JPH0575259B2 - - Google Patents

Description

[Detailed description of the invention] [Industrial application field] The present invention extracts the characteristics of the curing curve of a vulcanized compound as a test sample obtained by a torsional vibration curing test using several representative characteristic values. and compare the representative characteristic values with limit values such as upper and lower limits in the standard characteristic value group representing the specifications of the compound,
The present invention relates to a method for automatically determining whether or not a test sample meets the specifications based on the comparison results.

[Prior Art] Conventionally, the torque-curing time curve (so-called curing curve) drawn with a torsional vibration curing tester has been
It is recorded by the method shown in Figure A or B.

In other words, regarding Fig. 2A, there is a method of drawing the torque waveform detected by the torque detection section of the testing machine on the recording paper as it is, a method of drawing the entire waveform as shown in the figure, and a method of drawing the entire waveform as shown in the figure, and a method of drawing the torque waveform detected by the torque detection section of the testing machine, and a method of drawing the entire waveform as shown in the figure, and a method of effectively utilizing the chart width. A one-sided recording method in which the negative half of the torque waveform is omitted is commonly used. Or again,
As shown in FIG. 2B, there is also a peak hold method in which only the peaks of these waveforms are drawn, and this is commonly used for the purpose of preventing consumption of the pen tip and ink.

As a conventional and widely used method for quantifying hardening curves, the most popular method is to express the envelope of the torque waveform using five parameters, as shown in FIG. That is, first find the minimum torque M _L and maximum torque M _H of the hardening curve, then (M _H
-M _L ) is divided into 10/100, 50/100, and 90/100, and the curing times corresponding to the points are determined as shown in FIG. 3 and are defined as t10, t50, and t90, respectively. M _L relatively represents the elastic modulus of the sample before hardening, and M _H relatively represents the elastic modulus of the sample after sufficiently hardening, and at time t10,
t50 and t90 have a curing reaction of 10, 50, and 90, respectively.
%Represents the time required to complete.

However, with these methods, if the zero level of torque changes during measurement due to some reason, such as drift of the load cell amplifier or drift of the load cell itself, the amount of change is directly added to the torque axis direction of the hardening curve. It has the disadvantage of causing errors.

Furthermore, as is clear from the figure, this method
until the curing curve is saturated and horizontal (i.e.
The disadvantage is that parameters other than M _L cannot be determined unless measurements are taken until M _H is determined. This poses a major obstacle to the purpose of increasing measurement efficiency.
For example, in the quality control of unvulcanized compounded rubber, for which a curing tester is most effectively applied, the main purpose is to identify the test sample, so it is not necessary to measure up to M _H. However, as long as the conventional method described above is used,
The disadvantage of not measuring up to M _H is that parameters that can be used for judgment cannot be obtained.

As an expedient method to compensate for the inconvenience of such measurements, some methods are used that simply use the torque in a certain time or the time to reach a certain torque as a parameter without determining M _H. In that case, values such as the constant time and constant torque to be used must be artificially determined for each compounding composition, and the procedure is complicated and requires skill, which remains an unresolved problem.

[Problems to be Solved by the Invention] Therefore, an object of the present invention is to eliminate the above-mentioned conventional drawbacks, eliminate the need to artificially determine constants for each compounding composition, and calculate all parameters for each curing curve. It is an object of the present invention to provide a method for determining a vulcanized compound, which can uniquely determine the vulcanized compound and can significantly shorten the measurement time.

[Means for Solving the Problems] In order to achieve such an object, in the first embodiment of the present invention, a curing curve of a vulcanized compound as a test sample obtained by a torsional vibration curing test is Characteristics are extracted using several representative characteristic values, and the representative characteristic values are compared with limit values such as upper and lower limits in a group of standard characteristic values that represent the specifications of the compound, and the test results are determined based on the comparison results. Conditions for terminating data capture regarding the limits of pass/fail of vulcanized compounds and the representative characteristic values to be used in a method for automatically determining compliance of vulcanized compounds with specifications. is specified in advance, the absolute value M of the torque difference between two adjacent peak torques is captured every half cycle of the torque waveform that forms the hardening curve, and the data of the calculated peak torque difference M is stored once. Then, it is determined whether the stored data satisfies the conditions for terminating the data acquisition, and when the conditions are satisfied, the maximum value M _H and the minimum value are determined as representative characteristic values for the peak torque difference M. Value M _L , (M−
_{M _ _}
At _least one _{of the} time _t It is characterized by determining pass/fail.

