Method for synergistically regulating low coercive force of soft magnetic steel of electromagnetic valve by cold working residual stress

Abstract

The present application relates to soft magnetic steel residual stress regulation technical field, specifically relates to a kind of electromagnetic valve soft magnetic steel cold working residual stress low coercivity synergic regulation method.Cold finishing is carried out to soft magnetic steel blank, and cold working part with section area compression rate of 1.5%~4.0% is obtained;It is placed in insulating straight positioning fixture, short pulse group, long pulse group and bridging pulse group are sequentially applied along the axial direction, so that the three groups of pulses correspond to surface layer, core and transition region respectively, and the surface temperature after each discharge is lower than 320 DEG C;After processing, keep radial positioning, naturally cool to below 30 DEG C and stand for 12~24h.The method reduces the residual stress difference between the surface layer and the core of cold working part and reduces the coercivity by the sequential action of different electromagnetic penetration depth pulses and the matching of nominal electric action amount, while maintaining cold hardness and straightness.

Description

Technical Field

[0001] This invention relates to the field of residual stress control technology for soft magnetic steel, and specifically to a method for coordinated control of low coercivity of cold-worked residual stress in soft magnetic steel for electromagnetic valves. Background Technology

[0002] The soft magnetic steel parts of solenoid valves need to quickly engage and disengage during repeated on / off cycles, maintaining stable clearances and linearity of movement. After cold working processes such as turning, sizing, and straightening, the surface and core of the soft magnetic steel exhibit different degrees of plastic deformation, easily leading to unevenly distributed residual stress along the radial direction. This residual stress increases the resistance to magnetic domain movement and enhances coercivity; it may also continue to be released during subsequent assembly and cyclic use, causing bending or dimensional changes in the parts.

[0003] Existing soft magnetic materials typically employ annealing to reduce coercivity and processing stress. For example, patent CN120272678A discloses a heat treatment process that improves the magnetic properties of soft magnetic stainless steel through high-temperature holding and slow cooling. Patent CN112410532A discloses a process for treating ring-shaped parts by selecting electrical pulse, magnetic pulse, or electromagnetic coupling based on the rate of dimensional change after the release of coercivity and residual stress. Patent CN113502379A further discloses adjusting the pulse width by regulating the discharge circuit time constant to adjust the depth of electromagnetic force application and the residual stress distribution in different regions.

[0004] While high-temperature annealing can reduce stress and coercivity, it may simultaneously weaken the cold-worked state and alter the straightness and dimensional accuracy of slender parts. Existing electromagnetic pulse methods can reduce residual stress and have proposed adjusting the pulse depth, but they lack continuous matching for the stress difference between the surface, core, and transition region of cold-worked soft magnetic steel parts. Furthermore, excessively high pulse heat input can reduce hardness. Therefore, the constraints between low coercivity, cold-worked hardness, and assembly dimensional stability still need to be addressed. Summary of the Invention

[0005] The technical problem to be solved by this invention is that the residual stress of the soft magnetic steel parts of the cold-worked solenoid valve is unevenly distributed radially, which easily increases coercivity and causes dimensional changes; overall high-temperature annealing will weaken the cold work hardness and reduce the straightness, making it difficult to balance low coercivity and assembly dimensional stability.

[0006] Based on the above reasons, the present invention provides a method for synergistic control of cold-work residual stress and low coercivity of soft magnet steel in electromagnetic valves, comprising the following steps: S1, cold working and finishing of the soft magnetic steel blank to obtain cold-worked soft magnetic steel parts for solenoid valves with a cross-sectional area compression rate of 1.5%~4.0%; S2, the cold-worked part is placed into an insulated linear positioning fixture and its two ends are clamped between water-cooled copper electrodes. The current direction is along the axial direction of the cold-worked part, and short pulse groups, long pulse groups, and bridging pulse groups are processed sequentially. The short pulse group includes 1 to 3 pulses with a peak current density of 3800 to 5200 A / mm. 2 Damped oscillating pulses with a pulse width of 60~150μs are applied. The ratio of the equivalent electromagnetic penetration depth δs of the short pulse group to the radius R of the cold-worked part, δs / R, is 0.15~0.30. A long pulse group is applied 2~5s after the short pulse group ends. The long pulse group includes 1~3 pulses with a peak current density of 1200~2500A / mm². 2 Damped oscillating pulses with a pulse width of 0.4~1.2ms are used. The ratio of the equivalent electromagnetic penetration depth δl to R of the long pulse group is 0.80~1.10, and the ratio of the nominal electric effect of the short pulse group to the long pulse group is 0.75~1.25. A bridging pulse group is applied 2~5s after the end of the long pulse group. The bridging pulse group includes 1~2 pulses with a peak current density of 2200~3800A / mm. 2 Damped oscillatory pulses with a pulse width of 0.2~0.5ms are used. The ratio of the equivalent electromagnetic penetration depth δm to R of the bridging pulse group, δm / R, is 0.45~0.65. The ratio of the nominal electric effect of the bridging pulse group to the geometric mean of the nominal electric effects of the short pulse group and the long pulse group is 0.8~1.2. The nominal electric effect of each pulse group is calculated according to Σ(Jp 2 τ) is calculated, where Jp is the peak pulse current density, τ is the pulse width, and the axial mid-surface temperature of the cold-worked part measured 0.2s after each discharge is kept below 320°C; S3. After the last discharge, release the contact pressure of the water-cooled copper electrode on the end face of the cold-worked part, maintain the radial positioning state of the cold-worked part in the insulated linear positioning fixture, let it cool naturally to below 30°C and continue to stand for 12~24h, then release the insulated linear positioning fixture to obtain the cold-worked stabilized soft magnetic steel part.

[0007] Preferably, in step S1, the cold finishing includes passing the soft magnetic steel billet through a cemented carbide sizing die at a speed of 0.08 m / s at 20~30℃, and then performing 2~4 passes of alternating reverse straightening with three rollers; the cross-sectional area compression ratio η is calculated as η=(A0-A1) / A0×100%, where A0 is the cross-sectional area before sizing and A1 is the cross-sectional area after sizing.

