Study of PSpice-models of analog radioelements. Encyclopedia of radio electronics and electrical engineering

Encyclopedia of radio electronics and electrical engineering / Microcontrollers

 Comments on the article

In his article ("PSpice models for simulation programs"in "Radio" No. 5-8, 2000) the author spoke about the rules for constructing models of analog components for modeling programs based on the PSpice language. The proposed article continues this topic. It is devoted to methods for studying PSpice models and methods for constructing component models for This is very important, since only the use of reliable models of components allows one to obtain adequate simulation results.

Sooner or later, every radio amateur comes to the conclusion: before installing the radio element on the board during the manufacture of the device, you should first check its serviceability. This will save the device in the future from failure after power is applied or from lengthy searches for the cause of its inoperability. For this purpose, industrial enterprises organize partial or complete incoming control of radio elements, which is much easier than maintaining a large staff of highly qualified and highly paid equipment adjusters.

The approach should be similar when modeling electronic circuits. Using unverified models leads to wasted time looking at graphs that have nothing to do with reality. In this case, you can make a false conclusion about the health or inoperability of the device and make the wrong decision. Therefore, input control should be organized here as well. In the future, this will pay off in saving time and the reliability of the simulation results.

Sources for replenishing personal libraries can be models included in the libraries of the simulation software package used, from the libraries of other but compatible simulation programs - models that are abundantly presented on the Internet on the websites of firms of developers of simulation programs and manufacturers of electronic components, published in printed publications, and self-developed models. At the same time, one can only guess about their quality. Before using these models, it is desirable to test them. It is with this approach that there is confidence in the results obtained. It becomes clear - what can be and what cannot be.

The proposed article describes some methods for testing models of discrete analog radio elements, provides measurement schemes and texts of modeling tasks in the PSpice format. Tasks are configured for specific models of radio elements, the testing of which is described in the article. If any other elements are to be tested, the programs should be improved. It is not difficult. As a rule, all improvements come down to changing the limits for changing currents, voltages, analysis time, choosing a load, setting the required mode of the component model for direct current. If you get creative, some tests can be used to develop new tests for other models, including complex macro models.

DIODE RECOVERY MEASUREMENT

To evaluate the dynamic properties of a diode model, its reverse recovery time should be measured. Let's do this using the example of a model of a rectifier diode KD212A. It is known that after changing the polarity of the voltage applied to a real diode from direct to reverse, it closes not instantly, but with some delay. In this case, a large current can flow through the diode in the opposite direction for some time. For KD212A according to the reference book [1], the reverse recovery time is guaranteed at Uobr=200 V, Ir=2 A, not more than 300 ns.

Now let's check the model of this diode. Let's create measurement conditions close to those under which the parameters of the KD212A diode are given in the reference book. To do this, let's apply to the diode model (Fig. 1, Table 1) a multipolar voltage pulse with an amplitude of 200 V through a resistor with a resistance of 100 Ohm.

Let's start the simulation process and see how the diode current will change (Fig. 2).

Indeed, there is a characteristic surge of current in the opposite direction on the graph. Its duration is the reverse recovery time. The current peak when the diode is turned on is explained by the recharging of its barrier capacitance. Model diode current is measured in amps and voltage is measured in hundreds of volts. In order to build two curves (current and voltage) on one graph, the voltage should be divided by 100 using the graphics processor. It can be seen from the graphs that the reverse recovery time is approximately 33 ns. The results correspond to reality, although the reverse recovery time is much less than the passport 300 ns.

Here, in general, the problem of using information from domestic reference books to build models is clearly manifested. As a rule, all parameters set either "no more" or "no less" cannot be used to build mathematical models, since they mainly reflect the desire of developers to play it safe. Therefore, it is better to try to use models created by manufacturers, or to carry out some kind of independent measurements.

If this diode is used, for example, in a rectifier, then the presence of such surges leads to an increase in switching noise. This is usually dealt with by connecting a shunt capacitor in parallel with the diode (Fig. 3).

Let's see what it gives (Fig. 4).

It can be seen that the situation is changing, but not drastically. Obviously, the failure when switching to the direct state is associated with the recharging of the capacitor C1. The task for modeling (Table 2) is composed of two included one after the other.