In the second embodiment of the present invention, the characteristics of the curing curve of a vulcanized compound as a test sample obtained by a torsional vibration curing test are extracted using several representative characteristic values, and the representative characteristic values are In a method for determining a vulcanized compound, the test sample is compared with limit values such as an upper limit value and a lower limit value in a group of standard characteristic values representing the specifications, and the suitability of the test sample for the specifications is automatically determined based on the comparison result. Regarding the representative characteristic values to be used,
Specify in advance the conditions for terminating the data acquisition regarding the pass/fail limits of the vulcanized compound and the typical characteristic values to be used, and calculate the difference between two adjacent peak torques every half period of the torque waveform forming the curing curve. Find the absolute value M of the torque difference and the difference ΔM between the absolute value M of the two adjacent torque differences, store the data of the obtained peak torque difference M and the difference ΔM, and then use the stored data to import the data. When the conditions are satisfied, the maximum value ΔM _nax of ΔM and the time t〓 _nax are determined as representative characteristic values for the peak torque difference M and the difference ΔM.
and the value of M at that time M〓 _nax , the time t〓〓 when ΔM reaches a predetermined ratio β to ΔM _nax , the value of M at that time M〓, the time t _Y when ΔM reaches the predetermined value. At least one of the value M _Y of M at that time and the value ΔM _W of ΔM at a predetermined time t _W is determined from the stored data, and the determined representative characteristic value is compared with the pass/fail limit to determine the vulcanized compound. The characteristics of the curing curve of the vulcanized compound as the test sample obtained by the torsional vibration curing test, which is characterized by determining pass/fail, are extracted using several representative characteristic values,
The representative characteristic value is compared with limit values such as upper and lower limits in a group of standard characteristic values representing the specifications of the compound, and based on the comparison results, the suitability of the test sample to the specifications is automatically determined. In the method for determining sulfur compounds, the conditions for completing data capture regarding the limits of pass/fail of the vulcanized compound and the representative characteristic values to be used are specified in advance for representative characteristic values to be used, and a curing curve is formed. For each half cycle of the torque waveform, the absolute value M of the torque difference between two adjacent peak torques and the difference ΔM between the absolute value M of the two adjacent torque differences are determined, and the determined peak torque difference M and the difference ΔM are calculated.
When the data is stored, it is determined whether the stored data satisfies the conditions for terminating the data acquisition, and when the conditions are satisfied, the representative characteristics for the peak torque difference M and the difference ΔM are determined. As values, the maximum value ΔM _nax of ΔM, the time t〓 _nax and the value of M at that time M〓 _nax , and the time t〓〓 when ΔM reaches a predetermined ratio β to ΔM _nax
and the value of M at that time M〓, the time t _Y when ΔM reaches a predetermined value, the value M _Y of M at that time, and the value ΔM _W of ΔM at a predetermined time t _W. It is characterized in that it is determined from stored data and the determined representative characteristic value is compared with the pass/fail limits to determine pass/fail of the vulcanized compound.

In the third embodiment of the present invention, the characteristics of the curing curve of a vulcanized compound as a test sample obtained by a torsional vibration curing test are extracted using several representative characteristic values, and the representative characteristic values are In a method for determining a vulcanized compound, the test sample is compared with limit values such as an upper limit value and a lower limit value in a group of standard characteristic values representing the specifications, and the suitability of the test sample for the specifications is automatically determined based on the comparison result. Regarding the representative characteristic values to be used,
Specify in advance the conditions for terminating the data acquisition regarding the pass/fail limits of the vulcanized compound and the typical characteristic values to be used, and calculate the difference between two adjacent peak torques every half period of the torque waveform forming the curing curve. the absolute value M of the torque difference of Calculate loss torque M'', which is the absolute value of When the conditions are satisfied, the value M of M'' at the time when M _L and M _H are obtained is determined as the representative characteristic value for the peak torque difference M, the difference ΔM, and the loss torque M''. ″ _L and M″ _H , time points
M″ at tα, t _X , t〓 _nax , t〓〓, t _Y , t _W and t _Z
value, maximum value of M″ M″ _nax and peak torque difference M
and the storage torque M′ at the time when the loss torque M″ is obtained = √ ² − (″) ² and the loss tangent tanδ
The method is characterized in that at least one of =M''/M' is determined from stored data, and the determined representative characteristic value is compared with a pass/fail limit to determine pass/fail of the vulcanized compound.

