KR101582731B1 - Sense Amplifying Circuit of Resistance Changing Memory - Google Patents

Abstract

A sense amplifier circuit for sensing a resistance change of a resistive memory is disclosed. The sense amplifier circuit has a cell bias boosting part and a reference bias boosting part, and enlarges the operation margin of the common source amplifier through the diode-connected transistor. The common source amplifier can perform stable operation with increased operating margin. In addition, the small signal level is amplified through the two-stage amplification, and an effective sense amplification operation is performed.

Description

[0001] The present invention relates to a sense amplifier circuit for a resistive memory,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sense amplifier circuit of a memory, and more particularly, to a sense amplifier circuit of a resistive memory that detects a change in resistance of a cell and performs a read operation.

A resistive memory refers to a memory in which data write and read operations can be performed using a resistance change. That is, the data storage operation is performed through the change of the resistance state, and the change of the resistance can be converted into the change of the current or voltage so that the read operation can be performed.

Representative resistive memories include MRAM, ReRAM, and PRAM. The ReRAM is a device that changes the resistance state of the resistance-variable element to analyze whether the conductive filament is formed or not. Further, the PRAM is a device in which the resistance state changes according to the entry of the resistance change element into the crystalline state and the entry into the amorphous state.

In either case, the storing operation of the data is performed by changing the resistance state of the resistive memory. In addition, the read operation for the stored data supplies the current to the cell to be read and uses the voltage drop phenomenon caused by the supply of the current. In addition, a constant voltage is applied, and a resistance state is detected by detecting a difference in voltage distributed according to a change in resistance state.

1 is a circuit diagram of a sense amplifier circuit of a memory according to the prior art.

1 is a representative drawing of Korean Patent Application No. 2010-0134350 filed by the present applicant.

는 1.035V이고, PMOS의 문턱전압의 절대치

는 450mV이다. 따라서, 트랜지스터 Q1의 오버드라이버 전압은

이므로 0.585V가 된다. 따라서, 트랜지스터 Q1이 활성영역에서 동작하기 위해서는 오버드라이버 전압보다 큰 소스-드레인 간의 전압 Vsd가 필요하다. 이는 Q1의 드레인 단자의 전압이 0.615V 미만이 되어야함을 의미한다.Referring to FIG. 1, in the 180-nm CMOS process, the level of the power supply voltage VDD is set to 1.2V. Rcell represents the resistance of the MTJ, and the resistance value in the high resistance state is about 32K. The current supplied to read the resistance state of the MTJ, Ibias, remains low. When the Ibias having a high value is supplied, the amount of change in voltage due to the resistance increases during the read operation, so that it is impossible to detect the constant resistance state. For example, Ibias is set to 5 uA. In addition, the transistor MN1 turned on by the word line WL operates in a linear region. Transistor MN1 acts as a simple switch and has a lower resistance than the resistance of the MTJ. In the 180nm process, the drain-source resistance Rds of MN1 is 1K. Therefore, the voltage applied to the gate terminal of the transistor Q1 is 165 mV. In the above process, the absolute value of the gate-source voltage of the transistor Q1

Is 1.035 V, and the absolute value of the threshold voltage of the PMOS

Is 450 mV. Therefore, the overdrive voltage of the transistor Q1 is

It becomes 0.585V. Therefore, in order for the transistor Q1 to operate in the active region, a voltage Vsd between the source and the drain that is larger than the overdrive voltage is required. This means that the voltage at the drain terminal of Q1 should be less than 0.615V.

However, since the overdrive voltage of the transistor Q1 has a very high value, the currents of the transistors Q1 and Q3 operating in the active region have relatively high values. Therefore, the voltage at the drain terminals of the transistors Q1 and Q3 must also have a high value. That is, the voltage at the drain terminal of the transistor Q1 must have a low value to operate in the active region, and the voltage at the drain terminal of the transistor Q1 must have a high value due to the current generated in the active region. This causes transistor Q1 to fail to operate in the active region and to operate in the linear region. When transistor Q1 operates in a linear region, it acts as a simple switch and does not act as an amplifier. Therefore, the voltage variation in the Rcell is not amplified in the operation of the device, and the voltage difference applied to the sense amplifier is not sufficiently amplified.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a sense amplification circuit for resistive memory that is widened in the active region.