[0008] Preferably, the diameter of the cold-worked part obtained in step S1 is 4.00 mm and the length is 60.0 mm.

[0009] Preferably, in step S2, the interval between adjacent short pulses is 60~90s, the interval between adjacent long pulses is 45~75s, and the interval between adjacent bridging pulses is 30s.

[0010] Preferably, in step S2, the waveform square integral coefficient κ of each damped oscillation pulse is 0.18, and κ is calculated according to κ=∫J 2 dt / (Jp 2 τ) is calculated, where J is the current density that varies with time.

[0011] Preferably, in step S2, the equivalent electromagnetic penetration depth δ is calculated as δ = 2 × (ρτ / μe). 1 / 2 Calculate, where ρ is taken as 4.0 × 10 -7 Ω·m, τ is the pulse width converted to s, μe is the dynamic equivalent absolute permeability under the corresponding pulse waveform, and the unit of μe is H / m. The μe used in the short pulse group, long pulse group and bridging pulse group were measured on the same batch of test specimens under the same cold working condition without applied electrical pulse, after the short pulse group was completed, and after the short pulse group and long pulse group were completed, respectively.

[0012] Preferably, in step S2, the pulse peak current density Jp is calculated by dividing the measured peak current by the minimum cross-sectional area of ​​the cold-worked part, and the pulse width τ is the time from the first time the current reaches 0.1Jp to the last time it decays to 0.1Jp.

[0013] Preferably, the insulating linear positioning fixture includes two sets of alumina V-blocks and non-magnetic spring pressure plates, the center distance between the two alumina V-blocks is 40mm, and the radial clamping force is 15N; one end of the insulating linear positioning fixture restricts the axial displacement of the cold-worked parts, and the other end allows the cold-worked parts to expand and contract axially.

[0014] Preferably, in step S2, the end faces of the cold-worked part are precision machined to Ra0.8μm, the contact pressure of the water-cooled copper electrode to the end face is 10MPa, the cooling water temperature is 18℃, and the cooling water flow rate is 3.0L / min.

[0015] Preferably, the cross-sectional area compression ratio of the cold-worked part obtained in step S1 is 2.93%; in step S2, the short pulse group includes two pulses with a peak current density of 4400 A / mm². 2 A short pulse with a pulse width of 90 μs is applied, with an interval of 75 s between adjacent short pulses. A long pulse group is applied 3 s after the end of the short pulse group. The long pulse group consists of two pulses with a peak current density of 1800 A / mm². 2 A long pulse with a pulse width of 0.55ms is applied, with an interval of 45s between adjacent long pulses. A bridging pulse group is applied 3s after the end of the long pulse group. The bridging pulse group consists of two pulses with a peak current density of 2800A / mm². 2 A bridging pulse with a pulse width of 0.23ms is used, with an interval of 30s between adjacent bridging pulses; in step S3, the cold-worked parts are naturally cooled to 27°C and then left to stand for 18 hours.

[0016] The beneficial effects of this invention are: (1) Cold-worked parts are processed in the order of short pulse group, long pulse group and bridging pulse group so that the electric pulse action covers the surface, core and the transition area between the two in sequence. The bridging pulse group supplements the effect of medium electromagnetic penetration depth, which helps to reduce the stress adjustment difference between the surface and the core, makes the residual stress release degree at different radial positions more similar, and reduces DC coercivity on this basis.

[0017] (2) By limiting the ratio of the nominal electrical action of the short pulse group to the long pulse group, and matching the nominal electrical action of the bridging pulse group with the geometric mean of the nominal electrical action of the first two groups, the electrical pulse action on the surface or core can be prevented from being too weak or too strong. Combined with the upper limit of the surface temperature after each discharge, excessive heat input can be reduced while reducing residual stress and coercivity, thus maintaining the hardness and dimensional condition of cold-worked parts.

[0018] (3) After the electrical pulse ends, the electrode contact pressure is released, while the cold-worked part is kept in the radial positioning state in the insulated linear positioning fixture. Cooling and static placement are completed in the positioning state, which can limit the bending caused by stress redistribution. The resulting part has a small initial straightness deviation and can still maintain a low straightness increment after repeated on-off cycles.

[0019] (4) This method completes stress adjustment at a temperature lower than the overall high-temperature annealing temperature, without requiring overall recrystallization of the soft magnetic steel parts. The treated parts can retain the hardness formed by cold work strengthening, and shorten the release delay of the solenoid valve after power-off by reducing the residual stress difference and coercivity, thereby taking into account magnetic response, assembly accuracy and cyclic dimensional retention. Attached Figure Description

[0020] Figure 1 This is a process flow diagram for the coordinated control of residual stress and low coercivity of soft magnet steel during cold working in solenoid valves.

[0021] Figure 2 This is a comparison graph showing the power-off release delay between the example and the comparative example.

[0022] Figure 3 This is a comparison chart of the straightness increments of the example and the comparative example after 100,000 on / off cycles. Detailed Implementation

[0023] I. Raw Materials and Common Operating Instructions The cold-drawn round bars of Silicon Core Iron B-FM soft magnetic steel were supplied by Carpenter Technology (Shanghai) Distribution Company Ltd. The composition was 2.50% silicon, 0.40% manganese, 0.120% phosphorus, 0.03% carbon, with the balance being iron. The resistivity at 21°C was 40 μΩ·cm, and the diameter of the raw material round bars was 12.7 mm. Round bars of the same grade and in the same supply condition were used in all preparation examples.

[0024] Cold working finishing uses a carbide sizing die, and the three working rollers of the three-roll straightener are arranged at equal angles along the circumference. The cross-sectional area compression ratio is calculated as η=(A0-A1) / A0×100%, where A0 is the cross-sectional area before sizing and A1 is the cross-sectional area after sizing. The final diameter of each batch of cold-worked parts is 4.00mm, and the final length is 60.0mm.