[Operation] According to the present invention, various parameters including the time element are determined based on the peak torque difference M, the difference ΔM, and the loss torque M'',
Since these parameters are compared and contrasted with a predetermined reference value sequence, there is no need to artificially determine constants for each formulation, and all parameters can be uniquely determined for each curing curve. I can do it. Furthermore, in the present invention, the maximum value in the curing curve, that is, the saturation value
Since it is not necessary to measure until M _H is obtained, the measurement time can be shortened, and especially when dealing with ΔM, the measurement time can be significantly shortened. In addition, in the present invention, since the peak torque difference between two adjacent peaks is determined, the zero-drift error is canceled out between these adjacent peaks, and thus more accurate measurement data can be obtained. Since one point of data can be obtained per day, the density of measurement data is high.
From this point of view, it is advantageous for performing highly accurate measurements in a short measurement time, and is also advantageous for smoothing and averaging processing of data during data analysis.
Effective for tracking fast curing reactions.

Moreover, according to the present invention, the viscoelastic changes of the test sample during the curing process are tracked by separating them into two properties, viscous and elastic, so it is more reliable and more complete than conventional methods. It becomes possible to perform quality control.

[Example] The present invention will be described in detail below with reference to the drawings.

In the present invention, in order to eliminate the above-mentioned conventional drawbacks, the peak torque difference M and the difference ΔM between the torque differences M are determined every half cycle, as shown in FIGS. 1A to 1C. This method will be explained using FIGS. 1 and 4.

For convenience of explanation, FIG. 1 shows a torque waveform enlarged in the time axis direction. Now, assume that the peak torques that alternately appear upward and downward every half cycle of the waveform during curing are P ₁ , P ₂ , ..., P ₅ . Two adjacent peak torques are always stored using the peak hold method customary in light current circuit technology, and the two adjacent peak torques are stored in an external recorder or computer.
The torque difference M between peaks is sequentially outputted as M ₁ =|P ₁ −P ₂ | M ₂ =|P ₂ −P ₃ |.

Note that the above calculation may be performed using an adder circuit, or the peak signals P ₁ , P ₂ , . . . may be input to a computer, and the calculation may be performed using software.

As shown in FIG. 2, in the peak torque difference M obtained in this way, the above-mentioned zero drift error is canceled out by two adjacent P values, so that more accurate data can be obtained. Moreover, in terms of data density, in contrast to the one point per period that can be used in FIGS. 2A and B, the method of the present invention allows one point of data to be obtained every half period. This results in twice the data density, which is advantageous for data smoothing and averaging processing at the data analysis stage, and is also suitable for tracking fast curing reactions.

In the present invention, in addition to uniquely determining the parameters for each curing curve as described above, in order to further shorten the measurement time, it is possible to In addition to calculating only the peak torque difference M as a difference, the difference ΔM of the peak torque difference M may be calculated.

When a curve obtained by differentiating the hardening curve of FIG. 3 with respect to time is obtained, a chevron-shaped curve as shown in FIG. 5 is generally obtained. The peak of that mountain is the third
Corresponding to the inflection point of the curve in the figure, the time to reach this peak is usually shorter than the _M It is often less than 1/2.

Therefore, by analyzing parameters starting from this peak instead of M _H , the measurement time can be significantly shortened.

As mentioned above, the actual hardening curve is not a continuous curve but a set of discontinuous points, which are the peaks of the oscillating torque waveform, so the mathematical means to actually obtain the curve in Figure 5 is not differentiation but difference. become.

Returning to FIG. 4, to explain the procedure for calculating the difference ΔM, the length of the arrow labeled ΔM ₁ in the figure is the difference between P ₁ and P ₃ , but this also means that M ₁ and
equal to the difference in M ₂ (i.e., ΔM ₁ ). Therefore,
In the same way as finding M from the difference in peak torque every half cycle, ΔM is found from the difference in peak torque every half cycle, and by plotting that value against the curing time, we get the result shown in Figure 4. A new hardening curve for ΔM will be obtained.

The method described above allows M and ΔM to be
This method uses the highest data density of 2 points per cycle, but the curing speed is sufficiently slow compared to the vibration cycle and does not require a very high density, or the density may be lower due to the storage capacity of the computer being used. If it is necessary to lower the number, without necessarily using the above-mentioned difference analysis method, the number sequence Mi averaged or smoothed at an appropriate time interval coarser than the period of vibration and its first difference number sequence ΔMi (i is (natural number) may be substituted.

In that case, the mathematical means to be used for averaging or smoothing may be commonly used and known means such as the moving average method or the exponential smoothing method.