According to an aspect of the present invention, there is provided a semiconductor device comprising: a first cell voltage amplifier for amplifying a variation in cell resistance as a common source; A second cell voltage amplifier for receiving the amplified signal of the first cell voltage amplifier and amplifying the amplified signal from the common source; And a sense amplifier for receiving and amplifying the amplified signal of the first cell voltage amplifier and the amplified signal of the second cell voltage amplifier.

According to the present invention described above, the operation margin of the transistors performing the common source amplifying operation is enlarged. The conditions required to operate in the active region are enlarged and a stable small signal amplification operation can be performed. In addition, as the overdrive voltage decreases, the power consumption is reduced and the change in cell resistance is effectively amplified by the sense amplifier because the input to the sense amplifier through the two-stage amplification is used.

1 is a circuit diagram of a sense amplifier circuit of a memory according to the prior art.
2 is a circuit diagram showing a sense amplifier circuit of a resistive memory according to a preferred embodiment of the present invention.

The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It should be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

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

Example

2 is a circuit diagram showing a sense amplifier circuit of a resistive memory according to a preferred embodiment of the present invention.

Referring to FIG. 2, the sense amplifier circuit includes a first cell voltage amplification unit 100 and a second cell voltage amplification unit 200.

The first cell voltage amplification unit 100 has a cell bias up unit 110 and a first common source amplification unit 120. The first cell voltage amplifying unit 100 generates a voltage by the bias current Ibias applied from the first current source 310 and amplifies the voltage by a small signal level. The amplification of the small signal level is in accordance with the common source amplifier configuration.

The first current source 310 forms a bias current Ibias corresponding to the applied bias voltage Vbias, and the formed bias current Ibias flows through the cell resistance Rcell.

Further, the cell bias boosting unit 110 is connected between the cell resistance Rcell and the ground. The cell bias boosting unit 110 is turned on by the word line voltage Vwd and boosts the voltage applied to the cell resistance Rcell. The cell voltage is formed at the first node N1 by the bias current Ibias flowing through the cell resistor Rcell. The formed cell voltage is applied to the first common source amplification unit 120. [ The first common source amplifier 120 operates in the saturation region and amplifies the variation of the cell voltage. Also, the first common source amplifier 120 operates in the active region by the DC component of the cell voltage.

The second cell voltage amplifying unit 200 has a reference bias boosting unit 210 and a second common source amplifying unit 220. The second current source 320 forms a bias current Ibias corresponding to the applied bias voltage Vbias, and the formed bias current Ibias flows through the reference resistance Rref. Also, the reference bias boosting section 210 is connected between the reference resistor Rref and ground. The reference bias boosting section 210 is turned on by the word line voltage Vwd to boost the voltage applied to the reference resistor Rref. A reference voltage is formed at the third node N3 by the bias current Ibias flowing through the reference resistor Rref. The formed reference voltage is applied to the second common source amplifying part 220. The second common source amplifying unit 220 operates in the active region and amplifies the amplified signal from the first common source amplifying unit 120 and inputs the amplified signal to the sense amplifier 330.

The sense amplifier 330 receives the output signal of the first cell voltage amplification unit 100 and the output signal of the second cell voltage amplification unit 200 and amplifies the output signal.

The first cell voltage amplifying part 100 has a first switching transistor Nsw1 and a first booting transistor Nbot1 as a cell bias boosting part 110 and a first common source amplifying part 120 as a first transistor M1 and a 2 transistor M2.

The transistor Pbias1 forming the first current source 310 is connected between the positive power supply voltage VDD and the first node N1 and receives the bias voltage Vbias. The bias current Ibias is determined according to the bias voltage Vbias.

The cell resistance Rcell is connected between the first node N1 and the cell bias boosting unit 110. [ The cell resistance Rcell has a characteristic of a variable resistance in which the resistance changes depending on the state of the material constituting the cell resistance Rcell. Further, the bias current Ibias flows through the cell resistance Rcell in order to sense the resistance change of the cell resistance Rcell. The cell voltage, which is the voltage of the first node N1, is determined by the cell resistance Rcell and the bias current Ibias flowing through the cell bias boosting unit 110. [

The cell bias boosting unit 110 is connected between the cell resistance Rcell and the ground. The cell bias boosting unit 110 has a first switching transistor Nsw1 and a first boot transistor Nbot1. The first switching transistor Nsw1 is connected between the cell resistor Rcell and the first boot transistor Nbot1. The word line voltage Vwd is applied to the gate terminal of the first switching transistor Nsw1. When the word line voltage Vwd has a high level, the first switching transistor Nsw1 is turned on and operates in a linear region or a triode region. Therefore, the voltage drop at the drain-source terminal of the first switching transistor Nsw1 becomes a negligible level.

