CN110909516A - Modeling method considering influence of shape and size of active region in single event effect circuit simulation - Google Patents
Modeling method considering influence of shape and size of active region in single event effect circuit simulation Download PDFInfo
- Publication number
- CN110909516A CN110909516A CN201911058784.2A CN201911058784A CN110909516A CN 110909516 A CN110909516 A CN 110909516A CN 201911058784 A CN201911058784 A CN 201911058784A CN 110909516 A CN110909516 A CN 110909516A
- Authority
- CN
- China
- Prior art keywords
- well
- circuit
- single event
- active region
- event effect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Design And Manufacture Of Integrated Circuits (AREA)
- Insulated Gate Type Field-Effect Transistor (AREA)
Abstract
The invention relates to a modeling method considering the shape and size of an active region in single event effect circuit simulation, which solves the technical problem that the prior single event effect circuit simulation method does not completely support verification for circuit design reinforcement and layout design reinforcement, and the method mainly comprises the following implementation steps: 1) constructing a single event effect basic model considering the influence of the shape and the size of the active region; 2) calibrating the dependency relationship of parameters to be extracted in the single event effect basic model along with the change of layout information; 3) extracting layout shape information and circuit netlist information of a circuit to be analyzed; 4) selecting a heavy ion incident position in the circuit layout to be analyzed, and reconstructing a single event effect model of the circuit layout to be analyzed; 5) and updating the circuit netlist, and simulating and calculating the single event effect of the circuit.
Description
Technical Field
The invention relates to a single event effect simulation evaluation and reinforcement verification technology of a CMOS integrated circuit, in particular to a modeling method considering the influence of the shape and the size of an active region in single event effect circuit simulation.
Background
The single event effect of a semiconductor integrated circuit in a spacecraft can be caused by rays such as protons and heavy ions in a space radiation environment, and single event soft errors (single event upset and single event transient) become one of important factors causing the on-orbit failure of the spacecraft, so that the reliability and the service life of the spacecraft are seriously influenced.
In order to effectively improve the single event resistance of the integrated circuit, one idea is to set redundant nodes in the circuit design, for example, latches in an SRAM unit, a D-type trigger unit and the like are modified into double redundant node latches, and only a single node can be corrected by the circuit when the state is inverted; the other idea is to add modification in layout design, inhibit bipolar amplification charge collection by increasing the contact density of a trap and maintaining the potential of the trap in area on the basis of not changing a logic netlist, and the like. The single event effect circuit simulation is considered to be a means capable of predicting the single event resistance of the integrated circuit at the initial stage of design, and the single event resistance performance evaluation method for the integrated circuit aiming at circuit design reinforcement and layout design reinforcement is supposed to have the capability of evaluating the single event resistance of the integrated circuit for reasonably and reliably evaluating the effectiveness of the reinforced design.
In the existing single event effect circuit simulation method, different processes exist in the aspect of how to introduce the electrical interference caused by high-energy particles in the circuit into a transistor-level intensive model (SPICE model). The patent with the application number of 201310718181.7 and the name of 'a verification method and a system for resisting single event effect based on semiconductor circuit' utilizes device/circuit hybrid simulation to evaluate single event sensitivity response of a CMOS cascade circuit, and has the problems that only single node collection charge is considered to generate pulse disturbance, and charge sharing of different active intervals is not considered; the patent with the application number of 201610051734.1 and the name of 'a single event effect multi-bit upset circuit simulation method' proposes that double-exponential current source disturbance pulses are added to randomly selected paired nodes in a double-redundancy node latch reinforced by adopting circuit design, sensitive node pairs capable of causing upset are searched, and the problem is that sensitive evaluation information such as upset sections and LET threshold values cannot be obtained; the patent with the application number of 201611000399.9 and the name of 'a combined logic circuit single-particle multi-transient soft error sensitivity evaluation method considering layout information' proposes to adopt a Monte Carlo method to calculate the amount of electric charge deposited on the drain electrode of each transistor and further convert the electric charge into a double-exponential current source disturbance pulse.
Disclosure of Invention
In order to solve the technical problems of circuit design reinforcement and incomplete support verification of layout design reinforcement in the conventional single event effect circuit simulation method, the invention provides a method flow for extracting source-drain active region and well contact shape information based on layout analysis, constructing a resistance network with a heavy ion incident position as a center, dynamically calculating substrate potentials of all active regions and contributions brought by starting parasitic transistors, and reconstructing a single event effect disturbance current source item aiming at all active regions, and the method flow is used for accurately evaluating the effectiveness of single event reinforcement measures in a design stage.
