METHOD AND MOLD ARRANGEMENT WITH A SENSOR DEVICE FOR DETECTING THE PRESENCE OF A METAL MELT

Abstract

The invention relates to a mold arrangement and a method for detecting the presence of a molten metal in the mold arrangement. A sensor device (S1, S2) can measure the presence of the molten metal at a specific point by having a dielectric coating (M1) arranged on at least a part of the inside of the mold wall (2), wherein the sensor device (S1, S2) has an unpowered coaxial line (4) with an open end, wherein the coaxial line (4) has an inner conductor (5) arranged in a sensor receptacle (3) of the mold wall (2) and a dielectric (7) extending between the inner conductor (5) and an outer conductor (6), wherein the open end of the coaxial line (4) is adjacent to the dielectric coating (M1), wherein a transmit signal (18) with a frequency f is applied to the coaxial line (4) by means of the electronic unit, wherein the signal (19) reflected from the coaxial line (4) is evaluated by means of the electronic unit (16).

Description

[0001] The invention relates to a casting mold arrangement having the features of the preamble of patent claim 1 and a method having the features of the preamble of patent claim 8.

[0002] At least one sensor device is used to detect the presence of molten metal in the casting mold assembly. The casting mold assembly comprises a casting mold wall and at least one sensor device arranged in the casting mold wall. A transmission signal is sent to the sensor device and a reception signal is passed from the sensor direction to an electronic unit. The presence or absence of the molten metal is detected by means of a change in the reception signal. The sensor devices are also referred to as sensors below.

[0003] DE10 2018 116952A1 discloses a casting mold arrangement with multiple sensors in the form of measuring sections. Using the measuring sections, the flow of molten metal during casting can be monitored. The measuring sections are manufactured in advance and arranged in corresponding receptacles in the casting mold. The measuring sections have a transmitting coil and a receiving coil with a wire winding and are arranged spaced apart within an insulating material, for example, in compressed sand. The measuring sections are essentially ring-shaped. When the molten metal is poured into the casting mold, it passes through the measuring sections one after the other. As it passes through the respective measuring section, the melt is detected. The halves of the casting mold form an electrical waveguide of the measuring section. Thus, the transmitting coil and the receiving coil are each arranged within a waveguide.An electrical signal is applied to the transmitting coil, the carrier frequency of which is below the lowest cutoff frequency of the waveguide. This generates a highly locally concentrated electromagnetic field. The presence or absence of melt in the area of ​​the measuring section influences the received signal. When the melt enters the area of ​​the measuring section, the presence of the melt within the measuring section can be detected by the change in the received signal. The coils are severed when the mold is opened. This arrangement is suitable for permanent molds and sand molds for manual casting.

[0004] In practice, the interior of the mold can be sprayed with an anti-stick compound to allow the solidified workpiece to be removed from the die-casting tool. The two-part design of the die-casting tool, the use of the electrically insulating anti-stick compound with coils surrounding the sprue, and the automation of the casting process complicate the design of the measuring setup. Furthermore, the presence of molten metal cannot be measured at specific points, but only in the measuring sections. The use of measuring sections in casting molds with highly curved surfaces is complex.

[0005] From the publication by Siqi Fan, Katie Staebell, Devendra Misra entitled "Static Analysis of an Open-Ended Coaxial Line Terminated by Layered Media" (IEEE Trans. on Instrumentation and Measurement, pp. 435-437, Vol. 39, No. 2, April 1990), the electrical properties, namely the capacitance of an open coaxial conductor, are investigated. The open end of the coaxial conductor is adjacent to a first layer of a dielectric and a second layer is adjacent to the side of the first layer facing away from the coaxial conductor. The dependence of the capacitance of the arrangement for different dielectric materials for the first layer is investigated, namely water, dioxane and methanol. The second layer is intended to be formed by either an air layer or a metal layer. This publication is referred to below as paper [1].

[0006] From the book by Werner Bächtold, «Lineare Elemente der Höchstfrequenztechnik», 2nd edition, 1998, vdf, the parameters of a coaxial conductor are the inductance L' and the capacitance C' as well as the wave impedance Z W This publication is hereinafter referred to as book [2].

[0007] The invention is therefore based on the object of designing and developing the sensor device in such a way that the sensor device can measure the presence of the molten metal at specific points, can be used along the inner surface even on surfaces with strong 3D curvature, and has a design that is as simple and robust as possible in terms of its gate and shape due to the adverse environment. Furthermore, it is advantageous if the necessary electronics do not require expensive measuring devices, such as a network analyzer or spectrum analyzer.

[0008] This object underlying the invention is now achieved by a casting mold arrangement having the features of patent claim 1.

[0009] A dielectric coating is arranged on at least part of the inside of the casting mold wall. According to the invention, the sensor device has an open-circuit coaxial line with an open end, wherein the coaxial line has an inner conductor arranged in a sensor receptacle of the casting mold wall and a dielectric extending between the inner conductor and an outer conductor. The open end of the coaxial line borders the dielectric coating. The open end is also separated from the molten metal by the dielectric layer. The end of the coaxial cable borders this dielectric layer made of a first medium, so-called medium 1. Medium 1 is an initially liquid "anti-stick" solution that is injected into the mold before the die-casting process, solidifies, and is intended to prevent the solidified workpiece from sticking in the mold.

[0010] The open end is flush with the inner surface of the mold wall. The open end thus forms a very small part of the mold, allowing the cross-section of the coaxial cable to be kept very small.

[0011] The invention has several advantages. The sensor device can be designed to be small, allowing it to measure the presence of the molten metal at specific points. Due to its small dimensions along the inner surface, the sensor device can also be used on surfaces with strong 3D curvature. The sensor device is simple and robust. Furthermore, no expensive measuring devices, such as a network analyzer or a spectrum analyzer, are required.

[0012] The casting mold wall can also be referred to as a chill mold and is typically made of metal, thus forming a metal chill mold.

