ISSN 0967-0912, Steel in Translation, 2007, Vol. 37, No. 3, pp. 288–295. © Allerton Press, Inc., 2007. Original Russian Text © Yu.A. Bodyaev, A.V. Gorbunov, A.F. Radionov, V.K. Belov, D.O. Begletsov, E.V. Gubarev, 2007, published in “Stal’,” 2007, No. 3, pp. 52–57.
Electroerosive Machining of the Working Rollers for Trimming Auto-Industry Steel Sheet Yu. A. Bodyaev, A. V. Gorbunov, A. F. Radionov, V. K. Belov, D. O. Begletsov, and E. V. Gubarev OAO Magnitogorskii Metallurgicheskii Kombinat Magnitogorsk State Technical University DOI: 10.3103/S0967091207030291
It is difficult to ensure the required microtopography of working rollers in trimming mills for the production of cold-rolled auto-industry sheet steel with regulated surface characteristics [1–9]. In burnishing the working rollers, the surface microtopography changes significantly, as follows [6] Rarsbur ≈ KRaburRardbur; KRabur ≈ 0.75; Pcrsbur ≈ KPcburPcrdbur; HSCrsbur ≈ KHSCburHSCrdbur;
KPcbur ≈ 1.00; KHSCbur ≈ 1.00,
where the coefficients K characterize the change in the corresponding roughness parameter on burnishing. Thus, burnishing sharply reduces the amplitude characteristics of the profile, without changing the frequency characteristics. The degree to which the working-roller microtopography is impressed on the trimmed sheet surface corresponds to the following formulas Ratrimsh ≈ KRaimpRarsbur; KRaimp ≈ 0.45; Pctrimsh ≈ KPcimpPcrsbur; HSCtrimsh ≈ KHSCimpHSCrsbur;
KPcimp ≈ 0.90; KHSCimp ≈ 0.90.
Knowing these coefficients, the required surface characteristics of the workers after electroerosive treatment may be determined. If initially Ra = 2.5–3.5 µm, Pc = 55–60 cm–1, HSC = 60–75 cm–1, a sheet surface with the required parameters (Ra = 0.8–1.2 µm and Pc > 50 cm–1) may be obtained after dressing. Thus, the main problem is to select the texturing conditions that ensure the required surface microtopography of the roller prior to insertion in the rolling cell. Note that obtaining a roller surface with the required parameters will not necessarily eliminate unwanted impressions on the sheet. On trimming in working rollers with shot incisions, such defects are extremely rare. Accordingly, we may formulate the hypothesis that the probably of this defect will be reduced if the surface microtopography after electroerosive treatment is the same as after shot blasting. The trapezoidal profile must be characterized not only by the parameters Ra, Pc, or
HSC but also by the similarity of the phase portraits. Only this requirement may yield a surface with the specified microtopography and eliminate unwanted impressions. For the Sarclad-Hercules electric-discharge texturing (EDT) unit operating in OAO MMK sheet-rolling shop 5, the working parameters may be regulated over the following ranges: the times during which it is on (between pulses) τ1 and off τ2 in the range 5–200 µs (τ1 > τ2); current up to 62 A; constant reinforcement (10 V); the voltage threshold characterizing the distance between the electrodes and the treated surface usually exceeds 4 V to eliminate electrode welding; the rotary and translational