The technical solution of the invention is as follows:
the invention relates to a modeling method for considering the influence of the shape and the size of an active region in single event effect circuit simulation, which comprises the following steps:
1) constructing a single event effect basic model considering the influence of the shape and the size of an active region
Constructing a single event effect basic model consisting of three parts, wherein the single event effect basic model comprises a well which is contacted with each active region,A resistor network formed by resistors between each active region and the heavy ion incidence position, and a well contact current source item I between the N-type well contact and the P-type well contactwellAnd single event effect disturbance current source terms I numbered from 1, 2 to M of active regionsSET1、ISET2Up to ISETm(ii) a The resistor network comprises R1、R2Up to RnWhere n is the total number of resistors in the resistor network,
the resistance value in the resistance network is composed of a contact resistance and a body resistance, and the specific expression is as follows:
Ri＝f1/Ai+f2·di，i＝1,2,...,n (1)
wherein A isiDenotes the area of the well contact, diIndicating the spacing between the well contact and the active region; f. of1And f2Parameters to be extracted are obtained; (ii) a
Well contact current source term I between N-type well contact and P-type well contactwellThe method is described according to a double-exponential current source form, and the specific expression is as follows: :
wherein, Ipeak,wellRepresenting the peak current, tr,wellDenotes the rise time of the current, tf,wellRepresenting the current fall time, τr,well、τf,wellCurrent rise and fall time constants; the expression has four items of parameters to be extracted, Ipeak,well、τr,well、τf,well、(tf,well-tr,well) The four parameters are determined by the incident point of heavy ions and the distance value at the junction of the N trap and the P trap, the function form of the parameters accords with an exponential relation, and the linear terms and the power exponent are parameters which need to be extracted respectively;
single event effect disturbance current source item I numbered from 1, 2 to M active regionsSET1、ISET2Up to ISETmIs to collect the current source term I according to the driftdrifti(t) and diffusion recoveryCurrent collection source item Idiffusioni(t) is described in a form of superposition, and the specific expression is as follows:
ISETi(t)＝Idrifti(t)+Idiffusioni(t)，i＝1,2,...,m (3)
wherein, IdriftiAnd IdiffusioniThe current source is in a double-exponential current source form, and the specific expressions all follow expressions similar to the formula (2);
wherein, drift collects current source item Idrifti(t) and diffusion Collection Current Source term Idiffusioni(t) are all in the form of dual-exponential current sources, and the specific expressions all follow expressions similar to the formula (2); drift collection current source item Idrifti(t) and diffusion Collection Current Source term IdiffusioniThe parameters required to be extracted in the expression of (t) are respectively Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) And Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni) (ii) a The eight parameters are determined by the shape and size of the active area and the distance value between the active area and the heavy ion incidence point, the function form of the eight parameters accords with an exponential relation, and the linear term and the power exponent are parameters which need to be extracted respectively;
2) calibrating the dependency relationship of parameters to be extracted in the single event effect basic model along with the change of layout information
2.1) extracting process structure information by referring to a circuit design process library model, and constructing a single-tube device model available for numerical simulation by adjusting normal electrical characteristic parameters by using a semiconductor device simulation tool; the normal electrical characteristic parameters comprise transistor threshold voltage, static current and doping region resistance values, and channel and source drain region doping information is calibrated;
2.2) calculating resistance values corresponding to different trap contact areas and different distance combinations between the trap contacts and the active region by setting virtual nodes, and fitting to obtain the parameter f in the formula (1)1And f2Taking the value of (A);
2.3) setting different positions for heavy ion incidence,obtaining a well contact current source term I through a formula (2)wellCorresponding to Ipeak,well、τr,well、τf,well、(tf,well-tr,well) The value of the data group changes along with the change of a set heavy ion incidence point and the change of a distance value at the junction of the N trap and the P trap;
2.4) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
2.5) calculating the drift factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
wherein (x)0,y0) Represents the incident position of heavy ions, (x)i,yi) Representing the central coordinate, R, after discretization of the active area0A characteristic radius representing the lateral expansion of the heavy ion trajectory; the discretization step length should be small enough to ensure that the drift factor value obtained by integration when the discretization step length is continuously reduced does not change obviously.
2.6) calculating the current source term I collected by drift when heavy ions are incident at different positionsdriftiCorresponding to Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) A data set whose value changes with the change of the drift factor value;
2.7) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
2.8) calculating the diffusion factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
wherein (x)0,y0) Represents the incident position of heavy ions, (x)i,yi) Represents the center coordinates after discretization of the active area, (x)k,yl) Substitute for Chinese traditional medicineCenter coordinates, R, after discretization of the region around the point of incidence0A characteristic radius representing the lateral expansion of the heavy ion trajectory; the discretization step length should be small enough to ensure that the diffusion factor value obtained by continuously reducing the integration during the discretization step length does not change obviously.