[0013] The sensor receptacle is preferably formed by a bore. The sensor receptacle extends from the outside of the mold to the inside.

[0014] The metal mold is provided with a bore with a specific diameter ØB.

[0015] In an alternative embodiment, the diameter of the sensor receptacle and the diameter of the sensor device arranged therein widen from the outside to the inside. The sensor receptacle can be conical. This has the advantage that, during a die-casting process, the high pressures are supported on the inner walls of the sensor receptacle, thus improving the attachment of the sensor device.

[0016] In one embodiment, the outer conductor is formed as a separate component that is arranged, in particular, secured, in the sensor receptacle. The coaxial line is formed by a coaxial cable that is mounted in the sensor receptacle. This separate component is referred to as the metal shield. The outer diameter of the metal shield is dimensioned such that it can be easily inserted into the sensor receptacle. The outer diameter is smaller than or equal to the diameter of the sensor receptacle.

[0017] There are now different ways in which the metal shield can be connected to the mold.

[0018] In particular, a component with an internal thread is attached to the outside of the casting mold wall, with an external thread matching the internal thread being formed on the outside of the outer conductor, in particular of the metal shield, with the outer conductor being screwed into the internal thread of the component. The component can be designed as a nut welded to the casting mold wall. The metal shield can be screwed into the internal thread. As a component, a screw nut can be welded to the outside of the metal mold, and the outside of the metal shield can be provided with an external thread matching the screw nut.

[0019] The metal melt, the metal shield, the metal inner conductor, and the metal mold all have high electrical conductivity and are therefore considered good electrical conductors. This screw connection prevents the coaxial cable from being expelled by the metal melt. The coaxial cable consists of the metal inner conductor and the dielectric. The metal shield is ideally included in the coaxial cable.

[0020] In an alternative embodiment, the outer conductor is formed by the conductive mold wall. A structural unit containing the dielectric and the inner conductor is arranged in the sensor receptacle.

[0021] The coaxial cable is connected to an evaluation electronics unit. The evaluation electronics can detect the presence or absence of a metal melt. The complex reflection factor is preferably recorded as the measured variable. The stray capacitance C changes.V when filling the metal melt and thus also the reflection factor, as will be described in more detail below.

[0022] The method for detecting molten metal utilizes such a casting mold assembly with at least one sensor device arranged in the casting mold wall. A transmit signal is sent to the sensor device, and a receive signal is passed from the sensor direction to the electronics unit, whereby the presence or absence of the molten metal is detected by means of a change in the receive signal.

[0023] In the preferred embodiment, a signal, in particular a continuous wave signal (CW signal), is generated by means of a synthesizer, the signal being divided in a power divider into two partial signals s0(t) = Â · cos(2 · π · f · t) with the same power, the one partial signal s0(t) = Â · cos(2 · π · f · t) being guided via a circulator into the coaxial line and forming the incoming wave or the incoming signal, wherein after the reflection with the reflection factors r L or r V the partial signal is fed back into the electronic unit as a returning wave and the returning wave is separated from the forward wave by means of the circulator and is fed back as s R (t) = R̂ · cos([2 · π · f · t] + Δφ).

[0024] In particular, the signals can R(t) and s0(t) are mixed by means of a mixer, the mixed signal being filtered by means of a low-pass filter, the output signal thus obtained sD(t)=sD=D^⋅cos(Δφ) in particular by means of an oscilloscope.

[0025] The frequency f is preferably determined by the following calibration procedure: a) Calculate the maximum frequency f max , where the relative permittivity ε r1 of the dielectric coating M1 a maximum value is assumed, with a highest frequency f max by the inner radius r a of the screen, the thickness D and the relative permittivity ε r1 of the dielectric coating, with the smallest wavelength λ min (f max ,ε r1 ) through λmin(fmax,εr1)=c0fmax⋅εr1 where the phase velocity in vacuum is denoted by c0 and the following two conditions are met: λmin(fmax,εr1)10>2⋅ra λmin(fmax,εr1)10>D b) The frequency f of the synthesizer is set to the maximum frequency f max set. c) On an oscilloscope, the value for s DL (Empty: s D = s dL ) to be read and recorded d) Cover the opening of the coaxial cable with a conductor, especially an aluminum foil or aluminum sheet. e) on the oscilloscope the value for s DV (Full of metal - melt): s_D = s_DV) read and recorded f) Calculate Δs D = s DL - s DV g) If Δs D is too small: Reduce the frequency f and repeat steps from c). h) If for all frequencies f the values ​​for Δs Dare not satisfactory, replace the coaxial cable with a new one with a different length L Koax and repeat steps from b).

[0026] The speed of the molten metal can be measured using two spaced-apart sensor devices. A first sensor device provides a first output signal and a second sensor device provides a second output signal, whereby these two output signals are fed to the two channels of a multi-channel oscilloscope, namely channel CH1 and channel CH2. First, the molten metal reaches the first sensor and the signal on channel CH1 changes its value, whereby from now on the time T Mess is measured until the value of the signal on channel CH2 also changes, this being the time at which the metal melt has reached the second sensor S2, the speed v schmelze now knowing the distance L Mess is calculated as follows: vMelt=LMessTMess

[0027] The dimensioning of a coaxial cable and a corresponding sensor for detecting the molten metal in the mold will now be described as an example. A data sheet entitled "Coaxial Panel Connector" (Data Sheet 23_SMA-50-0-53 / 199_N, HUBER+SUHNER) describes a component with an inner conductor, a dielectric, and a mounting flange. This data sheet is referred to below as [3].

[0028] The paper [1] serves as the basis. 1 This publication deals with a coaxial cable whose outer radius of the inner conductor is a = 0.824 mm and the inner radius of the shield is b = 2.655 mm.