[0025] The electrical pulse processing employs a capacitor-stored damped oscillating power supply with a peak current of no less than 70kA and an adjustable pulse width of 50μs to 1.5ms. The current direction of the component is axial. Both end faces are precision machined to Ra 0.8μm and then clamped between water-cooled copper electrodes. The end face contact pressure is 10MPa, the cooling water temperature is 18℃, and the flow rate is 3.0L / min. The peak current density Jp is calculated by dividing the measured peak current by the minimum cross-sectional area of ​​the component. The pulse width τ is the time from the initial current reaching 0.1Jp to the final current decaying to 0.1Jp. Except for Comparative Example 8, the waveform square integral coefficient κ = ∫J 2 dt / (Jp 2 τ) was adjusted to 0.18.

[0026] Using the part radius R as the scale, the equivalent electromagnetic penetration depth is calculated according to δ=2√(ρτ / μe). In the formula, ρ is fixed at 4.0×10⁻⁶. -7 Ω·m, τ is determined according to the aforementioned 0.1Jp boundary and converted to s, μe is the dynamic equivalent absolute permeability under the corresponding pulse waveform, and the unit of μe is H / m; δ and R use the same unit of length. The nominal electric effects of the short pulse group, long pulse group and bridging pulse group are respectively calculated according to Σ(Jp) of each group. 2 τ) Calculation. 0.2s after each discharge, the axial mid-surface temperature of the part is measured using a dual-color infrared thermometer.

[0027] μe was determined as follows. For each pulse condition, three test pieces were taken from the same batch as the corresponding part, with the same diameter and cross-sectional area compression ratio. Test pieces used for determining the μe of the short pulse group were kept in a cold working state; test pieces used for determining the μe of the long pulse group had their corresponding short pulse group pre-completed; test pieces used for determining the μe of the bridging pulse group had their corresponding short and long pulse groups pre-completed. Each test piece used the same power supply, fixture, electrode pressure, and cooling conditions as the corresponding part, and four-terminal voltage measurement points with a spacing of 20.0 mm were set at the axial center. The current I(t) was measured using a Rogowski coil with a bandwidth of not less than 20 MHz, and the four-terminal voltage U(t) was measured using an isolated differential probe with a bandwidth of not less than 20 MHz. The two signals were acquired synchronously at a sampling rate of not less than 100 MS / s.

[0028] Under the same fixture and lead arrangement, the parasitic inductance L0 of the external measurement circuit is calibrated using a copper reference rod of the same size; the resistivity of the copper reference rod is determined using the four-terminal DC method. L0 is determined by subtracting the voltage drop across the resistance from the measured voltage of the reference rod, based on the least squares relationship between the obtained voltage and dI / dt. The correction voltage for the test piece is calculated as Us(t) = U(t) - L0dI / dt.

[0029] Treating the test specimen as a solid cylinder with a length much greater than its radius, solve the axisymmetric magnetic diffusion equation for the cylinder within the range 0 ≤ r ≤ R: .

[0030] The initial condition is Hθ(r,0)=0, and the boundary conditions are Hθ(0,t)=0 and Hθ(R,t)=I(t) / (2πR). The solution results are then used to calculate the boundary conditions. Calculate the axial electric field on the specimen surface, where the value is taken at r=R; then obtain the model voltage using Um(t)=0.0200m×Em(t). The radial surface is discretized using at least 200 equidistant elements, with a time step no greater than the sampling interval. (The last part, "1.0×10," appears to be an unrelated fragment and is omitted from the translation.) -6 ~3.0×10 -3 The normalized root mean square error Q = [Σ(Us,k-Um,k)] is calculated by varying μe within the range of H / m, from the initial current reaching 0.1 Jp to the final current decaying back to 0.1 Jp. 2 / ΣUs,k 2 ] 1 / 2 When the value reaches its minimum, the corresponding μe is taken as the dynamic equivalent absolute permeability of the test piece; the identification result is valid when Q is not greater than 0.05. When the coefficient of variation of μe of the three test pieces is not greater than 10%, the arithmetic mean is taken to calculate the δ of the pulse group; when the coefficient of variation is greater than 10%, the voltage leads and parasitic inductance are checked and calibrated before re-measurement.

[0031] The insulated linear positioning fixture consists of two sets of alumina V-blocks and non-magnetic spring pressure plates. The center distance between the two V-blocks is 40mm, and the radial clamping force is 15N. One end of the fixture restricts axial displacement, while the other end allows axial thermal expansion and contraction. The electrical pulse treatment is completed within the fixture. After the final discharge, the contact pressure of the copper electrodes is released, but the radial positioning state of the part is maintained.

[0032] II. Preparation Example Preparation Example 1: Step (1) Cut out cold-drawn round bars of Silicon Core Iron B-FM soft magnetic steel, turn them at 23℃ to form 60 blanks with a diameter of 4.04mm and a length of 58.9mm, and then finish turn the end faces at both ends.

[0033] Step (2) At 25℃, each blank is passed through a cemented carbide sizing die with a diameter of 4.00 mm at a speed of 0.08 m / s, with a cross-sectional area compression rate of 1.97%; then, two passes of three-roll alternating reverse straightening are performed, with reverse deflections of 0.18 mm and 0.11 mm respectively. The two ends are trimmed to a length of 60.0 mm and burrs are removed.

[0034] A cold-worked part with a cross-sectional area compression ratio of 1.97% was obtained.

[0035] Preparation Example 2: Step (1) Cut out cold-drawn round bars of Silicon Core Iron B-FM soft magnetic steel, turn them at 24℃ to form 60 blanks with a diameter of 4.06mm and a length of 58.4mm, and then finish turn the end faces at both ends.

[0036] Step (2) At 25℃, each blank is passed through a cemented carbide sizing die with a diameter of 4.00 mm at a speed of 0.08 m / s, with a cross-sectional area compression rate of 2.93%. Then, three passes of alternating reverse straightening with three rollers are performed, with reverse deflections of 0.24 mm, 0.18 mm, and 0.12 mm respectively. The two ends are trimmed to a length of 60.0 mm and burrs are removed.