2.9) calculating the current source item I of diffusion collection when heavy ions are incident at different positionsdiffusioniCorresponding to Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni) A data set whose values vary with the variation of the diffusion factor value;
2.10) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
3) extracting layout shape information and circuit netlist information of circuit to be analyzed
3.1) taking the circuit layout file as input, screening and extracting coordinates at the junction of the N well and the P well and well contact edge coordinate information;
3.2) taking the circuit layout file as input, screening and extracting the coordinate information of the source and drain active regions of all transistors;
3.3) taking the circuit layout file as input, screening and extracting a circuit netlist, extracting corresponding transistors one by one according to the sequence of the source-level and drain-level active region coordinate records, and recording node numbers one by one;
4) selecting the incident position of heavy ions in the circuit layout to be analyzed, and reconstructing the single event effect model of the circuit layout to be analyzed
4.1) calculating to obtain a trap contact area A according to the coordinates of the junction of the N trap and the P trap extracted in the step 3.1) and the trap contact edge coordinate informationiAnd the spacing d between the well contact and the active regioniTaking values;
4.2) and then the well contact area A obtained in the step 4.1)iAnd the spacing d between the well contact and the active regioniAnd the parameter value f established in step 2.2)1And f2Substituting the formula (1) to calculate the resistance value between the trap contact to each active region and each active region to the heavy ion incidence position, thereby obtaining the resistance valueA resistance network of the single event effect model of the circuit layout to be analyzed;
4.3) calculating to obtain a distance value between a heavy ion incident point and the junction of the N trap and the P trap according to the coordinate information of the junction of the N trap and the P trap extracted in the step 3.1);
4.4) combining the linear terms and power exponent values in each exponential function established in the step 2.4) and the distance values of the heavy ion incidence point and the junction of the N trap and the P trap obtained by calculation in the step 4.3), establishing a trap contact current source term I of the single event effect model of the circuit layout to be analyzed according to the formula (2)well；
4.5) calculating drift factors and diffusion factor values of all active regions according to the formula (4) and the formula (5) according to the source and drain active region coordinate information of all the transistors extracted in the step 3.2), and further re-establishing a drift collection current source item I according to the linear item and the power index value in each exponential function established in the step 2.7) and the step 2.10)driftiAnd diffusion collection current source item Idiffusioni；
4.6) calculating a disturbance current source item I of the single event effect model of the synthesized circuit layout to be analyzed according to the calculated connection relation between the well potential of each active region and the electrical nodeSET1、ISET2Up to ISETm；
4.6.1) according to the resistance network and the well contact current source item I constructed in the step 4.2) and the step 4.2)wellCalculating to obtain the real-time well potential of each active region;
4.6.2) calculating the contribution of the parasitic transistor on the disturbance current term of the active region, wherein the specific calculation formula is as follows:
Iparasitic∝exp(Vbs·q/kT) (6)
wherein, VbsRepresenting the real-time voltage difference between the transistor substrate and the source;
4.6.3) adding the disturbance current source items in the step 4.5) and the step 4.6.2) to obtain a value representing the disturbance current source item I of the single event effect model of the circuit layout to be analyzedSET1、ISET2Up to ISETm；
5) Updating circuit netlist, simulating single event effect of calculating circuit
5.1) adding the single event effect model of the circuit layout to be analyzed obtained in the step 4) on the basis of the circuit netlist obtained in the step 3.3) to obtain an updated circuit netlist;
5.2) performing transient circuit simulation, and recording the output of an actual circuit within 10ns after the heavy ion incidence moment;
5.3) analyzing the circuit output, and recording whether state inversion occurs or disturbance pulse which can be recovered automatically after a period of time is generated.
The semiconductor device simulation tool adopted in the step 2.1) is a TCAD simulation tool.
The simulation tool for adding the single event effect model in the circuit netlist adopted in the step 5.1) is SPICE, and the programming language adopted by the simulation tool is Verilog-A language.
The invention has the advantages that:
1. the modeling method considering the influence of the shape and the size of the active region in the single event effect circuit simulation provided by the invention not only can consider a multi-node collection mechanism that a plurality of circuit nodes simultaneously collect excessive carriers generated by radiation and generate disturbance current items, but also can consider the improvement effect of different well contact designs and different shapes and sizes of the active region on the single event sensitivity of the circuit, thereby realizing the reasonable verification of the effectiveness of the single event effect circuit design reinforcement and the layout design reinforcement;
2. the invention introduces two dimensionless parametric drift factors and diffusion factors, and simplifies the dependency relationship of a single-particle disturbance current source item on the shape and the size of a complex active region into the dependency relationship on a one-dimensional parametric drift factor and a diffusion factor, thereby realizing the evaluation of the influence of different active region shapes and sizes on a circuit disturbance item without the help of complex operation;
3. the invention can be conveniently called by a circuit simulation tool SPICE after being realized by using Verilog-A language, and can be conveniently embedded into a general circuit design flow.
Drawings
FIG. 1 is a diagram of a single event effect basic model;
FIG. 2 is a schematic diagram of calculating an active region drift factor;
FIG. 3 is a schematic illustration of calculating an active region diffusion factor;
FIG. 4 is a schematic diagram of a 40nm process node inverter;
fig. 5 is a graph of the change of the output voltage of the inverter with time calculated by using a circuit simulation method and TCAD software.
Detailed Description
The method mainly solves the defects of the existing method, when the circuit single event effect is subjected to analog calculation, the influence of the shape and the size of an active region including a trap contact on the circuit single event effect sensitivity is considered, the method is not limited to injecting constant disturbance current pulses into a single node, and the influence of different shapes and sizes of drain active regions on drift diffusion collection and the influence of different shapes and sizes of the trap contact on bipolar amplification collection can be described. A modeling method for considering the influence of the shape and the size of an active region in single event effect circuit simulation is successfully provided by constructing a resistance network of which a trap is in contact with each active region and a heavy ion incidence position and converting the shape and the size of the active region into values of a drift factor and a diffusion factor of each active region.
Preferred embodiments of the present invention will be further described with reference to the accompanying drawings.