[0029] Since the outer radius of the inner conductor a = 0.824 mm and the inner radius of the shield b = 2.655 mm do not necessarily correspond to the available dimensions of coaxial cables, expressions (8) and (9) from [1] must be analyzed. Furthermore, it is assumed that the coaxial cable has a wave impedance of Z W = 50Ω. In [2] 2 with the inductance L' and the capacitance C' the wave impedance of Z W calculated according to (1). Z_w=L′C′=[60Ω]⋅μrεr⋅In(rari) Whereby: µ r Relative permeability of the dielectric ε r Relative permittivity of the dielectric r a Inner radius of the umbrella, in [m] r i Outer radius of the inner conductor, in [m] Z w Wave impedance of the coaxial cable, complex quantity, in [Ω]

[0030] As can be seen in (1), the wave impedance Z Win addition to the material parameters of the dielectric also the ratio r a / r i A comparison between the data sheet [1] 1 for a “Coaxial Panel Connector” with wave impedance Z W = 50Ω and the information from [1] results in (2). [1]:ba=2.655mm0.824mm=3.22209[1]:rari=22⋅0.0041m0.00128m=3.20313

[0031] It should now be assumed that the conditions from (2) correspond sufficiently closely to (3). ba≈rari

[0032] Based on (3), the factors V a be defined according to (4): bra=Va=1.295ari=Vi=1.288

[0033] With the relationships in (4) it can be assumed that the coaxial cable used in [1] is comparable to the “Coaxial Panel Connector” [1].

[0034] It is therefore justified that the following expression (5) should apply to the factor V. bra=V=ari

[0035] Now the expression (5) is applied to the equation (8) from [1] for the capacitance C1, which will be shown again here (6). C1=2⋅ε0⋅εr1[In(ba)]2⋅∫ab∫ab∫0πcos(ϕ′)⋅dρ⋅dρ′⋅dϕ′[ρ]2+[ρ′]2−[2⋅ρ⋅ρ′⋅cos(ϕ′)]

[0036] With the substitution (7) we get (8) for the new capacity C 1V . ρ=V⋅P→dρ=V⋅dPρ′=V⋅P′→dρ′=V⋅dP′a=V⋅rib=V⋅ra

[0037] And thus C1=2⋅ε0⋅εr1[In(rari)]2⋅∫rira∫rira∫0πcos(ϕ′)⋅V⋅ dP⋅V⋅dP′⋅dϕ′[V⋅P]2+[V⋅P′]2−[2⋅V⋅P⋅V⋅P′⋅cos(ϕ′)] C1=2⋅ε0⋅εr1[In(rari)]2⋅V⋅∫rira∫rira∫0πcos(ϕ′)⋅dP⋅dP⋅dϕ′[P]2+[P′]2−[2⋅P⋅P′⋅cos(ϕ′)]

[0038] The triple integral in (8) does not need to be recalculated, but the values ​​are in the Fig. 2 and Fig. 3 already included in [1] (9). C1V(ri,ra,V,εr1)=2⋅ε0⋅εr1[In(rari)]2⋅∫rira∫rira∫0πcos(ϕ′)⋅dP⋅dP ′⋅dϕ′[P]2+[P′]2−[2⋅P⋅P′⋅cos(ϕ′)]C1V(ri,ra,V,εr1)=1V⋅C1(a,b,εr1)

[0039] Where the values ​​for C1(a, b, ε r1 ) in the Fig. 2 and Fig. 3 are already included in [1]. Analogously, the capacitance C2 can be calculated from [1]. In addition to (7), the substitution (10) also applies to the thickness d of the medium 1. d=V⋅D where D now represents a new thickness.

[0040] According to [1], the following applies to the capacity C2: C2=4⋅ε0⋅εr1[ln(ba)]2⋅∑n=1∞[1−εr1εr21+εr1εr2]2⋅[∫ab∫ab∫0 πcos(ϕ')⋅dρ⋅dρ'⋅dϕ'[ρ]2+[ρ']2+[2⋅n⋅d]2−[2⋅ρ⋅ρ'⋅cos(ϕ')]]

[0041] With the substitutions (7) and (10) the new capacity C 2V (12): C2=4⋅ε0⋅εr1[ln(rari)]2⋅V⋅∑n=1∞[1−εr1εr21+εr1εr2]2⋅[∫rira∫ri ra∫0πcos(ϕ')⋅dP⋅dP'⋅dϕ'[P]2+[P']2+[2⋅n⋅D]2−[2⋅P⋅P'⋅cos(ϕ')]]

[0042] And here too (12) applies. C2V=4⋅ε0⋅εr1[ln(rari)]2⋅∑n=1∞[1−εr1εr21+εr1εr2]2⋅[∫rira∫rira∫0πcos(ϕ')⋅dP⋅dP'⋅dϕ '[P]2+[P']2+[2⋅n⋅D]2−[2⋅P⋅P'⋅cos(ϕ')]]C2V(ri,ra,V,D,εr1,εr2)=1V⋅C2(a,b,d,εr1,εr2)

[0043] The sum of the two capacities C1(a, b, ε r1 ) and C2(a, b, d, ε r1 , ε r2 ) result in the total capacity C(a, b, d, ε r1 , ε r2 ), which in [1] with the Fig. 2 and Fig. 3 is shown graphically. Applied to (9) and (12), the final result (13) is obtained.

[0044] Fig. 2 and Fig. 3 in [1]: C(a, b, d, ε r1 , ε r2 ) = C, (a, b, ε r1 ) + C2(a, b, d, ε r1 , ε r2 ) with New total capacity: C V (r i , r a , V, D, ε r1 , ε r2 ) = C 1V (r i , r a , ε r1 ) + C 2V (r i , ra , V, D, ε r1 , ε r2 ) →C2V(ri,ra,V,D,εr1,εr2)=1V⋅C2(a,b,d,εr1)+1V⋅C2(a,b,d,εr1,εr2) =1V⋅[C1(a,b,εr1)+C2(a,b,d,εr1,εr2)]=1V⋅C(a,b,d,εr1,εr2)CV(ri,ra,V,D,εr1,εr2)=1V⋅C(a,b,d,εr1,εr2)

[0045] For the analysis of the detector proposed here, the two Fig. 1 and Fig. 2. These situations also correspond to the two Fig. 2 and Fig. 3 in [1].

[0046] There are numerous possibilities for designing and developing the mold assembly and the method. Preferred embodiments of the invention are explained in more detail below with reference to the drawing and the accompanying drawing. The drawing shows: Fig. 1 in a schematic, sectional detail view of a casting mold arrangement with a first embodiment of a sensor device without melt, Fig.2 in a schematic, sectional detail view of the mold arrangement with the first embodiment of a sensor device with melt, Fig. 3 an equivalent circuit diagram of an open-circuit coaxial cable, Fig. 4 a diagram showing the phase shift Δδ as a function of the frequency f for dioxanes, Fig. 5 a diagram showing the phase shift Δδ as a function of the frequency f for methanol, Fig. 6 in a schematic view an electronic unit for carrying out the method, Fig. 7 shows a schematic, sectional detailed view of the mold arrangement with a second embodiment of a sensor device, Fig. 8 shows a schematic, sectional detailed view of the mold arrangement with a third embodiment of a sensor device, Fig.9 shows a schematic, sectional detailed view of the mold arrangement with a fourth embodiment of a sensor device, and Fig. 10 shows a schematic representation of an arrangement for measuring the flow velocity.

[0047] In Fig. 1 and Fig. 2 shows a section of a casting mold assembly 1 with a casting mold wall 2 and a coating M1 inside the casting mold wall 2. The casting mold wall 2 can also be referred to as a mold. The coating M1 is formed by a dielectric anti-adhesion layer and can also be referred to as medium 1. Depending on whether the casting mold assembly 1 is filled with a melt or not, a layer with a medium M2 of air ( Fig. 1) or in the form of a molten metal ( Fig. 2) to the dielectric coating with the medium M1. µ r1 = 1 Relative permeability of the medium M1 ε r1 Relative permittivity of the medium M1 µ r2 = 1 Relative permeability of the medium M2 ε r2 Relative permittivity of the medium M2

[0048] The casting mold wall 2 has a sensor receptacle 3, wherein a sensor device is arranged in the sensor receptacle 3. The metal mold 2 is provided with a bore with a diameter of ØB, wherein the bore forms the sensor receptacle 3.

[0049] The sensor device can also be referred to as a sensor. The sensor device has an open-ended coaxial cable 4, wherein the coaxial cable 4 has an inner conductor 5 arranged in the sensor receptacle 3 of the mold wall 2 and a dielectric 7 extending between the inner conductor 5 and an outer conductor 6. The outer conductor 6 is formed by a metal shield 8. By rotating the metal shield 8 about its own axis, the coaxial cable can be guided into the bore until the end face of the coaxial cable reaches the end of the bore. The outer side of the metal shield 8 is provided with a thread 10.

[0050] The Fig. 1 and Fig.The sensor device shown in Figure 2 is designed to detect the presence of a metal melt M2. The metal melt M2, the metal shield 8, the metal inner conductor 5, and the metal mold 2 possess high electrical conductivity and are therefore considered good electrical conductors. The outer diameter of the metal shield 8 is dimensioned such that it can be easily inserted into the sensor receptacle 3, namely the bore.

[0051] Furthermore, a nut 9 is welded to the outside of the metal mold 2 by means of a weld seam 9.1, and the outside of the metal shield 8 is provided with a thread 10 matching the nut 9. This screw connection prevents the coaxial cable 4 from being ejected by the molten metal.

[0052] To secure the coaxial cable 4, a heat-resistant adhesive is applied between the metal shield 8 and the sensor receptacle 3 in a preferred embodiment. In addition to securing, the adhesive also functions as a sealant, preventing the molten metal from flowing toward the nut 9. The same adhesive is used to secure the metal inner conductor 5 to the dielectric 7, and the dielectric 7 to the inner surface of the metal shield 8, again performing two functions: securing and sealing.

[0053] By applying an adhesive between the metal inner conductor 8 and the hole of the dielectric, depending on the value of the relative permittivity ε r The cured adhesive will change the wave impedance. Since only the change in the phase of the reflection factor is significant, the wave impedance change is tolerable within certain limits.

[0054] It is sufficient that the phase changes clearly and measurably when the mold is empty and filled. Furthermore, the dielectric can be pressed into the coaxial cable 4 so that it is held in place purely by friction; in this case, no adhesive is required inside the coaxial cable 4.

[0055] The reflection factors r are used as a measurement value L and r V according to (14) and (15) according to [2] with the equivalent circuit of the sensor according to Fig. 3.

[0056] For the equivalent circuit diagram in Fig. 3 shall apply: Z_w=Zw=50ΩCapacitance in empty state according to Figure 1:CV=CVLCapacitance with metal melt according to Figure 2:CV=CVV r L Empty state after Fig. 1: r_L=1j⋅2⋅π⋅f⋅CVL−Z_w1j⋅2⋅π⋅f⋅CVL+Z_w=1−j⋅2⋅π⋅f⋅CVL⋅Zw1−j⋅2⋅π⋅f⋅CVL⋅Zw r V : With metal melt after Fig. 2: r_V=1j⋅2⋅π⋅f⋅CVV−Z_w1j⋅2⋅π⋅f⋅CVV+Z_w=1−j⋅2⋅π⋅f⋅CVV⋅Zw1−j⋅2⋅π⋅f⋅CVV⋅Zw

[0057] Where f denotes the frequency in Hz and j denotes the imaginary unit.