[0037] A cold-worked part with a cross-sectional area compression ratio of 2.93% was obtained.

[0038] Preparation Example 3: Step (1) Cut Silicon Core Iron B-FM soft magnetic steel cold-drawn round bars, turn them at 26℃ to form 60 blanks with a diameter of 4.08mm and a length of 57.8mm, and then finish turn the end faces at both ends.

[0039] Step (2) At 25℃, each blank is passed through a cemented carbide sizing die with a diameter of 4.00mm at a speed of 0.08m / s, with a cross-sectional area compression rate of 3.88%; then, four passes of alternating three-roll reverse straightening are performed, with reverse deflections of 0.30mm, 0.25mm, 0.18mm and 0.12mm respectively. The two ends are trimmed to a length of 60.0mm and burrs are removed.

[0040] A cold-worked part with a cross-sectional area compression ratio of 3.88% was obtained.

[0041] III. Examples

[0042] Example 1:

[0043] Step (1): Prepare 20 cold-worked parts according to the method of Preparation Example 1, and place each part into an insulated linear positioning fixture and clamp it between water-cooled copper electrodes. First, apply two peak current densities of 3950 A / mm². 2 A short pulse with a pulse width of 70μs, an interval of 60s between adjacent short pulses, and a δs / R of 0.16.

[0044] Step (2): Apply two peak current densities of 1700 A / mm² 3 seconds after the last short pulse ends. 2 A long pulse with a pulse width of 0.50 ms and an interval of 45 s between adjacent long pulses has a δl / R of 0.96; the nominal electrical action ratio between the short pulse group and the long pulse group is 0.76.

[0045] Step (3): Apply two peak current densities of 2200 A / mm² 3 seconds after the last long pulse ends. 2 The bridging pulses had a pulse width of 0.22 ms, with an interval of 30 s between adjacent bridging pulses, and a δm / R ratio of 0.46. The ratio of the nominal electrical activity of the bridging pulse group to the geometric mean of the nominal electrical activity of the short pulse group and the long pulse group was 0.85. The highest surface temperature measured during the process was 162℃.

[0046] Step (4) After the last discharge, keep the part in the insulated linear positioning fixture and let it cool naturally to 28°C. Continue to stand for 12 hours, then remove the fixture to obtain the cold-work stabilized soft magnetic steel part.

[0047] Example 2:

[0048] Step (1): Twenty cold-worked parts were prepared according to the method in Example 2. Each part was then placed into an insulated linear positioning fixture and clamped between water-cooled copper electrodes. Two peak current densities of 4400 A / mm² were applied first. 2 A short pulse with a pulse width of 90μs, an interval of 75s between adjacent short pulses, and a δs / R of 0.22.

[0049] Step (2): Apply two peak current densities of 1800 A / mm² 3 seconds after the last short pulse ends.2 The pulse width is 0.55ms, the interval between adjacent long pulses is 45s, and the δl / R is 0.95; the ratio of the nominal electrical action of the short pulse group to the long pulse group is 0.98.

[0050] Step (3): Apply two peak current densities of 2800 A / mm² 3 seconds after the last long pulse ends. 2 The bridging pulses had a pulse width of 0.23 ms, with an interval of 30 s between adjacent bridging pulses, and a δm / R ratio of 0.55. The ratio of the nominal electrical activity of the bridging pulse group to the geometric mean of the nominal electrical activity of the short pulse group and the long pulse group was 1.02. The highest surface temperature measured during the process was 229℃.

[0051] Step (4) After the last discharge, keep the part in the insulated linear positioning fixture and let it cool naturally to 27°C. Continue to stand for 18 hours, then remove the fixture to obtain the cold-work stabilized soft magnetic steel part.

[0052] Example 3:

[0053] Step (1): Twenty cold-worked parts were prepared according to the method in Example 3. Each part was then placed into an insulated linear positioning fixture and clamped between water-cooled copper electrodes. A peak current density of 5100 A / mm² was first applied. 2 A short pulse with a pulse width of 140μs and a δs / R ratio of 0.29.

[0054] Step (2): Apply two peak current densities of 1500 A / mm² 4 seconds after the short pulse ends. 2 The pulse width is 0.65ms, the interval between adjacent long pulses is 60s, and the δl / R is 0.83; the ratio of the nominal electrical action of the short pulse group to the long pulse group is 1.24.

[0055] Step (3) Apply a peak current density of 3200 A / mm² 4 seconds after the last long pulse ends. 2 The bridging pulse with a pulse width of 0.37 ms had a δm / R ratio of 0.64. The ratio of the nominal electrical activity of the bridging pulse group to the geometric mean of the nominal electrical activity of the short pulse group and the long pulse group was 1.16. The highest surface temperature measured during the process was 224℃.

[0056] Step (4) After the last discharge, keep the part in the insulated linear positioning fixture and let it cool naturally to 29°C. Continue to stand for 24 hours, then remove the fixture to obtain the cold-work stabilized soft magnetic steel part.

[0057] Example 4:

[0058] Step (1): Twenty cold-worked parts were prepared according to the method in Example 2. Each part was then placed into an insulated linear positioning fixture and clamped between water-cooled copper electrodes. Three peak current densities of 4600 A / mm² were applied first.2 A short pulse with a pulse width of 100μs, an interval of 90s between adjacent short pulses, and δs / R of 0.25.

[0059] Step (2): Apply three peak current densities of 2100 A / mm² 5 seconds after the last short pulse ends. 2 The pulse width is 0.53ms, the interval between adjacent long pulses is 60s, and the δl / R is 1.08; the ratio of the nominal electrical action of the short pulse group to the long pulse group is 0.91.

[0060] Step (3) Apply a peak current density of 3500 A / mm² 5 seconds after the end of the last long pulse. 2 The bridging pulse with a pulse width of 0.45 ms had a δm / R ratio of 0.59; the ratio of the nominal electrical activity of the bridging pulse group to the geometric mean of the nominal electrical activity of the short pulse group and the long pulse group was 0.83. The highest surface temperature measured during the process was 306℃.