The method comprises the following specific steps:
step 1: the method comprises the steps of constructing a single event effect basic model considering the influence of the shape and the size of an active region, wherein the single event effect basic model is composed of three parts as shown in figure 1, the single event effect basic model comprises a resistor network formed by resistors between a trap contact and each active region and between each active region and a heavy ion incidence position, and a trap contact current source item I between an N-type trap contact and a P-type trap contactwellAnd single event effect disturbance current source terms I numbered from 1, 2 to M of active regionsSET1、ISET2Up to ISETm(ii) a The resistor network comprises R1、R2Up to RnWhere n is the total number of resistors in the resistor network,
the resistance value in the resistance network is composed of a contact resistance and a body resistance, and the specific expression is as follows:
Ri＝f1/Ai+f2·di，i＝1,2,...,n (1)
wherein A isiDenotes the area of the well contact, diIndicating the spacing between the well contact and the active region; f. of1And f2Parameters to be extracted are obtained;
well contact current source term I between N-type well contact and P-type well contactwellThe method is described according to a double-exponential current source form, and the specific expression is as follows: :
wherein, Ipeak,wellRepresenting the peak current, tr,wellDenotes the rise time of the current, tf,wellRepresenting the current fall time, τr,well、τf,wellCurrent rise and fall time constants; the expression has four items of parameters to be extracted, Ipeak,well、τr,well、τf,well、(tf,well-tr,well)；
Single event effect disturbance current source item I numbered from 1, 2 to M active regionsSET1、ISET2Up to ISETmIs to collect the current source term I according to the driftdrifti(t) and diffusion Collection Current Source term Idiffusioni(t) is described in a form of superposition, and the specific expression is as follows:
ISETi(t)＝Idrifti(t)+Idiffusioni(t)，i＝1,2,...,m (3)
wherein, drift collects current source item Idrifti(t) and diffusion Collection Current Source term Idiffusioni(t) are all in the form of dual-exponential current sources, and the specific expressions all follow expressions similar to the formula (2);
drift collection current source item Idrifti(t) and diffusion Collection Current Source term IdiffusioniThe parameters required to be extracted in the expression of (t) are respectively Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) And Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni)；
Step 2: calibrating the dependency relationship of parameters needing to be extracted in the single event effect basic model along with the change of layout information;
step 2.1: extracting process structure information by referring to a circuit design process library model, and constructing a single-tube device model available for numerical simulation by adjusting normal electrical characteristic parameters by using a semiconductor device simulation tool; the normal electrical characteristic parameters comprise transistor threshold voltage, static current and doping region resistance values, and channel and source drain region doping information is calibrated;
step 2.2: by setting virtual nodes, resistance values corresponding to different trap contact areas and different distance combinations between the trap contacts and the active region are obtained, and the parameter f in the formula (1) is obtained through fitting1And f2Taking the value of (A);
step 2.3: setting different positions of heavy ion incidence, and obtaining a trap contact current source item I through a formula (2)wellCorresponding to Ipeak,well、τr,well、τf,well、(tf,well-tr,well) The value of the data group changes along with the change of a set heavy ion incidence point and the change of a distance value at the junction of the N trap and the P trap;
step 2.4: fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
step 2.5: calculating the drift factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
FIG. 2 is a schematic diagram illustrating the calculation of the active region drift factor, wherein (x)0,y0) Represents the incident position of heavy ions, R0Characteristic radius (typical values: 50nm or 100nm) representing the lateral extension of the heavy ion trajectory. Discretizing the active region according to a certain step length (such as 1nm step length), wherein the discretization step length is small enough to ensure that the drift factor value obtained by integration (referring to formula 1) does not obviously change when the discretization step length is continuously reduced, and the value (x) is obtainedi,yi) Representing the center coordinates after discretization of the active area. The drift factor value of each active region is calculated, and the larger the drift factor is, the more the charge amount collected by drift is.
Step 2.6: current source item I is collected in a drifting manner when heavy ions are incident to different positionsdriftiCorresponding to Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) A data set whose value changes with the change of the drift factor value;
step 2.7: fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
step 2.8: calculating the diffusion factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
FIG. 3 is a schematic diagram illustrating the calculation of the diffusion factor of the active region (x)0,y0) Represents the incident position of heavy ions, R0Representing the characteristic radius of the lateral expansion of the heavy ion trajectory. Discretizing (such as 1nm step length) the active region and the vicinity of the heavy ion incident point according to a certain step length, wherein the discretization step length is small enough to ensure that the diffusion factor value obtained by continuously reducing the discretization step length with reference to formula 2) is not obviously changed, (x) andi,yi) Represents the center coordinates after discretization of the active area, (x)k,yl) Representing the center coordinates after discretization of the area around the point of incidence. Calculating to obtain the diffusion factor value of each active region, wherein the diffusion factor is largerA large value represents a larger amount of charge collected by diffusion.