Generally, the torque waveform generated when a viscoelastic sample is subjected to torsional vibration has a slight phase shift in the time axis direction with respect to the torsion angle waveform, as shown in FIG. The reason for this is the contribution of the elastic term (also referred to as loss term) included in the elastic modulus of the sample. As shown in Fig. 6, the phase difference angle is δ, the absolute value of the difference every 1/2 period of the torque generated at the moment when the waveform of the torsion angle passes through the midpoint and peak of the amplitude is M″, M″ and as shown
Defining M′, M″ and M′ relatively represent the viscous term and the elastic term of the elastic modulus of the sample, respectively, M″ corresponds to the loss torque, and M′ corresponds to the storage torque. The ratio M″/M′ of the two is equal to the loss tangent tanδ. Also, M represents the relative value of the overall elastic modulus.

When both the torque and torsion angle waveforms are sinusoids, the following relationships (1) and (2) hold between these four parameters, so any two of the four parameters can be If actual measurements are taken, the remaining two can be calculated.

M″ ² +M′ ² =M ² M″/M′=tanδ (1) Therefore, tanδ=M″/M′ =M″/√ ² −(″) ^2A typical viscoelastic material such as rubber The stiffness (modulus of elasticity) of the material always has two properties: viscous and elastic properties, so like the conventional hardening test,
A unified method of simply observing the trajectory of peak torque cannot accurately grasp its properties.
In contrast, according to the present invention, the stiffness of the test sample during the curing process is observed as binary information divided into a viscous term and an elastic term, as described above, so it is more effective than conventional methods. More reliable and complete quality control information can be obtained.

In the case of rubber vulcanization, the vulcanization curve of M'' is usually obtained in the shape shown in Figure 7. Therefore, the various characteristic parameters determined for the curing curve of M, the value of M'' at the same time, and All tanδ values calculated from M and M'' at that time using equation (2) can be used as effective parameters for quality control, and it is worth setting control limits for them and managing them. .

In the present invention, as described above, as the original data of the hardening curve, the absolute value M of the torque difference between adjacent peaks in each half period of the oscillating torque waveform, the torque difference M between the peaks in each period, or the torque difference M in each period at least one of three series of the difference ΔM of the peak torque difference M of After extracting a sequence of data at equal time intervals (Mi, ΔMi, Mi'') and storing these data in a memory, various desired representative characteristic values are calculated as described later.

Here, the various data M, ΔM and
An example of a circuit configuration for obtaining M and M'' of M'' is shown in FIG.

In Fig. 8, 1 is a torsional vibration hardening tester;
2 is a strain gauge type load cell for measuring the vibrating torque waveform of the test sample, and 3 is a gauge power source for this load cell 2. A torque signal obtained by supplying the output of the load cell 2 to the differential amplifier 4, that is, a torque signal having a waveform as shown in FIG. 9, is supplied to sample/hold circuits 5 and 6 and sample/peak hold circuits 7 and 8. . These circuits 5 to 8 are connected to the optical sensor 9 at time points t1, t2, t3, t3, etc. as shown in FIG.
Timing pulses S1, S2, S occurring at t4
3, S4 is supplied and sampling is performed at that timing.

The optical sensor 9 detects a timing hole 12 formed in a timing plate 11 that rotates in synchronization with a drive shaft 10 for applying vibration to a test sample at a constant period,
Thus, timing pulses S1, S2, S3, S4, S1, S2, . . . are generated in synchronization with the waveform of the torque signal.

The sample/peak hold circuit 7 holds the maximum value B of the torque signal between timing pulses S1 and S2. The sample/peak hold circuit 8 holds the maximum value E of the torque signal between timing pulses S3 and S4. sample·
Hold path 5 holds the value A of the torque signal at the instant of timing pulse S1. The sample and hold circuit 6 holds the value D of the torque signal at the instant of the timing pulse S3. The contents of these held data are updated every cycle of the torque signal, and thereafter held for the duration of one cycle.

The outputs of the sample and hold circuits 5 and 6 are supplied to a differential amplifier 13, which outputs the absolute value of the difference between the signals A and D held as described above, that is, M''.

The respective outputs of the sample/peak-hold circuits 7 and 8 are supplied to a differential amplifier 14, which outputs the absolute value of the difference between the signals B and E held as described above, that is, M.

These outputs M″ and M are respectively A/D
The digital quantities M'' and M obtained by supplying the converters 15 and 16 to the central processing unit (CPU) 1
Supply to 7.

The central processing unit 17 inputs various initial values from the keyboard 18, such as sample name, test conditions,
Based on the reference value sequence code number, etc., and test termination conditions, and read-only memory (ROM)
According to the control procedure stored in 19, the above-mentioned data M and M'' are processed to determine whether the test sample is acceptable or not.

Note that such processing can be started, for example, based on a start signal synchronized with a start command of the testing machine 1.

20 is a random access memory (RAM) that temporarily stores data input to the central processing unit 17 and data processed therein.

The judgment results processed by the central processing unit are
The output can be performed by an output device 21, such as a CRT display, a printer, a digital display tube, a lamp, a buzzer, or the like. For output devices such as printers and recorders, M and
M″ measurement results can also be recorded.

Here, the difference ΔM with respect to the absolute value M of the torque difference between peaks is obtained as ΔM(i)=M(i+1)−M(i), and these three series of data, namely torque amplitude M, torque amplitude difference Based on ΔM and loss torque M'', in the present invention, the following various characteristic values can be used to determine the control limits as uniquely determined representative characteristic values of the hardening curve.

(1) Representative characteristic values obtained from Mi (1.1) As shown in FIG. 10, when the M curve is determined, the minimum value M _L and maximum value M _H are determined from the curve.

(1.2) The time t when (Mi−M _L ) reaches a predetermined ratio α (%) of (M _H −M _L ), for example, the first
As shown in Figure 1, the times tα=t10, t50, and t90 for reaching α=10, 50, and 90% are determined.

(1.3) Time until Mi reaches the predetermined value _{M
t X} , for example, as shown in _Fig . 12, M
(for example 15Kgf・cm), M _X3 (for example 20Kg
f・cm), etc. t _X1 , t _X2 ,
t Find _X3 , etc.

(1.4) Find the value M _Z of M at a predetermined time t _Z.
For _example , _as shown in FIG.
The value of M M _Z1 at t _Z3 (for example, 90 seconds),
Find M _Z2 and M _Z3 .

(2) Representative characteristic values obtained from Mi and ΔMi (2.1) As shown in Figure 14, the maximum value ΔM _nax of ΔM and the values of t _nax and M at that time
M〓 Find _nax .

(2.2) Find the value M of M when ΔM reaches a predetermined ratio β (%) to Δ _nax and the time t〓 at that time. For example, as shown in Figure 15, MΔ40 when β = 40, 60, 80%,
Find MΔ60, MΔ80, tΔ40, tΔ60, tΔ80.

(2.3) As shown in FIG. 16, find the time t _Y when ΔM reaches a predetermined value ΔM _Y and the value M _Y of M at that time.

(2.4) As shown in FIG. 17, find the value ΔM _W of ΔM at a predetermined time t _W. However, t _W is t _Z
May overlap with

(3) Representative characteristic values obtained from Mi, ΔMi, and M″ (3.1) At the time when M _L and M _H are obtained
Find the values M _L ″ and M _H ″ of M″.

(3.2) At each point in time t〓, t _Z , t〓 _nax , t〓〓, t _Y , t
_W
and find the value of M″ at time t _Z.

(3.3) Find the maximum value M″ _nax of M″.

(3.4) For the combination of values of torque amplitude M and loss torque M'' at the same time as (3.1), (3.2), and (3.3), calculate the storage torque M′ and loss tangent tanδ, which are related by the following equation. .

M′=√ ² −(″) ² tanδ=M″/M′ (4) Any one or more of the various characteristic values listed in items (1) to (3) above, or A desired characteristic value derived from a combination of the characteristic value and device constants such as the vibration frequency and the shape factor of the measuring section can also be used for the determination of the sample.

Next, an example of a control procedure for determining the various characteristic values shown in items (1) to (3) above and further determining whether the characteristic value belongs to pass/fail is shown in FIGS. 18A and B. . This control procedure is stored in ROM 19 (8th
), and according to the control procedure, the central processing unit 17 controls the A/D converters 15 and 1.
Processing of data M'' and M from 6 and the above-mentioned pass/fail judgment are performed.

In FIGS. 18A and 18B, in step S1, various initial values such as the sample name, test conditions, code number of the reference value series, and number of lots are input.

Here _{, the} reference value sequence is _a sequence ( _{n 3 ,}
n ₁ , n ₂ , n ₄ ), and the values specifically determined in advance for the various representative characteristic values mentioned above are stored in the ROM.
I will draw it on 19. Note that when these reference values are compared with the above-mentioned area, the results are shown in FIG. 19. Here, areas smaller than the value n ₃ and areas larger than the value n ₄ are defined as failure areas.

Next, in step S2, the measurement end condition K
Enter. The measurement termination condition K determines what condition must be satisfied to terminate the measurement. For example, K = 1 to 6 is determined in advance as follows, and the ” and any numerical value from “1” to “6” to set one of the measurement end conditions.

K=1: The measurement ends when the measurement time ti reaches a predetermined time point t〓.K=2: The measurement ends when M reaches the maximum value M _H.

K=3: The measurement ends when ΔM reaches the local maximum value ΔM _nax .

K=4: ΔM is the value β・ΔM _nax of the predetermined ratio β of ΔM _nax
The measurement ends when the temperature drops to .

K=5: The measurement ends when M reaches the predetermined value _M.

K=6: The measurement ends when ΔM reaches the predetermined value _ΔMY .

In the next step S3, it is determined whether a start signal generated in response to the start-up of the measuring instrument 1 has been input. Only after the start signal is input does the process proceed to the next step S4, where the peak-to-peak torque signal M(i) and
M''(i) is fetched and the data is temporarily stored in the RAM 20 in step S5.

In the next step S6, the difference ΔM of M is calculated according to ΔM(i)=M(i+1)−M(i). The resulting difference ΔM
is temporarily stored in the RAM 20 in step S7.

Next, the central processing unit 17 takes in the following measurement values.

(1) The minimum value M _L of M and the instant t _L at that time are [M _L ,
t _L ].

(2) The maximum value M _H of M and the instant t _H at that time are [M _H ,
_tH ].

(3) Take in the maximum value ΔM _nax of ΔM and the instant t〓 _nax at that time as [ΔM _nax , t〓 _nax ].

Next, in step S9, these captured data are temporarily stored in the RAM 20.

Next, the process proceeds to step S10, where it is determined which measurement end condition K is. If K=1, the process proceeds to step S11; if K=2, the process proceeds to step S11.
The process proceeds to S12, S13 when K=3 or 4, S16 when K=5, and S17 when k=6.

In step S11, the current time t _i is set to a prespecified t〓 (for example, t10, t50,
t90, etc.), and if the time has not been exceeded, the process proceeds to step S18, where the time point is advanced from t _i to t _i+1 , and then returns to step S4. When t _i ≧t〓, it is determined that all data acquisition has been completed at this point, the measurement is ended, and the process proceeds to step S19, where parameters are analyzed.

In step S12, it is determined whether or not the current time t _i has exceeded t _H , and until the current time t i has exceeded t H, the time point is advanced from t _i to t _i+1 via step S18, as in the case of step S11. Repeat the measurement, t _i ≧t _H
It is determined that the measurement is completed only when the following is satisfied, and the process proceeds to the next step S19.

In step S13, it is determined whether t _i exceeds t〓 _nax or not, and the same measurement loop is repeated through step S18 as described above until it exceeds t〓 nax. When t _i ≧t〓 _nax holds, the next step is
The process advances to step S14, where it is determined whether K is 3 or 4, and when K=3, it is determined that the measurement has ended and the process proceeds to step S19.
On the other hand, when K=4, the process proceeds to step S15, where ΔM is set to the value β・ of the predetermined ratio β of ΔMmax.
It is determined whether the value has decreased to ΔMmax (for example, β=40, 60, 80%). Then, the step continues until ΔM becomes lower than β・ΔMmax.
The same process is repeated after S18. ΔM≦β・
When ΔMmax is established, the measurement is ended and the process proceeds to step S19.

In step S16, it is determined whether M has reached a predetermined value _M.sub.X , and the same processing is repeated via step S18 until _M.sub.X is reached. If M≧ _M

In step S17, it is determined whether ΔM has reached a predetermined value _ΔMY , and if not, the process proceeds to step S18.
Repeat the same process via. When ΔM≦ _ΔMY is established, the measurement is ended and the process proceeds to step S19.

In this manner, in the present invention, the process proceeds to step S19 only after all measurements scheduled using the parameter K have been completed, and the parameter is analyzed here.

(4) tα: Specific ratio α of (M − M _L ) to (M _H − M _L )
Find the time tα to reach (see Figure 11).

(5) Find the time _tX at which M _i reaches the predetermined value M _X (see Figure 12).

(6) Find the value M _Z of M and the difference ΔM _Z of M at a predetermined time t _Z (see FIG. 13).

(7) Maximum value of ΔM ΔM Time t of _nax 〓 Maximum value of M at _nax 〓 Find _nax (see Figure 14).

(8) Find the value M〓 of M when ∆M reaches the predetermined ratio β of ∆M _nax and the time t〓〓 (first
(See Figure 5).

(9) Find the time t _Y at which ΔM reaches the specific value ΔM _Y and M _Y at that time (see Figure 16).

