Thermocouple signal synchronous sampling anti-interference method for strong electromagnetic interference environment

Abstract

This invention discloses a method for synchronous sampling and anti-interference of thermocouple signals in environments with strong electromagnetic interference, relating to the field of semiconductor manufacturing technology. The method synchronously acquires thermocouple temperature measurement signals and the operating signal of a plasma excitation source. It extracts the fundamental frequency and harmonics of interference by performing spectral analysis on the operating signal and calculates the interference intensity factor. Based on the fundamental frequency, an adaptive notch filter is dynamically configured and a cascaded low-pass filter is used to suppress RF main frequency and harmonic interference. Furthermore, the temperature change rate is calculated, and the moving average window length is adaptively adjusted in conjunction with the interference intensity factor to balance response speed and smoothness. Simultaneously, a time-division sampling control signal is generated using the fundamental frequency to acquire the cold junction temperature near the zero-crossing point of the synchronization signal, avoiding electromagnetic coupling interference. Finally, cold junction compensation is performed to output the final temperature. This invention can track RF frequency and power changes in real time, achieving high-precision, fast-response temperature measurement in environments with strong electromagnetic interference.

Description

Technical Field

[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to an anti-interference method for synchronous sampling of thermocouple signals in environments with strong electromagnetic interference. Background Technology

[0002] In semiconductor wafer manufacturing processes, temperature is a key parameter affecting process outcomes such as etching rate, film uniformity, and deposition quality. As feature sizes continue to shrink, the requirements for wafer temperature uniformity and stability have reached sub-Celsius levels. Thermocouples, due to their simple structure, fast dynamic response, and wide temperature measurement range, are widely integrated into temperature-sensing wafers or embedded temperature sensors for real-time monitoring of wafer surface temperature.

[0003] However, modern semiconductor process equipment commonly uses radio frequency (RF) power sources (typically 13.56MHz, 2MHz, 400kHz) or microwave sources (2.45GHz) to generate plasma. These strong electromagnetic fields couple into the thermocouple signal loop through spatial radiation and conduction, superimposing interference electromotive forces with amplitudes ranging from tens to hundreds of millivolts, severely obscuring the true temperature signal. Existing technologies include thermocouple filtering circuits for plasma environments, employing a combination of RC low-pass filters and common-mode chokes. However, their filter parameters are fixed and cannot adapt to the dynamic changes in RF frequency and power during different process steps. Furthermore, there are RF interference suppression methods based on digital signal processing, but these do not consider the problem of electromagnetic coupling interference affecting the thermocouple cold junction compensation loop, leading to significant fluctuations in cold junction temperature readings and ultimately failing to meet the requirements of advanced processes. Simultaneously, existing methods experience large temperature change rates during process switching (such as etching → cooling), and fixed-window averaging results in severe response lag, failing to capture true temperature transients. In summary, there is an urgent need for a wafer temperature measurement anti-interference method that can adaptively suppress wideband radio frequency interference while ensuring fast response and accurate cold junction compensation. Summary of the Invention

[0004] The purpose of this invention is to provide a method for synchronous sampling and anti-interference of thermocouple signals in environments with strong electromagnetic interference, which solves the problems in the prior art where fixed filter parameters cannot adapt to changes in radio frequency, cold junction compensation is affected by electromagnetic coupling interference, and fixed window averaging leads to response lag.

[0005] To achieve the above objectives, the present invention provides a method for synchronous sampling and anti-interference of thermocouple signals in environments with strong electromagnetic interference, comprising the following steps: S1. Synchronously acquire thermocouple temperature measurement signals and plasma excitation source operating signals, and record the sampling time sequence; wherein, the operating signals include operating current or voltage signals; S2. Perform spectrum analysis on the working signal to extract the fundamental frequency of the interference. and the frequencies of each harmonic, and calculate the interference intensity factor. ; S3, Based on the fundamental frequency of the interference The parameters of an adaptive notch filter are dynamically configured, and a first-order low-pass filter is cascaded to suppress high-frequency harmonics and broadband noise. This adaptive filtering is then applied to the thermocouple temperature measurement signal to obtain the filtered temperature signal. ; S4, from the filtered temperature signal Calculate the current rate of temperature change And combined with the interference intensity factor obtained in step S2 The weighting coefficients are determined based on the radio frequency interference level, and the average window length of the sliding window is adaptively adjusted. The filtered temperature signal is then subjected to variable window averaging to obtain the window-averaged temperature signal. , The temperature is indicated by the temperature difference between the hot and cold ends. S5. Utilizing interference fundamental frequency Generate a time-division sampling control signal, based on the synchronization signal extracted from the working signal, before and after the zero-crossing point of the synchronization signal. The cold junction temperature sensor signal is acquired within the time window to obtain the cold junction temperature after time-division sampling. ;in, This is the offset time from the zero-crossing point to the center of the sampling window; S6. Based on the cold junction temperature obtained from the time-sharing sampling in step S5. The window-averaged temperature signal obtained in step S4 Cold junction compensation is performed to obtain the final temperature output. .

