. Manuf.
,
15
(
6
), pp.
1207
1213
.
10.
Perkins
,
T. A.
,
Kronenberger
,
T. J.
, and
Roth
,
J. T.
,
2007
, “
Metallic Forging Using Electrical Flow as an Alternative to Warm/Hot Working
,”
ASME J. Manuf. Sci. Eng.
,
129
(
1
), pp.
84
94
.
11.
Roth
,
J.
,
Loker
,
I.
,
Mauck
,
D.
,
Warner
,
M.
,
Golovashchenko
,
S.
, and
Krause
,
A.
,
2008
, “
Enhanced Formability of 5754 Aluminum Sheet Metal Using Electric Pulsing
,”
Trans. North Am. Manuf. Res. Inst. SME
,
36
, pp.
405
412
.https://www.tib.eu/en/search/id/BLCP%3ACN070826031/Enhanced-Formability-of-5754-Aluminum-Sheet-Metal/
12.
Nguyen-Tran
,
H.
,
Oh
,
H.
,
Hong
,
S.
,
Han
,
H. N.
,
Cao
,
J.
,
Ahn
,
S.
, and
Chun
,
D.
,
2015
, “
A Review of Electrically-Assisted Manufacturing
,”
Int. J. Precis. Eng. Manuf.-Green Technol.
,
2
(
4
), pp.
365
376
.
13.
Ruszkiewicz
,
B. J.
,
Grimm
,
T.
,
Ragai
,
I.
,
Mears
,
L.
, and
Roth
,
J. T.
,
2017
, “
A Review of Electrically-Assisted Manufacturing With Emphasis on Modeling and Understanding of the Electroplastic Effect
,”
ASME J. Manuf. Sci. Eng.
,
139
(
11
), p.
110801
.
14.
Kinsey
,
B.
,
Cullen
,
G.
,
Jordan
,
A.
, and
Mates
,
S.
,
2013
, “
Investigation of Electroplastic Effect at High Deformation Rates for 304SS and Ti–6Al–4V
,”
CIRP Ann. Manuf. Technol.
,
62
(
1
), pp.
279
282
.
15.
Baranov
,
S.
,
Staschenko
,
V.
,
Sukhov
,
A.
,
Troitskiy
,
O.
, and
Tyapkin
,
A.
,
2011
, “
Electroplastic Metal Cutting
,”
Russ. Electr. Eng.
,
82
(
9
), pp.
477
479
.
16.
Jones
,
E.
,
Jones
,
J. J.
, and
Mears
,
L.
,
2013
, “
Empirical Modeling of Direct Electric Current Effect on Machining Cutting Force
,”
ASME
Paper No. MSEC2013-1229.
17.
Ulutan
,
D.
,
Pleta
,
A.
, and
Mears
,
L.
,
2015
, “
Electrically-Assisted Machining of Titanium Alloy Ti–6Al–4V and Nickel-Based Alloy IN-738: An Investigation
,”
ASME
Paper No. MSEC2015-9465.
18.
Egea
,
A. J. S.
,
Rojas
,
H. A. G.
,
Montaña
,
C. A. M.
, and
Echeverri
,
V. K.
,
2015
, “
Effect of Electroplastic Cutting on the Manufacturing Process and Surface Properties
,”
J. Mater. Process. Technol.
,
222
, pp.
327
334
.
19.
Merchant
,
M. E.
,
1945
, “
Mechanics of the Metal Cutting Process—Part I: Orthogonal Cutting and a Type 2 Chip
,”
J. Appl. Phys.
,
16
(
5
), pp.
267
275
.
20.
Tay
,
A.
,
Stevenson
,
M.
,
de Vahl Davis
,
G.
, and
Oxley
,
P.
,
1976
, “
A Numerical Method for Calculating Temperature Distributions in Machining, From Force and Shear Angle Measurements
,”
Int. J. Mach. Tool Des. Res.
,
16
(
4
), pp.
335
349
.
21.
Vedantam
,
K.
,
Bajaj
,
D.
,
Brar
,
N.
, and
Hill
,
S.
,
2006
, “
Johnson-Cook Strength Models for Mild and Dp 590 Steels
,”
AIP Conf. Proc.
,
845
(1), p.
775
.
22.
Hameed
,
S.
,
Rojas
,
H. A. G.
,
Egea
,
A. J. S.
, and
Alberro
,
A. N.
,
2016
, “
Electroplastic Cutting Influence on Power Consumption During Drilling Process
,”
,
87
(
5–8
), pp.
1835
1841
.
23.
Davidson
,
S. R.
, and
James
,
D. F.
,
2003
, “
Drilling in Bone: Modeling Heat Generation and Temperature Distribution
,”
ASME J. Biomech. Eng.
,
125
(
3
), pp.
305
314
.
24.
Kalpakjian
,
S.
, and
Schmid
,
S. R.
,
2008
,
Manufacturing Processes for Engineering Materials
, 5th ed.,
Prentice Hall
,