JP5724161B2 - Material testing machine - Google Patents

Description

The present invention relates to a material testing machine equipped with a de-I digital filter.

  When using the finite impulse response type digital filter when processing the measurement signal obtained from the sensor of the material testing machine, the window function such as the Hamming window or the Blackman window is used to suppress the cast phenomenon in the time domain. A digital filter used as a coefficient is used (Patent Document 1). That is, in the conventional finite impulse response type digital filter, convolution integration for calculating the output of the filter is performed by multiplying the delay data stored in the delay element and the filter coefficient and taking the sum of them.

JP-A-10-145185

However, in order to realize a digital filter based on the conventional method, a memory that stores the filter coefficient and a multiplier that multiplies the delay data stored in the delay element and the filter coefficient are necessary.
As a result, in order to realize a digital filter, a circuit of an appropriate scale is required. In particular, as the number of taps of the digital filter increases, the number of memories storing filter coefficients increases. It was.

The invention described in claim 1 is a material testing machine including a noise removing digital filter built in at least one of a test force amplifier and an extensometer amplifier, wherein the digital filter includes four basic blocks which are the same circuit. A digital filter for cascade connection, wherein the basic block includes a predetermined number of delay means cascaded to sequentially store discrete input data, and an average value of data respectively output from the predetermined number of delay means Average value calculation means for calculating the average value data and outputting average value data, and when cascading the four basic blocks, the average value data output from the basic block on the preceding stage is used as delay means on the subsequent stage side. As a configuration to obtain the average value data output from the basic block of the fourth stage, which is the final stage, as a filter output while inputting That.
According to the digital filter in the material testing machine having such a configuration, the characteristics similar to those of the digital filter using the window function can be simplified even though the conventionally used multiplier and filter coefficient memory are unnecessary. Can be realized. That is, only four stages of basic blocks having the same circuit configuration are connected in cascade, and the number of taps in each stage is set to the same number so that the Blackman-Harris can be used without using a multiplier and a filter coefficient memory. As with the window, a good impulse response can be obtained.
3. The digital filter in the material testing machine according to claim 2, wherein an adding means for adding data respectively output from the predetermined number of delay means, and a dividing means for dividing the addition result output from the adding means by the predetermined number. Thus, since the average value calculation means included in each basic block is configured, it is possible to execute a filter process with a moving average calculation at high speed .

According to the digital filter in the material testing machine according to the present invention, the conventionally used multiplier and filter coefficient memory are not required, so that the circuit configuration can be greatly simplified regardless of the number of taps. That is, only four stages of basic blocks having the same circuit configuration are connected in cascade, and the number of taps in each stage is set to the same number so that the Blackman-Harris can be used without using a multiplier and a filter coefficient memory. As in the case of the window, it is possible to obtain a special effect that a good impulse response can be obtained.
The material testing machine according to the present invention incorporates the above-mentioned digital filter in at least one of the test force amplifier and the extensometer amplifier, so that the desired filter processing can be performed without increasing the scale of the circuit configuration. Can do.

It is a block diagram which shows the digital filter by embodiment. It is a diagram which shows the impulse response of the digital filter shown in FIG. It is a block block diagram of a material testing machine including a digital filter to which the present invention is applied in a test force amplifier. It is a whole block diagram of the material testing machine containing the block structure shown in FIG.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing an FIR (finite impulse response) filter to which the present invention is applied. The illustrated digital filter has a configuration in which four filter circuits are connected in cascade (cascade connection). That is, one digital filter is configured by cascading the same filter circuits in four stages.

Discrete data X to be filtered (X is time-sequentially generated x ₀ , x ₁ , x ₂ ,...) Is sequentially input to the input terminal IN shown in FIG. Delay data output from each of the _n delay elements D _{0 to} D _n−1 is input to the adder ADD. The addition result output from the adder ADD is input to the divider DIV, and a division operation (÷ n) is performed. Then, the first stage averaged data is output from the divider DIV. In this embodiment, a case where n = 256 is set as an example will be described. Therefore, the division operation (÷ 256) is performed in the divider DIV.

  Thus, the first stage filter circuit does not include a multiplier. In other words, since all the filter coefficients are 1, a multiplier is unnecessary. The averaged data output from the first stage filter circuit is input to the second stage filter circuit. This second-stage filter circuit has the same circuit configuration as the first-stage filter circuit. The averaged data output from the second stage filter circuit is input to the third stage filter circuit.

  The averaged data output from the third stage filter circuit is input to the fourth stage filter circuit. The third-stage filter circuit and the fourth-stage filter circuit also have the same circuit configuration as the first-stage filter circuit. The final filter output is obtained from the fourth-stage filter circuit.