[0006] Preferably, the interference intensity factor in step S2 The calculation expression is: ; in, The fundamental frequency interference amplitude (unit: mV) represents the voltage amplitude induced in the thermocouple circuit by the radio frequency fundamental frequency component. This is an estimate of the effective amplitude of the thermocouple signal (unit: mV), representing the amplitude of the thermoelectric potential corresponding to the true temperature. For the first Subharmonic interference amplitude (unit: mV); The highest harmonic order to be considered is usually 3 to 5.

[0007] Preferably, the transfer function of the adaptive notch filter in step S3 is... for: ; in, The normalized angular frequency determines the center frequency of the notch filter. , Sampling frequency (unit: Hz); , is the notch width factor (dimensionless). To transform complex variables; Transfer function of cascaded first-order low-pass filters for: ; Among them, the cutoff frequency of the low-pass filter (Unit: Hz), sampling period (Unit: s).

[0008] Preferably, the current temperature change rate in step S4 From the filtered temperature signal Calculated using the central difference: ; in, For the first Filtered temperature values ​​(unit: °C) at each sampling point; For the first Filtered temperature values ​​(unit: °C) at each sampling point.

[0009] Preferably, the adaptive window length in step S4 Determined by the following formula: ; in, The reference window length (dimensionless); The weighting coefficients for the interference intensity factor. This is the weighting coefficient for the rate of temperature change; Interference intensity factor weighting coefficient According to radio frequency interference level calibration: when Time to take ,when Time to take Linear interpolation is used when the range is between 0 and 1. This is the absolute value of the rate of temperature change (unit: °C / s). This is the interference intensity factor (dimensionless).

[0010] Preferably, in step S5, the time-division sampling control signal is based on a pulse modulation signal of the RF power supply synchronized with the phase-locked loop or the power frequency zero-crossing point, and the sampling window width is [missing information]. ,in , The synchronization signal frequency (unit: Hz) is usually 50Hz (power frequency) or 1kHz~10kHz (radio frequency pulse modulation frequency). The offset time from the zero crossing point to the center of the sampling window (unit: s).

[0011] Preferably, the cold junction compensation in step S6 specifically includes: based on the cold junction temperature obtained from time-division sampling... The corresponding cold junction thermoelectric potential can be obtained by referring to the thermocouple calibration table. The averaged temperature signal obtained in step S4 is then used to calculate the average temperature signal obtained in step S4. Converted to thermoelectric potential ;Will and The total thermoelectric potential is obtained by adding them together. Then consult the thermocouple calibration table to find the correct values. Convert to final temperature output .

[0012] Preferably, the notch width factor With interference intensity factor Negative correlation, when When it increases Decreasing the notch depth increases the cutoff frequency of the first-stage low-pass filter; The effective temperature signal is attenuated by more than three orders of magnitude above the highest effective frequency of the wafer temperature signal (approximately 1 kHz).

[0013] Preferably, adaptive window length With temperature change rate and interference intensity factor They are inversely proportional, which allows for automatic switching to a short window to maintain response speed during process changes or strong interference.

[0014] Preferably, the interference fundamental frequency and interference intensity factor Simultaneously used in adaptive filtering, variable window averaging, and time-division sampling, it achieves coordinated optimization of anti-interference measures; the notch center frequency of adaptive filtering is based on... Dynamically set; the synchronization signal frequency for time-division sampling is extracted from the working signal via a phase-locked loop. Specifically: when the RF power supply is in continuous mode, the power frequency zero-crossing signal is extracted as the synchronization signal; when the RF power supply is in pulse modulation mode, the pulse modulation signal is extracted as the synchronization signal. The notch depth of the adaptive filter and the window length of the variable window averaging are both based on... Dynamic adjustment.

