Time amplifier based on current scaling technology

By dividing the time amplifier circuit into reset, amplification and slow discharge stages, and using the current scaling technology of MOS tubes and comparators, the problems of gain deviation and delay time saturation when processing small time difference signals are solved, and the further improvement is achieved High gain stability and resolution.

CN119070752APending Publication Date: 2024-12-03BEIJING UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411115333.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional time amplifiers are prone to enter metastable state when processing small time difference signals, causing the gain to deviate from the expected design value, making it difficult to improve the effective resolution, and delay time saturation will occur when the gain is too high, affecting the overall circuit performance.

Method used

The time amplifier circuit based on current scaling technology is adopted. By dividing the amplification process into three stages: reset, amplification and slow discharge, the MOS tube and comparator are used to achieve more stable time difference amplification. The amplification stage uses large current discharge and the slow discharge stage uses Small current discharge and convert voltage signals into time difference signals through comparator to avoid delay time saturation of cross-coupling structures.

Benefits of technology

It improves the gain stability of the time amplifier and the performance of the overall analog-to-digital conversion circuit, enhances the amplification capability of small-time difference signals, and reduces the impact of power supply voltage jitter and noise.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119070752A_ABST
    Figure CN119070752A_ABST
Patent Text Reader

Abstract

The invention discloses a time amplifier based on a current scaling technology. The time amplifier is composed of nine MOS tubes, two phase inverters, two capacitors and two comparators. The working logic of the time amplifier is divided into three stages, the first stage is a reset stage, and the voltage of an upper pole plate of a load capacitor is reset to power supply voltage; the second stage is an amplification stage, the amplification stage is started after a rising edge appears, and a load capacitor is discharged with large current; the third stage is a slow discharge stage, when the two rising edges appear, the slow discharge stage is started, the load capacitors corresponding to the input ends for inputting the signals of the two rising edges discharge with the same small current, and when the voltage of the upper pole plates of the two load capacitors drops to the threshold voltage of the comparator, the comparator is started; the comparator outputs two rising edges with a certain time difference. Therefore, amplification of the time difference signal is completed. The analog-to-digital conversion circuit is beneficial for further improving the gain of the time amplifier and improving the performance of the whole analog-to-digital conversion circuit.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a time amplifier circuit based on current scaling technology, belonging to the technical field of data converters. Background Art

[0002] In modern life, the signals people are exposed to are usually analog signals. As people's understanding of the real world continues to deepen, various electronic devices that rely on rapid processing of digital signals make people's lives more convenient. In order to enable people to interact better with nature, it is an important research direction for analog-to-digital converters to maintain high resolution while reducing the delay time required for conversion. Faced with the current situation that the effective resolution of time-domain analog-to-digital converters is difficult to improve when performing high-speed quantization, time amplifiers have the important task of amplifying time information and thus improving effective resolution. Based on the performance requirements for analog-to-digital converters, time amplifiers that can amplify time information have become the mainstay.

[0003] With the continuous improvement of the performance requirements for analog-to-digital converters, high-speed, medium-resolution analog-to-digital converters have gradually become a research trend. The multi-stage time domain analog-to-digital converter can make full use of the characteristics of the time signal based on pipeline operation, and bring the speed of the analog-to-digital converter to a new level. While ensuring high speed, interpolation technology or cursor technology is usually used to reduce the minimum time scale to improve the effective resolution. At present, it has entered the stage where the minimum time scale is hundreds of femtoseconds. Since the time signal is very sensitive to power supply voltage changes and noise, the smaller the minimum time scale, the greater the impact of power supply voltage jitter and noise on the effective resolution. The function of the time amplifier is to amplify the time difference signal input at its input end by an integer multiple and output it. In order to improve the effective resolution, the time amplifier can be used to amplify the residual time signal of the previous stage and transmit it to the next stage for quantization. In this way, the function of the time amplifier can be used to amplify the minimum time difference that the analog-to-digital converter needs to distinguish, and suppress the influence of power supply voltage jitter and noise.

[0004] The basic working principle of the current time amplifier is: the time signal to be amplified is contained in the time interval between the two rising edges of the input terminal, and when the rising edge comes, the corresponding load capacitor begins to be discharged, and the cross-coupling structure is used to make the discharge current of the load capacitor corresponding to the input terminal with the leading rising edge greater than the discharge current of the load capacitor corresponding to the other input terminal. Two threshold voltage detectors composed of inverters are used to detect the upper plate voltages of the two load capacitors respectively, and the rising edge is output when the voltage observed by the threshold voltage detector drops to its threshold voltage. In this way, by adjusting the size relationship of the discharge current, two rising edge signals with a larger time interval can be obtained, thereby realizing the amplification of the time difference signal.

