1 Introduction

Micro-slit array is a typical structure in various high-tech products such as X-ray phase-contrast grating and progressive die for integrated circuit lead frame. This kind of structure generally comprises of arranged micro-slits ranging from several microns to several hundred microns in width. And difficult to machine materials are usually adopted to obtain specific operating requirements and broad environmental suitability. Such characteristics above as well as its rigorous demand on surface integrity bring it a considerable challenge to achieve efficient and precision fabrication of micro-slit array.

Wire electrochemical micromachining (WECMM) employs a micro metallic wire as the cathode and electrochemically removes material via controlled anodic reactions in an electrolyte cell when an appropriate voltage is applied [1]. A slit would be produced along the pre-programmed trajectory between the wire electrode and the workpiece. The principle of electrochemical dissolution allows WECMM process to proceed regardless of material mechanical properties and yield slits absence of stress, recast layer, tool wear, etc.. Typical micro structures such as micro-groove [2], gear [3], and square helix [4] had been successfully fabricated. WECMM could be a feasible alternative for precision fabrication of micro-slit array via flexibly changing machining stations.

In WECMM, material dissolution takes place in a high aspect-ratio slit with a side gap of tens of microns. It is difficult for electrolyte renewal and product removal. Products including metal hydroxide sludge and bubbles adversely influence electrolyte conductivity and thereby affect material removal rate. Thus, WECMM is still limited in efficiency to meet the industrial production requirement. Literature shown that much efforts have been made from several aspects to improve material removal rate and the process efficiency. Electrolyte flushing is the conventional way to supply fresh electrolyte to the machining gap and expel products away from the gap. Maeda et al. studied the effects of electrolyte flushing velocity, nozzle diameter on electrode feeding rate. Reduction of nozzle diameter was recommended to prevent probable wire vibrations caused by electrolyte flushing [5]. Wang et al. pumped the electrolyte flow at 0.75 m/s to the machining zone and fabricated 5-mm-thick micro structures at 0.5 μm/s [6]. Qu et al. ejected electrolyte out from the nozzle at 87 m/s and significantly enhanced the electrode feeding rate to 30 μm/s in WECMM of 1.8-mm-thick titanium alloy (Ti-2Al-1.5Mn) [7]. He et al. used axial electrolyte flushing and pulsed current in WECMM, machined 10-mm-thick γ-TiAl alloy (Ti-42Al-6V-1Cr) at the rate of 3.0 μm/s [8].

From another aspect, various tool-electrode movements were assisted in the machining process to accelerate electrolyte renewal. A vibration of low frequency and small amplitude was generated by a piezoelectric actuator and adopted on the wire cathode tool. When the wire electrode vibrates in its axial direction, waste electrolyte carrying electrolysis products is dragged out from the machining gap and fresh electrolyte is pulled into the gap on account of the surface tension existing between the wire and electrolyte. It has been experimentally verified that the process stability [9], slit homogeneity [10], and machining efficiency [11] could be enhanced a lot. To intensify the mass transport process in WECMM with tool vibration, micro wires with hydrophilic surface [12] or textured surface [13] were introduced. As the linear velocity of a wire electrode is limited in a micro-amplitude vibration, a reciprocated travelling wire was introduced to machine high aspect-ratio structures [14]. A slit with a width of 177 μm and an aspect ratio of 113 was successfully fabricated when the wire electrode was travelled at 0.35 m/s. To eliminate frequent changes of travelling direction, a monodirectional travelling wire ring was employed as the cathode [4, 15]. Besides, structured electrodes such as helical electrode [16, 17], ribbed electrode [18], and edged electrode [19] were also developed for WECMM. Swirling flow generated by the rotation of structured electrodes was proven effective to accelerate the electrolytic refreshment and the products removal. To sum up, a travelling wire electrode could provide a higher slipping velocity to vibration and rotation of a wire electrode.

However, to increase production efficiency in fabrication of arrayed structures, the direct way is utilizing multiple tool electrodes to work in different stations simultaneously. Multiple electrodes have been well applied for electrochemical drilling [20, 21] and electrochemical machining of blisk channels [22]. In WECMM, wire electrodes are usually mounted to a specific fixture in parallel and fed together to the workpiece [23]. Axial vibrations have been employed to enhance the mass transport process [24, 25]. This paper proposed a novel method for multiple slit electrochemical micromachining (multi-slit ECMM) using a single wire, which is spooled through a specific pulley with multiple grooves to generate a multiple-wire face. Two ends of the wire are fixed on a wire storage cylinder, with the rotation of which the wire is travelled. A specific apparatus was developed to study the method performances. Finally, technological and economical comparison of machining performances in multi-slit ECMM, HS-WEDM, and LS-WEDM are carried out.