speeds are selected empirically so that no defects (networks, spiral twists, etc.) are seen on the surface. Thus, to select the operating conditions, the variation in amplitude and frequency characteristics of the profile with variation in τ1 must be established, when the other parameters of the EDT unit are constant. The analogous variation of τ2, the current, the electrode polarity, and the number of passes is studied. The roughness of the rollers and of the strip before and after trimming is determined by means of equipment developed by the Mikrotopografiya Scientific-Research Center (Moscow State Technological University). The measurement results are averaged over three tracks in six surface zones of the working rollers. With variation in τ1, there are considerable changes in surface microtopography, as is evident in Fig. 1. The other characteristics are fixed (τ2 = 60 µs, reinforcement 10 V, threshold 6 V, and current 40 A): Time τ1 for which unit is on, µs Mean deviation Ra, µm Mean square deviation Rq, µm Asymmetry coefficient Rsk Excess coefficient Rku Maximum peak height Rp, µm Maximum depth Rv, µm Oil capacity of profile Rz, µm
288
6 15 30 0.75212 1.7232 2.5266 0.9781 2.0843 3.2651 0.053855 –0.54055 –0.21806 4.217 2.6353 3.4203 4.6248 5.0234 9.5604 –4.0716 –7.4401 –9.916 0.55873 0.59654 0.50912 8.627 12.391 19.4058
ELECTROEROSIVE MACHINING OF THE WORKING ROLLERS (a)
(b)
Profile of interest
Histogram of distribution
0.02 0.01 0 –0.01 –0.02 0
1
2
3
4
200 150
0.2
original windowed
50 40 30 20 10 0 250 200 150 100 50
0 –0.5
original windowed
–1.0 –1.5 –2.0
0
0
1
2
3
4
5 0
0.2
0.4
dY/dX
1.0 0.5 Number of HSC peaks per cm, as a function of % Rt
Histogram of distribution
0.02 0.01 0 –0.01 –0.02
0.4
k=2
2.0 1.5
100 50 0 500 400 300 200 100
5 0
Profile of interest
dY/dX
Number of Pc peaks per cm, as a function of % Rt
289
–0.01 0
0.01 0.02 Y(X)
20 40 60 80 100
0
Number of Pc peaks per cm, as a function of % Rt original windowed
2.0
k=2
1.5 1.0 0.5
Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5 –2.0
20 40 60 80 100
–0.01 0
0.01 0.02 Y(X)
(c) Profile of interest
Histogram of distribution
0.02 0.01 0 –0.01 –0.02 0 70 60 50 40 30 20 10 0 140 120 100 80 60 40 20 0
1
2
3
4
5 0
dY/dX 2.0
Number of Pc peaks per cm, as a function of % Rt original windowed
0.2
0.4
k=2
1.5 1.0 0.5
Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5 –2.0
20 40 60 80 100
–0.01 0 0.01 0.02 Y(X)
Fig. 1. Surface profilogram of the working rollers treated on a Sarclad-Hercules EDT unit when τ1 = 6 (a), 15 (b), and 30 (c) µs.
The variation in mean Ra and HSC as a function of τ1 is shown in Fig. 2. It is evident that, with increase in τ1, the amplitude parameters sharply increase, while the STEEL IN TRANSLATION
Vol. 37
No. 3
2007
frequency parameters decrease. The variation in τ2 has little influence on the microtopography (Fig. 3). The other characteristics are fixed
290
BODYAEV et al. Ra, µm 3
300
2
200
1
100
0
HSC, cm–1
10
20
30
40 τ1, µs
0
10
20
30
40 τ1, µs
Fig. 2. Dependence of the surface roughness (Ra) and asymmetry coefficient (HSC) on the time for which the unit is on (τ1).