Step 2.9: diffusion collection current source item I when heavy ions are incident to different positionsdiffusioniCorresponding to Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni) A data set whose values vary with the variation of the diffusion factor value;
step 2.10: fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
and step 3: extracting layout shape information and circuit netlist information of circuit to be analyzed
Step 3.1: the circuit layout file is used as input, the coordinate of the junction of the N trap and the P trap and the trap contact edge coordinate information are screened and extracted, and the information is stored according to the splicing form of a plurality of rectangular areas as follows:
Nwell tap:(1435,4920)(1855,5030)；(2680,4695)(3100,5030)；(5415,4920)(5835,5030)；(8250,4920)(8670,5030)；(0,5030)(10560,5250)；(9740,4920)(10160,5030)；
Pwell tap:(970,210)(1390,320)；
step 3.2: the circuit layout file is used as input, the coordinate information of source and drain active regions of all transistors is screened and extracted, and the coordinate information is stored according to the splicing form of a plurality of rectangular regions as follows:
MN0 source(370,1450)(850,1910)；
MN0 drain(1030,1450)(1390,1910)；(1170,1230)(1390,1450)；(1170,930)(1490,1230)；(1170,770)(1390,930)；(970,320)(1390,770)；
......
step 3.3: taking a circuit layout file as input, screening and extracting a circuit netlist, extracting corresponding transistors one by one according to the sequence of the source level and drain level active region coordinate records, and recording node numbers one by one;
M0 9 1 3 9 nMOS
......
and 4, step 4: selecting the incident position of heavy ions in the circuit layout to be analyzed, and reconstructing the single event effect model of the circuit layout to be analyzed
Step 4.1: calculating to obtain a well contact area A according to the coordinates of the junction of the N well and the P well and the coordinate information of the well contact edge extracted in the step 3.1iAnd the spacing d between the well contact and the active regioniTaking values;
step 4.2: then the well contact area A obtained in the step 4.1 is usediAnd the spacing d between the well contact and the active regioniAnd the parameter value f established in step 2.21And f2Substituting the resistance values of the trap, which is contacted with each active region and between each active region and the heavy ion incidence position, into the formula (1) to obtain a resistance network of a single event effect model of the circuit layout to be analyzed;
step 4.3: calculating to obtain a heavy ion incidence point and a distance value between the junction of the N trap and the P trap according to the coordinate information of the junction of the N trap and the P trap extracted in the step 3.1;
step 4.4: and (3) determining a trap contact current source item I of the single event effect model of the circuit layout to be analyzed according to a formula (2) by combining the linear item and the power exponent value in each exponential function determined in the step (2.4) and the distance value of the heavy ion incidence point and the junction of the N trap and the P trap calculated in the step (4.3)well；
Step 4.5: calculating drift factors and diffusion factor values of all active regions according to formula (4) and formula (5) and according to the source and drain active region coordinate information of all transistors extracted in step 3.2, and further re-establishing a drift collection current source item I according to linear items and power index values in each exponential function established in step 2.7 and step 2.10driftiAnd diffusion collection current source item Idiffusioni；
Step 4.6: calculating a disturbance current source item I of the single event effect model of the synthesized circuit layout to be analyzed according to the calculated connection relation between the well potential of each active region and the electrical nodeSET1、ISET2Up to ISETm；
Step 4.6.1: according to the resistance network and the well contact current source item I constructed in the step 4.1 and the step 4.2wellCalculated to obtain eachReal-time well potential of the active region;
step 4.6.2: calculating the contribution of the parasitic transistor on an active region disturbance current term, wherein the specific calculation formula is as follows:
Iparasitic∝exp(Vbs·q/kT) (6)
wherein, VbsRepresenting the real-time voltage difference between the transistor substrate and the source;
step 4.6.3: adding and solving the disturbance current source items in the step 4.5 and the step 4.6.2 to obtain a value representing a disturbance current source item I of the single event effect model of the circuit layout to be analyzedSET1、ISET2Up to ISETm；
And 5: updating circuit netlist, simulating single event effect of calculating circuit
Step 5.1: on the basis of the circuit netlist extracted in the step 3.3, adding the single event effect model of the circuit layout to be analyzed obtained in the step 4 to obtain an updated circuit netlist;
step 5.2: performing transient circuit simulation (simulation software adopted here is SPICE), and recording the output of an actual circuit within 10ns after the heavy ion incidence moment;
step 5.3: and analyzing the circuit output, and recording whether state inversion occurs or disturbance pulse which can be recovered automatically after a period of time.
According to the steps, the influence of the change of the shape and the size of the active region on the sensitivity of the circuit single event effect can be calculated. Fig. 4 and 5 show a set of practical examples. FIG. 4 is a schematic diagram of a 40nm process node inverter, where the notation shows the P-well contact width, and the method steps according to the present invention are given in the example, the P-well contact width is reduced from 1.44 μm to 0.3 μm, and the time variation of the inverter output voltage is calculated when heavy ions are incident on the P-well region, as shown in FIG. 5.
As can be seen from fig. 5, as the well contact width gradually increases, the transient pulse width of the circuit single event effect increases. In addition, the instantaneous output (solid line) obtained by calculation according to the steps of the method provided by the invention is better in accordance with the result (discrete point) obtained by calculation of a device simulation tool, and the influence of the instantaneous output (solid line) on the single event effect sensitivity of the circuit can be reflected by using the modeling method provided by the invention when the layout design changes.