2 Principle of multiple wire electrochemical micromachining using a single metallic wire

Figure 1 presents the schematic diagram of multi-slit ECMM using a single metallic wire. The wire is spooled through specific pulleys with multiple grooves to generate a multiple-wire face. Two ends of the wire are fixed on a wire storage cylinder. The arrayed multi-wires act as the cathode and produce arrayed structures when a pre-programmed tool trajectory are carried out. When the wire storage cylinder rotates, the multi-wires are travelled at a constant velocity u in its axial direction.

$$ u=\omega r $$
(1)

where ω and r are the angular velocity and the radius of the wire storage cylinder, respectively. The wire travelling direction alters periodically with the rotation direction of the storage cylinder.

Fig. 1
figure 1

Schematic diagram of multi-slit ECMM using a single metallic wire

Full size image

In the machining process, electrolyte is dragged out from the machining gap at a velocity v, which could be described in Eq. (2) [6].

$$ v\left(y,z\right)=-\frac{S^2}{2\mu}\frac{dp}{dz}\left[1-{\left(\frac{y}{S}\right)}^2\right]+\frac{u}{2}\left(\frac{y}{S}\right)+\frac{u}{2} $$
(2)

where μ is the electrolyte viscosity, p is the local electrolyte pressure, and S is the side gap between the wire electrode and workpiece, as shown in Fig. 2. It illustrates that the electrolyte velocity v depends mainly on the wire electrode travelling velocity u. In this proposed method, the multiple wires are actually a single wire travelled by the only storage cylinder. The wire travelling velocity and direction in each wire machining branch are the same. According to Eq. (2), product removal rate and electrolyte renewal rate in each branch should be the same with each other. It is of great benefits to obtain high homogeneity–arrayed structures in multiple wire machining.

Fig. 2
figure 2

Flow field and electric model of multi-slit ECMM

Full size image

According to the circuit theory, the multi-slit ECMM system is a parallel loop. In each branch, the workpiece, the electrolyte, and the wire electrode comprise as a module to be part of the loop. In the local machining gap, there are electrolyte resistance Re, resistance and capacitance of the double layer between the cathode and the electrolyte Rce, Cce, and resistance and capacitance of the double layer between the anode and the electrolyte Rae, Cae. Ignoring the effects of distributed capacitances which are mostly taken into account in cases with ultra-short voltage pulses, each machining branch could be considered as an equivalent resistance Req. The equivalent resistance of the whole loop would decrease proportionally to the wire electrode amount, n, which is defined as the number of all wire electrodes taking part in electrochemical machining as shown in Fig. 2. On account of the pulse generator’s internal resistance, potential dropped between multiple wires and workpiece namely inter-electrode voltage should also decrease with the increase of wire electrode amount. In the following sections, experiments are conducted to explore the relationship between actual inter-electrode potential and wire electrode amount. Thus, it could be adjusted with wire electrode amount to keep constant.

Therefore, each wire machining branch has the same electric supply and the same wire travelling status, as well as the same flow field. The proposed novel method should produce arrayed structures with high consistency.

3 Experimental procedures

A specific experimental apparatus presented in Fig. 3 was developed to study the process performances of the proposed method. This system consists an X-Y-Z moving platform, a pulse current generator, a current acquisition device, and a wire electrode travelling module, which is the core of the proposed method. The wire electrode travelling module comprises of a wire storage cylinder, a motor, and a pulley holder as presented in Fig. 3. The specific pulley is made of Si3N4 and its structure parameters are listed in Table 1. A specific connecting is designed to separately supply electric energy for each machining branch, thus the machining current in each branch could be detected and recorded independently. In the machining process, the workpiece is fed along a programmed tool path. The wire electrode periodically changes its travelling direction with the storage cylinder controlled by a position switch.

Fig. 3
figure 3

Experimental apparatus for multiple wire ECMM

Full size image
Table 1 Structure parameters of specific pulleys
Full size table

Dimensional consistency is the key point to evaluate the process feasibility and machining accuracy of multiple wire ECMM. Three-millimeter-thick plates made of stainless steel 304 were prepared as specimens. The kerf length was 2 mm. The machined slits were measured using a scanning electron microscope according to Fig. 4. Then the averaged slit widths as well as its standard deviation were calculated. The main machining parameters for WECM were chosen through experimental trials as follows: a pulsed voltage of 10 kHz in frequency and 50% in duty cycle, 5 g/l sodium nitrate aqueous solution at 28 °C. The molybdenum wire of 0.12 mm in diameter is employed as cathode and travelled at a speed of 0.32 m/s.