(a) 0.02 0.01 0 –0.01 –0.02
Profile of interest
0
1
2
3
4 dY/dX 2.0 1.5
Number of Pc peaks per cm, as a function of % Rt original windowed
100 80 60 40 20 0 400 300 200 100
Number of HSC peaks per cm, as a function of % Rt original windowed
(b) Histogram of distribution
5 0
0.2
0.4
k=2
200 150 100
1.0 0.5 0 –0.5 –1.0
50 0 1000 800 600 400 200
–1.5 –2.0
0
20 40 60 80 100
0.02 0.01 0 –0.01 –0.02
–0.01 0
0.01 0.02 Y(X)
0
Profile of interest
0
1
2
3
Histogram of distribution
4 dY/dX 2.0 1.5
Number of Pc peaks per cm, as a function of % Rt original windowed
5 0
0.2
0.4
k=2
1.0 0.5 Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5 –2.0
20 40 60 80 100
–0.01 0
0.01 0.02 Y(X)
(c) 0.02 0.01 0 –0.01 –0.02 140 120 100 80 60 40 20 0 500 400 300 200 100 0
Profile of interest
0
1
2
3
Histogram of distribution
4 5 0 0.2 dY/dX k=2 2.0 1.5
Number of Pc peaks per cm, as a function of % Rt original windowed
0.4
1.0 0.5 Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5 –2.0
20 40 60 80 100
–0.01 0
0.01 0.02 Y(X)
Fig. 3. Surface profilogram of the working rollers when τ2 = 20 (a), 80 (b), and 170 (c) µs. STEEL IN TRANSLATION
Vol. 37
No. 3
2007
ELECTROEROSIVE MACHINING OF THE WORKING ROLLERS
291
HSC, cm–1 250
Ra, µm 1.2 1.0 0.8 0.6 0
200 150 50
100
150
200 τ2, µs
100
0
50
100
150
200 τ2, µs
Fig. 4. Dependence of Ra and HSC on the time τ2 for which the unit is off.
with the same working parameters and τ1 = 9 µs: Time that unit is off τ2, µs
The electrode polarity has a significant influence on the surface microtopography (Fig. 7, τ1 = 25 µs, τ2 = 40 µs):
20
80
170
Ra, µm
1.657
1.3575
1.2449
Electrode polarity
Rq, µm
2.108
1.3576
1.4732
Rsk
–0.19942
–0.58636
0.14974
Rku
3.2166
4.1735
2.1875
Rp, µm
5.0962
3.8203
5.0553
Rv, µm
–7.4857
–6.4271
–3.2643
Oil capacity of profile
0.59494
0.6272
0.39236
Rz, µm
12.4799
10.178
7.9419
It is evident from the variation in mean Ra and HSC (Fig. 4) that, with increase in τ2, the amplitude parameters are practically unchanged, while the frequency parameters decline slightly. Current variation has little effect in the given range (Fig. 5): Current, A
42
50
58
Ra, µm
2.4934
2.4406
2.3623
Rq, µm
3.3402
3.3082
3.1471
Rsk
0.71442
0.33433
–0.011896
Rku
4.7387
3.9949
3.4659
Rp, µm
15.1032
10.7731
9.8892
Rv, µm
–8.6541
–9.9454
–8.8122
Oil capacity
0.36427
0.48003
0.47122
Rz, µm
23.7438
20.6768
18.6802
It is evident from Fig. 6 that the current has practically no influence on the amplitude parameters Ra, while there is a slight increase in the frequency characteristics HSC. The large negative Ra values indicate an asymmetric distribution of ordinates of the profile. STEEL IN TRANSLATION
Vol. 37
No. 3
2007
+
–
Ra, µm
0.20777
0.10432
Rq, µm
0.28519
0.13052
Rsk
0.97075
0.15049
Rku
6.3116
3.2621
Rp, µm
1.4591
0.50593
Rv, µm
–0.79978
–0.37372
Oil capacity
0.35406
0.42484
Rz, µm
2.2585
0.87773