The second task is just a copy of the first, to which the capacitor C1 is then added, connected in parallel with the diode. It is convenient to do this, since all graphs after the calculation will be shown simultaneously.

VOLT-FARAD CHARACTERISTICS OF THE VARICAP MODEL

Another important characteristic of a diode is the dependence of the capacitance of the p-n junction on the voltage applied in the opposite direction. For devices such as varicaps, this is the main dependence. Let's build the capacitance-voltage characteristic for the 2V104A varicap model. Let us apply to the diode model (Fig. 5) a voltage linearly increasing at a rate of 10 V/μs with an amplitude of 50 V applied in the opposite direction. In this case, the p-n junction will be closed, and the current through the diode, due to the very large reverse resistance, will practically be purely capacitive and will be determined by the equation ld \u10d CdV'(t), where V'(t) is the rate of voltage increase (107 V /μs=XNUMX V/s).

We solve this equation for Сd, we get Сd=Id/V'(t).

From here we get the formula for the capacitance of the diode: Cd \u107d Id / XNUMX.

Or finally, taking into account the dimension, Sd (pF) \u0,1d XNUMX Id (μA).

Let's compose and run the simulation task (Table 3), then let's see how the diode current will change with time (Fig. 6).

The current will be very small, and to see it at the same time as the voltage, its values ​​\u1000b\u1bmust be multiplied by the GPU by 10. Since the dependence of the applied voltage on time is linear, we will replace the time on the X axis with the voltage of the source V7. Then we divide the current values ​​​​by XNUMX. As a result, we obtain the capacitance-voltage characteristic of the diode (Fig. XNUMX), where along the axis the current value in microamps will be numerically equal to the capacitance of the diode in picofarads.

The handbook [1] indicates that with a reverse voltage of 4 V, the capacitance of the varicap is in the range from 90 to 120 pF. According to the graph for the model, we get 108 pF. And this suggests that the model under study in this parameter corresponds to the properties of a real varicap.

SATURATION CHARACTERISTICS OF THE BIPOLAR TRANSISTOR MODEL

When designing contactless switches, it is important to know the saturation mode characteristics of the transistor. These parameters are decisive for the selection of a switching transistor in pulse converters and load switching devices.

For such a device to have a high efficiency. The switching transistor must either be in the Fully Open or Fully Closed state and switch from one state to the other as quickly as possible. In the Fully Open state, the transistor should be saturated. The power dissipated on it is determined by the product of the collector current and the saturation voltage of the collector-emitter section at a given collector current, plus some additional power, determined by the base current, which is required to maintain the transistor in saturation. It is equal to the product of the base saturation voltage and the base current. Sometimes the additional power spent on driving the transistor is quite significant. This is a significant disadvantage of bipolar transistors.

In reference books, saturation voltage is interpreted ambiguously. Usually it is indicated at a certain base and collector current, or graphs of the saturation voltage (Ukenas and Ubenas) on the base current at a fixed collector current are plotted, or dependences of Ukenas and Ubenas on the collector current are plotted with a saturation coefficient of Knas=10 for low-power transistors (for powerful ones - Knas= 2).

Let us construct the dependence of the saturation voltage of the collector-emitter and base-emitter on the base current for the model of a powerful bipolar transistor KT838A, widely used in pulsed secondary power supplies, the parameters of which largely depend on the quality indicators of the switching transistor. Reference [2] lists its parameters: Ubenas (at Ik=4,5 A; Ib=2 A) - no more than 1,5 V; Ukenas (at Ik=4,5 A; Ib=2 A; T=+25 °C) - no more than 1,5 V; Ukenas (at Ik = 4,5 A; Ib = 2 A; T = -45 ° C and T = + 100 ° C) - no more than 5 V.

Using the measurement scheme (Fig. 8, Table 4), we calculate these dependencies.

The results obtained (Fig. 9) do not contradict the reference data. Obviously, a sharp increase in the collector-emitter voltage with a decrease in the base current is due to the exit of the transistor from the saturation mode.

Now let's build the dependence of the saturation voltage of the collector-emitter and base-emitter of models of powerful bipolar transistors KT838A and more modern KT8121A2 on the collector current at a fixed saturation factor equal to two. In the handbook [2] for the KT838A transistor, unfortunately, there is no such characteristic, but there is for KT8121A2. Let's compare transistor models by this indicator.