Fig. 4
figure 4

Measurement of the machined slit width

Full size image

4 Results and discussions

4.1 Feasibility of multi-slit ECMM using a single wire

The wire is spooled through pulleys to generate a three-wire face and three slits were machined simultaneously, as presented in Fig. 5. The applied voltage is set at 17 V and the wire electrode feeding rate is 2.6 μm/s. Machined slit widths are listed in Table 2. Dimensional deviations of slit width in the same position are less than 2 μm. The results indicated that the proposed method is feasible to fabricate slit array in high consistency.

Fig. 5
figure 5

Three slits machined simultaneously

Full size image
Table 2 Measurement results of slits the italics numbers are the top and bottom slit width.
Full size table

In addition, the process stability with more wire electrodes was tested. Actually, the whole process stability depends on each branch machining stability, which may be disturbed by insufficient product removal or wire electrode shake. Figure 6 presents all the branch machining currents in a five-wire machining case. The current keeps constant roughly but jumps slightly when the wire travelling direction changes. And the total machining currents in WECM with different wire electrode amounts were recorded, as shown in Fig. 7. It indicated that current fluctuation and abnormal signals increased with the wire electrode amount. As analyzed in Section 2, when the process parameters are same, the actual inter-electrode voltage as well as the machined slit width will decrease with the increase of wire amount. These make product removal and electrolyte renewal more difficult and disturb the branch process stability. Figure 8 presents the slits machined via WECM with different wire electrode amounts. It is obvious that the slit straightness got worse when the electrode amount increased to seven and nine. And the slit’s contorted positions accord with where the wire travelling direction changes. With the increase of wire amount, the twined wire turns in the pulley as well as the frictional force are increased. Thus, the metallic wire would be stretched and extended. When the travelling direction of the pulley changes, the wire will shake and slide laterally in the pulley groove because of the gap and the increased friction. Therefore, the part of a wire electrode in the machining gap is like the center of a simple-supported beam which has the amplifying deflection and vibration. In this case, the slit width would be affected. The situation could be modified with a precision pulley and a stable wire travelling system.

Fig. 6
figure 6

Machining current of each branch in WECM with five wire electrodes

Full size image
Fig. 7
figure 7

Total machining current in WECM with different wire electrode amounts.

Full size image

4.2 Effects of wire amount on inter-electrode voltage and slit width

Experiments were conducted to investigate the effects of wire amount on inter-electrode voltage and slit width. The output of pulse generator was 17 V in amplitude and 10 kHz in frequency. The electrode feeding rate was 2.6 μm/s. Inter-electrode voltage waveforms in machining process were recorded and presented in Fig. 9. When a wire electrode was employed, the actual voltage is 14.3 V. With the electrode amount increased to nine, the actual inter-electrode voltage is only 12.8 V as listed in Table 3. Accordingly, the averaged slit width decreased from 236.2 to 217.2 μm, as shown in Fig. 10. As analyzed in Section 2, the equivalent resistance of the whole machining loop would decrease proportionally to the wire electrode amount and the actual voltage namely inter-electrode voltage should also decrease. And a lower voltage would draw a narrower slit.

Fig. 8
figure 8

Slits machined via WECM with different wire electrode amounts

Full size image
Table 3 Display voltage vs. inter-electrode voltage
Full size table
Fig. 9
figure 9

Inter-electrode voltage waveform in WECM with different electrode amounts

Full size image
Fig. 10
figure 10

Averaged slit width via WECM with different wire electrode amounts

Full size image

Then, an idea was proposed that the output of the pulse generator could be adjusted to a suitable value depending upon the wire amount. Experiments were conducted to obtain these values. The output of pulse generator was 10 kHz in frequency and the electrode feeding rate was 2.6 μm/s. The results of suitable adjusted display voltage were listed in Table 3. When the electrode amount increased from one to nine, the supplied voltage was adjusted from 17 to 18.9 V. And inter-electrode voltage waveforms in machining with different wire amounts were recorded and presented in Fig. 11. In this way, the actual inter-electrode voltage could be kept constant at 14.3 V. Accordingly, the averaged slit width was kept around 235 μm, as shown in Fig. 12. In this way, the effects of wire amount on slit width were modified with adjusted applied voltages.

Fig. 11
figure 11

Inter-electrode voltage waveform in WECM with different electrode amounts

Full size image
Fig. 12
figure 12

Averaged slit width in WECM with different wire electrode amounts

Full size image

4.3 Machining efficiency of multi-slit ECMM in different voltage modes

This paper proposed a novel method for multiple wire electrochemical micromachining to increase machining efficiency in fabrication of arrayed structures. Here, the efficiency is evaluated via a largest possible electrode feeding rate, at which a slit of 2 mm in kerf can be machined stably without short circuits.