[0061] Step (4) After the last discharge, keep the part in the insulated linear positioning fixture and let it cool naturally to 28°C. Continue to stand for 16 hours, then remove the fixture to obtain the cold-work stabilized soft magnetic steel part.

[0062] Example 5:

[0063] Step (1): Twenty cold-worked parts were prepared according to the method in Example 2. Each part was then placed into an insulated linear positioning fixture and clamped between water-cooled copper electrodes. A peak current density of 4800 A / mm² was first applied. 2 A short pulse with a pulse width of 120μs and a δs / R ratio of 0.18.

[0064] Step (2): Apply two peak current densities of 1300 A / mm² 3 seconds after the short pulse ends. 2 The pulse width is 0.75ms, the interval between adjacent long pulses is 75s, and the δl / R is 1.02; the ratio of the nominal electrical action of the short pulse group to the long pulse group is 1.09.

[0065] Step (3): Apply a peak current density of 2500 A / mm³ 3 seconds after the last long pulse ends. 2 A bridging pulse with a pulse width of 0.38 ms was used, with a δm / R ratio of 0.50. The ratio of the nominal electrical activity of the bridging pulse group to the geometric mean of the nominal electrical activity of the short pulse group and the long pulse group was 0.90. The highest surface temperature measured during the process was 173℃.

[0066] Step (4) After the last discharge, keep the part in the insulated linear positioning fixture and let it cool naturally to 26°C. Continue to stand for 20 hours, then remove the fixture to obtain the cold-work stabilized soft magnetic steel part.

[0067] IV. Comparative Example Comparative Example 1: Except for not applying short pulses, long pulses, and bridging pulses, and keeping the cold-worked parts obtained in Example 2 in an insulated linear positioning fixture for 18 hours, the other part specifications and fixture conditions are the same as in Example 2, resulting in cold-worked soft magnetic steel parts that have not been treated with electrical pulses.

[0068] Comparative Example 2: The short pulse group, long pulse group, and bridging pulse group in Example 2 are replaced with three peak current densities of 2500 A / mm. 2 A long pulse with a pulse width of 0.57 ms and a δ / R ratio of 1.02 was used, with the interval between adjacent pulses set to 3 s, and all other conditions were the same as in Example 2. The difference between the sum of the nominal electrical effects of these three long pulses and the sum of the nominal electrical effects of the three sets of pulses in Example 2 was 0.3%, resulting in a soft magnetic steel part treated with a single penetration depth electrical pulse.

[0069] Comparative Example 3: Except for the removal of the bridging pulse in Example 2, the other raw materials, pulse parameters, pulse sequence, fixture holding and static conditions are the same as in Example 2, resulting in a soft magnetic steel part without the application of a bridging pulse.

[0070] Comparative Example 4: Except for changing the pulse sequence in Example 2 to long pulse group, short pulse group and bridging pulse group, and the interval between adjacent pulse groups is 3s, the peak current density, pulse width, number of pulses, interval within the group, clamp holding and static conditions of the other pulse groups are the same as in Example 2, and a soft magnetic steel part with changed pulse sequence is obtained.

[0071] Comparative Example 5: Except for changing the long pulse width in Example 1 from 0.50ms to 0.54ms, the other raw materials, pulse parameters, pulse sequence, intra-group intervals, fixture holding, and stationary conditions are the same as in Example 1. The ratio of the nominal electrical action of the short pulse group to the long pulse group changes from 0.76 to 0.70, and the ratio of the nominal electrical action of the bridging pulse group to the geometric mean of the nominal electrical action of the short pulse group and the long pulse group is 0.82, resulting in a soft magnetic steel part with a ratio of the nominal electrical action of the short pulse group to the long pulse group lower than the lower limit.

[0072] Comparative Example 6: Except for the short pulse peak current density in Example 3, which was 5100 A / mm 2 Change to 5200A / mm 2Apart from the above, the other raw materials, pulse parameters, pulse sequence, intra-group intervals, fixture holding, and static conditions are the same as in Example 3. The ratio of the nominal electrical action of the short pulse group to the long pulse group changes from 1.24 to 1.29, and the ratio of the nominal electrical action of the bridging pulse group to the geometric mean of the nominal electrical action of the short pulse group and the long pulse group is 1.14, resulting in a soft magnetic steel part with a ratio of the nominal electrical action of the short pulse group to the long pulse group that is higher than the upper limit.

[0073] Comparative Example 7: Except for immediately releasing the insulating linear positioning fixture after the last bridging pulse in Example 2, allowing the part to cool naturally and stand for 18 hours without radial positioning, the other raw materials and electrical pulse parameters are the same as in Example 2, resulting in a soft magnetic steel part that is not cooled and stood in the positioning state.

[0074] Comparative Example 8: Except for adjusting the damping resistor of the electrical pulse power supply to change the waveform square integral coefficient κ of each pulse from 0.18 to 0.28, the peak current density, pulse width, number of pulses, equivalent electromagnetic penetration depth, pulse sequence, intra-group interval, fixture holding, and static conditions in Example 2 remained unchanged. The highest surface temperature measured during the process was 344°C, resulting in soft magnetic steel parts with a surface temperature exceeding 320°C.

[0075] Comparative Example 9: Step (1) Prepare 20 cold-worked parts according to the method of Preparation Example 2. Without applying an electrical pulse, place the cold-worked parts in a quartz tube furnace with an inner diameter of 50 mm. The protective gas consists of hydrogen and nitrogen, with hydrogen comprising 5% and nitrogen comprising 95%, and a dew point of -40°C. The protective gas is introduced at 3.0 L / min and purged for 30 min, then the temperature is increased to 850°C at 5°C / min and held for 120 min.

[0076] Step (2) After cooling to 200°C at 3°C / min, remove the part, cool it to 25°C in air and let it stand for 18 hours to obtain a soft magnetic steel part that has undergone conventional high-temperature magnetic annealing.