[0058] In the following, the arguments φ L = arg(r L ) and φ V = arg(r V ) of the complex reflection factors r L and r V be considered (16), (17). (16) φ L = arg(r L ) : Empty state after Fig. 1: r_L=1⋅e−j⋅φL−j⋅φL=e−j⋅2⋅φLφL=arg(r_L)=arctan(2⋅π⋅f⋅CVL⋅Zw);0 rad≤φL≤π / 2 φ V = arg(r V ): With metal melt after Fig. 2: r_V=1⋅e−j⋅φ−j⋅φV=e−j⋅2⋅φVφV=arg(r_V)=arctan(2⋅π⋅f⋅CVV⋅Zw);0 rad≤φV≤π / 2

[0059] The difference angles Δφ are now important L and Δφ V according to (18) and (19). ΔφL=φ0−[2⋅φL] ΔφV=φ0−[2⋅φV]

[0060] Where the phase angle φ0 should serve as a reference.

[0061] The phase jump Δδ according to (20) is to be considered as the figure of merit. Key figure:Δδ=ΔφV−ΔφL=[2⋅φL]−[2⋅φV]=2⋅[φL−φV]

[0062] This parameter describes the phase shift during the pouring of the molten metal. The choice of frequency ensures that no ambiguities arise, i.e., condition (21) must be met: Condition:φL≠φV

[0063] As an example, the data from [1] are used. It is assumed that the relative permittivity of the “anti-stick” solution lies between that of methanol (ε r1 = 33.1) and that of dioxane (ε r1 = 2.1). Furthermore, a layer thickness (thickness of the "anti-adhesion" solution film) of D = 0.1 mm is assumed.

[0064] Using the calculation rules (5) and (10) the values ​​of the capacity C(a, b, d, ε r1 , ε r2 ) from the Fig. 2 and Fig.3 from [1] was determined graphically. Table 1 summarizes the results. The dimensions of the coaxial cable should be taken from the data sheet [1]. Taking mechanical tolerances into account, the factor V can now be derived (22). bra=V=ari≈1.29→Fixing V:V=1.29

[0065] From this, with the help of (10), the thickness d, which is used as the x-axis in the Fig. 2 and Fig. 3 from [1], determined (23). d=V⋅D=1.29⋅0.1mm=0.129mm Table 1: Summary of values Summary Dioxane Methanol e r1 2.1 33.1 D 0.1mm 0.1mm V 1.29 d according to (10): d = V · D 0.129mm 0.129mm Fig. 2, [1] for air: 0.04 · 10 -12 F 0.2 · 10 -12 F C(a, b, d, e) r1 , eh r2 ) Fig. 3, [1] for metal: 0.6 · 10 -12 F 10 · 10 -12 F C(a, b, d, e) r1 , eh r2 )

[0066] To determine the appropriate frequency range for the frequency f, the graphical representations Fig. 4 and Fig. 5. The highest frequency f max is determined by the inner radius r a of the screen, the thickness D and the relative permittivity ε r1 of the medium 1. The smallest wavelength λmin (f max ,ε r1 ) In medium 1 is determined to (24). λmin(fmax,εr1)=c0fmax⋅εr1

[0067] With the phase velocity c0 in vacuum.

[0068] In order to use the capacity C calculated in the statics, all considerations must be carried out in the quasi-statics. This leads to conditions (25) and (26): λmin(fmax,εr1)10>2⋅ra λmin(fmax,εr1)10>D

[0069] Calculated, this results in: Dioxanes:λmin(fmax=109Hz,εr1=2.1)=0.2m Methanol:λmin(fmax=109Hz,εr1=33.1)=0.05m

[0070] Using the values ​​from [1] we get: Dioxanes:λmin(fmax,εr1)10=0.02>2⋅ra=0.0041m>D=0.0001m Methanol:λmin(fmax,εr1)10=0.005>2⋅ra=0.0041m>D=0.0001m

[0071] This results in the highest frequency f max to : f max = 1GHz.

[0072] Dioxane INPUT: Dioxanes V= 1.29 [-] [1], Fig. 2 and Fig. 3: d = 0.129 [mm] [1], Fig. 2 LUFT: C(a,b,d,Epsilon_r1, Epsilon_r2) = 4E-14 [F] [1], Fig. 3 METALL: C(a,b,d,Epsilon_r1, Epsilon_r2) = 6E-13 [F] max. frequency: f_max = 1000000000 [Hz] Wave impedance (real), Zw = 50 [Ω] Ebsilon_r1 = 2.1 [-] CALCULATION: Dioxanes D= 0.1 [mm] Empty: C_VL = 3.10078E-14 [F] With metal: C_VV = 4.65116E-13 [F] π = 3.141592654 [-] Empty: 2 * π * C_VL * Zw = 9.74137E-12 [s] With metal: 2 * π * C_VV * Zw = 1.46121E-10 [s] Methanol INPUT: Methanol V = 1.29 [-] [1], Fig. 2 and Fig. 3: d = 0.129 [mm] [1], Fig. 2 LUFT: C(a,b,d,Epsilon_r1, Epsilon_r2) = 2E-13 [F] [1], Fig. 3 METALL: C(a,b,d,Epsilon_r1, Epsilon_r2) = 1E-11 [F] max. frequency: f_max = 1000000000 [Hz] Wave impedance (real), Zw = 50 [Ω] Ebsilon_r1 = 33.1 [-] CALCULATION: Methanol D = 0.1 [mm] Empty: C_VL = 1.55039E-13 [F] With metal: C_VV = 7.75194E-12 [F] π = 3.141592654 [-] Empty: 2 * π * C VL * Zw = 4.87069E-11 [s] With metal: 2 * π * C_VV * Zw = 2.43534E-09 [s]