Claims (3)
1. A modeling method for considering the influence of the shape and the size of an active region in single event effect circuit simulation is characterized by comprising the following steps:
1) constructing a single event effect basic model considering the influence of the shape and the size of an active region
Constructing a single event effect basic model consisting of three parts, wherein the single event effect basic model comprises a resistance network consisting of resistances between a trap contact and each active region and between each active region and a heavy ion incidence position, and a trap contact current source item I between an N-type trap contact and a P-type trap contactwellAnd single event effect disturbance current source terms I numbered from 1, 2 to M of active regionsSET1、ISET2Up to ISETm(ii) a The resistor network comprises R1、R2Up to RnWhere n is the total number of resistors in the resistor network,
the resistance value in the resistance network is composed of a contact resistance and a body resistance, and the specific expression is as follows:
Ri＝f1/Ai+f2·di，i＝1,2,...,n (1)
wherein A isiDenotes the area of the well contact, diIndicating the spacing between the well contact and the active region; f. of1And f2Parameters to be extracted are obtained;
well contact current source term I between N-type well contact and P-type well contactwellThe method is described according to a double-exponential current source form, and the specific expression is as follows:
wherein, Ipeak,wellRepresenting the peak current, tr,wellDenotes the rise time of the current, tf,wellRepresenting the current fall time, τr,well、τf,wellCurrent rise and fall time constants; the expression has four items of parameters to be extracted, Ipeak,well、τr,well、τf,well、(tf,well-tr,well)；
Single event effect disturbance current source item I numbered from 1, 2 to M active regionsSET1、ISET2Up to ISETmIs to collect the current source term I according to the driftdrifti(t) and diffusion Collection Current Source term Idiffusioni(t) is described in a form of superposition, and the specific expression is as follows:
ISETi(t)＝Idrifti(t)+Idiffusioni(t)，i＝1,2,...,m (3)
wherein, drift collects current source item Idrifti(t) and diffusion Collection Current Source term Idiffusioni(t) are all in the form of dual-exponential current sources, and the specific expressions all follow expressions similar to the formula (2);
drift collection current source item Idrifti(t) and diffusion Collection Current Source term IdiffusioniThe parameters required to be extracted in the expression of (t) are respectively Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) And Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni)；
2) Calibrating the dependency relationship of parameters to be extracted in the single event effect basic model along with the change of layout information
2.1) extracting process structure information by referring to a circuit design process library model, and constructing a single-tube device model available for numerical simulation by adjusting normal electrical characteristic parameters by using a semiconductor device simulation tool; the normal electrical characteristic parameters comprise transistor threshold voltage, static current and doping region resistance values, and channel and source drain region doping information is calibrated;
2.2) by setting virtual nodes, obtaining the combinations of different trap contact areas and the distances between different trap contacts and the active regionResistance value of the resistor is fitted to obtain a parameter f in the formula (1)1And f2Taking the value of (A);
2.3) setting different positions of heavy ion incidence, and obtaining a trap contact current source item I through a formula (2)wellCorresponding to Ipeak,well、τr,well、τf,well、(tf,well-tr,well) The value of the data group changes along with the change of a set heavy ion incidence point and the change of a distance value at the junction of the N trap and the P trap;
2.4) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
2.5) calculating the drift factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
wherein (x)0,y0) Represents the incident position of heavy ions, (x)i,yi) Representing the central coordinate, R, after discretization of the active area0A characteristic radius representing the lateral expansion of the heavy ion trajectory;
2.6) calculating the current source term I collected by drift when heavy ions are incident at different positionsdriftiCorresponding to Ipeak,drifti、τr,drifti、τf,drifti、(tf,drifti-tr,drifti) A data set whose value changes with the change of the drift factor value;
2.7) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
2.8) calculating the diffusion factor of the active region aiming at the selected heavy ion incidence position, wherein the specific calculation formula is as follows:
wherein (x)0,y0) Represents the incident position of heavy ions, (x)i,yi) Representing the discretization of the active areaCenter coordinate (x)k,yl) Representing the central coordinate, R, of the region around the point of incidence after discretization0A characteristic radius representing the lateral expansion of the heavy ion trajectory;
2.9) calculating the current source item I of diffusion collection when heavy ions are incident at different positionsdiffusioniCorresponding to Ipeak,diffusioni、τr,diffusioni、τf,diffusioni、(tf,diffusioni-tr,diffusioni) A data set whose values vary with the variation of the diffusion factor value;
2.10) fitting the data set according to the exponential function, and respectively solving the values of a linear term and a power exponent in the exponential function;
3) extracting layout shape information and circuit netlist information of circuit to be analyzed
3.1) taking the circuit layout file as input, screening and extracting coordinates at the junction of the N well and the P well and well contact edge coordinate information;
3.2) taking the circuit layout file as input, screening and extracting the coordinate information of the source and drain active regions of all transistors;