[0005] By designing the size ratio of the discharge branch transistor, an accurate amplification factor can be achieved. Due to the cross-coupling structure of the traditional time amplifier, when the signal to be amplified has a small time difference, the cross-coupling structure will enter a metastable state, so that the gain of the time amplifier for the small time difference signal deviates from the expected design value. It is difficult to increase the gain with the traditional structure. If the gain is too high, the load capacitor on one side will discharge quickly to the threshold voltage of the threshold voltage detector connected thereto, while the load capacitor on the other side will discharge slowly, causing the delay time to be saturated, thereby affecting the effective resolution of the overall circuit. The present invention can overcome the shortcomings of the prior art, further improve the circuit of the time amplifier and improve the performance of the time amplifier. Summary of the invention

[0006] The present invention proposes a new amplification method for the time amplifier circuit, and uses a simpler and more stable method to complete the time amplification of the input signal time difference. The present invention divides the working logic of the time amplifier into three stages. The first stage is the reset stage, in which the upper plate voltage of the load capacitor is reset to the power supply voltage; the second stage is the amplification stage, which enters the amplification stage after a rising edge appears, and discharges the load capacitor with a large current; the third stage is the slow discharge stage, which enters the slow discharge stage after two rising edges appear, and the load capacitor corresponding to the input end of the input two rising edge signals is discharged with the same small current. When the upper plate voltage of the two load capacitors drops to the threshold voltage of the comparator, the comparator outputs two rising edges with a certain time difference. In this way, the amplification of the time difference signal is completed.

[0007] The above purpose is achieved through the following technical solutions:

[0008] A time amplifier circuit based on current scaling technology, which consists of a discharge branch, a load capacitor and a comparator. It involves repeated conversion between the time domain and the voltage domain. After the amplification stage, the two rising edges of the input containing the time difference information to be amplified generate a voltage signal with a certain voltage difference on the upper plates of the two load capacitors, and enter the slow discharge stage. When the voltage on the upper plate of the load capacitor drops to a certain threshold voltage, the two slowly discharged voltage signals with a certain voltage difference are converted into two rising edge signals with a corresponding time difference through the comparator.

[0009] The time amplifier is composed of nine MOS tubes, two inverters, two capacitors and two comparators.

[0010] The capacitor C1 and the capacitor C2 each have two discharge branches. The two discharge branches of the capacitor C1 are formed by stacking and cascading M1, M5 and M3, M9, respectively. The upper M1 and M3 in the discharge branch are controlled by the signal P, while the lower M5 and M9 in the discharge branch are controlled by the signal N through the cascade inverter INV1 and the signal CLK.

[0011] The two discharge branches of the capacitor C2 are formed by stacking and cascading M2, M6 and M4, M9 respectively. The upper M2 and M4 in the discharge branch are controlled by signal N, while the lower M6 and M9 in the discharge branch are controlled by signal P through cascade inverter INV2 and CLK signal respectively.

[0012] The MOS tubes M7 and M8 are controlled by CLK to charge and reset the upper plates of capacitors C1 and C2, namely the OP and ON nodes. The negative input terminals of the comparators COMP1 and COMP2 are connected to the OP node and the ON node respectively and are compared with the VTH voltage to output TP and TN signals.

[0013] The discharge mode adopted by the present invention is different from that of the traditional time amplifier. In the traditional time amplifier, each discharge branch does not overlap. The present invention uses MOS tube M9 to merge one discharge branch of each discharge branch of capacitors C1 and C2, and allows M3 and M4 to evenly divide the current of M9 when entering the slow discharge stage, which is conducive to further improving the gain of the time amplifier. A comparator is used to replace the threshold voltage detector composed of multiple inverters cascaded in the traditional time amplifier, so that the threshold voltage of TP and TN flipping is more accurate, which can effectively improve the stability of the gain of the time amplifier and improve the performance of the overall analog-to-digital conversion circuit. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a circuit principle structure block diagram of the present invention. DETAILED DESCRIPTION

[0015] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0016] like Figure 1As shown, the technical solution adopted by the present invention is a time amplifier circuit based on current scaling technology, wherein the CLK signal and the P signal and N signal at the input end maintain a certain correlation in phase, and the P signal and the N signal become low level earlier and high level later than the CLK signal, so as to avoid the path from the power supply to the ground, and generate high current and cause extra power consumption. In the reset stage, CLK is low level, and the P signal and the N signal are also low level. At this time, only the MOS tubes M7 and M8 are turned on, and the power supply charges the upper plates of the capacitors C1 and C2, namely the OP and ON nodes, to the power supply voltage. When CLK becomes high level, the time amplifier enters the standby state, waiting for the rising edge of the P signal and the N signal. When the P signal rises earlier than the N signal, in the time interval between the rising edge of the P signal and the rising edge of the N signal, the P signal is high, the N signal is low, M1, M5, M3 and M9 are turned on, and M2, M4 and M6 are turned off, so the two discharge branches of capacitor C1 are turned on and capacitor C1 exclusively enjoys the discharge current of M9. At this time, the discharge current of C1 is (A+1)·I. After the rising edge of the N signal, M5 is turned off, M4 and M2 are turned on, and the slow discharge stage is entered. The discharge branches of capacitors C1 and C2 only have the common branch controlled by M9, and capacitors C1 and C2 share the discharge current of M9. That is, the discharge current of C1 and C2 is I / 2 at this time. Under the action of comparators COMP1 and COMP2, TP and TN output high level when the voltage of OP and ON is less than VTH. Because C1 discharges in advance with a current of (A+1)·I and lasts for the time interval Δt between two rising edges, when the voltage of the OP node drops to VTH during the slow discharge phase, the ON node needs to continue to discharge for 2·(A+1)·Δt to drop to the VTH voltage, that is, TP leads the rising edge of TN by 2·(A+1)·Δt. After both output signals TP and TN have rising edges, the P signal and the N signal become low, and then CLK becomes low to enter the reset phase, waiting for the next cycle of amplification.