The amplitude and frequency parameters are greater for positive polarity than for negative polarity (Fig. 8). It is well known that Pc declines with increase in Ra. With change in polarity from positive to negative, besides the relative decrease in Ra, there is a smaller increment in the number of peaks. This indicates radical difference in surface microtopography in the two cases, as illustrated by the phase portraits of the profiles (Fig. 7). As a function of the number of passes, the surface microtopography undergoes significant changes (Fig. 9, τ1 = 25 µs, τ2 = 40 µs): Pass
1
2
3
Ra, µm
3.2668
5.8145
5.5798
Rq, µm
4.1746
7.182
7.0196
Rsk
–0.24684
0.6082
0.66752
Rku
3.1747
2.9922
3.1691
Rp, µm
10.2481
25.935
22.2214
Rv, µm
–14.3706
–11.9565
–13.5844
Oil capacity
0.58373
0.31555
0.37939
Rz, µm
24.5909
37.873
35.7996
292
BODYAEV et al. (a) Profile of interest
0.05
Histogram of distribution
0 –0.05
0
1
2
3
4 dY/dX 2.0
Number of Pc peaks per cm, as a function of % Rt
80
original windowed
60
1.0
20
0.5
0.2
0.4
k=2
0
Number of HSC peaks per cm, as a function of % Rt
–0.5
original windowed
100
–1.0
50
–1.5 –2.0
0
0
1.5
40
0 150
5
–0.04
0
0.04 Y(X)
20 40 60 80 100 (b) Profile of interest
(c) Histogram of distribution
0.05 0 –0.05 100 80 60 40 20 0 150
50 0
Histogram of distribution
0 0
1
2
3
Number of Pc peaks per cm, as a function of % Rt original windowed
4 dY/dX 2.0
5 0
0.2
0.4
k=2
1.5
20 0 200 150 100 50
0.5 Number of HSC peaks per cm, as a function of % Rt
0 –0.5 –1.0 –1.5
–2.0 –0.04 20 40 60 80 100
–0.05 80 60 40
1.0
original windowed
100
Profile of interest
0.05
0 0
0
1
2
3
4 5 0 0.2 dY/dX k=2 2.0
Number of Pc peaks per cm, as a function of % Rt original windowed
0.4
1.5 1.0 0.5
Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5
–2.0 –0.04 20 40 60 80 100
0
0.04 Y(X)
0.04 Y(X)
Fig. 5. Profilogram of the treated rollers treated on a Sarclad-Hercules EDT unit when the current is 42 (a), 50 (b), and 58 (c) A.
The amplitude and frequency parameters are practically unchanged with 1–3 passes (Fig. 9), but it is evident that the profile peaks are thinner on the surface profilograms and phase portraits.
On the basis of the results, we may formulate the following recommendations for the selection of electroerosive treatment conditions so as to ensure surface microrelief of the cold-rolled sheet with Ra = 0.8–1.2 µm and STEEL IN TRANSLATION
Vol. 37
No. 3
2007
ELECTROEROSIVE MACHINING OF THE WORKING ROLLERS
293
Ra, µm 0.35
HSC), τ1 must be increased; to reduce Ra and increase Pc (or HSC), τ1 must be reduced.
0.30
Typical surface profiles of upper and lower working rollers after electroerosive treatment in the recommended conditions are shown in Fig. 10:
0.25 0.20
Upper roller
Lower roller
Ra, µm
4.7154
4.5845
Rq, µm
5.8629
5.6542
Rsk
0.45612
0.14029
Rku
2.8827
2.9844
Rp, µm
19.452
21.1047
Rv, µm
–15.9726
–15.0265
Oil capacity
0.45089
0.41589
Rz, µm
34.8564
35.4706
0.15 HSC, cm–1 175 150 125 100 40
45 50 Current, A
55
60
Fig. 6. Dependence of Ra and HSC on the current.