3.3) taking the circuit layout file as input, screening and extracting a circuit netlist, extracting corresponding transistors one by one according to the sequence of the source-level and drain-level active region coordinate records, and recording node numbers one by one;
4) selecting the incident position of heavy ions in the circuit layout to be analyzed, and reconstructing the single event effect model of the circuit layout to be analyzed
4.1) calculating to obtain a trap contact area A according to the coordinates of the junction of the N trap and the P trap extracted in the step 3.1) and the trap contact edge coordinate informationiAnd the spacing d between the well contact and the active regioniTaking values;
4.2) and then the well contact area A obtained in the step 4.1)iAnd the spacing d between the well contact and the active regioniAnd the parameter value f established in step 2.2)1And f2Substituting the resistance values of the trap, which is contacted with each active region and between each active region and the heavy ion incidence position, into the formula (1) to obtain a resistance network of a single event effect model of the circuit layout to be analyzed;
4.3) calculating to obtain a distance value between a heavy ion incident point and the junction of the N trap and the P trap according to the coordinate information of the junction of the N trap and the P trap extracted in the step 3.1);
4.4) combining the linear terms and power exponent values in each exponential function established in the step 2.4) and the distance values of the heavy ion incidence point and the junction of the N trap and the P trap obtained by calculation in the step 4.3), establishing a trap contact current source term I of the single event effect model of the circuit layout to be analyzed according to the formula (2)well；
4.5) calculating drift factors and diffusion factor values of all active regions according to the formula (4) and the formula (5) according to the source and drain active region coordinate information of all the transistors extracted in the step 3.2), and further re-establishing a drift collection current source item I according to the linear item and the power index value in each exponential function established in the step 2.7) and the step 2.10)driftiAnd diffusion collection current source item Idiffusioni；
4.6) calculating a disturbance current source item I of the single event effect model of the synthesized circuit layout to be analyzed according to the calculated connection relation between the well potential of each active region and the electrical nodeSET1、ISET2Up to ISETm；
4.6.1) according to the resistance network and the well contact current source item I constructed in the step 4.2) and the step 4.2)wellCalculating to obtain the real-time well potential of each active region;
4.6.2) calculating the contribution of the parasitic transistor on the disturbance current term of the active region, wherein the specific calculation formula is as follows:
Iparasitic∝exp(Vbs·q/kT) (6)
wherein, VbsRepresenting the real-time voltage difference between the transistor substrate and the source;
4.6.3) adding the disturbance current source items in the step 4.5) and the step 4.6.2) to obtain a value representing the disturbance current source item I of the single event effect model of the circuit layout to be analyzedSET1、ISET2Up to ISETm；
5) Updating circuit netlist, simulating single event effect of calculating circuit
5.1) adding the single event effect model of the circuit layout to be analyzed obtained in the step 4) on the basis of the circuit netlist obtained in the step 3.3) to obtain an updated circuit netlist;
5.2) performing transient circuit simulation, and recording the output of an actual circuit within 10ns after the heavy ion incidence moment;
5.3) analyzing the circuit output, and recording whether state inversion occurs or disturbance pulse which can be recovered automatically after a period of time is generated.
2. The modeling method for considering the influence of the shape and the size of the active region in the single event effect circuit simulation according to claim 1, wherein the modeling method comprises the following steps: the semiconductor device simulation tool adopted in the step 2.1) is a TCAD simulation tool.
3. The modeling method for considering the influence of the shape and the size of the active region in the single event effect circuit simulation according to claim 1, wherein the modeling method comprises the following steps: and 5.1) adding a simulation tool of the single event effect model in the circuit netlist adopted by the step is SPICE, and the adopted programming language is Verilog-A language.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911058784.2A CN110909516B (en) | 2019-11-01 | 2019-11-01 | Modeling method considering influence of shape and size of active region in single event effect circuit simulation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911058784.2A CN110909516B (en) | 2019-11-01 | 2019-11-01 | Modeling method considering influence of shape and size of active region in single event effect circuit simulation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110909516A true CN110909516A (en) | 2020-03-24 |
| CN110909516B CN110909516B (en) | 2023-03-24 |
Family
ID=69815926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911058784.2A Active CN110909516B (en) | 2019-11-01 | 2019-11-01 | Modeling method considering influence of shape and size of active region in single event effect circuit simulation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110909516B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114818566A (en) * | 2022-05-26 | 2022-07-29 | 西北核技术研究所 | Modeling method for well potential modulation caused by single event effect |
| CN116306448A (en) * | 2023-05-17 | 2023-06-23 | 深圳安森德半导体有限公司 | Design method of digital-analog hybrid integrated circuit based on BCD (binary coded decimal) process |
| CN116882356A (en) * | 2023-09-06 | 2023-10-13 | 深圳铨力半导体有限公司 | Prediction method for influencing low-voltage signals by high-voltage level conversion in layout |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050127971A1 (en) * | 2003-12-12 | 2005-06-16 | Hoff James R. | Redundant single event upset supression system |