Using the measurement circuit (Fig. 10), we take the ratio of the collector current to the base current equal to two, using for this a dependent current source controlled by the current F1 with a transfer coefficient of 0,5.

The control will be the current through the voltage source V1 with zero voltage (this is the requirement of PSpice). By varying the source current I1 in the range from 0,1 to 10 A (and hence the base current from 0,05 to 5 A), we calculate how the voltage at the base and collector of the transistor will change. Let's use the capabilities of the .DC directive for this.

The task for modeling (Table 5) consists of two, connected in series one after another, for the KT838A and KT8121A2 transistors. In this case, the characteristics of both devices will appear simultaneously on one screen (Fig. 11).

It can be seen from the graphs that the KT8121A2 transistor has better characteristics in saturation mode than the KT838A. With a collector current of 4,5 A, the saturation voltage of the KT838A collector-emitter is about 2,1 V, and the KT8121A2 is about 0,5 V. Thus, it is preferable to use the KT8121A2 transistor to build powerful switches, since less power will be dissipated on it.

VOLT-AMPERE CHARACTERISTICS OF A POWERFUL FIELD TRANSISTOR MODEL

Tables of analogues of domestic and imported transistors are given in abundance in various printed sources and on the Internet. A quite obvious question arises - is it possible to use analogue models by assigning them the names of domestic transistors? In table. 6 shows imported analogues of powerful field-effect transistors. This table is good because the models of many analogues can be found in the OrCAD-9.2 libraries. Such transistors are mainly used in switching power supplies for televisions, VCRs, and monitors.

The author was interested in the KP805A transistor, since the BUZ2541 transistor failed in the power supply of his SONY KV-E90 TV. Let's try to compare at least approximately the main parameters of KP805A with the characteristics of models of imported analogues from the table. The MTP6N60E transistor model was found on the tntusoft website, the BUZ90 transistor model was found in the siemens.lib library, and the IRFBC40 transistor model was found in the pwmos.lib library. Despite the fact that transistors are presented in the table as analogues, their models look very different.

The MTP6N60E and BUZ90 transistor models are represented by very complex macro models (Fig. 12, Fig. 13), and the IRFBC40 transistor model is the simplest, built on the basis of the built-in model. Let's see, at the same time, how this will affect their parameters.

(click to enlarge)

First, let's build a family of output current-voltage characteristics of the models of these transistors connected according to a common-source circuit (Fig. 14).

The output characteristic of a field-effect transistor is the dependence of the drain current on the drain voltage at a fixed gate voltage. A family of output characteristics is formed by plotting graphs for several values ​​of the gate voltage. Let's create a task for modeling (Table 7) and run it. As the gate voltage varies, the curve will characteristically change (Fig. 15 - 17), forming a family of output parameters.

To plot the characteristics of different transistors, you should manipulate the "*" (asterisk) sign in the program in the connection lines of transistor models. Comparing the dependences, it can be noted that the MTP6N60E transistor model has a lower amplification (at least twice) and reflects the electrical breakdown phenomenon at the declared voltage Uc and max=600 V, while in the IRFBC40 transistor model, the electrical breakdown phenomenon does not appear. In the sense of taking into account the phenomenon of electrical breakdown, the first model is more in line with reality. However, it is too early to state that the models of these transistors give close characteristics. The only thing they have in common is that with the declared current Ic = 6 A and voltage U3i = 10 V, their drain-source voltages are approximately equal, amounting to approximately 6 V for the MTP60N5,6E, and about 40 V for the IRFBC5,8.

The BUZ90 transistor model from the siemens.lib library, apparently, is not very successful and is normally calculated when the drain voltage changes only up to 100 V. If you expand the interval above 120 V, you cannot obtain normal output characteristics (Fig. 17), and the calculation process is very drags on in time. And this is despite the fact that the model is included in the proprietary siemens.lib library, which comes with the OrCAD distribution. The use of such a model in the future may lead to problems with obtaining results. It is customary to believe in branded libraries, so it will not be easy to explain the behavior of the simulated device. This suggests the conclusion that any model, even from a reliable source, must be tested before being used.

Let us now build the transient characteristics of the MTP6N60E, IRFBC40, BUZ90 transistors. The measurement scheme is shown in fig. 14, and the task for modeling - in table. 8.