[0015] Therefore, the present invention employs the above-described anti-interference method for synchronous sampling of thermocouple signals in environments with strong electromagnetic interference, which has the following beneficial effects: (1) The parameters of the digital notch filter can be adjusted in real time according to the operating frequency of the RF power supply and its harmonic components. No matter how the process steps change, it can always maintain efficient suppression of interference signals and significantly improve the anti-interference consistency under different operating conditions.

[0016] (2) By cascading a first-order low-pass filter after the notch filter, radio frequency harmonics and broadband random noise are effectively attenuated. At the same time, the cutoff frequency is set reasonably to ensure that the rapid response of the temperature signal is not affected, thus achieving a good balance between high-frequency suppression and dynamic response.

[0017] (3) By using a time-division sampling strategy, the cold junction temperature signal is accurately selected in the trough region of the radio frequency interference waveform, which fundamentally avoids the pollution of the cold junction measurement circuit by electromagnetic coupling, greatly improves the cold junction compensation accuracy, and effectively eliminates the interference of room temperature reading fluctuations on the final temperature measurement result.

[0018] (4) The average window length is automatically adjusted according to the real-time temperature change rate and interference intensity. When the temperature changes rapidly, the window is automatically shortened to ensure the response speed. When the temperature is steady, the window is automatically extended to smooth the residual noise, thus taking into account both transient capture capability and steady-state measurement accuracy.

[0019] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of the thermocouple signal synchronous sampling anti-interference method for strong electromagnetic interference environment according to the present invention; Figure 2 This is a schematic diagram of the adaptive notch filter structure according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the time-division sampling timing according to an embodiment of the present invention; Figure 4 This is a graph showing the relationship between the adaptive window length and the rate of temperature change in an embodiment of the present invention. Figure 5 This is a comparison chart of cold junction temperature fluctuations under continuous radio frequency interference environment according to an embodiment of the present invention. Detailed Implementation

[0021] The following detailed description of embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0022] This embodiment uses semiconductor wafer etching as an application scenario. The process equipment employs a 13.56MHz RF power supply in 500W continuous output mode. A K-type thermocouple is embedded on the wafer surface. The target wafer absolute temperature is 145°C, and the ambient cold junction temperature is approximately 25°C. At this point, the temperature difference between the hot and cold junctions is 120°C. The system sampling frequency... .

[0023] The specific implementation process of the present invention is described in conjunction with the accompanying drawings: like Figure 1 As shown in the overall flowchart, the following steps are performed: Step S1: Synchronously acquire the thermocouple temperature measurement signal and the working current signal of the plasma excitation source. At this time, radio frequency interference with an amplitude of about 80mV is superimposed on the thermocouple signal, and the actual thermoelectric potential is about 4.8mV (corresponding to a temperature difference of 120℃).

[0024] Step S2: Perform spectrum analysis on the operating current signal. For example... Figure 2 As shown, the fundamental frequency is extracted. The second harmonic amplitude was detected at 30mV, and the third harmonic amplitude at 12mV. Fundamental interference amplitude... Estimation of effective amplitude of thermocouple signal Take the highest harmonic order Calculate the interference intensity factor: ; Step S3: According to Calculate the notch width factor Normalized angular frequency Configure an adaptive notch filter, such as... Figure 2 As shown, the transfer function is: ; Cascaded first-order low-pass filter, cutoff frequency Sampling period Transfer function: ; Filtered output The radio frequency fundamental frequency is attenuated by about 30dB.

[0025] Step S4: From Calculate the rate of temperature change. Heating phase. steady state phase .Pick (Equivalent value of effective signal enhancement after filtering), reference window length Weighting coefficient , According to radio frequency interference level calibration: hour Calculate the adaptive window length: Warming phase: ; Steady-state phase: ; like Figure 4 As shown in the relationship curve, the greater the temperature change rate, the shorter the window. A variable-window averaging is performed on the filtered signal to obtain... This value indicates the temperature difference between the hot and cold ends (i.e., the temperature value corresponding to the assumption that the cold end is 0°C).

