JPH0737933B2 - Waveform signal processing device for particle analyzer - Google Patents

Description

The present invention relates to a waveform signal having a valley in a particle detection signal of a particle analyzer for detecting the size of particles such as blood cells (hereinafter referred to as a valley waveform signal). The present invention relates to a trough waveform signal processing device of a particle analysis device for processing a.

(Prior Art) Among the particle detection signals, the conductive type has a pair of electrodes immersed in a suspension of measurement particles suspended in an electrolyte solution,
As shown in FIG. 32, both electrodes are isolated by an insulating wall 101 having minute pores 100, and a potential difference is applied between both electrodes to form pores 1.
It is configured such that a current flows between the electrodes only through 00. In this particle detection device, when a water pressure difference is applied between the liquids on both sides to allow the particles to flow through the pores 100 together with the suspension, the particles can be passed through by appropriately selecting the diameter of the pores 100 with respect to the particle diameter. A current change proportional to the particle volume is generated between the electrodes.

The waveform of the particle detection signal at this time is the pore 10 shown in FIG.
For example, the passage of the particles 102 at 0 is as shown in FIG. However, in the figure, the passage of particles is shown in an overlapping manner, and in practice, the particle concentration of the suspension is adjusted so that each state occurs separately.

In FIG. 33, the waveform a'is the case of the passage route a in which the particles 102 pass very close to the wall surface of the pore 100, and the pore 1
There are sharp peaks at the 00 entrance and exit, with a gentle valley between the two peaks. That is, the signal of this waveform a'is a valley waveform signal.

The corrugation b'is the case of the passage route b in which the two particles 102 closely pass near the center of the pore 100, and there is a deep valley between the two peaks.

Waveform C'is the case of the passage C in which the particles 102 pass exactly through the center of the pore 100, and shows a symmetrical and clean waveform with one peak. The signal of this waveform C'is hereinafter referred to as a single peak waveform signal.

Waveform D ′ is a case of a passage D that obliquely passes near the wall surface of the pore 100 at the entrance and near the center at the exit, and shows a sharp peak only near the entrance.

The reason why the particle 102 shows a sharp peak when the particles 102 pass near the wall surfaces of the inlet and outlet of the pore 100 is that the current density in this vicinity (so-called edge portion of the pore 100) is high. Further, the width of the corrugation is longer when the particles 102 pass near the wall surface of the pores 100 than when they pass through the center of the pores 100 because the flow velocity is slow near the hanging surface, This is because it takes longer for the particles 102 to pass than when they have passed.

From the above, even if the particles 102 having the same size pass through the pores 100, the peak value of the peak of the waveform will be different due to the difference in the passage route. Therefore, in a strict sense, a detection signal having a size proportional to the volume of the particle cannot be obtained, and the particle size distribution drawn from the peak value of this peak contains an error.

To solve the problem of the distortion of the detection waveform of particles, the following three types of measures have been typically taken.

That is, the first means is to form a so-called "sheath flow" by hydrodynamically narrowing the flow path that passes through the pores of the particles to a narrow range in the center. In the first means, the waveform of the particles that have passed through the pores is only the one that passes through the passage c in FIG. 32, so only the shape like the waveform c ′ is obtained, and it is strictly proportional to the volume of the particle. The obtained detection signal is obtained. Therefore, according to this first means, a detection waveform without any distortion can be obtained, which is ideal in principle, but there is a drawback that the apparatus becomes expensive because a complicated fluid mechanism is required.

The second means is that the particle detection signal is analyzed, and only the signal of the waveform c ', which is not distorted when passing through the vicinity of the center of the pore, is adopted, and the signals of other waveforms having distortion are ignored. Waveform selection is performed, and it is shown in Japanese Patent Publication Nos. 54-26919 and 55-12980.

However, in this second means, since a considerable number of particle detection signals are ignored by the waveform selection, the number of particles to be measured is substantially reduced and the accuracy is deteriorated. Further, in order to cover the decrease in the number of particles, it is necessary to measure a larger number of particles and there is a problem that the measurement time becomes long.

The third means is to use a signal such as a waveform c ′ passing through the passage c in FIG. 32, that is, a single peak waveform signal, without changing the peak value of the peak, and to obtain a waveform a ′ passing through the passage c. The signal, i.e., the valley waveform signal, employs the peak value of the valley portion, and is shown in, for example, Japanese Patent Laid-Open Nos. 58-83234, 60-257342, and 46-7315. This third means utilizes the property that when the length of the pore is longer than its diameter, the peak value of the center of the particle detection waveform is exactly proportional to the volume of the particle, regardless of the passage path of the particle. And about this property, Kachel, V
(1982) Sizing of Cells by the Elecrical Res
instance Pulse Tecnique, in: Cell Analysis Vol.1
(N. Catsimpoolas, ed.), Plenum Press, New York, pp.
195-331.

According to the third means, it is not necessary to form a so-called "sheath flow" as compared with the first means, and a complicated fluid mechanism is unnecessary, so that the structure of the device is simple and inexpensive, and Compared with the second means, it is not necessary to ignore the waveform by selecting the waveform, and the number of measured particles is not substantially reduced, so that the measurement accuracy is not deteriorated.

(Problems to be Solved by the Invention) The third means described in JP-A-58-83234 and JP-A-60-257342 has a problem in processing a valley waveform signal. That is, phase shift correction is performed to correctly detect the peak value of the valley portion of the valley waveform signal.
It is difficult to match the timing of this phase shift. Further, the phase shift correction easily causes phase shift distortion of the waveform. Although a sample and hold circuit is used, this circuit generally requires the use of expensive elements.

Conventionally, there is a peak hold circuit as shown in FIG. That is, this sample and hold circuit consists of operational amplifiers OP50, OP51, capacitors C50, diodes D50, D51, resistors.
It is composed of R50, switches SW1 and SW2.

Switches SW1 and SW2 are normally closed, in which case the output is zero volts.

When it is determined that the particles are particles, the switches SW1 and SW2 are opened to enter the peak hold mode. In this case, the switches SW1, SW
Opening and closing of 2 is performed by a signal from the discrimination circuit, but there is a time delay. Therefore, in order to match the timing with the input waveform, the waveform passed through the delay circuit is input. However, when passing through a delay circuit, the waveform tends to be distorted due to group delay.

To reset the peak hold circuit, switches SW1 and SW2 may be closed, but it takes some time to completely discharge the electric charge of capacitor C50. The larger the held peak value, the longer the switch time. SW1, SW2
Need to close. For this reason, when the next particle is determined during discharge, the peak of the particle detection signal may not be held. That is, when the peak of the next signal is already approaching after the reset period has elapsed, the peak cannot be faithfully held. On the contrary,
The high-speed operational amplifiers OP50 and OP51 can be held, but the operational amplifiers OP50 and OP51 are expensive.

On the other hand, the one described in Japanese Examined Patent Publication No. 46-7315 does not directly detect the valley position of the valley waveform signal and does not hold the peak value of the valley, so the timing for detecting the peak value of the valley is set. Is getting harder.

Therefore, an object of the present invention is that there is no need to adjust the timing for detecting the peak value of a valley, a phase shift circuit or the like is not necessary, there is no phase shift distortion, and the circuit element and configuration are simple and inexpensive. It is an object of the present invention to provide a valley waveform signal processing device of a particle analyzer capable of performing the above.

(Means for Solving Problems) A valley waveform signal processing device of a particle analyzer according to the present invention comprises a first peak hold means for holding a first peak value of a valley waveform signal as a first peak hold value, First difference signal generating means for generating the first difference signal by subtracting the valley waveform signal from the first peak hold value, and the first peak value of the first difference signal as the second peak hold And a second difference signal generating means for generating a second difference signal by subtracting the second peak hold value from the first peak hold value. It is a thing.

(Operation) According to the configuration of the present invention, the first peak hold value, which is the first peak of the valley waveform signal, and the first peak of the difference signal between the valley side of the valley waveform signal and the first peak hold value. Since the crest value of the valley of the valley waveform signal is obtained by the difference from the certain second peak hold value, a signal having a crest value that is exactly proportional to the volume of the particle can be obtained. Therefore, compared to the conventional example, there is no need to adjust the timing for detecting the peak value of the valley, and a phase shift circuit etc. is not required, so the circuit configuration is simple and phase shift distortion is caused. And there is no need for expensive elements in the circuit for holding the peak value.

(Embodiment) An embodiment of the present invention will be described with reference to FIGS. 1 to 31. That is, the trough waveform signal processing device of this particle analysis device includes a first peak hold means 1 for holding the first peak value of the trough waveform signal as a first peak hold value in FIG. A first difference signal generating means 2 for generating a first difference signal by subtracting the valley waveform signal from a peak hold value, and a first peak value of the first difference signal is held as a second peak hold value. And a second difference signal generation means 4 for generating a second difference signal by subtracting the second peak hold value from the first peak hold value. .

Therefore, as shown in FIG. 1, the first peak hold means 1 obtains the first peak hold value M2 which is the first peak of the valley waveform signal M1, and the first difference signal generation means 2
The first difference signal M4 which is the difference between the first peak hold value M2 and the valley waveform signal M1 is obtained by the second peak hold means 3 and the second peak which is the first peak of the first difference signal M4 is obtained by the second peak hold means 3.
Peak hold value M5 is obtained, and the second difference signal generating means 4 obtains a second difference signal M6 which is the difference between the first peak hold value M2 and the second peak hold value M5. The difference signal M6 of 2 becomes the peak value of the valley M3 of the valley waveform signal M1.

Therefore, according to this embodiment, the first peak hold value M2, which is the first peak of the valley waveform signal M1, and the first difference signal between the valley side of the valley waveform signal M1 and the first peak hold value M2. Since the crest value of the trough M3 of the trough waveform signal M1 is obtained by the difference from the second peak hold value M5 which is the peak, a crest value signal that is accurately proportional to the volume of the particle can be obtained. Therefore, compared to the conventional example, there is no need to adjust the timing for detecting the peak value of the valley, and a phase shift circuit etc. is not required, so the circuit configuration is simple and phase shift distortion is caused. And there is no need for expensive elements in the circuit for holding the peak value.

The entire particle analyzer to which the valley waveform signal processing device of this embodiment is applied will be described below. In FIG. 2, 11 is a container
A detector (insulating wall) constituting the detection means 1 in which 13 particles are immersed in a suspended liquid 12 has pores 14 near the lower end thereof. The pores 14 are located below the liquid surface of the liquid 12,
Electrodes 15 and 16 are arranged inside and outside thereof, respectively. A liquid control device 17 is for absorbing the liquid 12 through the pores 14 in the detector 11.

A detection circuit 18 generates a pulse-like signal proportional to the size of the particle each time the particle passes through the pore 12 based on the difference in electrical impedance between the particle and the liquid 12.

Reference numeral 9 denotes a waveform processing device that performs a waveform process on the particle detection signal detected by the detection means 1, and includes the valley waveform signal processing device of the present invention.

Reference numeral 10 denotes an arithmetic processing unit, which analyzes a signal sent from the waveform processing unit 9 and showing a crest value proportional to the volume of particles, and performs various arithmetic processes such as drawing a particle size distribution chart and counting the number of particles. Is.

Next, the waveform processing device 9 will be described. The waveform processing device 9 has a preprocessing block C as shown in FIG.
, A control block B, and a bottom output block A. The pre-processing block C has a DC reproducing function of clamping the DC level of the particle detection signal sent from the detection circuit 18 to a predetermined level and a function of differentiating the particle detection signal. Among the signals output from the preprocessing block C, the DC reproduction signal of the particle detection signal is sent to the bottom output block A and the control block B, and the differential signal is sent to the control block B. The control block B receives the DC reproduction signal and the differential signal and outputs a control signal for controlling the bottom output block A. The bottom output block A receives the DC reproduction signal from the pre-processing block C and the control signal from the control block B, and has a function of outputting the crest value of the valley when the DC reproduction signal is a valley waveform signal. .

First, the pre-processing block C will be described in detail. FIG. 4 is a circuit diagram showing an embodiment of the preprocessing block C. The pre-processing block C is mainly the first DC regeneration block.
K1, a first differential block DIF1, a second direct current regeneration block K2, a second differential block DIF2, and a particle detection signal (single peak waveform signal in the figure) of the detection means 1 via the operational amplifier OP1 of the voltage follower. Is entered. The pre-processing block C includes the reference signal generation block REF and generates the reference signals Ref1 and Ref2, but the reference signal generation block REF may be arranged in the control block B.

The first DC recovery block K1 sets the DC level of the particle detection signal to a predetermined level, and the operational amplifier OP2,
It is composed of OP3, resistors R1 and R2, capacitor C1, and diodes D1 and D2. The crest value of the direct current reproduction signal DC1 in which the particle detection signal is direct current reproduced in the first direct current reproduction block K1 is detected by the waveform processing device 9, and the arithmetic processing device 10 shown in FIG.
Parsed in. In addition, the DC reproduction signal DC1 is supplied to the operational amplifier OP4.
The inverted signal {circle around (1)} is compared with the differential signal V1 in the control block B, which will be described later, and is applied to detect particles that have passed through the pores 14. R3 and R4 are resistors.

The first differentiation block DIF1 is for differentiating the particle detection signal, and includes operational amplifiers OP5, OP6, resistors R5 to R8, a capacitor C2,
Composed of C3.

The differential signal V1 differentiated by the first differential block DIF1 is
By comparing with the first reference signal Ref1 in the control block B which will be described later, the return signal and noise with a low peak value are removed, and the peak and valley positions of the particle detection signal are detected by comparing with the zero potential. can do.
That is, the return signal is obtained when the particles 20 and 21 have returned to the vicinity of the pores 14 as shown by the broken line in FIG. In the case of the particle 21, the return signal having the waveform as shown in FIG. 6A and in the case of the particle 20 the signal having the waveform as shown in FIG. 6B is generated in the detection means 1. . Since the magnitude of such a return signal is much smaller than the magnitude of the true particle detection signal generated when the particles pass through the pores 14, it is necessary to measure only particles of similar size. Does not affect the particle measurement because the reflected signal can be discriminated by the magnitude of the signal. However, when the large particles and the small particles are mixed in the suspension, the return signal of the large particles becomes almost equal to the detection signal of the small particles. Therefore, the detection signal of a small particle and the return signal of a large particle cannot be discriminated from each other depending on the size of the signal, which adversely affects the measurement of a small particle.

That is, when the return signals of FIGS. 6 (a) and 6 (b) are differentiated, a waveform as shown in FIG. 6 (c) is obtained with respect to FIG. 6 (a), and a waveform as shown in FIG. The waveform is as shown in d). Then, by comparing with the reference signal Ref1, the return signal having a low peak value in FIG. 9A can be removed. However, it is not possible to remove the sowing return signal of FIG. 6B, which has a high peak value.

Further, since the first differential block DIF1 is provided with the integral characteristic in the high frequency region, in the case of a signal including noise as shown in FIG. 7 (a), the differential signal V1 becomes the signal shown in FIG. 7 (b). , By comparing with the reference signal Ref1, noise can be removed.

Further, as shown in FIG. 8 (a), points P1 to P3, B1 at which the differential signal V1 in FIG. 8 (b) obtained by differentiating the particle detection signals of the single peak waveform signal and the valley waveform signal have zero potential are Since it corresponds to the peak P or the valley B of the original signal, it is possible to detect the position of the peak and the valley of the particle detection signal by comparing the differential waveform with the zero potential and detecting the position where the differential waveform crosses zero. You can

The second direct current regeneration block K2 outputs the direct current regeneration signal DC2 for direct current regeneration of the particle detection signal and removal of the return signal having a high peak value, and the operational amplifiers OP8, OP9 and the resistor R
It consists of 8, R9, capacitor C4, and diodes D3, D4. First DC regeneration block K1 and second DC regeneration block K2
Generally have different cutoff frequencies, and the circuit constants are set so as to have the optimum characteristics according to the purpose.

The second differential block DIF2 is for differentiating the particle detection signal and outputting a differential signal V2 for comparison with the direct current reproduction signal DC2, and is composed of operational amplifiers OP11, OP11, resistors R10 to R13, capacitors C5, C6. ing. The first differential block DIF1 and the second differential block DIF2 also generally have different differential characteristics.

By comparing the DC reproduction signal DC2 and the differential signal V2, the return signal which could not be removed in the processing of the differential signal V1 in FIG. 6 because the peak value is high can be removed.
That is, as shown in FIG. 9 (a), in the case of a normal particle detection signal generated by the passage of particles through the pores 14, the differential signal V2 is higher than the DC reproduction signal DC2 (V2> DC2), so the differential signal V2 crosses the DC playback signal DC2, but in the case of a return signal, the differential signal V2 becomes lower than the DC playback signal DC2 (V2 <DC2) as shown in Fig. 7B, so the differential signal V2 is DC. Never cross the playback signal DC2. In this way, it becomes possible to discriminate between the particle detection signal and the return signal.

When the measurement particles are blood cells, and when measuring platelets in the blood cells, this return signal removal function is essential. This is because blood contains red blood cells and white blood cells in addition to platelets unless special treatment is applied. That is, these red blood cells and white blood cells are much smaller than platelets, and the peak value of the return signal generated when red blood cells and white blood cells fly back is the wave of the particle detection signal generated when platelets pass through the pores. This is because it is almost equal to the high price. Therefore, the second DC regeneration block DC2 and the second differential block DIF2 are indispensable in the platelet measurement, but they are not necessary when measuring red blood cells and white blood cells.

The reference signal generation block REF is an operational amplifier OP7, R17 to R21,
It is composed of a capacitor C8 and a Zener diode ZD.

Next, the control block B will be described in detail. Figure 3 is the third
FIG. 11 is a block configuration diagram in which the control block B in the figure is further divided into smaller blocks, and FIG. 11 is a circuit diagram thereof. Of the signals input to the control block B, the signals V1, V2, DC1, ▲ ▼, DC2, Ref1, Ref2 excluding the count start signal are all sent from the pre-processing block C shown in FIG. is there. In each block shown in Fig. 11, COM1
COM4 is a comparator, DQ1 to DQ5 is a D latch, I1 to I13 are inverters, NAND is a NAND circuit, NO is a normally open analog switch, and V1 and V2 are multivibrators.

In this example, a case where large particles and small particles are mixed in the suspension, for example, red blood cells, white blood cells and platelets are mixed, and only small platelets are measured from the mixed particles will be described.

Peak / valley detection block L1 shown in FIGS. 10 and 11.
Is the differential signal V1 and the reference signal Ref1 by the comparator COM1.
In comparison with FIG. 6 and FIG. 7, the return signal and noise having a low peak value are removed, and the differential signal V1 is compared with the zero potential to compare the peak and valley of the DC reproduction signal DC1. The signal a for detecting the position of is output. That is, when the differentiated signal V1 is smaller than the reference signal Ref1, the signal a is at a low level (hereinafter "L") as shown in FIG. 12 (b).
Is written. ), On the contrary, when it becomes large, it becomes a high level (hereinafter referred to as “H”), and the signal a does not respond to a return signal or noise having a low peak value. Differential signal V1 is reference signal Ref1
When it is larger than that, if the b input from the input inhibition block L6 to the center pulse generation block L5 is H, the Q output c of the D latch DQ1 becomes H as shown in FIG. 12 (d). At this time, at the same time, the analog output NO of the peak / valley detection block L1 is closed by the Q output c of the D latch DQ1 of the center pulse generation block L5, and the reference signal Ref1 falls to zero potential as shown in 12th (a). Therefore, this time, the differential signal V1 and the zero potential are compared by the comparator COM1, so that the signal a becomes a signal for detecting the positions of the peak P and the valley B as described in FIG.

The passage detection block L2 has the direct current reproduction signal DC1 as described above.
The passage of particles is detected by comparing the inversion signal (1) obtained by inverting and the differential signal V1 in the comparator COM2. That is, when the differentiated signal V1 becomes smaller than the inverted signal ∇, the output d becomes H and the T input of the D latch DQ1 of the center pulse generation block L5 becomes L, so the Q output c becomes L. As shown in FIG. 12, when the Q output c becomes L, the peak of the particle detection signal has already passed, so the Q output c and the output d are signals corresponding to the passage of one particle. Since the output becomes H, the analog switch NO is opened and the reference signal Ref1 is input to the comparator COM of the peak / valley detection block L1. (Fig. 12 (a)).

FIG. 12 shows the case of a valley waveform signal, FIG. 13 shows the case of a single peak waveform signal, and FIG. 14 shows the case of two consecutive single peak waveform signals.

Thus, since the passage detection block L2 is a method of detecting passage of particles by comparing the differential signal V1 and the inverted signal ▲ ▼, in the case of the valley waveform of FIG. 12 and the single peak waveform of FIG. 13, There is only one pulse in the signals c and d, and it is judged that one particle has passed, and in the case of the waveform when two particles pass closely as shown in Fig. 14, the signals c and d have It can be judged that two pulses occur and two particles pass. Also, since both the differential signal V1 and the inverted signal V change depending on the size of the particle detection signal, the time position of passage detection of the particle detection signal is substantially constant regardless of the size of the particle.

In addition, the resistor R22 (3.3K
Ω) is connected, but this 6th pin is called a strobe terminal, and the output becomes H regardless of the input while drawing a current of 3 to 5 mA. As a result, the movement of the comparator is stopped and the generation of noise is prevented when it is not necessary.

As described above, the feedback loop block L3 compares the differential signal V2 of the particle detection signal and the DC reproduction signal DC2 with the comparator COM.
3, the output e is output which does not respond to the return signal having a high peak value but responds to the reproduction signal DC2 of the detection signal of particles such as platelets. That is, as shown in FIG. 15, when a particle such as platelet N1 or red blood cell N2 passes, the DC reproduction signal DC2 of the particle detection signal is crossed with the differential signal V2, so that the output e (FIG. 15) is generated. (C))
Becomes H. In the case of the return signal N, the differential signal V2 does not cross the DC reproduction signal DC2, so the comparator COM3 is not inverted and the output e of the return-rejection block remains L.

The large signal elimination block L4 includes a DC reproduction signal DC1 and a reference signal R
Comparing ef2 with the comparator COM4, it is determined that the signal exceeding the reference signal Ref2, which is the large signal discrimination level, is a signal due to large particles such as red blood cells and white blood cells.
When the DC reproduction signal DC1 exceeds the reference signal Ref2, the output f of the comparator COM4 becomes L as shown in FIG. 15 (d).

The peak hold control block L7 outputs a control signal ▲ ▼ / T that enables peak hold. Q showing the passage of one particle in the center pulse generation block L5
While the output c is H, when the comparator COM3 of the loopback block L3 is inverted and the T input e of the D latch DQ2 rises, the D input c of the D latch DQ2 appears at the Q output g and the Q output g becomes H. As a result (FIG. 15 (e)), the peak output described later can be performed while the Q output g is H. D latch DQ
2 is reset by the Q output of the reset pulse generation block L11, but is also reset when the output f of the large signal removal block L4 detects a large signal and becomes L (first
Figure 15 (e)). Therefore, in the case of a large signal, the g signal becomes H once but is not held until the particle passage detection time,
Therefore, the peak value of the large signal is not output as will be described later. The output g of the D latch DQ2 of the peak hold control block L7 outputs the control signal ▲ ▼ / T via the inverter I9 (Fig. 16 (o)), and is sent to the bottom output block A of Fig. 3 to It controls the normally closed analog switches NC1 and NC2 shown in Fig. 19.

Incidentally, the D latch DQ2 of the peak hold control block L7
Needs to set the timing as follows. That is, when the D input that receives the Q output c of the latch DQ1 is H and the T input that receives the output e of the block L3 rises, the Q output g becomes H, but after the output e rises first, Even when the Q output c becomes H, g remains L, so that the differential signal V1, the reference signal Ref1, and the differential signal V2 are set so that the Q output c rises before the output e.
It is necessary to set the frequency characteristics and signal size of the DC reproduction signal DC2.

The peak control block L9 outputs a control signal PC that holds the first peak of the DC reproduction signal DC1.
Peak hold control block L7 D latch DQ3 that receives the Q output g of peak hold control block L when the D input is H
Receiving the inverted signal of the signal a which detected the position of the peak / valley of 1 When the T input rises, the output is L and the control signal PC
Becomes H. (Fig. 16 (h) to (j)). The control signal PC controls the normally open type analog switch NO1 shown in FIG. 19 of the bottom output block A, and holds the first peak when the control signal PC is H.

The valley control block L8 holds the valley of the DC reproduction signal DC1 and outputs control signals BC1, ▲ ▼ for distinguishing the valley waveform signal from the single peak waveform signal, and the Q output of the peak hold control block L7. When the D input of the D latch DQ4 receiving g is H and the T input receiving the signal a'inverted twice from the signal a rises, the Q output becomes H.
Then, the output becomes L. Control signal by H of Q output ▲
▼ becomes L, and the control signal BC1 becomes H due to the output L
become. (Fig. 16 (h), (k) to (m)). The falling edge of the control signal {circle around (5)} is delayed a little from the rising edge of the control signal BC1 by the delay circuit composed of the resistor R23 and the capacitor C9 to prevent floating as described later. Both the control signals BC1 and ▲ ▼ are sent to the bottom output block A, the control signal BC1 controls the normally open type analog switches NO2 and NO3, and the control signal ▲ ▼ controls the normally closed type analog switch NC3.

The output control block L10 outputs the control signal ▲ ▼ for controlling the output signal of the bottom control block A, and when the B input of the monostable multivibrator V1 receiving the Q output g of the peak hold control block L7 is H. Center pulse generation block that detects the passage of a single particle
When the A input that receives the Q output c of L5 falls, a pulse having a time width of ts is output from the Q output, and the control signal ▲ ▼ in FIG. 16 (n) is output. This output ▲
▼ is also sent to the bottom output block A, and controls the normally closed analog switch NC4 shown in FIG.
During the ts time of the signal, the output signal indicating the peak value is output in pulses from the bottom output block A.

In the input inhibition block L6, the output b is set to L because the D latch DQ5 is set by the output of the output control block L10.
Therefore, the a input to the center pulse generation block L5 is prohibited, and the peak value output is prevented from being erroneously operated by the near passage pulse when the particles continuously pass through the pores. If the output b of the D latch DQ5 of the input prohibition block L6 is L, this circuit will not operate, so it must be set to H before the counting is started. In this case, the D input of the D latch DQ5 is input. Pull down
If the T input is connected to a counting start signal (rising signal) from the outside, the signal b always becomes H at the start of counting, and the state where no inhibition is applied can be achieved.

The reset pulse generation block L11 is the output control block L
When the output time ts has elapsed due to the Q output of 10, a reset pulse is output from the Q output and output of the monostable multivibrator V2, and the peak hold control block L7
D latch DQ2, peak control block L9 D latch DQ3,
It is sent to the D latch DQ4 of the valley control block L8 and the D latch DQ5 of the input inhibition block L6 to reset them.

As another embodiment of the control block B, the case of measuring red blood cells and white blood cells is shown in FIG. That is, control block B
11 is a block obtained by removing the block L3 without return from FIG. When measuring red blood cells and white blood cells, it is not necessary to remove the return signal.
L3 is no longer needed.

Next, the bottom output block A in FIG. 3 will be described in detail. FIG. 18 is a configuration diagram in which the bottom output block A is further divided into smaller blocks, and FIG. 19 is a circuit diagram thereof.
Of the signals input to the bottom output block A, the DC reproduction signal DC1 to be subjected to waveform processing is sent from the preprocessing block C in FIG. 3, and the other PC, ▲
The control signals represented by ▼ / T, BC1, ▲ ▼, ▲ ▼ are sent from the control block B.

The first peak hold block KL1 is operational amplifier OP13, OP
14, resistors R24 to R26, capacitors C10 to C12, diodes D5, D
6. It consists of a normally open analog switch NO1 and a normally closed analog switch NC1. The DC reproduction signal DC1 is input to the operational amplifier OP13 via the operational amplifier OP21 of the voltage follower,
The first peak hold value P / H is output from the operational amplifier OP14. In this case, if the normally closed analog switch NC1 is closed by the control signal ▲ ▼ / T, the peak hold value is not obtained and the output follows the input (tracking mode). When the normally closed analog switch NC1 is opened by the control signal ▲ ▼ / T, the output follows the input until the peak comes, but when the peak comes, the peak value is held (peak hold mode). FIG. 20 is a diagram showing, as an example, how the first peak hold block KL1 follows the near pass pulse SP when particles continuously pass through the pores. The normally closed analog switch NC1 is opened / closed by the control signal ▲ ▼ / T, the peak hold mode is entered when the control signal ▲ ▼ / T is L, and the tracking mode is entered when the control signal ▲ ▼ / T is H. In Figure 20, the first peak P /
After holding H, move to tracking mode to quickly follow the input waveform and move to the next peak P / H
You can see that you are holding faithfully.

On the other hand, the normally open type analog switch NO1 holds the first peak of the valley waveform signal and becomes H at the same time as the control signal.
It is closed by the PC (Fig. 16 (j)). For this reason,
As shown in Fig. 21, the positive input of the direct current reproduction signal DC1 input to the operational amplifier OP13 is forcibly dropped to zero potential, so that the first peak value is higher even if the second peak value of the valley waveform is higher. To be held. As a result, the first peak hold block KL1 outputs the first peak hold value P / H shown in FIG. This corresponds to the symbol M2 in FIG.

In this way, in the case of peak hold in this embodiment, unlike the conventional example, only one normally closed analog switch NC
It is controlled only by 1, and if the normally closed analog switch NC1 is opened, the peak hold mode can be entered smoothly, so no delay circuit is required. Further, it is not necessary to completely discharge the hold charge of the capacitor C12, and if the normally closed analog switch NC1 is closed, the current waveform is immediately followed.

The first difference signal generation block KL2 is an operational amplifier OP15, a resistor
It is composed of R27 to R30. The direct current reproduction signal DC1 is subtracted from the first peak hold value P / H, and the first difference signal B1 is output as shown by the broken line in FIG. This is the code in Figure 1
Corresponds to M4. The first difference signal B1 is zero when the first peak hold block KL1 is performing the tracking operation.

The second peak hold block KL3 is operational amplifier OP16, OP
17, resistors R31 to R33, capacitors C13 to C15, diodes D7, D
It consists of a normally open analog switch NO2 and a normally closed analog switch NC2. The basic operation of this second peak hold block KL3 is the first peak hold block K
It is the same as L1, and after the first peak of the first difference signal B1 is held, the positive input of the operational amplifier OP16 is forcibly dropped to zero potential by the control signal BC1 which becomes H at the position of the valley of the valley waveform signal. Therefore, the second peak hold block KL3 outputs the second peak hold value B / H shown in FIG. This corresponds to the code M5 in FIG. This second
The height h ₂ of the peak hold value B / H is equal to the depth h ₁ of the valley of the DC reproduced waveform D1.

The decision block KL6 is composed of an operational amplifier OP18, a normally-open analog switch NO3, a normally-closed analog switch NC3, and a resistor R34, and the normally-open analog switch NO3 is controlled by a control signal BC1 for detecting a valley. The closed analog switch NC3 is also controlled by the control signal ▲ ▼, so that in the case of the valley waveform signal as shown in FIG. 24 (a), the second peak hold value B / H is the second difference signal generation block KL4. However, in the case of a single peak waveform signal as shown in FIG. 7B, the positive input of the operational amplifier OP18 becomes zero potential.

The normally open analog switch NO3 and the normally closed analog switch NC3 are opened and closed at the timing shown in FIG. 25 by delaying the control signal ▲ ▼ from the control signal BC1 as described above, and the operational amplifier OP18 has a positive voltage. The second difference signal generation block KL
The output of 4 is temporarily prevented from becoming unstable.

The second difference signal generation block KL4 includes an operational amplifier OP19 and a resistor.
It is composed of R35 to R38, and in the case of a valley waveform signal, subtracts the second peak hold value B / H from the first peak hold value P / H,
The second difference signal B2 is output as shown in FIG.
This corresponds to the code M6 in FIG. Further, in the case of the single peak waveform signal of FIG.
The peak hold value P / H is output.

The output block KL5 is composed of an operational amplifier OP20, a resistor R39, and a normally closed type analog switch NC4, and outputs a second difference signal B2 having a desired peak value in a pulse waveform for a certain period of time in response to a signal.

26 to 30 show signal waveforms at various parts of the bottom output block A. The waveform of the dotted line in the figure shows the DC reproduction signal DC1.

FIG. 26 shows the case where the DC reproduction signal CD1 is a valley waveform signal. In this case, since the control signals BC1, ▲ ▼ are generated, the second peak hold value B / H is input by the determination block KL6, and the first peak hold value P / H and the second peak hold value P / H are input by the second difference signal generation block KL4. The peak hold value B / H is subtracted, and the peak value of the valley is output as described above.

Note that the waveform d ′ of the particle passage route d shown in FIGS. 32 and 33 is also processed in the same manner as the valley waveform signal as long as there is a flat portion in the portion corresponding to the valley of the valley waveform, Is output as the peak value.

FIG. 27 shows the case where the DC reproduction signal DC1 is a single peak waveform signal. In this case, the control signals BC1, ▲ ▼ are not generated, and the second difference signal generation block KL4 outputs the first peak hold value P / H, that is, the peak value of the first peak of the DC reproduction signal DC1.

FIG. 28 shows a case where the direct current reproduction signal DC1 does not have a particle detection signal and has a waveform of only noise. In this case, since the noise is discriminated in the control block B, ▲
▼ / T signal is not generated and the peak is not held. Further, since the ▲ ▼ signal is not generated, the peak value is not output.

FIG. 29 shows the case where it is determined that the DC reproduced signal DC1 exceeds the reference signal Ref2 in the large signal, that is, the control block B. For example, this is the case when a large signal due to red blood cells or white blood cells is input in the platelet measurement. In this case, the control signal ▲ ▼ / T once falls to L and the first peak hold block KL1 enters the peak hold mode, but the DC reproduction signal DC1 exceeds the reference signal Ref2 before reaching its peak, so the control signal ▲ Since ▼ / T immediately returns to H and enters the tracking mode, the peak of the DC reproduction signal DC1 is not held. In addition, the crest value is not output because the ▲ ▼ signal is not generated. In this way, unnecessary red blood cell and white blood cell signals are removed, for example, when measuring platelets.

FIG. 30 shows the case where the DC reproduction signal DC1 is a return signal, but as described above, the peak is not held and the peak value is not output in this case as well.

FIG. 31 shows an example in which the particle size distribution is measured by the waveform processing device 9 of this embodiment. In FIG. 31, the horizontal axis represents the peak value and the vertical axis represents the frequency. The particles for measurement are latex particles having a normal distribution type particle size distribution. The solid line curve in the figure is the particle size distribution curve obtained by performing the waveform processing described in the present embodiment, and the broken line curve is the particle size distribution curve obtained only from the peak value of the signal peak as in the prior art. It can be seen that when the waveform processing of this example is performed, the particle size distribution curve is significantly improved and exhibits a normal distribution type.

According to this embodiment, the crest value of the valley is output when the original waveform has a valley, and the peak value is output when the original waveform has a single peak. A signal with a peak value proportional to is obtained. When outputting the crest value of the valley, the first peak value and the depth of the valley of the original waveform are once held as the first peak hold value and the second peak hold value, respectively, and then the first peak hold value. The second peak hold value is subtracted from the value,
This is the peak value of the valley. Therefore, unlike the prior art, it is not necessary to perform the timing adjustment for detecting the crest value of the valley, and the phase circuit and the like are not required. Therefore, the circuit configuration is simplified and the problem of causing phase distortion is eliminated. Also, it does not require expensive sample and hold elements or circuits.

(Effect of the Invention) According to the valley waveform signal processing device of the particle analyzer of the present invention, the first peak hold value that is the first peak of the valley waveform signal, the valley side of the valley waveform signal, and the first peak hold. The peak value of the valley of the valley waveform signal is obtained by the difference with the second peak hold value, which is the first peak of the signal. Therefore, a signal with a peak value that is exactly proportional to the volume of the particle is generated. can get. Therefore, compared to the conventional example, there is no need to adjust the timing for detecting the peak value of the valley, and a phase shift circuit etc. is not required, so the circuit configuration is simple and phase shift distortion is caused. In addition, there is an effect that an expensive element is not required in the circuit for holding the peak value.

[Brief description of drawings]

1 is a schematic explanatory view of the present invention, FIG. 2 is an explanatory view of a particle analyzer to which an embodiment of the present invention is applied, FIG. 3 is a block diagram of a waveform processing device, and FIG. 4 is a preprocessing block. FIG. 5 is a circuit diagram, and FIG. 5 is an explanatory diagram for explaining the cause of the generation of the return signal.
6 is a waveform diagram of the return signal and its differential signal, FIG.
Fig. 8 is a waveform diagram of the particle detection signal containing noise and its differential signal. Fig. 8 is a waveform diagram of the single peak waveform signal and valley waveform signal and their differential signals. Fig. 9 is a large signal of large particles Waveform diagram of return signal and differential signal,
Fig. 10 is a block diagram of the control block, Fig. 11 is a circuit diagram thereof, Fig. 12 is a waveform diagram of a partial block in the case of a valley waveform signal, and Fig. 13 is a waveform diagram of a partial block in the case of a single peak waveform signal. Waveform diagram, FIG. 14 is a waveform diagram of a partial block when a single peak waveform signal is continuous, FIG. 15 is a waveform diagram of a partial block when the platelets, erythrocytes and particles return, and 16
FIG. 17 is a waveform diagram of each part in the case of a valley waveform signal, FIG. 17 is a block diagram of another embodiment in which the return-removing block of the control block is removed, and FIG. 18 is a block diagram of the bottom output block,
FIG. 19 is a circuit diagram thereof, FIG. 20 is an explanatory diagram showing a peak hold state, FIG. 21 is an explanatory diagram explaining a state in which the first peak is held, and FIG. 22 is a state in which the first difference signal is detected. FIG. 23 is an explanatory view illustrating a state in which the peak of the first difference signal is held, FIG. 24 is an explanatory view illustrating an output waveform of the second difference signal generation block, and FIG. 25. Is an explanatory view explaining the operation of the judging means, FIG. 26 is a waveform diagram of each part of the valley waveform signal, FIG. 27 is a waveform diagram of each part of the single peak waveform signal, and FIG. 28 is a waveform of each part in the case of noise. Fig. 29 is a waveform diagram of each part in case of particle detection signal of large particles,
FIG. 30 is a waveform diagram of each part in the case of a return signal, FIG. 31 is a characteristic diagram of frequency with respect to the peak value of the signal output from the waveform processing device, and FIG. 32 illustrates a process in which particles pass through pores. FIG. 33 is a waveform diagram of a particle detection signal corresponding to each particle, and FIG. 34 is a peak hold circuit diagram of a conventional example. DESCRIPTION OF SYMBOLS 1 ... 1st peak hold means, 2 ... 1st difference signal generation means, 3 ... 2nd peak hold means, 4 ... 2nd difference signal generation means

Claims (1)

[Claims] 1. A first peak hold means for holding a first peak value of a valley waveform signal as a first peak hold value, and a first peak hold value subtracting the valley waveform signal from the first peak hold value to obtain a first peak hold value. First difference signal generating means for generating a difference signal, second peak holding means for holding the first peak value of the first difference signal as a second peak hold value, and the first peak hold value And a second difference signal generating means for generating a second difference signal by subtracting the second peak hold value from the above.