[0073] For the exact frequency determination, the structure is Fig. 6. The Fig. Figure 6 shows the signal processing concept. A synthesizer 11, a power divider 12, a circulator 13, a mixer 14, and a low-pass filter 15 are located in a housing 16 and form the electronics unit 16. The electronics unit 16 is firmly attached to the mold 2 at a defined distance via a rigid connection 17. Care should be taken to ensure that the heat flow from the mold 2 to the electronics unit is as minimal as possible. The synthesizer 11 generates a continuous wave (CW) signal, which is split into two signals with the same power in the power divider 12. One signal is fed via the circulator 13 into the coaxial line 4 and forms the outgoing wave or the incoming signal 18. After reflection with the reflection factors rL or r V The signal is fed into the electronic unit 16 as a returning wave 19. The circulator 13 separates the returning wave from the forward wave and is called s R (t) denotes (27), where t is the time. sR(t)=R^⋅cos([2⋅π⋅f⋅t]+Δφ) with: R̂ amplitude, real, in volts f frequency, in Hz t time, in s Δφ phase angle, in rad : Δφ=ΔφL=φ0(f)−[2⋅φL] or Δφ=ΔφV=φ0(f)−[2⋅φV]

[0074] φ0(f): Phase angle, in radians, relative to the signal s0(t) = · cos (2 · π · f · t) To ensure that the phase angle φ0(f) remains constant in the empty state and with molten metal, the coaxial line or the coaxial cable from the electronics unit 16 to the mold 2 must be rigid, ie, the shield 8 consists of a solid tube and not a flexible braid. The phase angle φ0(f) depends on the frequency f and the coaxial cable length L. Koax , as well as the elements in the path. Furthermore, the coaxial cable length L Koax be as short as possible.

[0075] The signals s R (t) and s0(t) are fed to a mixer 14, followed by a low-pass filter 15. The output signal s D (t) according to (28) is led out of the housing via an electrical connection and fed to an oscilloscope 20. sD(t)=sD=D^⋅cos(Δφ) with D Amplitude, real, in volts Δφ phase angle, in rad : △φ=ΔφL=φ0(f)−[2⋅φL] or Δφ=ΔφV=φ0(f)−[2⋅φV]

[0076] The oscilloscope 20 monitors the course of the signal s D (t) = s D . In the empty state, the signal is measured according to (29). Empty:sD(t)=sDL=D^⋅cos(φ0(f)−[2⋅φL])

[0077] And when the mold 2 is filled with the metal melt, the signal changes to (30) With metal melt:sD(t)=sDV=D^⋅cos(φ0(f)−[2⋅φV])

[0078] The presence of a metal melt is thus only detected by a signal change Δs D (31) detected. ΔsD=sDL−sDV≠0

[0079] From equation (28) it can be seen that ambiguities (32a) and (32b) can occur. With n = ···,-3,-2,-1, 0, 1, 2, 3, ... and m, p = 1, 2, 3, ... we get: sD(t)=sD=D^⋅cos(φ0(ƒ)−[2⋅φV])=D^⋅cos([2⋅n⋅π]±[φ0(ƒ)−[2⋅φL]]) φ0(ƒ)−[2⋅φV]=[2p±1]⋅π2 and φ0(ƒ)−[2⋅φL]=[2m±1]⋅π2

[0080] From the ambiguities (32a) and (32b), the conditions A ... C, (33a) ... (33c) can be derived. From (32a) the following conditions follow: Condition A:Δδ2⋅π=φL−φVπ≠n Condition B:φ0(ƒ)−[φV+φL]π≠nn=⋯,−3,−2,−1,0,1,2,3,⋯ and from (32b): Condition C:2⋅φ0(ƒ)−[2⋅φV]π≠2p±1 and 2⋅φ0(ƒ)−[2⋅φL]π≠2m±1p=1,2,3,…m=1,2,3,…

[0081] The condition A (33a) can now be applied to the Fig. 4 and Fig. 5. All values ​​must be converted to ° instead of radians. This yields (34). n≠Δδ,[°]360°Δδ,[°]=180°π⋅Δδ

[0082] As in the Fig. 4 and Fig. As can be seen from Fig. 5, (34) is satisfied for frequencies less than or equal to 1 GHz, since Δδ, [°] < 360°.

[0083] Conditions B and C can only be met in practice, since the phase angle φ0(f) is known depending on the setup. To determine the frequency f, the following calibration procedure can be used to find the optimal frequency f: 1. Calculate the maximum frequency f according to (24), (25) and (26) max , where the relative permittivity ε r1 the “anti-stick” solution assumes a maximum value. 2. The frequency f of the synthesizer is set to the maximum frequency f max to set. 3. On the oscilloscope, the value for s DL to read and record 4. Cover the opening of the coaxial cable with aluminum foil or aluminum sheet. 5. On the oscilloscope, the value for s DV to read and record 6. Calculate Δs D = s DL - s DV 7. If Δs Dis too small: Reduce the frequency f and repeat steps 3 and up.

[0084] If for all frequencies f the values ​​for Δs D are not satisfactory, replace the coaxial cable with a new one with a different length L Koax and repeat the steps from number 2.

[0085] Fig. Figure 7 shows a second embodiment of the sensor device. A sensor receptacle 3 in the form of a threaded hole has been machined into the mold 2. The sensor receptacle 3 here has an internal thread 22. The internal thread 22 can extend over the entire length of the sensor receptacle. The metal shield 8 has a continuous external thread 10. By screwing, the coaxial cable is screwed into the mold 2 until the end face reaches the end of the threaded hole. The coaxial cable 4 is secured with the lock nut 9, thus firmly connecting it to the mold 2.

[0086] No adhesive is required for fixation. However, an adhesive can again serve as a seal, as with the first version of the sensor device (see description of Fig. 1 and Fig. 2). The same remarks regarding the wave impedance change in the presence of an adhesive apply as explained above.