(click to enlarge)

Let's differentiate these dependencies and get graphs of the slope change (Fig. 18 - 20).

At a current of 2 A, we have S(MTP6N60E)=3000 mA/V; S(IRFBC40)=2040mA/V; S(BUZ90)=2050 mA/V. According to the handbook [2], KP805A has a characteristic slope of 2500 mA/V. The values ​​seem to be close. But that's only at one point!

What conclusions can be drawn from this? Judging by the current-voltage characteristics of the MTP6N60E, IRFBC40, BUZ90 transistor models, it is difficult to assume that these are the same devices. However, the real experience of replacement during equipment repair confirms their interchangeability in switching power supplies. As for the use of analogue models as a model of the domestic KP805A transistor, this cannot be done directly, since there is a significant difference in their current-voltage characteristics.

The MTP6N60E and IRFBC40 transistor models proved to be efficient and, in general, reflect the properties of some typical high-power MOS transistors and are suitable for simulation. It is their models, as the most successful ones, that can be used in the future as prototypes for creating models of domestic field-effect transistors. The simplest way is to select model parameters with subsequent testing and comparison with the characteristics of a real device from a reliable reference. A simple KP805A model (using the IRFBC40 model as a prototype) can be created using the PART MODEL EDITER program, which is part of the OrCAD package. And if you take into account the electrical breakdown in it by connecting the diode, you get a completely "workable" model.

DEPENDENCE OF THE CHANNEL RESISTANCE OF A FIELD-FET TRANSISTOR MODEL ON THE GATE VOLTAGE

By analogy with the previous example, we construct the output current-voltage characteristics of the KP312A transistor (Fig. 21, Table 9). It can be seen from the graphs that field-effect transistors have a controlled resistance region that is very symmetrical about zero at a low drain voltage |Usi |<|Usu us | /2.

The FET channels behave almost like linear resistors, the resistance of which depends on the gate voltage. If the polarity of the drain voltage is reversed, the linearity of the resistor is not violated. Therefore, on a field-effect transistor, it is possible to implement a variable electrically controlled resistor operating on direct and alternating current. This interesting property is often used in various automatic control systems. However, it should be remembered that for field-effect transistors with a control p-n junction, the condition |Uzi|<|Usi|+0,5 V must be met. Otherwise, when exposed to reverse drain voltage, the section of the control p-n junction near the drain will be so open that in In the drain circuit, significant gate forward current will flow, destroying the linearity of the resistor. The forward voltage on the silicon pn junction, not exceeding 0,5 V, does not create a significant forward current.

In this regard, the dependence of the transistor channel resistance on the gate voltage is of interest. Let's build it. The peculiarity of such an experiment is that it is impossible to display the graph of the resistance of the field-effect transistor channel directly on the screen of the PSpice graphic postprocessor, but you can get its electrical equivalent. Divide the drain voltage by the drain current RDS=UD(J2)/ID(J2) to get the resistance. This method is universal and can be used to measure resistance in other models, including macro models. Thus, you will need a voltage divider with an A / V function and a current-to-voltage converter.

Now we will draw up a measurement scheme (Fig. 22). The current-voltage converter, made on the basis of a voltage source controlled by the current H1 (INUT), is connected by the measuring input in parallel to the zero voltage source, which is connected to the drain circuit of the field-effect transistor. This is PSpice's requirement when measuring current. By changing the gate voltage (voltage source V1) and setting different values ​​of the drain voltage (voltage source V3), we obtain the corresponding family of channel resistance characteristics of the KP312A field-effect transistor (voltage divider output A / B).

When compiling a task for modeling (Table 10), let's design the divider (Fig. 23) as a separate macromodel .SUBCKT DIVIDE A B A/B, where A and B are the inputs of the divider; A/B is its output. This will allow us to reuse the divider in various experiments in the future.

We will measure the resistance in the transient analysis mode according to the .TRAN directive. In this case, the voltage of the source V1 will increase in proportion to time and, accordingly, the drain current of the transistor. The drain voltage according to the directive .STEP V3 LIST -0.5 0.5 1 1.5 2 will change according to the list specified in it in the region of controlled resistance (see Fig. 21).

We apply the drain voltage to input A of the divider, and the voltage from the INUT output, proportional to the drain current, to input B. At the output of the divider, we get a voltage proportional to the resistance of the field-effect transistor channel. In this case, the voltage in volts corresponds to the resistance in ohms, and in kilovolts - to the resistance in kiloohms.

By running the simulation task, we obtain the required family of characteristics (Fig. 24).

From the graphs, it can be seen that the channel resistance increases as the gate voltage approaches the cutoff voltage, which for this model is -5 V. And this is understandable, because the transistor turns off. In the range from 0 to -1,5 V, a relatively linear region of resistance change can be distinguished. The drain voltage also affects the channel resistance, with an increase in the drain voltage, it increases. This is in good agreement with the theoretical and practical characteristics of field-effect transistors [3, 4]. In some reference books, instead of resistance graphs, conductivity dependences are given. Obviously, if we swap the inputs A and B of the divider, we will get conductivity graphs.

DEPENDENCE OF THE RESISTANCE OF THE CHANNEL OF THE FIELD-FIETD TRANSISTOR MODEL ON THE DRAIN CURRENT

Using the previous experiment, we plot the dependences of the channel resistance of the field-effect transistor model on the drain current. Let's draw up an appropriate measurement scheme (Fig. 25). Here everything is the same as in the previous case, only we will include a source of linearly increasing current I1 in the drain circuit.

Resistance measurement is carried out in the transient analysis mode according to the .TRAN directive. In this case, the current of the current source I1 will increase in proportion to time and, accordingly, the drain current of the field-effect transistor. Of course, the drain voltage will also change. We apply the drain voltage to input A of the divider, and the voltage from the INUT output, proportional to the drain current, to input B. At the output of the divider, we get a voltage proportional to the resistance of the field-effect transistor channel. Voltage in volts corresponds to resistance in ohms, and in kilovolts to resistance in kiloohms.

By running the simulation task (Table 11), we get the curves (Fig. 26) - this is the desired result.

It can be seen from the graphs that with an increase in the closing voltage at the gate of the field-effect transistor, the channel resistance increases, obviously, as it should be. At the same time, in the gate voltage range from 0 to -0,5 V, it practically does not depend on the drain voltage, so the FET channel under such conditions behaves like a linear resistor.

NOISE CHARACTERISTICS OF THE FIELD TRANSISTOR

When designing amplifying devices, it is important to take into account the noise properties of the components, since after amplification it is necessary to obtain a good signal-to-noise ratio. It is known that active elements make the main contribution to noise. The noise of the amplifying device will turn out to be small if the least noisy active device is installed in the first stage. Field-effect transistors are often used for these purposes.

The inherent noise of a field-effect transistor can be conditionally divided into thermal, excess and shot. Thermal noise is caused by the chaotic movement of charge carriers, creating current and voltage fluctuations. At medium operating frequencies of the FET, this noise source is the main one.

Excessive noise (or 1/f noise) dominates in the low frequency region, its intensity increases approximately inversely with frequency. The source of this noise is arbitrary local changes in the electrical properties of materials and their surface states. It largely depends on the perfection of technology and the quality of raw materials, but it cannot be completely eliminated in principle. For modern field-effect transistors with a control p-n junction, excess noise exceeds thermal noise only at frequencies below 100 Hz, for MOS transistors it is more intense and begins to noticeably manifest itself from frequencies below 1 ... 5 MHz.

Shot noise is generated by gate leakage current. For field-effect transistors, it is relatively small, so it is usually not taken into account, however, at high frequencies, when the gate capacitance begins to play a significant role, it can be noticeable.

Let's give an example of comparing the noise properties of models of field-effect transistors with a control pn junction: Japanese J2N3824 and domestic KP312A. In the measurement circuit (Fig. 27), the transistor is connected to a common source and operates on a load with a resistance of 1 kOhm.

Using the capabilities of the .AC and .NOISE directives, we will compose a modeling task (Table 12), with the help of which we will calculate the spectral density of the output noise voltage Su out (f), V2 / Hz.

From the graphs (Fig. 28) it can be seen that the transistors are close in noise properties, therefore, from this point of view, the KP312A transistor is a full-fledged replacement for the J2N3824.

When calculating the internal noise level, the names of the output variables have a standard form:

• INOISE - equivalent level of noise voltage or current at the input, equal to (Sin equiv(f))1/2;
• ONOISE - noise voltage level at the output, equal to (Su out(a))1/2;
• DB(INOISE) - equivalent level of noise voltage or current at the input in decibels;
• DB(ONOISE) - output noise voltage level in decibels.

In the Probe program, the square root of the voltage and current spectral density of the internal noise is displayed as V(INOISE), I(INOISE), V(ONOISE).

In order to plot both curves on the same graph, it is easiest to put two tasks one after the other in the modeling task by simply copying through the buffer and substitute the name of the model of interest in each part.

OUTPUT VOLT-AMPERE CHARACTERISTICS OF BSIT

MOSFETs have characteristics close to ideal for a switch, for which they are widely used. However, in modern power conversion devices, the requirements for switches are very stringent. They must operate at high frequency, at high current, and be economical. The main disadvantage of MOSFETs is the relatively low allowable drain-to-source voltage. In addition, the resistance of an open transistor increases in proportion to the square of this voltage. In the best instances of high-power high-voltage field-effect transistors, the saturation voltage at rated current reaches several volts, respectively, they dissipate more power. In this regard, bipolar transistors are significantly superior to field ones.

Of course, the idea arose to combine the properties of these devices in one package. As a result, a MOS controlled bipolar transistor was created, called IGBT (Insulated Gate Bipolar Transistor - insulated gate bipolar transistor). In domestic literature, it is called BSIT - bipolar statically induced transistor.

Structurally, the LSIT is a bipolar transistor, which is controlled by a low-voltage MOSFET (Fig. 29). The result is a device that combines the advantages of field-effect and bipolar transistors. LSITs have practically no input current, they have excellent dynamic characteristics up to frequencies of 20...50 kHz. Losses in them grow in proportion to the current, and not to the square of the current, as in field-effect transistors. The maximum voltage on the LSIT collector is limited only by technological breakdown.

Today, BSITs are produced for a rated voltage of 2000 V or more. At rated current, their saturation voltage does not exceed 2 ... 3 V. In table. 13 shows the electrical characteristics of some common BLIT transistors, and for comparison, the last line shows the parameters of a powerful BUZ384 field-effect transistor.

Let's build a family of output characteristics of models of a bipolar statically induced transistor APT30GT60 and a powerful field-effect transistor BUZ384:

On fig. 30, 31 shows the measurement schemes, and in table. 14, 15 the text of the task for modeling is given. The gate voltage of transistors is a parameter that forms the CVC family. It is changed in the range from 4,5 to 6 V in increments of 0,5 V, and the voltage on the collector (and, accordingly, the drain) is in the range from 0 to 50 V.

As a result, we obtain the output characteristics of the APT30GT60 LSIT model (Fig. 32) and the BUZ384 field effect transistor model (Fig. 33).

The graphs show that the models really reflect the properties of real devices and demonstrate the superiority of LSIT over field-effect transistors when both devices operate in switching mode. So at a current of 10 A, the saturation voltage for the APT30GT60 LSIT is approximately 2,4 V, and for the BUZ384 field-effect transistor it is 5,6 V. The values ​​differ by about 2,3 times, respectively, in the open state at a current of 10 A, the APT30GT60 transistor will dissipate 2,3 times less power.

SWITCHING CHARACTERISTICS OF BSIT

Often, bipolar statically induced transistors are used to operate in switching modes. Let's check (Fig. 34) how it works with an inductive load.

We will apply a trapezoidal pulse with a steep front and a gentle decay to the input. The task for modeling is given in Table. 16, and the results are shown in fig. 35.

The resulting graphs once again confirm that a transistor operating on an inductive load should be selected with a voltage margin.

CREATING MICROWAVE MODELS OF COMPONENTS

PSpice models of electronic components can be divided into static and dynamic, low-frequency and high-frequency, for small and large signals. Such a classification makes it possible to organize a hierarchical series of models that differ in computational costs and allow the transition from one model to another in the course of modeling. Obviously, the most accurate and versatile in this series is the dynamic high-frequency model of a large signal.

Dynamic models of a large signal are described by nonlinear equations and require increased computational time. In PSpice, such models are used mainly for calculating DC modes and analyzing transients.

Models for small signals are much simpler. They are described by linear equations. Usually they are used in calculations when small increments of the signal are applied, when the CVC of the device can be considered linear in the vicinity of the operating point. In PSpice, such models are used for calculations in the frequency domain, as well as for calculating the sensitivity and transfer functions for direct current at small signals.

Built-in PSpice models of passive and active components - dynamic large signal models. They are valid for not very high frequencies. However, radio amateurs have long mastered the microwave range, so it is quite logical to learn how to create models of discrete components that "operate" at higher frequencies - high-frequency dynamic models of a large signal.

Calculations at frequencies above 100 MHz require taking into account various parasitic effects (lead inductance, lead-to-lead capacitance, etc.). For discrete resistors of small resistance, it is necessary, first of all, to take into account the inductance of the leads. In the first approximation, it can be calculated by the formula Lv \u2d 4h[ In (0,75h / d) -1], where h and d are the lead length and diameter, respectively, in cm, Lv is the lead inductance, in nH. Often, in calculations, it is assumed that the linear inductance of the leads is approximately equal to 200 nH / mm. At frequencies above 10 MHz, the inductive reactance of the leads is more than 36 ohms, which can be significant if the nominal resistance of the resistor is small. For resistors of high resistance, the parameters are seriously affected by the inter-terminal capacitance St. The complete high-frequency model of a discrete resistor is shown in fig. XNUMX.

Parasitic capacitance must be taken into account in the film resistors of hybrid circuits and in the diffusion resistors of integrated circuits at high frequency. If the diffusion resistor is isolated by a p-n junction, this is the non-linear capacitance of the isolating junction. In this case, at elevated temperatures, it may also be necessary to take into account the reverse current of the transition. Finally, in some cases, one should also take into account the rectifying properties of the transition, if at some moments it can open.

Strictly speaking, at high frequencies, the resistor behaves like a distributed RC line. However, it is hardly advisable to use multisection models of long lines. Very good - concentrated U-shaped model (Fig. 37, 38). Here C is the total capacitance of the insulation. It is divided into two half-capacitor capacitors. Diodes D1 and D2 are the same. The area of ​​each of them is equal to half the area of ​​the insulating pn junction. P - output of the substrate.

In high-frequency models of discrete capacitors, one should take into account the loss resistance r and the lead inductance Le, and in some cases, when the capacitor is used in timing circuits, also the leakage resistance Ry (Fig. 39) In integrated circuits, capacitors are usually implemented by reverse-biased p-n junctions . When modeling them, you should use diode models.

In the high-frequency model of a discrete inductor, it is necessary to take into account the active resistance of the winding r and the interturn capacitance C (Fig. 40).

Built-in transistor models are usually valid up to frequencies of 30 ... 100 MHz. On fig. 41 shows the equivalent circuit of a non-linear high frequency model of a bipolar transistor. Here C1-C3, R1-R3 are the equivalent capacitance and leakage resistance between the terminals of the transistor. These elements are included only if the transistor is made in a housing. LE0, LC0, LB0 - equivalent inductance of the outputs, respectively, of the emitter, collector and base. They are calculated using the above formula for calculating the output inductance of a discrete resistor.

At frequencies of several hundred megahertz, at least the inductance LE0 must always be taken into account, since at high current the emitter resistance of the transistor is about the same or even less.

LE and LB, which make up nano-henry units, are the inductance of the internal conductors connecting the emitter and base to the external leads. CCE and CCB - internal capacitance between the contact pads, respectively, of the emitter and base and the collector contact.

Such equivalent circuits, which take into account high-frequency effects, are designed as a macro model and used instead of conventional component models. I believe that readers who are familiar with the article "Pspice-models for simulation programs" in "Radio" No. 5-8 for, 2000, will not be difficult to write the texts of macro-models of such components on their own. In table. 17, as an example, shows a macro model of the microwave transistor NE68135 from CEL, valid up to a frequency of about 5 GHz.

Literature

1. Semiconductors: Diodes. Directory. Ed. N. N. Goryunova. - M.: Energoatomizdat, 1985.
2. Semiconductors: Transistors of medium and high power. Directory. Ed. A. V. Golomedova. - M.: Radio and communication, 1989.
3. Ignatov A. N. Field-effect transistors and their application. - M.: Radio and communication, 1984.
4. Lobachev LN Field-effect transistors. - M.: Radio and communication, 1984.

Author: O. Petrakov, Moscow

See other articles Section Microcontrollers.