| US20080072199A1 (en) * | 2006-06-14 | 2008-03-20 | Kyoji Yamashita | Method for designing semiconductor integrated circuit |
| US20080077376A1 (en) * | 2006-09-25 | 2008-03-27 | Iroc Technologies | Apparatus and method for the determination of SEU and SET disruptions in a circuit caused by ionizing particle strikes |
| CN103577643A (en) * | 2013-11-06 | 2014-02-12 | 中国空间技术研究院 | SRAM type FPGA single event upset effect simulation method |
| CN106503392A (en) * | 2016-11-14 | 2017-03-15 | 哈尔滨工业大学 | A kind of many transient state soft-error sensitivity appraisal procedures of single-particle of the combinational logic circuit of consideration laying out pattern information |
| CN108508351A (en) * | 2018-03-30 | 2018-09-07 | 西北核技术研究所 | A kind of single-particle direct fault location emulation mode based on double-two fingers number current source |
-
2019
- 2019-11-01 CN CN201911058784.2A patent/CN110909516B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050127971A1 (en) * | 2003-12-12 | 2005-06-16 | Hoff James R. | Redundant single event upset supression system |
| US20080072199A1 (en) * | 2006-06-14 | 2008-03-20 | Kyoji Yamashita | Method for designing semiconductor integrated circuit |
| US20080077376A1 (en) * | 2006-09-25 | 2008-03-27 | Iroc Technologies | Apparatus and method for the determination of SEU and SET disruptions in a circuit caused by ionizing particle strikes |
| CN103577643A (en) * | 2013-11-06 | 2014-02-12 | 中国空间技术研究院 | SRAM type FPGA single event upset effect simulation method |
| CN106503392A (en) * | 2016-11-14 | 2017-03-15 | 哈尔滨工业大学 | A kind of many transient state soft-error sensitivity appraisal procedures of single-particle of the combinational logic circuit of consideration laying out pattern information |
| CN108508351A (en) * | 2018-03-30 | 2018-09-07 | 西北核技术研究所 | A kind of single-particle direct fault location emulation mode based on double-two fingers number current source |
Non-Patent Citations (3)
| Title |
|---|
| 周昕杰等: "TCAD结合SPICE的单粒子效应仿真方法", 《电子与封装》 * |
| 李若飞等: "单粒子仿真方法研究", 《微处理机》 * |
| 陈睿等: "单粒子效应TCAD数值仿真的三维建模方法", 《核电子学与探测技术》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114818566A (en) * | 2022-05-26 | 2022-07-29 | 西北核技术研究所 | Modeling method for well potential modulation caused by single event effect |
| CN114818566B (en) * | 2022-05-26 | 2024-06-11 | 西北核技术研究所 | Modeling method for well potential modulation caused by single event effect |
| CN116306448A (en) * | 2023-05-17 | 2023-06-23 | 深圳安森德半导体有限公司 | Design method of digital-analog hybrid integrated circuit based on BCD (binary coded decimal) process |
| CN116306448B (en) * | 2023-05-17 | 2023-08-11 | 深圳安森德半导体有限公司 | Design method of digital-analog hybrid integrated circuit based on BCD (binary coded decimal) process |
| CN116882356A (en) * | 2023-09-06 | 2023-10-13 | 深圳铨力半导体有限公司 | Prediction method for influencing low-voltage signals by high-voltage level conversion in layout |
| CN116882356B (en) * | 2023-09-06 | 2023-12-22 | 深圳铨力半导体有限公司 | Prediction method for influencing low-voltage signals by high-voltage level conversion in layout |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110909516B (en) | 2023-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110909516B (en) | Modeling method considering influence of shape and size of active region in single event effect circuit simulation | |
| Black et al. | HBD layout isolation techniques for multiple node charge collection mitigation | |
| US20080077376A1 (en) | Apparatus and method for the determination of SEU and SET disruptions in a circuit caused by ionizing particle strikes | |
| CN106503392B (en) | A kind of more transient state soft-error sensitivity appraisal procedures of single-particle for the combinational logic circuit considering laying out pattern information | |
| US20080281572A1 (en) | Integrated circuit (ic) design method and method of analyzing radiation-induced single-event upsets in cmos logic designs | |
| CN108363893B (en) | Single-particle pulse current source modeling method under complex condition | |
| Roche et al. | SEU response of an entire SRAM cell simulated as one contiguous three dimensional device domain | |
| Cao et al. | A layout-based soft error vulnerability estimation approach for combinational circuits considering single event multiple transients (SEMTs) | |
| CN110991072B (en) | SRAM single-particle transient effect simulation analysis method and system | |
| Balbekov et al. | Circuit-level layout-aware modeling of single-event effects in 65-nm CMOS ICs | |
| US8813006B1 (en) | Accelerated characterization of circuits for within-die process variations | |
| Yue et al. | Modeling and simulation of single-event effect in CMOS circuit | |
| Leveugle et al. | Laser-induced fault effects in security-dedicated circuits | |
| Correas et al. | Prediction of multiple cell upset induced by heavy ions in a 90 nm bulk SRAM | |
| Gerrer et al. | Interplay between statistical reliability and variability: A comprehensive transistor-to-circuit simulation technology | |
| US7200825B2 (en) | Methodology of quantification of transmission probability for minority carrier collection in a semiconductor chip | |
| Rao et al. | A detailed characterization of errors in logic circuits due to single-event transients | |
| Kauppila | Layout-aware modeling and analysis methodologies for transient radiation effects on integrated circuit electronics | |
| Azimi et al. | Micro latch-up analysis on ultra-nanometer vlsi technologies: a new monte carlo approach | |
| Stratigopoulos et al. | A general method to evaluate RF BIST techniques based on non-parametric density estimation | |
| US10339240B2 (en) | Adaptive high sigma yield prediction | |
| Asenov et al. | Unified approach for simulation of statistical reliability in nanoscale CMOS transistors from devices to circuits | |
| Beroulle et al. | Laser-induced fault effects in security-dedicated circuits | |
| US8429592B2 (en) | N/P configurable LDMOS subcircuit macro model | |
| Malherbe et al. | Investigating the single-event-transient sensitivity of 65 nm clock trees with heavy ion irradiation and Monte-Carlo simulation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |