SKD11 steel has excellent properties such as high strength, high hardness, and high wear resistance, is widely used in molds. Electrode wire of slow wire cutting wire is one-way wire. It does not need to be threaded in processing of concave mold parts, and it can realize multiple cuttings to ensure good processing quality. Therefore, slow wire cutting technology has an irreplaceable role in processing of mold parts.
Main process indicators of slow wire cutting are cutting speed, surface roughness, and processing accuracy. When wire cutting meets processing requirements, the higher cutting speed and the fewer times of cutting, the more production costs can be reduced. Scholars at home and abroad have conducted a lot of research on online cutting process law and parameter optimization, but most of them are concentrated on medium and fast wire cutting machine tools, cutting speed and surface roughness are mostly used as process indicators. There are few reports on research of important machining accuracy process indicators.
Following will take into account cutting speed, surface roughness, and processing accuracy process indicators on slow wire-cutting machine tool, study SKD11 processing law and optimal process parameters under processing requirements, provide theoretical support for actual production of enterprise.

On slow wire-cutting machine tool, cutting test of SKD11 die steel was carried out with φ0.25 mm electrode wire. Selecting process parameters such as pulse width, pulse interval, servo voltage, servo speed, wire speed and other process parameters as test variables, an orthogonal test with 5 factors and 4 levels was designed to explore influence on process indicators such as cutting speed, surface roughness, machining accuracy, etc. Value of process parameter and test arrangement are shown in Table 1.

 

Table 1 Experimental design and results
In order to facilitate measurement of processing accuracy of each set of experiments, cutting sample design in experiment is shown in Figure 1. Cutting thickness is 20 mm, cutting length is sum of sample side length 20 mm and cutting amount 1 mm, which is 21 mm. Cutting speed is equal to ratio of cutting area to cutting time used in each set of tests, where cutting time used in each set of tests is time from beginning of sample to completion of sample, which can be read directly from machine; surface roughness value is average value obtained after measuring 3 positions on same cutting surface of each sample with TR210 handheld roughness meter; processing accuracy mainly considers dimensional accuracy, distance between the two sides. Use a micrometer to measure distance between the two sides without a cut mark, take average of dimensions measured from upper, middle, and lower positions of sample. Offset parameter of wire cutting is set to 0.159 mm.

 

Figure 1 Cutting sample

According to above-mentioned test method, 16 sets of tests were carried out. Test results are shown in Table 1, test samples are shown in Fig. 2. For measurement of 3 process indicators, measurement of cutting speed and processing accuracy (cutting size) is more accurate, especially measurement of cutting speed, which is directly guaranteed by machine tool processing time record, each measurement of cutting size has a very small error; Relatively speaking, measurement error of surface roughness of each cut surface is relatively large. When performing data processing, multiple measurements are also performed at same position, and a relatively stable value is taken as measurement result. Range analysis method is introduced to process test result data, and range analysis results are shown in Table 2.

 

Figure 2 Test sample

 

Table 2 Range analysis of test results
Note: Ki represents average value of sum of test results at each level in each process index, i takes value 1, 2, 3, 4; R1, R2, R3 represent extreme difference of cutting speed, cutting size, and surface roughness, respectively.

According to test results in Table 1, difference between maximum and minimum cutting speeds of 16 groups of tests reached 24.96 mm2/min. It can be seen that combination of different process parameters has a greater impact on cutting speed. According to range analysis of test results in Table 2, relationship curve of average cutting speed for each level of each process parameter is shown in Figure 3.

 

Figure 3 Average cutting speed of each level of each process parameter
It can be seen from Figure 3 that process parameters that have a significant impact on cutting speed are mainly pulse width and pulse interval, influence of pulse width is greater than that of pulse interval; influence of servo voltage, servo speed, and wire speed on cutting speed is basically same and small. As pulse width increases, it can be seen from principle of wire cutting that pulse discharge energy increases and electro-erosion ability increases, so cutting speed increases. As pulse interval increases, discharge energy per unit time decreases, cutting speed also decreases. According to instructions of wire cutting machine tool, when average electrode gap voltage is less than set value of servo voltage, electrode wire retreats, so as servo voltage setting value increases, cutting speed becomes slower, and the larger servo speed setting value is, the cutting speed becomes faster. Stable environment of wire cutting discharge area is conducive to cutting process. As wire feeding speed increases, it can promote discharge of electrolytic corrosion products, but excessively fast wire feeding speed will cause electrode wire to vibrate, causing electrode wire to not contact conductive block momentarily, interfere with stable discharge state, and reduce cutting speed.

For ideal cutting size of 5 mm, from measurement results of upper, middle and lower positions of test piece in Table 1, it can be seen that: upper (position 1) cut size of test piece <middle (position 2) cut size of test piece <lower (position 3) cut size of test piece. Reasons for this phenomenon may be: ①Verticality of electrode wire may be wrong; ②When test piece is cut, wire electrode wire direction at each position is from top to bottom, and upper discharge area can have more and cleaner working fluid, which can promote discharge and have a large amount of erosion.   Take average value measured at upper, middle, and lower 3 positions of test piece as cutting size of each group. According to test results, there are 10 sets of tests with a cutting accuracy of less than 5 μm, only 13th and 14th tests with a cutting accuracy of more than 10 μm. Influence of various process parameters on cutting accuracy is obvious. According to range analysis of test results in Table 2, relationship curve of average cutting size of each level of each process parameter is made, as shown in Figure 4. Pulse width has the greatest influence on cutting accuracy, pulse interval and wire speed have basically no influence on cutting accuracy. The larger pulse width, the greater discharge energy, and the stronger erosion ability. Therefore, as pulse width increases, cutting size becomes smaller, especially when pulse width is set to 14 μs, error reaches 10 μm; pulse interval and wire speed have nothing to do with discharge energy, so it has no effect on cutting accuracy. Influence trend of servo voltage and servo speed on cutting size is basically same as influence trend on cutting speed. This is because when discharge energy is constant, the slower cutting speed, the more energy allocated per unit time on cutting surface, the greater amount of electro-erosion removal, and the smaller cutting size.

 

Figure 4 Average cutting size of each level of each process parameter

It can be seen from Table 2 that degree of influence of each process parameter on the surface roughness is pulse width, pulse interval, servo voltage, servo speed, and wire speed in order. Average surface roughness relation curve of each level of each process parameter is shown in Figure 5. Surface roughness range under pulse width factor is about 0.3 μm, surface roughness range under pulse interval factor is about 0.2 μm. As pulse width increases, discharge energy increases, and galvanic pits increase, so surface roughness becomes larger and pulse interval increases, which is beneficial to discharge and fullness of galvanic corrosion products, and sufficient deionization, which is conducive to improving surface quality. Influence of other factors on surface roughness is relatively small, and there is a certain error in measurement of surface roughness of specimen, so curve trend is not too obvious. Treatment of this phenomenon can weaken analysis of impact of these three process parameters on surface roughness, select process parameters from perspective of "cost reduction and quality assurance" in actual production of enterprise.

 

Figure 5 Average surface roughness of each level of each process parameter

According to analysis of influence of process parameters on process indicators, process parameters corresponding to maximum cutting speed and minimum surface roughness are inconsistent. After studying influence of process parameters on cutting accuracy, only need to adjust offset setting to ensure cutting accuracy under processing requirements. Therefore, optimization of combination of process parameters under multiple process indicators is set to optimize combination of parameters under process indicators of cutting speed and surface roughness.
Through gray correlation analysis method, surface roughness and cutting speed are dimensionalized by formula (1) and (2), then gray correlation coefficients of two process indicators are calculated according to formula (3), gray correlation degree of two process indicators is calculated by formula (5), calculation results are shown in Table 3.
 
 
Among them, picture takes value 1, 2,..., 16; X1J and X2J respectively represent dimensioning treatment of surface roughness and cutting speed process index; y1j and y2j respectively represent jth under surface roughness and cutting speed process index test.
 