Read and write useful comments on this article.

<< Back

Latest news of science and technology, new electronics:

A New Way to Control and Manipulate Optical Signals 05.05.2024

The modern world of science and technology is developing rapidly, and every day new methods and technologies appear that open up new prospects for us in various fields. One such innovation is the development by German scientists of a new way to control optical signals, which could lead to significant progress in the field of photonics. Recent research has allowed German scientists to create a tunable waveplate inside a fused silica waveguide. This method, based on the use of a liquid crystal layer, allows one to effectively change the polarization of light passing through a waveguide. This technological breakthrough opens up new prospects for the development of compact and efficient photonic devices capable of processing large volumes of data. The electro-optical control of polarization provided by the new method could provide the basis for a new class of integrated photonic devices. This opens up great opportunities for ... >>

Primium Seneca keyboard 05.05.2024

Keyboards are an integral part of our daily computer work. However, one of the main problems that users face is noise, especially in the case of premium models. But with the new Seneca keyboard from Norbauer & Co, that may change. Seneca is not just a keyboard, it is the result of five years of development work to create the ideal device. Every aspect of this keyboard, from acoustic properties to mechanical characteristics, has been carefully considered and balanced. One of the key features of Seneca is its silent stabilizers, which solve the noise problem common to many keyboards. In addition, the keyboard supports various key widths, making it convenient for any user. Although Seneca is not yet available for purchase, it is scheduled for release in late summer. Norbauer & Co's Seneca represents new standards in keyboard design. Her ... >>

The world's tallest astronomical observatory opened 04.05.2024

Exploring space and its mysteries is a task that attracts the attention of astronomers from all over the world. In the fresh air of the high mountains, far from city light pollution, the stars and planets reveal their secrets with greater clarity. A new page is opening in the history of astronomy with the opening of the world's highest astronomical observatory - the Atacama Observatory of the University of Tokyo. The Atacama Observatory, located at an altitude of 5640 meters above sea level, opens up new opportunities for astronomers in the study of space. This site has become the highest location for a ground-based telescope, providing researchers with a unique tool for studying infrared waves in the Universe. Although the high altitude location provides clearer skies and less interference from the atmosphere, building an observatory on a high mountain poses enormous difficulties and challenges. However, despite the difficulties, the new observatory opens up broad research prospects for astronomers. ... >>

Random news from the Archive Comparator Texas Instruments TLV3691IDCKR 09.02.2015 The new TLV3691IDCKR comparator from Texas Instruments is a Nanopower comparator, drawing an extremely low current of 75 nA and yet capable of operating at supply voltages as low as 0,9V. for portable battery devices. The advantage of the comparator is a wide range of input signals - their value can be 100 mV higher than the levels of the power lines. The TLV3691IDCKR can be used as a zero-crossing detector, as a windowed comparator (2xTLV3691), as an over-current sensor, and for over- and under-power protection in the device. The TLV3691 has an internal hysteresis of 17 mV to reduce the sensitivity of the comparator to the noise of the monitored signal. An additional increased hysteresis can be arranged in the classical way using a bias resistor. The TLV3691 comparator operates in a wide industrial temperature range of -40...125°C.

Other interesting news:

▪ 24-bit 256-channel ADC for ADAS1131 tomographs

▪ Smartphone Micromax Canvas XP 4G

▪ Electric cars in driving schools

▪ Human stomach grown in vitro

▪ Electronics work inside the body

News feed of science and technology, new electronics

Interesting materials of the Free Technical Library:

▪ site section Low frequency amplifiers. Article selection

▪ article Disadvantages and defects of vision. Encyclopedia of visual illusions

▪ How did the educational and scientific processes take place in Medieval universities? Detailed answer

▪ geranium article. Legends, cultivation, methods of application

▪ Article Open! IR! Encyclopedia of radio electronics and electrical engineering

▪ article It's not just about the raincoats. physical experiment

Leave your comment on this article:

All languages ​​of this page

Arabic	Hebrew	Polish
Bengali	Hindi	Portuguese
Chinese	Italian	Spanish
English	Japanese	Turkish
French	Korean	Ukrainian
German	Malay	Vietnamese

Home page | Library | Articles | Website map | Site Reviews

www.diagram.com.ua
2000-2024