Pc = 50 cm–1, with a low probability of unwanted impressions in trimming: electroerosive treatment of the working roller must be conducted in a single pass at positive polarity; and the amplitude and frequency parameters must be regulated by the time for which the unit is switched on. To increase Ra and reduce Pc (or ×10–3 0.5 1.0 0 –0.5 –1.0 0
(a)
1
2
3
4
Pc, cm–1 80 60 40 20 0
Prior to burnishing, the roughness of the working rollers is Ra = 4.2–5.0 µm, with Pc = 50–60 cm–1. After burnishing, the microtopographic parameters correspond to their values before trimming (Fig. 11). It fol×10–3 0.5 1.0 0 –0.5 –1.0 0
(b)
1
2
3
4
60 40 20
5 10 15 20 25 30 35 G1, µm
dY/dX k=2 0.08 0.04 0 –0.04 –0.08 –0.8 –0.4 0 0.4 0.8 Y(X) ×10–3
0
5 10 15 20 25 30 35 G1, µm k=2
0.08 0.04 0 –0.04 –0.08 –0.8 –0.4 0 0.4 0.8 Y(X) ×10–3
Fig. 7. Profilograms and phase portraits of profiles obtained with positive (a) and negative (b) electrode polarity. STEEL IN TRANSLATION
Vol. 37
No. 3
2007
294
BODYAEV et al. Ra, µm 5 4 1 3 2 2 1 0 1 2 3 Number of passes
lows from Fig. 10 that, after trimming, Ra for the roller profiles is reduced by ~25%, while the spacing Pc is practically unchanged. The variation in surface microtopography of the strip on trimming is illustrated in Fig. 12: Ra declines by ~10%, while Pc is practically unchanged and does not fall below 50 cm–1. The results suggest the adoption of the following variation in surface microtopography of the working
Fig. 8. Dependence of Ra on the number of passes with positive (1) and negative (2) polarity.
1 0.010 0.005 0 –0.005 –0.010 –0.015 0
2
3 0.02
0.02
1
2
3
4
0.01
0.01 0 –0.01 0
Number of Pc peaks per cm, as a function of % Rt original windowed
0 –0.01 1
2
3
4
0
Number of Pc peaks per cm, as a function of % Rt
1
2
3
4
Number of Pc peaks per cm, as a function of % Rt
dY/dX k = 2 dY/dX k = 2 dY/dX k = 2 3000 2500 original original 1.0 1.0 1.0 2000 windowed windowed 2000 0.8 0.8 0.8 1500 0.6 0.6 0.6 1000 1000 0.4 500 0.4 0.4 0 0 0.2 0.2 0.2 Number of HSC peaks per cm, Number of HSC peaks per cm, Number of HSC peaks per cm, 0 0 0 as a function of % Rt as a function of % Rt as a function of % Rt 80 100 –0.2 –0.2 –0.2 original original original 120 windowed 80 windowed windowed 60 –0.4 –0.4 –0.4 60 80 –0.6 –0.6 –0.6 40 40 –0.8 –0.8 –0.8 40 20 20 –1.0 –1.0 –1.0 –0.008 0 0.008 –0.008 0 0.008 –0.008 0 0.008 0 20 40 60 80100 0 20 406080100 0 20 40 60 80 100 Y(X) Y(X) Y(X)
6000 4000 2000 0
Fig. 9. Variation in phase portraits and surface profilograms of rollers treated on the EDT unit as a function of the number of passes (1–3) with positive electrode polarity.
(a)
(b)
Profile of interest
0.05
Histogram of distribution
0
Histogram of distribution
–0.05 0
0
0.05 0
–0.05 100 80 60 40 20 0 250 200 150 100 50
Profile of interest
1
2
3
4 dY/dX 2.0 1.5
Number of Pc peaks per cm, as a function of % Rt original windowed
5 0
0.2
k=2
0.5
original windowed
0 –0.5 –1.0 –1.5 –2.0
20 40 60 80 100
–0.03
0
0 60 50 40 30 20 10 0 200 150 100 50
1.0
Number of HSC peaks per cm, as a function of % Rt
0.4
0.03 Y(X)
0
1
2
3
Number of Pc peaks per cm, as a function of % Rt original windowed
4 dY/dX 2.0 1.5
5 0
0.2
0.4
k=2
1.0 0.5 Number of HSC peaks per cm, as a function of % Rt original windowed
0 –0.5 –1.0 –1.5 –2.0
–0.03
20 40 60 80 100
0
0.03 Y(X)
Fig. 10. Surface profiles of the upper (a) and lower (b) working rollers after taking account of the recommendations. STEEL IN TRANSLATION
Vol. 37
No. 3
2007
ELECTROEROSIVE MACHINING OF THE WORKING ROLLERS 4.0 3.5 3.0 2.5 2.0
trimming up to 2400 t of sheet, no unwanted impressions are observed. The microtopographic parameters of the sheet after trimming the first 200 t are in the required range: Ra = 0.8–1.2 µm, Pc > 50 cm–1.