(10) Find the value ΔM _W of ΔM at a predetermined time t _W (see Figure 17).

(10) Various time points t _L , t _H , t〓, t _Z , t _X , t〓〓, t _W , t _Y
Find the loss torque M″ and loss tangent tanδ at .

In addition, in the case of K=3 to 6, the corresponding one of the above-mentioned item (2) and the above-mentioned items (4) and (11) is omitted.

Next, proceed to step S20. Here, all or any desired part of the parameters (1) to (3) obtained in step S8 as described above and the parameters (4) to (11) obtained in step S19 are explained. The data is compared with the reference value sequence group inputted in advance in step S1. Thereby,
Determine which section of the reference value sequence each parameter belongs to.

Furthermore, in the next step S21, the result of this determination is displayed or recorded by the output device 21. By this output device 21, M, ΔM, M″,
The value of tan δ at each instant and the various parameters determined by items (1) to (11) may also be displayed or recorded.

Next, proceed to step S22, where step
In S1, it is determined whether or not the determination process regarding the planned number of lots inputted has been completed, and if it has not been completed, the process returns to step S3. On the other hand, when the test is completed, it is assumed that all tests for the current lot have been completed, and the process returns to step S1 to wait for the test for the next lot.

[Effects of the Invention] As is clear from the above, according to the present invention, various parameters including the time element can be determined based on the peak torque difference M, the difference ΔM, and the loss torque M''. , those parameters,
Since it is compared with a predetermined reference value series, there is no need to artificially define constants for each formulation, and all parameters can be uniquely determined for each curing curve. Moreover, in the present invention, it is not necessary to measure until the maximum value in the curing curve, that is, the saturation value M _H , is obtained, so the measurement time can be shortened, and especially when dealing with ΔM, the measurement time can be significantly shortened. I can do it. In addition, in the present invention, since the peak torque difference between two adjacent peaks is determined, the zero drift error is canceled out between these adjacent peaks, and thus more accurate measurement data can be obtained. Since one point of data can be obtained per unit, the density of the measurement data is high, and from this point of view it is advantageous for performing highly accurate measurements in a short measurement time, as well as for smoothing and averaging of data during data analysis. It is also advantageous for processing and is effective for tracking fast curing reactions.

Moreover, according to the present invention, the viscoelastic changes of the test sample during the curing process are tracked by separating them into two properties, viscous and elastic, so it is more reliable and more complete than conventional methods. It becomes possible to perform quality control.

[Brief explanation of the drawing]

Figure 1 is an explanatory diagram of the basic concept of data extraction in the present invention, Figure 2 A and B are hardening curve diagrams,
Figure 3 is an explanatory diagram of the numerical representation of the hardening curve, and Figure 4 is the peak torque difference M and its difference ΔM according to the present invention.
FIG. 5 is a hardening curve diagram showing the time change of torque gradient, FIG. 6 is an explanatory diagram showing the relationship between torque waveform and torsion angle waveform, and FIG. 7 is peak torque difference M and loss. 8 is a block diagram illustrating an example of an apparatus for carrying out the present invention; FIG. 9 is a signal waveform diagram illustrating the relationship between torque waveforms, torsion angle waveforms and timing pulses; and 10. The figure is an explanatory diagram of the procedure for determining M _L and M _H. Figure 11 is an explanatory diagram of the procedure for determining t〓.
The figure is an explanatory diagram of the procedure for determining t _X , Figure 13 is an explanatory diagram of the procedure for determining M _Z , and Figure 14 is ΔM _nax
Figure 15 is an explanatory diagram of the procedure for determining M〓 _nax and t〓 _nax .
An explanatory diagram of the procedure for finding M〓, t〓〓, Figure 16 shows M _Y ,
t An explanatory diagram of the procedure for determining _Y , FIG. 17 is an explanatory diagram of the procedure for determining ΔM _W , FIG. 18 A and B is a flowchart showing an example of the control procedure in the embodiment of the present invention, and FIG. 19 is a reference value sequence FIG. 1... Torsional vibration measuring machine, 2... Strain gauge type load cell, 3... Gauge power supply, 4... Differential amplifier, 5, 6... Sample/hold circuit, 7, 8
... Sample/peak hold circuit, 9... Optical sensor, 10... Drive shaft, 11... Timing board,
12... Timing hole, 13, 14... Differential amplifier, 15, 16... A/D converter, 17...
Central processing unit, 18... Keyboard, 19... Read only memory, 20... Random access memory, 21... Output device such as a display, recorder, printer, etc.

Claims (1)

[Claims] 1. The characteristics of the curing curve of a vulcanized compound as a test sample obtained in a torsional vibration curing test are extracted using several representative characteristic values, In a method for evaluating a vulcanized compound, the test sample is compared with limit values such as an upper limit value and a lower limit value in a group of standard characteristic values representing the specifications, and the suitability of the test sample for the specifications is automatically determined based on the comparison result. Regarding the representative characteristic values to be used, the limit of pass/fail of the vulcanized compound and the conditions for terminating the data acquisition for the representative characteristic values to be used are specified in advance, and the half period of the torque waveform forming the curing curve is determined in advance. For each time, the absolute value M of the torque difference between two adjacent peak torques is captured, the data of the obtained peak torque difference M is stored, and the stored data sets the condition for ending the data capture. It is determined whether the conditions are satisfied, and when the conditions are satisfied, the maximum value M _H is determined as a representative characteristic value for the peak torque difference M.
and the minimum value M _L , the time t for (M - M _L ) to reach a predetermined ratio α to (M _H - _{M L} ), and the time t for the peak torque difference M to reach a predetermined value M _{X and} a predetermined At least one of the peak torque differences M _Z at time t _Z is determined from the stored data, and the determined representative characteristic value is compared with the pass/fail limit to determine pass/fail of the vulcanized compound. A method for determining characteristic vulcanized compounds. 2. Extract the characteristics of the curing curve of the vulcanized compound as the test sample obtained by the torsional vibration curing test using several representative characteristic values, and use the representative characteristic values as standard characteristics representing the specifications of the compound. About typical characteristic values to be used in the evaluation method for vulcanized compounds, which compares limit values such as the upper and lower limits in a value group and automatically determines whether the test sample meets the specifications based on the comparison results. , specifying in advance the conditions for terminating the data acquisition regarding the pass/fail limit of the vulcanized compound and the representative characteristic value used, and for each half cycle of the torque waveform forming the curing curve, adjacent 2 Calculate the absolute value M of the torque difference between two peak torques and the difference ΔM between the absolute value M of the two adjacent torque differences, store the data of the determined peak torque difference M and the difference ΔM, and store the stored data. It is determined whether or not the data obtained satisfies the condition for terminating the data acquisition, and when the condition is satisfied, the maximum value ΔM of ΔM is determined as a representative characteristic value for the peak torque difference M and the difference ΔM. _nax and the time t〓 _nax and the value of M at that time M〓 _nax , the time t when ΔM reaches a predetermined ratio β to ΔM _nax and the value of M at that time M〓, ΔM is a predetermined value At least one of the time point t _Y when tY reaches , the value of M at that time M _Y , and the value ΔM _W of ΔM at a predetermined time t _W is determined from the stored data, and the determined representative characteristic value is determined as the pass/fail value. 1. A method for evaluating a vulcanized compound, the method comprising: determining pass/fail of the vulcanized compound by comparing it with a limit of . 3 Extract the characteristics of the curing curve of the vulcanized compound as the test sample obtained by the torsional vibration curing test using several representative characteristic values, and use the representative characteristic values as standard characteristics representing the specifications of the compound. About typical characteristic values to be used in the evaluation method for vulcanized compounds, which compares limit values such as the upper and lower limits in a value group and automatically determines whether the test sample meets the specifications based on the comparison results. , specifying in advance the conditions for terminating the data acquisition regarding the pass/fail limit of the vulcanized compound and the representative characteristic value used, and for each half cycle of the torque waveform forming the curing curve, adjacent 2 Absolute value M of the torque difference between two peak torques, Absolute value M of the torque difference between two adjacent torques
The torque loss is the absolute value of the torque difference for each half period of the instantaneous value at which the torsional vibration deformation applied to the vulcanized compound crosses the midpoint of its amplitude.
M'', store the data of the calculated peak torque difference M, difference ΔM, and loss torque M'', and check whether the stored data satisfies the conditions for terminating the data acquisition. When the above conditions are satisfied, the values of M" at the time when M _L and M _H are obtained are determined as representative characteristic values for the peak torque difference M, the difference ΔM, and the loss torque M _". M″ _H , time tα, _tX ,
The value of M″ at t〓 _nax , t〓〓, t _Y , t _W and t _Z ,
Maximum value of M″ M″ _nax and the peak torque difference M
and the storage torque M′ at the time when the loss torque M″ is obtained = √ ² − (″) ² and the loss tangent tanδ
= M''/M' is determined from the stored data, and the determined representative characteristic value is compared with the pass/fail limit to determine pass/fail of the vulcanized compound. Method for determining sulfur compounds.