[0026] Step S5: Extract the 50Hz zero-crossing signal from the operating current using a phase-locked loop as a synchronization signal. Offset time. .like Figure 3 As shown in the time-division sampling timing diagram, the cold junction temperature is collected within a 5ms window before and after the zero crossing to obtain a stable temperature. .

[0027] Step S6: Based on the cold junction temperature obtained from time-sharing sampling The cold junction thermoelectric potential can be obtained from the K-type thermocouple calibration table. . The averaged temperature signal obtained in step S4 is considered as the temperature difference indicator between the hot and cold ends, and is converted into a thermoelectric potential. Referring to the K-type calibration table, the thermoelectric potential corresponding to 120℃ is 4.803mV (4.806mV for linear interpolation), i.e., ... and The total thermoelectric potential is obtained by adding them together. Then, referring back to the K-type thermocouple calibration table, 5.808mV corresponds to an absolute temperature of approximately 145.1℃, which is the final temperature output. It remained stable between 145.0℃ and 145.2℃.

[0028] Example 2 This embodiment compares the conventional method with the method of the present invention. The conventional method uses a fixed RC low-pass filter (cutoff frequency 1MHz), which attenuates the 13.56MHz fundamental frequency by only about 20dB, and the residual interference has an equivalent temperature fluctuation of about ±5℃; continuous sampling at the cold end causes the reading to fluctuate between 23.0℃ and 27.2℃, with a final temperature error of ±6℃; the fixed window has an average response delay of about 2.5 seconds during the heating phase. After adopting the method of the present invention, as... Figure 5 As shown, the cold junction temperature fluctuation is only ±0.4℃, while the traditional method fluctuates by ±2.1℃; the overall temperature output fluctuation is reduced to ±0.1℃ (around 145.1℃), and the response time is shortened to 0.4 seconds. When the RF power supply is switched to pulse modulation mode (1kHz), Similarly, excellent results were achieved. This embodiment verifies the significant advantages of the present invention in environments with strong electromagnetic interference.

[0029] Therefore, this invention employs the aforementioned method for synchronous sampling and anti-interference of thermocouple signals in environments with strong electromagnetic interference. It synchronously acquires thermocouple temperature measurement signals and the working signal of the plasma excitation source, performs spectral analysis on the working signal to extract the fundamental frequency and harmonic components of the interference, and calculates the interference intensity factor. Based on this, it dynamically configures the center frequency of an adaptive notch filter and cascades a first-order low-pass filter to achieve deep suppression of radio frequency fundamental frequency, harmonics, and broadband noise. Furthermore, it calculates the temperature change rate from the filtered temperature and adaptively adjusts the sliding window length based on the interference intensity factor. During rapid temperature changes or strong interference, the window is automatically shortened to maintain response speed, while in steady state, the window is extended to smooth residual noise. Addressing the problem of electromagnetic coupling interference in the cold junction measurement circuit, it utilizes the synchronous signal to sample the cold junction temperature near the zero-crossing point in a time-division manner, fundamentally avoiding the interference peak region. Finally, it performs cold junction compensation to output accurate temperature. This method simultaneously applies the interference analysis results to the notch filtering, variable window averaging, and time-division sampling stages, achieving synergistic optimization and significantly improving the anti-interference capability, dynamic response speed, and steady-state accuracy of thermocouple temperature measurement in environments with strong electromagnetic interference.