[0077] V. Performance Testing After the specified settling time was completed, the parts obtained in each embodiment and comparative example were conditioned for 2 hours in an environment of 23±2℃ and relative humidity of 45%~65%. Residual stress, Vickers hardness, and DC coercivity were measured using 3 parallel specimens each; straightness, roundness, power-off release delay, and straightness increment after cycling were measured using 5 parallel specimens each.

[0078] 1. Axial residual stress in the surface and core layers Refer to EN15305:2008 and adopt sin 2Axial residual stress was measured using the ψ method. CrKα radiation and α-Fe{211} diffraction were used, with a tube voltage of 30 kV, a tube current of 6 mA, and a collimated spot diameter of 0.5 mm. The ψ angles were successively set to 0°, ±15°, ±30°, and ±45°. The average of positive and negative ψ measurements under the same |ψ| was taken. A least-squares linear fit was performed on the interplanar spacing d and sin²ψ; the fitting slope M and sin²ψ were then used as the basis for the measurement. 2 The intercept d0 at ψ=0 is used to calculate the axial stress according to m=E×M / [(1+ν)d0], where the elastic modulus E is taken as 210GPa and Poisson's ratio ν is taken as 0.29. Compressive stress is recorded as negative and tensile stress as positive. The axial residual stress of the surface layer is measured at the midpoint of the axial direction of the part, and the average value is taken at three positions 120° apart in the circumferential direction. This average value is also recorded as the initial reading m0 before layer-by-layer removal.

[0079] The residual stress in the core was removed using a combination of layer-by-layer electrolytic removal and X-ray measurement. Separate parts from the same group were tested at 50°C using an electrolyte prepared from a 4:1 mass ratio of 65% phosphoric acid aqueous solution and 98% sulfuric acid, at a flow rate of 0.20 A / cm. 2 A uniform electrolysis was performed on the circumference of the part using a current density at a rotation speed of 60 r / min. After removing 0.20 mm radially each time, the axial residual stress was measured at three positions 120° apart along the circumference until the remaining diameter was 1.20 mm. Stress redistribution corrections were applied to each reading based on the removal of cross-sectional area, elastic modulus of 210 GPa, and Poisson's ratio of 0.29. Using the original radius R and the radius of the current measurement position r, the final three correction values ​​(r / R) of 0.50, 0.40, and 0.30 were applied to (r / R). 2 Perform a least-squares linear fit, taking (r / R) 2 The intercept at which the stress is zero is taken as the axial residual stress in the core. The stress difference between the surface and the core is taken as the absolute value of the difference between the two.

[0080] 2. Cross-sectional Vickers hardness The Vickers hardness (HV 0.5) was tested according to ISO 6507-1:2023, with a test force of 4.903 N and a holding time of 10 s. The part was cut, inlaid, and polished at low speed along its axial center. One indentation was made at the center of the cross-section, one indentation in each of the four directions with r / R = 0.50, and one indentation in each of the four directions with r / R = 0.85. The average value of the nine indentations was taken as the average Vickers hardness of the cross-section, the average value of the four outer indentations was taken as the surface hardness, and the value of the center indentation was taken as the core hardness. The difference between the surface and core hardness was taken as the absolute value of the difference between the two.

[0081] 3. DC coercivity The DC coercivity was measured under open magnetic circuit conditions according to IEC60404-7:2019. Before testing, the part was demagnetized using a 50Hz alternating magnetic field with progressively decreasing amplitude. The part was placed axially in the coaxial region of the excitation coil and the detection coil. The maximum magnetic field strength was 4000 A / m, and the magnetic field scanning rate was 80 A / (m·s). Three complete hysteresis loops were measured continuously, and the coercivity of the third loop was taken.

[0082] 4. Straightness and roundness Straightness was measured according to ISO 12780-2:2011. A coordinate measuring machine was used to scan along four generatrices spaced 90° apart on the circumference of the part. Areas 5mm from each end were removed. The maximum axial deviation was obtained using the least-squares centerline as a reference and converted to mm / 100mm. Roundness was measured according to ISO 12181-2:2011. A full circumferential scan was performed at three cross-sections 10mm, 30mm, and 50mm from one end. The roundness was evaluated using the least-squares circle method, and the average of the roundness values ​​at each cross-section was taken as the roundness of the part.

[0083] 5. Power outage release delay The component, used as an armature, is placed within a non-magnetic guide sleeve with a radial clearance of 0.10 mm. The initial working air gap between the armature end face and the 4.0 mm diameter low-carbon steel pole shoe is 0.20 mm. The excitation coil has an inner diameter of 5.0 mm, an outer diameter of 11.0 mm, and an effective length of 30 mm. It is wound with 1200 turns of 0.20 mm diameter enameled copper wire, and has a DC resistance of 28.5 Ω at 23°C. The return spring stiffness is 0.40 N / mm, and the initial axial force is 0.80 N. A 24 V DC voltage is applied for 1.0 s using a metal-oxide-semiconductor field-effect transistor. The turn-off circuit is connected to a 68 V transient suppressor without a freewheeling diode in parallel.

[0084] Each part undergoes 5 pre-cycles. During the 6th power-off cycle, the coil current and armature displacement are simultaneously recorded at a sampling frequency of 20kHz. The time from when the coil current drops to 10% of the steady-state value to when the armature axial displacement reaches 0.20mm is defined as the power-off release delay.

[0085] 6. Straightness increment after 100,000 cycles Using the same electromagnetic component as the power-off release delay, 100,000 on / off cycles were performed at a frequency of 1Hz and a duty cycle of 50%, with an ambient temperature of 25±2℃. After the cycle, the part was conditioned at 23±2℃ for 2 hours, and then the straightness was measured again using the aforementioned method. The straightness after the cycle was subtracted from the straightness before the cycle, and the difference was taken as the straightness increment after 100,000 cycles.