[0087] In one embodiment of this second sensor device, the adhesive can be omitted. In this case, the molten metal can penetrate into the gap between the inner surface of the metal shield and the outer surface of the dielectric, as well as into the gap between the inner surface of the dielectric and the outer surface of the metal inner conductor. The melt will cool and thus solidify and remain in the gaps. During subsequent casting processes, the metal residues may possibly change their state of aggregation again and possibly flow even more into the gaps. Since a very thin layer of the solidified / liquid metal melt compared to the inner radius of the shield r a and the outer radius of the inner conductor r iSince the calculation is based on a molten metal and thus a highly electrically conductive material, the wave impedance change is small. After a certain number of casting processes, the penetration depth should no longer change, and the wave impedance jump between sections with and without the penetrated molten metal remains constant over time. Here, too, only the phase change of the reflection factor is important, i.e., only the measurements before and after filling with the molten metal. Thus, the wave impedance change from measurement to measurement can be taken into account.

[0088] Fig.Figure 8 shows a third embodiment of the sensor device. Version 3 uses a design as described in [1]. Here, a structural unit consisting of the inner conductor 5, the dielectric 7, and a connecting flange 23 with the inner conductor 5 and the dielectric 7 is arranged in the sensor receptacle 3. The connecting flange 23 is attached to the outside of the mold wall 2 outside the sensor receptacle 3. The sensor receptacle 3 is again formed by a bore. The diameter of the sensor receptacle 3 corresponds to the outer radius of the dielectric 7. As shown in Fig.As can be seen in Figure 8, the inner wall of the sensor receptacle 3, i.e., the bore in the mold 2, assumes the function of the metal shield (not further designated here). This also creates a coaxial cable 4. The connecting flange 23 has an external thread 24, to which the electronics unit or another rigid connection can be connected. The connecting flange 23 further has a flange plate 25, which rests against the outside of the mold 2 and is screwed thereto with at least one screw 26.

[0089] Fig.Figure 9 shows a fourth embodiment of the sensor device. This fourth embodiment is suitable for applications in which the molten metal is forced into the mold 2 at high pressure. The diameter of the sensor holder 3 and also the diameter of the sensor device arranged therein expand from the outside to the inside of the mold 2. The sensor holder 3 is conical, and the coaxial cable 4 has a truncated cone or conical shape within the sensor holder 3. The conical design of the coaxial cable 4 distributes the forces across the metal mold 2 and prevents the coaxial cable 4 from being forced out.

[0090] The aim is that the wave impedance Z wThe shape of the section between the cone and the cylindrical coaxial cable minimizes impedance jumps, resulting in the lowest possible reflection coefficient due to the transition. It is generally assumed that the characteristic impedance in the conical coaxial cable differs from the characteristic impedance of the cylindrical coaxial cable. The fact that the characteristic impedance changes along the length of the coaxial cable does not contribute to the functionality, since only the signal change between the "empty" and "full" states is important.

[0091] An additional nut 9 fixes the coaxial cable 4 and prevents it from being torn out when the cast workpiece is removed.

[0092] Given the expanding shape in the form of a cone or another geometry, i.e., the expansion of the coaxial cross-section / end face facing the molten metal is larger than the cross-section of the coaxial cable outside the metal mold, the coaxial cable 4 is pressed into the tapered inner surface of the sensor receptacle 3 of the metal mold 2 by turning the nut 9. This fixes the coaxial cable 4. It is conceivable that an adhesive is used. The adhesive that may be used only serves as a seal. The same comments apply to the adhesive and to the penetration of the molten metal into the gaps as to the first two versions of the sensor device.

[0093] Fig.Figure 10 shows an arrangement for measuring the speed of the melt with two sensor devices S1 and S2 spaced apart in the direction of flow. With the arrangement of the sensor devices S1 and S2, the speed of the metal melt can be measured. Sensor device S1 provides a first signal. Sensor device S2 provides a second signal. These two output signals are fed to the two channels of a multi-channel oscilloscope, channel CH1 and channel CH2. First, the metal melt reaches sensor S1. The signal on channel CH1 changes its value. From now on, the time T Mess measured until the value of the signal on channel CH2 also changes. This is the point in time at which the molten metal reaches the sensor S2. The speed v Schmelze can now, knowing the distance L Mess can be calculated using (35). vMelt=LMessTMess List of reference symbols: 1 mold arrangement 2 mold wall / mold 3 Sensor recording 4 Coaxial line / coaxial cable 5 inner conductors 6 outer conductors 7 Dielectric 8 metal shade 9 Mother 9.1 Weld seam 10 threads of the metal shade 11 synthesizers 12 power dividers 13 Circulator 14 mixers 15 low-pass filters 16 Housing / electronics unit 17 rigid connection 18 Transmitting signal / incoming signal 19 returning signal 20 Oscilloscope 21 Energy supply 22 internal threads 23 Connection flange 24 external threads 25 Flange plate 26 Screw M1 dielectric coating / medium 1 M2 Layer of air or molten metal / Medium 2 S1 first sensor / first sensor device S2 second sensor / second sensor device CH1 first channel of the oscilloscope CH2 second channel of the oscilloscope QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2018 116952A1

[0003] Cited non-patent literature

[0000] Siqi Fan, Katie Staebell, Devendra Misra with the title "Static Analysis of an Open-Ended Coaxial Line Terminated by Layered Media", (IEEE Trans. on Instrumentation and Measurement, pp 435-437, Vol. 39, No. 2, April 1990

[0005] Werner Bächtold, «Linear Elements of High Frequency Technology», 2nd edition, 1998

[0006] Coaxial Panel Connector” (Data Sheet 23_SMA-50-0-53 / 199_N, HUBER+SUHNER

[0027]