Among them, value of i is 1, 2; the value of j is 1, 2,...,16; Xi0 represents ideal value of process index of picture. ζ is resolution coefficient, which is determined by formula.
 
Among them, m and n respectively represent number of i and j, that is, m=2 and n=16.

 

Table 3 Calculation results of gray correlation between surface roughness and cutting speed
ε is calculated by formula (4) = 0.4066; when Δmax ≤ 3Δ, resolution coefficient ε value is 1.5ε <ζ ≤ 2ε, that is: 0.609 9 <ζ ≤ 0.813 2, and ζ is 0.8.
 
According to gray correlation calculation results in Table 3, test under 15th group of parameter combination has the largest gray correlation value, and cutting size at this time is 4.994 mm. It is only necessary to change offset of electrode wire from 0.159 mm to 0.162 mm can obtain ideal surface roughness, cutting speed, cutting accuracy and other process indicators. In order to further seek a better combination of process parameters, average value of gray correlation degree under each process parameter level is calculated. Results are shown in Table 4. Optimal process parameter combination is 14 μs pulse width, 20 μs pulse interval, and servo voltage 18 V, 7-speed servo speed, 11 m/min wire feeding speed. It should be pointed out that there is a small difference between mean gray correlation degree of servo speed level 2 and mean gray correlation degree of level 4. Due to certain errors in measurement of surface roughness, there may be some deviations in mean gray correlation degree under these two levels. In theory, the larger servo speed value, the faster cutting speed. In order to objectively seek optimal combination of process parameters, optimal servo speed is tentatively set to 7 or 9 gears, which will be determined by subsequent test verification; Difference between mean gray correlation degree of speed level 1 and mean gray correlation degree of level 2 is smaller. From perspective of actual production of enterprise, because the slower wire speed, the lower cost, so whether it is from theoretical calculation analysis or actual production considerations, optimal wire speed is 11 m/min.

 

Table 4 Mean value of gray correlation degree of each process parameter at each level
It can be seen from Figure 4 that impact of pulse width on cutting speed is main one. Therefore, when ensuring cutting accuracy and process indicators, only impact of pulse width on cutting speed is considered. When pulse width is set to 14 μs, average cutting size at this level is 4.990 mm. After changing offset of electrode wire from 0.159 mm to 0.164 mm, better cutting accuracy can be obtained.

Combination of process parameters after analysis and optimization is pulse width 14 μs, pulse interval 20 μs, servo voltage 18 V, servo speed 7 or 9 gears, wire speed 11 m/min, electrode wire offset 0.164 mm. Surface roughness, cutting speed, and cutting size obtained in the first group (servo speed 7 gears) and second group (servo speed 9 gears) parameter test are Ra2.589 μm, 84.85 mm2/min, 4.999 mm and Ra2. 612 μm, 86.90 mm2/min, 5.001 mm; gray correlation values of corresponding surface roughness and cutting speed are 0.770 and 0.802. After experimental verification, it is determined that servo speed in optimized process parameter combination is set to 9 gears, optimized process parameters can be used for processing to obtain a more ideal processing effect.

After experimental research on process law of slow walking wire cutting SKD11 and optimization of process parameters based on comprehensive process index consideration, following conclusions are drawn.
(1) Pulse width has the greatest and most significant influence on cutting speed, surface roughness, and processing accuracy index; secondly, pulse interval has a strong influence on cutting speed and surface roughness index, and basically has no effect on processing accuracy; servo voltage, servo speed, and wire speed have relatively weak influence on these three process indicators.
(2) Optimized combination of process parameters is pulse width 14 μs, pulse interval 20 μs, servo voltage 18 V, servo speed 9 gears, wire speed 11 m/min, electrode wire offset 0.164 mm, which can achieve 86.90 mm2 /min cutting speed, Ra2.612 μm surface roughness, and SKD11 wire cutting with a machining accuracy within 5 μm, provide theoretical and technical support for enterprises to reduce costs and improve efficiency in actual production.