[0086] Results Analysis In Examples 1-5, the absolute values ​​of axial residual stress in both the surface and core layers were no higher than 80 MPa, the absolute value of the surface-to-core stress difference was 8-26 MPa, the DC coercivity was 72-109 A / m, the average cross-sectional hardness was 194-218 HV0.5, the absolute value of the surface-to-core hardness difference was 8-17 HV0.5, and the straightness was 0.018-0.029 mm / 100 mm. These results indicate that, with a cold-working amount of 1.97%-3.88% and different combinations of electromagnetic penetration depths, sequential processing with short pulses, long pulses, and bridging pulses can make the stress release degree at different radial locations approximately similar, while preserving the cold-working hardness and the straightened dimensional state.

[0087] Under the same conditions of raw materials and cold working, the absolute value of the surface-core stress difference in Example 2 decreased from 300 MPa to 14 MPa compared to Comparative Example 1, and the DC coercivity decreased from 238 A / m to 84 A / m. At the same time, the average cross-sectional hardness remained at 212 HV0.5, and the straightness decreased from 0.038 mm / 100 mm to 0.018 mm / 100 mm. This shows that the layered electrical pulse treatment is beneficial to reducing the radial stress mismatch and coercivity of cold-worked parts, without causing the overall cold-worked strengthening state to disappear.

[0088] The total nominal electrical action of Example 2 is close to that of Comparative Example 2. However, Comparative Example 2 only uses a single long pulse penetration depth, and its surface-core stress difference absolute value is 113 MPa and DC coercivity is 146 A / m, both of which are higher than 14 MPa and 84 A / m of Example 2. This indicates that the effect obtained is not only determined by the total electrical pulse action, but is related to the different penetration depth pulse action that the surface, core and middle regions are subjected to in sequence.

[0089] In Example 2, compared to Comparative Example 3 where the bridging pulse was removed, the absolute value of the stress difference between the surface and the core decreased from 52 MPa to 14 MPa, the DC coercivity decreased from 123 A / m to 84 A / m, and the absolute value of the hardness difference between the surface and the core decreased from 22 HV0.5 to 9 HV0.5. This indicates that setting a bridging pulse with a medium penetration depth after the short and long pulses is beneficial to reducing the difference in the transition zone between the surface and the core.

[0090] Compared to Comparative Example 4, which changed the pulse sequence, Example 2 showed that the absolute value of the stress difference between the surface and the core decreased from 91 MPa to 14 MPa, the DC coercivity decreased from 151 A / m to 84 A / m, and the straightness decreased from 0.043 mm / 100 mm to 0.018 mm / 100 mm. This indicates that the pulse sequence of first acting on the surface, then on the core, and finally on the middle region is beneficial in avoiding the subsequent deep stress adjustment from amplifying the surface-core difference.

[0091] In Example 1, the ratio of the nominal electrical activity of the short pulse group to the long pulse group was 0.76. In Comparative Example 5, only the long pulse width was changed from 0.50 ms to 0.54 ms, reducing this ratio to 0.70. The absolute value of the surface-core stress difference increased from 23 MPa to 42 MPa, and the DC coercivity increased from 104 A / m to 119 A / m. In Example 3, this ratio was 1.24. In Comparative Example 6, only the peak current density of the short pulse was changed from 5100 A / mm². 2 Change to 5200A / mm 2 This raises the ratio to 1.29, increasing the absolute value of the stress difference between the surface and the core from 26 MPa to 43 MPa. This indicates that when the ratio is below 0.75 or above 1.25, the stress adjustment between the surface and the core no longer matches.

[0092] The residual stress and DC coercivity of Example 2 and Comparative Example 7 are similar. However, after cooling and settling without positioning, the straightness of Comparative Example 7 is 0.054 mm / 100 mm, and the straightness increment after 100,000 cycles is 0.025 mm / 100 mm, which are higher than 0.018 mm / 100 mm and 0.004 mm / 100 mm of Example 2, respectively. This indicates that maintaining radial positioning and completing cooling and settling after the electrical pulse ends is beneficial for fixing the shape state after stress rearrangement.

[0093] Comparative Example 8 had a maximum surface temperature of 344°C, with its DC coercivity decreasing to 68 A / m. However, the average cross-sectional hardness decreased to 174 HV0.5, and the straightness was 0.047 mm / 100 mm. The straightness increment after 100,000 cycles was 0.031 mm / 100 mm. Compared to Example 2, while temperatures exceeding 320°C could further reduce coercivity, they also weakened cold work hardness and dimensional retention. Therefore, limiting the surface temperature after each discharge to below 320°C is beneficial for maintaining a balance between improved magnetic properties and cold work strengthening.

[0094] Comparative Example 9, which underwent conventional high-temperature magnetic annealing, achieved a DC coercivity of 58 A / m and an absolute value of the surface-to-core stress difference of 5 MPa, but its average cross-sectional hardness was only 147 HV0.5, and its straightness was 0.066 mm / 100 mm. Although Example 2 had a higher coercivity than Comparative Example 9, its hardness and straightness remained at 212 HV0.5 and 0.018 mm / 100 mm, respectively, indicating that this method can achieve a balance between low coercivity, cold work hardness, and assembly dimensions without using overall high-temperature annealing.

[0095] Depend on Figure 2As can be seen, the power-off release delay of Examples 1-5 is 4.8-7.4 ms, which is lower than the 17.2 ms of Comparative Example 1 without electrical pulse treatment and also lower than the 10.8 ms of Comparative Example 2 with single penetration depth treatment. This result corresponds to the lower DC coercivity of the examples, indicating that the treated parts can release the magnetic retention state more quickly after power-off.

[0096] Depend on Figure 3 As can be seen, the straightness increment of Examples 1-5 after 100,000 on / off cycles is 0.004~0.009 mm / 100 mm; the increments for Comparative Example 7 (which did not maintain positioning), Comparative Example 8 (which was annealed at a temperature exceeding 320°C), and Comparative Example 9 (which underwent high-temperature annealing) are 0.025 mm / 100 mm, 0.031 mm / 100 mm, and 0.036 mm / 100 mm, respectively. These results demonstrate that the low-stress mismatch state formed by this method exhibits good dimensional retention under repeated on / off cycles.