The first booting transistor Nbot1 is connected between the first switching transistor Nsw1 and the ground, and is diode-connected. Thus, the first boot transistor Nbot1 operates in the active region and has a voltage difference between the drain and source above the threshold voltage. That is, the voltage at the drain terminal of the first booting transistor Nbot1 is equal to or higher than the threshold voltage, and its value is determined by the bias current Ibias.

The first common source amplifier 120 includes a first transistor M1 and a second transistor M2. The first transistor M1 is connected between the positive power supply voltage VDD and the second node N2, and the second transistor M2 is connected between the second node N2 and the ground. At the small signal level, the first transistor M1 acts as a common source amplifier and the second transistor M2 acts as an active load. The small signal voltage at the second node N2 is applied to one input of the sense amplifier 330. [

When a change occurs in the cell resistance Rcell, the voltage of the first node N1 fluctuates. Assuming that the bias current Ibias has a constant value, the variation of the cell resistance Rcell is represented by the variation of the cell voltage of the first node N1. The variation of the voltage at the first node N1 is amplified by the first transistor M1 serving as a common source amplifier. For example, when the resistance of the cell resistance Rcell is reduced, the voltage at the first node N1 is reduced and the voltage at the gate terminal of the first transistor M1 is also reduced. Therefore, the current flowing through the first transistor M1 increases. Therefore, the voltage of the second node N2 rises by the second transistor M2 serving as the active load.

In order for the first transistor M1 to function as a common source amplifier and the diode-connected second transistor M2 to act as an active load, the two transistors must operate in the active region.

The second cell voltage amplifying part 200 has the second switching transistor Nsw2 and the second booting transistor Nbot2 as the reference bias boosting part 210 and the third common source amplifying part 220 has the third transistor M3 and the fourth transistor M4.

The second bias transistor Pbias2 constitutes a second current source 320, and is connected between the positive power supply voltage VDD and the third node N3. And the second bias transistor Pbias2 forms the bias current Ibias by the applied bias voltage Vbias.

The reference resistor Rref is connected between the third node N3 and the second switching transistor Nsw2. The voltage of the third node N3 is determined by the reference resistance Rref and the bias current Ibais flowing through the reference bias boosting section 210. [ However, it is preferable that the reference resistor Rref has a fixed value. Therefore, it is preferable that the voltage of the third node N3 also has a fixed value that does not vary.

The second switching transistor Nsw2 is connected between the reference resistor Rref and the second boot transistor Nbot2. The word line voltage Vwd is applied to the gate terminal of the second switching transistor Nsw2. By the application of the word line voltage Vwd, the second switching transistor Nsw2 is turned on and operated in the linear region. Therefore, the voltage difference between the drain and the source of the second switching transistor Nsw2 is negligible. A second boot transistor Nbot2 is connected between the second switching transistor Nsw2 and the ground. The second boot transistor Nbot2 is diode-connected. In addition, the second boot transistor Nbot2 operates in the active region because it has a diode-connected structure. Therefore, the voltage at the drain terminal of the second booting transistor Nbot2 is determined by the bias current Ibias.

The third transistor M3 and the fourth transistor M4 constitute the second common source amplifying part 220. First, the fourth transistor M4 is connected between the positive power supply voltage VDD and the fourth node N4. Further, the voltage of the third node N3 is applied to the gate terminal of the fourth transistor M4. However, since the voltage level of the third node N3 becomes a fixed state, the amplifying action in the fourth transistor M4 is substantially not generated. The third transistor M3 is connected between the fourth node N4 and the ground, and receives the voltage of the second node N2 at the gate terminal. The third transistor M3 amplifies the voltage of the applied second node N2 by a small signal and applies it to the other input terminal of the sense amplifier 330. [ Thus, in the second common source amplifier section 220, the third transistor M3 acts as a common source amplifier, and the fourth transistor M4 acts as an active load. For this, the third transistor M3 and the fourth transistor M4 must operate in the active region.