1 2 0
50
70 60 50 40
100
150
200
ACKNOWLEDGMENTS
1
The authors thank V. N. Yakimenko, D. A. Shashkin, A. P. Polyak, I. Yu. Kurochkin, V. S. Banshikov, K. S. Novikov, D. V. Lantushenko, and D. V. Sherstobitov for their participation in this research.
2 0
50
100
150
200
Fig. 11. Variation in Ra and Pc for the upper (1) and lower (2) rollers after burnishing in the course of strip trimming.
2.0 1.8 1.6 1.4 1.2 1.0 70 60 50 40
1 2
1 2 0
50
100
150
200
Fig. 12. Variation in Ra and Pc of cold-rolled strip in the course of strip trimming for the upper (1) and lower (2) rollers.
rollers and the trimmed sheet (KRabur = 0.75; KRabur = 0.45): Before At the beAfter electroeAfter Parameter ginning of trimming rosive burnishing trimming 200 t treatment Ra, µm Pc, cm–1 HSC, cm–1
2.5–3.5 55–60 60–75
295
1.8–2.6 55–60 60–75
0.9–1.4 55–60 60–75
0.8–1.2 50–55 55–70
The manufacture of experimental batches of coldrolled sheet confirms the effectiveness of this system. In
STEEL IN TRANSLATION
Vol. 37
No. 3
2007
REFERENCES 1. Gusev, Yu. B., Kosonogova, S. A., and Dubovov, D. A., Evaluating the Surface Microgeometry of Sheet for Auto Bodies, Stal, 2002, no. 1, pp. 48–49. 2. Garber, E. A., Gorelik, P. B., Diligenskii, E. V., et al., Influence of Cold-Rolling Conditions and Microgeometry of the Rollers on the Roughness of Cold-Rolled Strip, Proizv. Prokata, 1999, no. 6, pp. 7–10. 3. Usenko, Yu. I., Ivanov, V. I., Nesterenko, T. N., and Sereda, B. P., Technology and Equipment for Applying Roughness to Cold-Rolled Strip, Proizv. Prokata, 1999, no. 6, pp. 7–10. 4. Simao, I. M., Aspinwall, D. K., Wise, M. L. H., and ElMenshawy, F., Electrical Discharge Texturing of Cold Mill Work Rolls Using Different Tool Electrode Materials, Iron Steel Engineer, 1996, no. 3, pp. 43–47. 5. Zlov, V. E., Antipenko, A. I., Nosov, V. P., and Kochneva, T. M., Prospects for the Development of ColdRolled Sheet Production at OAO MMK, Stal, 2002, no. 1, pp. 48–49. 6. Bodyaev, Yu. A., Gornunov, A. V., Radionov, A. F., et al., Influence of Electroerosively Treated Working Rollers in a Trimming Mill on the Surface Microtopography of the Sheet, Stal, 2006, no. 5, pp. 90–94. 7. Gorbunkov, S. G., Dolzhenkov, A. Yu., Traino, A. I., et al., Formation of the Surface Microrelief in Texturing Rollers and Producing a Dull Finish of Precision Strip, Stal, 2003, no. 1, pp. 77–79. 8. Raimbekov, A. M., Teve, V. I., Zakolyukin, S. B., and Isaeva, T. I., Evaluating the Effectiveness of Electroerosive Texturing of Working Rollers, Stal, 2006, no. 2, pp. 38–41.