  As described above, in the digital filter (256 taps × 4 stages) according to the present embodiment, the first-stage average data is the second-stage input data, and the second-stage average data is the third-stage input data. Simple moving average calculation is performed by using the third stage averaged data as the fourth stage input data. However, it can be seen that the impulse response obtained as a result has characteristics close to those of a complex digital filter having a coefficient of 1021 Blackman-Harris windows as shown in FIG. FIG. 2 is a diagram in which the impulse response of a digital filter (256 taps × 4 stages) according to the present embodiment and the impulse response of a 1021 tap Blackman-Harris window are superimposed. The horizontal axis represents the step number of the impulse response, and the vertical axis represents the magnitude of the response. In the case of an FIR type digital filter, the magnitude of the impulse response is equal to each coefficient of the digital filter. As described above, the calculation method of the digital filter (256 taps × 4 stages) according to the present embodiment simply repeats four times that the sum of each stage is simply obtained and divided by the number of taps (that is, the average value is obtained). However, the result that appears at the end of the fourth stage is the result obtained by calculation of a 1021 tap FIR digital filter using the size of the Blackman-Harris window as a coefficient (multiplying and multiplying and dividing by the total number). A very close value is obtained. The impulse response of FIG. 2 expresses this visually. By the way, the “impulse response” referred to in digital computation is a system (filter, etc.) when a waveform (= impulse) whose magnitude is 1 only at time 0 and whose magnitude is 0 at other times is input. , System or black box) output (= response).

  FIG. 3 is a block diagram of a material testing machine including the digital filter shown in FIG. 1 in the test force amplifier AMP. This figure shows a circuit configuration for displaying an analog signal output from the load cell LC. The analog signal output from the load cell LC is supplied to the preamplifier 2. The output signal from the preamplifier 2 is input to the anti-aliasing processing analog filter 4 in order to prevent aliasing during sampling. The signal output from the anti-aliasing processing analog filter 4 is input to an A / D converter 6 that performs oversampling. The digital signal output from the A / D converter 6 is input to the digital filter 8. The digital filter 8 is the digital filter described with reference to FIG. 1 and is used to remove noise due to the oversampling.

  The output signal of the digital filter 8 configured according to the present embodiment is input to the offset removal circuit 10 for removing the offset component and making the measurement value zero when the load cell LC is unloaded. The offset removal circuit 10 is connected to an offset setting unit 12 for setting an offset value. The output signal of the offset removal circuit 10 is input to a multiplication circuit 14 that performs gain adjustment so that a full-scale measurement value is obtained when a rated actual load is applied to the load cell LC. The multiplication circuit 14 is connected to a gain setting unit 16 that sets the multiplication rate of the multiplication circuit 14. In addition, when applying the rated actual load to the load cell LC, it is possible to adjust the gain by actually applying a weight or the like to the load cell LC, or by applying a simulated load cell output change (resistance value change) to the preamplifier 2. is there. Thus, the signal processing before nonlinearly correcting the load cell output is completed. The output signal from the multiplication circuit 14 is input to a nonlinear correction circuit 18 configured by hardware.

  The test force amplifier AMP includes the preamplifier 2 to the nonlinear correction circuit 18 described above. The output of the nonlinear correction circuit 18 is input to the delay filter 20. The delay filter 20 is a filter for sending a measurement value to the display unit at the subsequent stage. The signal output from the delay filter 20 is input to the FIFO memory 22. The data output from the FIFO memory 22 is sent to the display 24 for visual display. In other words, the FIFO memory 22 functions to transfer data to the display 24 attached (or externally attached) to the control panel 42 (described in FIG. 4).

  Note that a servo motor control circuit for controlling the position of the crosshead 32 (described in FIG. 4) is not directly related to the present invention and will not be described. When an extensometer (not shown) is used instead of the load cell LC, the extensometer amplifier is configured by the preamplifier 2 to the nonlinear correction circuit 18.

  4 is an overall configuration diagram of a material testing machine including the block configuration shown in FIG. The load cell LC that detects the test force loaded on the test piece TP is placed on the crosshead 32. A signal from the load cell LC is sent to the control panel 42 via the cable unit CU. The control panel 42 includes blocks after the preamplifier 2 shown in FIG.

  A pair of support columns 31A and 31B are erected from the base 34, and their upper portions are fixedly connected by a cross yoke 36. A ball screw (not shown) that is rotated by a motor (not shown) is housed inside the pair of columns 31A and 31B. The crosshead 32, which is placed between the two ball screws and screwed together, moves up and down in accordance with the rotation of the ball screw. The upper grip 38 is fixedly connected to the crosshead 32 via the load cell LC, and the lower grip 40 is fixedly connected to the base 34. The upper grip 38 and the lower grip 40 face each other, and the test piece TP is gripped by these two grips 38 and 39. An extensometer KK for detecting the elongation of the test piece TP is directly connected to the test piece TP, and its signal is sent to the control panel 42. The signal line of the extensometer KK is not shown. The control panel 42 includes not only a load mechanism (not shown) but also various interface circuits (not shown). The material testing machine 44 is configured by the above components.

<Operations and effects according to the embodiment>
According to the present embodiment, the following actions and effects can be achieved.
(1) and the delay element _D 0 _{to D n-1} of n connected in cascade to sequential storage discrete input data (n = 256), n number of delay elements _D 0 respectively from _{to D n-1} When the basic block having the average value calculator (ADD, DIV) that calculates the average value of the output data and outputs the average value data is cascaded in four stages, the average output from the basic block on the previous stage side By inputting the value data to the delay element on the subsequent stage side, the average value data output from the fourth basic block is obtained as the filter output, so that the conventionally used multiplier and filter coefficient memory are not required. Nevertheless, the same characteristics as those of the digital filter using the window function can be easily realized.

(2) An adder ADD that adds the data output from each of the _n delay elements D _{0 to} D _n−1, and a divider that divides the addition result output from the adder ADD by n (n = 256) Since the average value calculator included in each basic block is configured by DIV, filter processing with moving average calculation can be executed at high speed.

  (3) The test force amplifier AMP (FIG. 3) of the material testing machine has an extremely simple configuration in order to remove noise due to oversampling after oversampling A / D conversion of the analog signal output from the load cell LC. Nevertheless, digital filter processing using a window function can be performed.

<Other variations>
(1) Although the digital filter shown in FIG. 1 has a cascade connection of four stages, it is the same as a digital filter using a window function by using a cascade connection of two or more stages from the viewpoint of performing a moving average calculation. A filter can be configured. However, since the shape of the window function changes depending on the number of taps n of each basic block and the number of cascaded stages K (K is an integer of 2 or more), it is necessary to select the number of taps n and the number of stages K as necessary. .

(2) By using dedicated hardware or a microcomputer using an FPGA or the like, average value data can be obtained directly from the data stored in each of the delay elements D _{0 to} D _n−1. Calculation is possible.

  (3) Although the test force amplifier has been described in the above-described embodiment, it can also be used as a noise removing filter built in the extensometer amplifier.

  (4) In the embodiments described so far, the digital filter applied to the material testing machine has been described. However, the present invention is not limited to the material testing machine. For all electronic devices that filter digital signals, Applicable.

  (5) In order to realize the digital filter described so far, whether it is realized by software or hardware can be appropriately selected according to the application.

The above description is merely an example, and the present invention is not limited to the above-described embodiments and modifications unless the features of the present invention are impaired.
It is also possible to combine the embodiment and one of the modified examples, or to combine the embodiment and a plurality of modified examples.
It is possible to combine the modified examples in any way.
Furthermore, other forms conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention.

D _{0 to} D _n-1 delay element ADD Adder DIV Divider AMP Test power amplifier CU Cable unit LC Load cell KK Extensometer TP Test piece 2 Preamplifier 4 Analog filter 6 for anti-aliasing processing A / D converter 8 Digital filter 10 Offset removal circuit 12 Offset setting unit 14 Multiplication circuit 16 Gain setting unit 18 Nonlinear correction circuit 20 Delay type filter 22 FIFO memory 24 Display 32 Crosshead 34 Base 36 Cross yoke 38 Upper gripper 40 Lower gripper 42 Control panel 44 Material testing machine

Claims (2)

In a material testing machine equipped with a noise removing digital filter built in at least one of a test force amplifier and an extensometer amplifier,
The digital filter is a digital filter that cascades four basic blocks that are the same circuit,
The basic block calculates a mean value of the data output from the predetermined number of delay means connected in cascade to sequentially store discrete input data and the predetermined number of delay means, and outputs the average value data. Average value calculating means to
When cascading the four basic blocks, the average value data output from the basic block on the front stage is input to the delay means on the rear stage,
A material testing machine characterized in that average value data output from the fourth basic block of the last stage is obtained as a filter output. The material testing machine according to claim 1,
The average value calculating means included in the basic block includes an adding means for adding data output from the predetermined number of delay means, and a dividing means for dividing the addition result output from the adding means by the predetermined number. And a material testing machine .

Images