[0017] The working mode of the N input terminal's early rising edge and the P input terminal's early rising edge is the same, and the amplification factor of the time difference signal is 2·(A+1). The value of A is controlled by adjusting the size ratio of the MOS tube in the discharge branch. A larger gain can be achieved by setting a smaller size ratio, which is beneficial to the matching between MOS tubes and improves the robustness of the time amplifier gain.

[0018] The present invention strictly distinguishes the discharge modes in the amplification stage and the slow discharge stage, that is, the large current discharge only occurs in the amplification stage, while the capacitors C1 and C2 are discharged with the same small current in the slow discharge stage. This can avoid the situation where the delay time saturation caused by the current on one side being always greater than the current on the other side causes the time amplifier gain to be less than the expected design value.

[0019] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A time amplifier based on current scaling technology, characterized in that: It consists of nine MOS tubes, two inverters, two capacitors and two comparators; The capacitor C1 and the capacitor C2 each have two discharge branches; the two discharge branches of the capacitor C1 are formed by stacking and cascading M1, M5 and M3, M9, respectively. The upper M1 and M3 in the discharge branches are controlled by a signal P, while the lower M5 and M9 in the discharge branches are controlled by a signal N through a cascade inverter INV1 and a signal CLK. The two discharge branches of the capacitor C2 are formed by stacking and cascading M2, M6 and M4, M9 respectively. The upper M2 and M4 in the discharge branch are controlled by the signal N, while the lower M6 and M9 in the discharge branch are controlled by the signal P through the cascade inverter INV2 and the CLK signal respectively. The MOS tubes M7 and M8 are controlled by CLK to charge and reset the upper plates of the capacitors C1 and C2, namely the OP and ON nodes respectively; the negative input terminals of the comparators COMP1 and COMP2 are respectively connected to the OP node and the ON node and are compared with the VTH voltage to output TP and TN signals.

2. The time amplifier based on current scaling technology according to claim 1, characterized in that: The CLK signal and the P signal and N signal at the input end are associated in phase, and the P signal and the N signal become low level earlier and high level later than the CLK signal; in the reset stage, the CLK signal is low level, and the P signal and the N signal are also low level. At this time, only the MOS tubes M7 and M8 are turned on, and the power supply charges the upper plates of the capacitors C1 and C2, i.e., the OP and ON nodes, to the power supply voltage; when the CLK signal becomes high level, the time amplifier enters the standby state, waiting for the rising edge of the P signal and the N signal to come; when the P signal has a rising edge earlier than the N signal, when the P signal has a rising edge In the time interval when the rising edge of the N signal appears, the P signal is at a high level, the N signal is at a low level, M1, M5, M3 and M9 are turned on, and M2, M4 and M6 are turned off, the two discharge branches of capacitor C1 are turned on and capacitor C1 exclusively enjoys the discharge current of M9, and the discharge current of capacitor C1 is (A+1)·I; after the rising edge of the N signal appears, M5 is turned off, M4 and M2 are turned on, and the slow discharge stage is entered. The discharge branches of capacitors C1 and C2 only have the common branch controlled by M9, and capacitors C1 and C2 evenly share the discharge current of M9, that is, at this time the discharge current of capacitors C1 and C2 is I / 2.

3. The time amplifier based on current scaling technology according to claim 2, characterized in that: Under the action of comparators COMP1 and COMP2, TP and TN output high levels when the voltages of OP and ON are less than VTH; when the voltage of the OP node drops to VTH in the slow discharge stage, the ON node needs to continue discharging for 2·(A+1)·Δt before it can drop to the VTH voltage, that is, TP leads the rising edge of TN by 2·(A+1)·Δt; after the two output signals TP and TN both have rising edges, the P signal and the N signal become low levels, and then the CLK signal becomes low level and enters the reset stage, waiting for the next cycle amplification.

4. The time amplifier based on current scaling technology according to claim 3, characterized in that: The amplification factor of the time difference signal at the N input terminal is 2·(A+1); the value of A is controlled by adjusting the size ratio of the MOS tube in the discharge branch.