Claims

1. A method for synergistic control of cold-work residual stress and low coercivity of soft magnet steel in an electromagnetic valve, characterized in that, Includes the following steps: S1, cold working and finishing of the soft magnetic steel blank to obtain cold-worked soft magnetic steel parts for solenoid valves with a cross-sectional area compression rate of 1.5%~4.0%; S2, the cold-worked part is placed into an insulated linear positioning fixture and its two ends are clamped between water-cooled copper electrodes. The current direction is along the axial direction of the cold-worked part, and short pulse groups, long pulse groups, and bridging pulse groups are processed sequentially. The short pulse group includes 1 to 3 pulses with a peak current density of 3800 to 5200 A / mm. 2 Damped oscillating pulses with a pulse width of 60~150μs are applied. The ratio of the equivalent electromagnetic penetration depth δs of the short pulse group to the radius R of the cold-worked part, δs / R, is 0.15~0.

30. A long pulse group is applied 2~5s after the short pulse group ends. The long pulse group includes 1~3 pulses with a peak current density of 1200~2500A / mm². 2 Damped oscillating pulses with a pulse width of 0.4~1.2ms are used. The ratio of the equivalent electromagnetic penetration depth δl to R of the long pulse group is 0.80~1.10, and the ratio of the nominal electric effect of the short pulse group to the long pulse group is 0.75~1.

25. A bridging pulse group is applied 2~5s after the end of the long pulse group. The bridging pulse group includes 1~2 pulses with a peak current density of 2200~3800A / mm. 2 Damped oscillatory pulses with a pulse width of 0.2~0.5ms are used. The ratio of the equivalent electromagnetic penetration depth δm to R of the bridging pulse group, δm / R, is 0.45~0.

65. The ratio of the nominal electric effect of the bridging pulse group to the geometric mean of the nominal electric effects of the short pulse group and the long pulse group is 0.8~1.

2. The nominal electric effect of each pulse group is calculated according to Σ(Jp 2 τ) is calculated, where Jp is the peak pulse current density, τ is the pulse width, and the axial mid-surface temperature of the cold-worked part measured 0.2s after each discharge is kept below 320°C; S3. After the last discharge, release the contact pressure of the water-cooled copper electrode on the end face of the cold-worked part, maintain the radial positioning state of the cold-worked part in the insulated linear positioning fixture, let it cool naturally to below 30°C and continue to stand for 12~24h, then release the insulated linear positioning fixture to obtain the cold-worked stabilized soft magnetic steel part.

2. The coordinated regulation method according to claim 1, characterized in that, In step S1, the cold finishing includes passing the soft magnetic steel billet through a cemented carbide sizing die at a speed of 0.08 m / s at 20~30℃, and then performing 2~4 passes of three-roll alternating reverse straightening; the cross-sectional area compression ratio η is calculated as η=(A0-A1) / A0×100%, where A0 is the cross-sectional area before sizing and A1 is the cross-sectional area after sizing.

3. The coordinated regulation method according to claim 1, characterized in that, The diameter of the cold-worked part obtained in step S1 is 4.00 mm and the length is 60.0 mm.

4. The coordinated regulation method according to claim 1, characterized in that, In step S2, the interval between adjacent short pulses is 60~90s, the interval between adjacent long pulses is 45~75s, and the interval between adjacent bridging pulses is 30s.

5. The coordinated regulation method according to claim 1, characterized in that, In step S2, the waveform square integral coefficient κ of each damped oscillation pulse is 0.18, and κ is calculated according to κ=∫J 2 dt / (Jp 2 τ) is calculated, where J is the current density that varies with time.

6. The coordinated regulation method according to claim 1, characterized in that, In step S2, the equivalent electromagnetic penetration depth δ is calculated as δ = 2 × (ρτ / μe). 1 / 2 Calculate, where ρ is taken as 4.0 × 10 -7 Ω·m, μe is the dynamic equivalent absolute permeability obtained by inverting the measured current waveform of the corresponding pulse, based on the four-terminal voltage response of the test specimens in the same batch and under the same cold working condition, and the axisymmetric magnetic diffusion equation of the cylinder. The unit of μe is H / m.

7. The synergistic regulation method according to claim 1, characterized in that, In step S2, the pulse peak current density Jp is calculated by dividing the measured peak current by the minimum cross-sectional area of ​​the cold-worked part, and the pulse width τ is the time from the first time the current reaches 0.1Jp to the last time it decays to 0.1Jp.

8. The coordinated regulation method according to claim 1, characterized in that, The insulated linear positioning fixture includes two sets of alumina V-blocks and non-magnetic spring pressure plates. The center distance between the two alumina V-blocks is 40mm, and the radial clamping force is 15N. One end of the insulated linear positioning fixture restricts the axial displacement of the cold-worked parts, while the other end allows the cold-worked parts to expand and contract axially due to thermal stress.

9. The coordinated regulation method according to claim 1, characterized in that, In step S2, the end faces of the cold-worked parts are precision machined to Ra0.8μm. The contact pressure of the water-cooled copper electrode to the end face is 10MPa, the cooling water temperature is 18℃, and the cooling water flow rate is 3.0L / min.

10. The coordinated regulation method according to claim 1, characterized in that, The cross-sectional area compression ratio of the cold-worked part obtained in step S1 is 2.93%; in step S2, the short pulse group includes two peak current densities of 4400 A / mm². 2 A short pulse with a pulse width of 90 μs is applied, with an interval of 75 s between adjacent short pulses. A long pulse group is applied 3 s after the end of the short pulse group. The long pulse group consists of two pulses with a peak current density of 1800 A / mm². 2 A long pulse with a pulse width of 0.55ms is applied, with an interval of 45s between adjacent long pulses. A bridging pulse group is applied 3s after the end of the long pulse group. The bridging pulse group consists of two pulses with a peak current density of 2800A / mm². 2 A bridging pulse with a pulse width of 0.23ms is used, with an interval of 30s between adjacent bridging pulses; in step S3, the cold-worked parts are naturally cooled to 27°C and then left to stand for 18 hours.

Images