Under the mode of a constant display voltage at 17 V, the largest feeding rate decreased from 4.9 to 3 μm/s when the electrode amount increased from one to nine, as shown in Fig. 13 a. The total feeding rate did not linearly increase with the wire amount as assumed. When nine electrodes were employed, the total feeding rate was 27 μm/s, which is only 5.5 times of that using a single electrode. As analyzed in Section 4.1, the actual inter-electrode voltage as well as the slit width decreases with the electrode amount. As a result, the limited product removal rate and electrolyte renewal rate would limit the largest possible electrode feeding rate.

Fig. 13
figure 13

Variations between electrode feeding rate and wire electrode amount

Full size image

In Section 4.2, it has verified that the constant inter-electrode voltage mode is effective for diminishing the adverse effects of electrode amount. Here, we employed the adjusted voltages listed in Table 3 to multi-slit ECMM. Under the mode of a constant inter-electrode voltage at 14.3 V, the largest feeding rate decreased from 4.9 to 4.6 μm/s when the electrode amount increased from one to nine. And the total feeding rate increased from 4.9 to 41.4 μm/s, which has been improved by 744.9%.

5 Technological comparison of multi-slit ECMM with WEDM

Wire electricdischarge machining (WEDM) has widely used in various industrial applications. Here, experiments were carried out to compare multi-slit ECMM with high-speed WEDM (Sanguang, DK7732P, China) and low-speed WEDM (Ecut 2F, AGIE, Switzerland) in machining efficiency and surface integrity. The task is to produce a slit array with 15 slits of 3 mm in kerf. In multi-slit ECMM, five slits were fabricated simultaneously and then the tool path is parallelly moved to the next five. The applied voltage was set at 17 V and the electrode feeding rate was 2.6 μm/s, which means 13 μm/s in total. The other conditions were the same as described in Section 3. The specimens were observed through a scanning electron microscope (S-3400N, Hitachi, Japan), as shown in Fig. 14. It was obvious that LS-WEDM has the highest machining accuracy while the slits encountered heavy thermal deformation in HS-WEDM. From the enlarged top view as listed in Fig. 15, it could be seen that the slit surface was quite different from machining methods. Afterwards, surface integrity of the machined slits including roughness and recast layers was tested via a surface profilometer (Micro XAM, ADE, USA) and observed via a metallurgic microscope (GX71, Olympus, Japan), respectively.

Fig. 14
figure 14

Slits machined with different methods

Full size image
Fig. 15
figure 15

Top view of slits machined with different methods

Full size image

The technological comparison results of multi-slit ECM with WEDM are listed in Table 4. First, both HS-WEDM and LS-WEDM have a higher machining efficiency than multi-slit ECMM. However, it could be improved from 13 to 24 μm/s via employing a constant inter-electrode voltage as discussed in Section 4.3. In this way, the machining efficiency are in the same range with few differences. Second, as shown in Fig. 16, both HS-WEDM and LS-WEDM generate a recast layer on the machined surface, while none recast layers were found on the ECMed surface. Third, LS-WEDM and wire ECMM have a smoother surface than HS-WEDM, as shown in Fig.17. Now, surface roughness and recast layers in WEDM could be enhanced via subsequent trimming [26, 27].Therefore, such aforementioned characteristics of multi-slit ECMM enable itself to be a feasible alternative for machining arrayed structures requiring good surface integrity and high efficiency.

Table 4 Technical comparison results of multi-slit ECMM with WEDM
Full size table
Fig. 16
figure 16

Recast layer on slit surfaces machined with different methods

Full size image
Fig. 17
figure 17

Surface morphology of slits machined with different methods

Full size image

6 Conclusions

This paper proposed a novel method for multi-slit ECMM using a single wire to enhance its efficiency in machining arrayed structures and a specific apparatus was developed to study its performances. The conclusions are:

  1. 1.

    The flow and electrical analysis illustrated that each wire machining branch in multi-slit ECMM has the same electric supply and the same wire travelling status, as well as the same flow field, which is of great benefits for producing arrayed structures with high consistency.

  2. 2.

    Experiments verified that the actual inter-electrode voltage as well as the slit width decreased with the increase of electrode amount and the maximum electrode feeding rate was limited. As a parallel circuit, the equivalent resistance of the machining zone decreases proportionally to the wire electrode amount and the potential dropped between wire electrode and workpiece decreases consequently.

  3. 3.

    To maintain a constant inter-electrode voltage, the output of the pulse generator was adjusted to a suitable value depending upon the wire amount. Compared with a constant display voltage mode, the largest electrode feeding rate increased from 3 to 4.6 μm/s and the total feeding rate increased from 27 to 41.4 μm/s when the electrode amount was nine.

  4. 4.

    Technological and economical comparison has shown that multi-slit ECMM is a feasible alternative for machining arrayed structures requiring good surface integrity and high efficiency.