Instead, when the bias voltage Vbias is applied at the call level, the first bias transistor Pbais1 and the second bias transistor Pbais2 generate the bias current Ibias, respectively. Further, the first switching transistor Nsw1 and the second switching transistor Nsw2 are turned on by the word line voltage Vwd. Assuming that the cell resistance Rcell and the reference resistance Rref are equal to each other, the voltage of the first node N1 and the voltage of the third node N3 have the same value. Therefore, the currents flowing through the first transistor M1 and the fourth transistor M4 have the same value. Also, since the second transistor M2 is diode-connected, the voltage of the second node N2 is determined by the current flowing through the first transistor M1. However, the voltage of the fourth node N4, which is the voltage of the drain terminal of the third transistor M3, may be set differently from the voltage of the second node N2. This is because the third transistor M3 is not in a diode-connected state. However, since the third transistor M3 forms a current mirror with the second transistor M2, the currents flowing through both transistors are equal to each other. The voltage of the fourth node N4 may be determined according to the current-voltage characteristics of the third transistor M3 and the fourth transistor M4. However, it is preferable that the voltages of the second node N2 and the fourth node N4 are equal to each other.

For example, when the sense amplifier 330 has a virtual short circuit therein, the call levels may be set equal to each other instead of the second node N3 and the fourth node N4.

The analysis at the small signal level is when the DC component is set to ground and an analysis of the variation of the signal is performed. In particular, since no signal fluctuation occurs at the third node N3, the gate terminal of the fourth transistor M4 is interpreted as being grounded. Also, a signal fluctuation occurs at the first node N1. This is due to the variation of the cell resistance Rcell. Thus, when the small signal level of the first node N1 increases, the first transistor M1 acts as a common source amplifier and amplifies the small signal level of the first node N1. The second transistor M2 acts as an active load for determining the gain. For example, at the second node N2, a small signal resistance between the drain and the source of the second transistor M2 appears, thereby determining the gain of the common source amplifier. The amplified signal at the second node N2 is applied to one input terminal of the sense amplifier, and is input to the gate terminal of the third transistor M3.

The third transistor M3 has the configuration of a common source amplifier. The gain of the third transistor M3 is determined by the small signal resistance between the drain and the source of the fourth transistor M4. A signal amplified by the third transistor M3 at the fourth node N4 is applied to the other input terminal of the sense amplifier.

If the small signal gain of the first transistor M1 as a common source amplifier is K1, the small signal gain of the third transistor M3 is K2, and the small signal voltage at the first node N1 is v1, The small signal voltage v2 is determined by the following equation (1).

In Equation (1), the small signal gain K1 has a negative value due to the configuration of the common source amplifier. This means that the output v2 is inverted in phase relative to the input v1.

If the small signal voltage at the fourth node N4 is v3, the small signal voltage v3 is determined by the following equation (2).

In Equation (2), K2 has a negative value because the third transistor M3 operates as a common source amplifier. Therefore, the small signal voltage v3 of the fourth node N4 has the same phase value as v1. On the other hand, the small signal voltage v2 of the second node N2 has an opposite phase value due to the small signal gain K1 having a negative value.

Therefore, the small signal voltages v2 and v3 amplified by the common source amplifier have negative and positive values, respectively, and can be input to the sense amplifiers to amplify the difference of the small signal voltage v1, which is the input signal.

In the present invention, the first common source amplifying unit 120 of the first cell voltage amplifying unit 100 can operate in the active region in a stable state. The DC component at the first node N1 is raised by the cell bias boosting part 110. [ This reduces the overdrive voltage of the first transistor M1 of the first common source amplifier section 120. [ This is clear in comparison with the prior art of FIG.

That is, the voltage of the first node N1 increases by the threshold voltage Vthn of the first boot transistor Nbot1. Therefore, the overdrive voltage of the first transistor M1 also decreases. The condition for the first transistor M1 to operate in the active region is that the voltage difference Vsd between the source and the drain must be larger than the overdrive voltage. In order for the first transistor M1 to operate in the active region, the condition of the voltage Vsd decreases from that of FIG. This means that even if the level of the drain voltage rises, the first transistor M1 operates in the active region. That is, since the first transistor M1 operates in the active region, it is possible to perform the small signal amplification operation, and the margin of the drain voltage of the first transistor M1 to perform this operation is increased.

Also, the direct current component flowing through the first transistor M1 and the second transistor M2 is reduced by the DC voltage of the increased first node N1. That is, the power required for the first transistor M1 to operate in the active region is reduced.

The same applies to the second cell voltage amplifying part 200 as well. The second cell voltage amplifying part 200 has a reference bias step-up part 210 including a diode-connected transistor. Therefore, the DC voltage of the third node N3 increases, and the voltage margin for operating the fourth transistor M4 in the active region is widened. Also, the DC current of the fourth transistor M4 generated in the active region by the voltage of the third node N3 is reduced, and the power consumption is reduced.

In particular, the first transistor M1 and the fourth transistor M4 have a symmetrical connection relationship with respect to the circuit configuration, the first transistor M1 is used as a common source amplifier, and the fourth transistor M4 is used as an active load. The second transistor M2 and the third transistor M3 have the configuration of a current mirror. The second transistor M2 acts as an active load of the first transistor M1, which is a common source amplifier. The third transistor M3 acts as a common source amplifier , And the fourth transistor M4 as an active load.

In the present invention, the margin for performing the small signal amplification operation is widened and the power consumption is reduced with respect to the change of the cell resistance, which is the input signal to be applied. In addition, the second common source amplifying unit 220 of the second cell voltage amplifying unit 200 performs the two-stage common source amplifying operation with respect to the small signal voltage according to the change of the cell resistance Rcell. That is, the signal amplified by the first transistor M1 is amplified by the third transistor M3, and the amplified signal is applied to the sense amplifier 330. Therefore, a sense amplifier circuit having an increased sensing margin and a low consumption power can be obtained.

100: first cell voltage amplifying unit 110: cell bias boosting unit
120: first common source amplification unit 200: second cell voltage amplification unit
210: reference bias boosting unit 220: second common source amplifying unit
310: first current source 320: second current source
330: sense amplifier

Claims (7)

A first cell voltage amplifying unit for amplifying variations in cell resistance as a common source;
A second cell voltage amplifying unit for receiving the amplified signal from the first cell voltage amplifying unit and amplifying the amplified signal from the common source; And
And a sense amplifier for receiving and amplifying the amplified signal of the first cell voltage amplification unit and the amplified signal of the second cell voltage amplification unit. The apparatus of claim 1, wherein the first cell voltage amplifying unit comprises:
A cell bias boosting unit connected between the cell resistance and ground for raising a direct current component of the cell voltage at a first node between the cell resistance and the first current source according to a bias current generated by the first current source; And
And a first common source amplifier for receiving the cell voltage of the first node and for amplifying the cell voltage of the first node to a second node through common source amplification. 3. The semiconductor device according to claim 2, wherein the cell-
A first switching transistor connected to the cell resistance and performing a switching operation by a word line voltage; And
And a first boot transistor coupled between the first switching transistor and ground and diode-connected to operate in an active region. The semiconductor integrated circuit according to claim 2, wherein the first common-
A first transistor connected between the positive power supply voltage and the second node and having a common source configuration by amplifying the cell voltage by a small signal and for inputting the amplified cell voltage to one input terminal of the sense amplifier; And
And a second transistor connected between the first transistor and the ground and diode-connected to act as an active load to the first transistor. 3. The apparatus of claim 2, wherein the second cell voltage amplifying unit comprises:
A reference bias step-up unit connected between the reference resistor and the ground, for raising a reference voltage at a third node between the reference resistor and the second current source according to a bias current generated by the second current source;
And a second common source amplification unit for receiving the reference voltage and outputting the amplified signal of the first cell voltage amplification unit to the fourth node through common source amplification and inputting the amplified signal to the other input terminal of the sense amplifier. Sense amplifier circuit. 6. The semiconductor device according to claim 5, wherein the reference bias step-
A second switching transistor connected to the reference resistor and performing a switching operation by a word line voltage; And
And a second boot transistor connected between the second switching transistor and ground and diode-connected to operate in an active region. 6. The semiconductor device according to claim 5, wherein the second common-
A third transistor for amplifying a common signal of the small signal of the second node and outputting it to the fourth node and inputting the signal of the fourth node to the other input terminal of the sense amplifier; And
And a fourth transistor coupled between the positive supply voltage and the fourth node, the fourth transistor being operative as an active load for the third transistor, and receiving a signal at the third node.

Images