[0030] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for synchronous sampling and anti-interference of thermocouple signals in environments with strong electromagnetic interference, characterized in that, Includes the following steps: S1. Synchronously acquire thermocouple temperature measurement signals and plasma excitation source operating signals, and record the sampling time sequence; wherein, the operating signals include operating current or voltage signals; S2. Perform spectrum analysis on the working signal to extract the fundamental frequency of the interference. and the frequencies of each harmonic, and calculate the interference intensity factor. ; S3, Based on the fundamental frequency of the interference The parameters of an adaptive notch filter are dynamically configured, and a first-order low-pass filter is cascaded to suppress high-frequency harmonics and broadband noise. This adaptive filtering is then applied to the thermocouple temperature measurement signal to obtain the filtered temperature signal. ; S4, from the filtered temperature signal Calculate the current rate of temperature change And combined with the interference intensity factor obtained in step S2 The weighting coefficients are determined based on the radio frequency interference level, and the average window length of the sliding window is adaptively adjusted. The filtered temperature signal is then subjected to variable window averaging to obtain the window-averaged temperature signal. , The temperature is indicated by the temperature difference between the hot and cold ends. S5. Utilizing interference fundamental frequency Generate a time-division sampling control signal, based on the synchronization signal extracted from the working signal, before and after the zero-crossing point of the synchronization signal. The cold junction temperature sensor signal is acquired within the time window to obtain the cold junction temperature after time-division sampling. ;in, This is the offset time from the zero-crossing point to the center of the sampling window; S6. Based on the cold junction temperature obtained from the time-sharing sampling in step S5. The window-averaged temperature signal obtained in step S4 Cold junction compensation is performed to obtain the final temperature output. .

2. The method for synchronous sampling and anti-interference of thermocouple signals in a strong electromagnetic interference environment according to claim 1, characterized in that, Interference intensity factor in step S2 The calculation expression is: ； in, This represents the amplitude of the fundamental frequency interference. For estimating the effective amplitude of thermocouple signals; For the first Subharmonic interference amplitude; The highest harmonic order is considered.

3. The method for synchronous sampling and anti-interference of thermocouple signals in a strong electromagnetic interference environment according to claim 2, characterized in that, The transfer function of the adaptive notch filter in step S3 for: ； in, For normalized angular frequency, , The sampling frequency; , This is the notch width factor; To transform complex variables; Transfer function of cascaded first-order low-pass filters for: ； Among them, the cutoff frequency of the low-pass filter Sampling period .

4. The method for synchronous sampling and anti-interference of thermocouple signals in a strong electromagnetic interference environment according to claim 3, characterized in that, Current temperature change rate in step S4 From the filtered temperature signal Calculated using the central difference: ； in, For the first Filtered temperature values ​​at each sampling point; For the first Filtered temperature values ​​from each sampling point.

5. The method for synchronous sampling and anti-interference of thermocouple signals in a strong electromagnetic interference environment according to claim 4, characterized in that, Adaptive window length in step S4 Determined by the following formula: ； in, The baseline window length; The weighting coefficients for the interference intensity factor. This is the weighting coefficient for the rate of temperature change; Interference intensity factor weighting coefficient According to radio frequency interference level calibration: when Time to take ,when Time to take Linear interpolation occurs when the distance is between 0 and 1.

6. The anti-interference method for synchronous sampling of thermocouple signals in a strong electromagnetic interference environment according to claim 5, characterized in that: In step S5, the time-division sampling control signal is based on a phase-locked loop synchronized with the pulse modulation signal of the RF power supply or the power frequency zero-crossing point, and the sampling window width is... ,in , This is the frequency of the synchronization signal.

7. The method for synchronous sampling and anti-interference of thermocouple signals in a strong electromagnetic interference environment according to claim 6, characterized in that, The cold junction compensation in step S6 specifically includes: based on the cold junction temperature obtained from time-division sampling. The corresponding cold junction thermoelectric potential can be obtained by referring to the thermocouple calibration table. The averaged temperature signal obtained in step S4 is then used to calculate the average temperature signal obtained in step S4. Converted to thermoelectric potential ;Will and The total thermoelectric potential is obtained by adding them together. Then consult the thermocouple calibration table to find the correct values. Convert to final temperature output .

8. The anti-interference method for synchronous sampling of thermocouple signals in a strong electromagnetic interference environment according to claim 7, characterized in that: Notch width factor With interference intensity factor Negative correlation, when When it increases Decreasing the notch depth increases the cutoff frequency of the first-stage low-pass filter; The frequency is more than three orders of magnitude higher than the highest effective frequency of the wafer temperature signal to ensure that the effective temperature signal is not attenuated.

9. The anti-interference method for synchronous sampling of thermocouple signals in a strong electromagnetic interference environment according to claim 8, characterized in that: Adaptive window length With temperature change rate and interference intensity factor They are inversely proportional, which allows for automatic switching to a short window to maintain response speed during process changes or strong interference.

Images