Claims

[1] Casting mold arrangement (1) with a casting mold wall (2) and at least one sensor device for detecting the presence of a molten metal in the casting mold arrangement (1), characterized by that a dielectric coating (M1) is arranged at least on a part of the inside of the casting mold wall (2), wherein the sensor device (S1, S2) has an open-circuit coaxial line (4) with an open end, wherein the coaxial line (4) has an inner conductor (5) arranged in a sensor receptacle (3) of the casting mold wall (2) and a dielectric (7) extending between the inner conductor (5) and an outer conductor (6), wherein the open end of the coaxial line (4) adjoins the dielectric coating. [2] Casting mold arrangement according to claim 1, characterized bythat a component with an internal thread is fastened to the outside of the casting mold wall (2), wherein an external thread (10) matching the internal thread is formed on the outside of the outer conductor (8), wherein the outer conductor (6) is screwed into the internal thread of the component. [3] Casting mold arrangement according to one of the preceding claims, characterized by that the component is designed as a nut (9) welded to the casting mold wall (2). [4] Casting mold arrangement according to one of the preceding claims, characterized by that the sensor holder (3) is formed by a bore. [5] Casting mold arrangement according to one of the preceding claims, characterized by that the diameter of the sensor holder (3) and also the diameter of the sensor device arranged therein widens from the outside to the inside of the mold wall (2). [6] Casting mold arrangement according to one of the preceding claims, characterized bythat the coaxial line (4) is designed as a coaxial cable (4) with a metal shield (8), wherein the metal shield (8) is arranged and in particular fastened in the sensor receptacle (3). [7] Casting mold arrangement according to one of the preceding claims, characterized by that a structural unit consisting of the inner conductor (5), the dielectric (7) and a connecting flange with the inner conductor (5) and the dielectric (7) is arranged in the sensor receptacle (3), wherein the outer conductor is formed by the conductive mold wall (2). [8] Method for detecting a molten metal in a casting mold arrangement, in particular according to one of the preceding claims, wherein the casting mold arrangement has a casting mold wall (2) and a sensor device (S1, S2) arranged in a sensor receptacle (3) of the casting mold wall (2), wherein a transmission signal is sent to the sensor device (S1, S2) and a reception signal is passed from the sensor direction to an electronic unit (16), wherein the presence or absence of the molten metal is detected by means of a change in the reception signal, characterized byin that the sensor device (S1, S2) has an idle coaxial line (4) with an open end, wherein the coaxial line (4) has an inner conductor (5) arranged in a sensor receptacle (3) of the mold wall (2) and a dielectric (7) extending between the inner conductor (5) and an outer conductor (6), wherein the open end of the coaxial line (4) is adjacent to the dielectric coating (M1), wherein a transmission signal (18) with a frequency f is applied to the coaxial line (4) by means of the electronics unit (16), wherein the signal (19) returning from the coaxial line (4) is evaluated by means of the electronics unit (16). [9] Method for detecting a molten metal in a casting mold arrangement, in particular according to one of the preceding claims, wherein the casting mold arrangement has a casting mold wall and at least one sensor device (S1, S2) arranged in the casting mold wall, wherein a transmission signal is sent to the sensor device (S1, S2) and a reception signal is passed from the sensor direction to an electronic unit (16), wherein the presence or absence of the molten metal is detected by means of a change in the reception signal, characterized bythat a signal, in particular a continuous wave signal (CW signal) is generated by means of a synthesizer (11), the signal being divided in a power divider (12) into two partial signals s0(t) = Â. cos(2 · π · f · t) with the same power, the one partial signal s0(t) = Â · cos(2 · π · f · t) being guided via a circulator (13) into the coaxial line (4) and forming the incoming wave or the incoming signal (18), wherein after the reflection with the reflection factors r L or r V the partial signal is fed back into the electronic unit (16) as a returning wave (19) and the returning wave is separated from the forward wave by means of the circulator (13) and is fed back as s R (t) = R̂ · cos([2 · π · f · t] + Δφ). [10] Method according to the preceding claim, characterized by that the signals s R(t) and s0(t) are mixed by means of a mixer (14), and the mixed signal is filtered by means of a low-pass filter (15), the output signal thus obtained sD(t)=sD=D^⋅cos(Δφ) in particular by means of an oscilloscope (20). [11] Method according to one of the preceding claims, characterized by that the frequency f is determined by the following calibration rule: a) Calculate the maximum frequency f max , where the relative permittivity ε r1 of the dielectric coating M1 a maximum value is assumed, with a highest frequency f max by the inner radius r a of the screen, the thickness D and the relative permittivity ε r1 of the dielectric coating, with the smallest wavelength λ min (f max ,ε r1 ) through λmin(fmax,εr1)=c0fmax⋅εr1 where the phase velocity in vacuum is denoted by c0 and the following two conditions are met: λmin(fmax,εr1)10>2⋅ra λmin(fmax,εr1)10>D b) The frequency f of the synthesizer is set to the maximum frequency f max set. c) On an oscilloscope, the value for s DL to read and record d) Cover the opening of the coaxial cable with a conductor, especially an aluminum foil or aluminum sheet. e) on the oscilloscope the value for s DV read and recorded f) Calculate Δs D = s DL - s DV g) If Δs D is too small: Reduce the frequency f and repeat steps from c). h) If for all frequencies f the values ​​for Δs D are not satisfactory, replace the coaxial cable with a new one with a different length L Koaxand repeat steps from b). [12] Method according to one of the preceding claims, characterized by that by means of two spaced apart L Mess arranged sensor devices (S1, S2) designed according to one of the preceding claims, the speed of the molten metal is measured. [13] Method according to the preceding claim, characterized by that the sensor device (S1) provides a first output signal and the sensor device (S2) provides a second output signal, these two output signals being fed to the two channels of a multi-channel oscilloscope, namely channel CH1 and channel CH2, the metal melt first reaching the first sensor S1 and the signal on the first channel CH1 changing its value, from now on the time T Messis measured until the value of the signal on the second channel CH2 also changes, this being the time at which the metal melt has reached the second sensor S2, the speed u Schmelze now knowing the distance L Mess is calculated as follows: vMelt=LMessTMess

Images