RU2282861C1 - Device for registration of pulse electric signals - Google Patents

Abstract

FIELD: measurement engineering.

SUBSTANCE: device has information retard microstrip, made onto dielectric plate having relative dielectric constant of ε1. Taps of microstrip are connected with sample and store elements through switches. Control inputs of switches are connected with strobe-signal microstrip through communication elements. Strobe-signal microstrip is made onto dielectric plate having relative dielectric constant of ε2; ε1≫ε2. Plates are disposed one above the other. Sides of plates, having continuous metal coating, are connected together electrically and mechanically. Communication elements are made in form of slots brought into coincidence at back metal coating of any plate. Slots in strobe-signal microstrip are made directly above line of strobe-signal. Slots are made under row of contact areas in information retard microstrip, which plates are disposed at ΔY distance from information retard microstrip; the distance is specified by mounting dimensions of switches.

EFFECT: widened bandwidth; increased digitization frequency; improved manufacturability.

2 dwg

Description

The proposed device relates to the field of measurement technology.

Measurement of the shape of single pulsed signals of nano- and subnanosecond duration can be carried out both by traditional methods (for example, using high-speed oscillographic recorders, or by more modern digital methods - direct analog-to-digital conversion (ADC) and time-scale conversion (MVP) ([1 , 2] - development of the applicant).

When conducting measurements in multichannel systems, for example, when studying laser systems where high-speed recording of up to 192 pulsed electrical signals is required [3], a simple inexpensive, reliable multichannel digital device is required. The simplicity of one channel will enhance this quality in a multi-channel system. Known [3, 4] devices that use a time-scale conversion with dynamic storage of the recorded signal on a sectional delay line are simple. They use a coaxial or meander microstrip line (MPL) with taps as a sectional delay line and, more recently, a combination of a meander MPL for the signal under study and a direct MPL for the strobe signal [3, 4].

Known digital recorders of single signals [2-4] use non-simultaneous reading of samples of a signal distributed along the line. This is achieved by sequentially supplying N strobe mixers (which are used with high-speed Schottky diodes) gating signals obtained, for example, by sequential sampling of strobe pulses to each mixer, from a strobe pulse propagating in the line generated by the start method from the recorded signal and the recorded signal distributed along the meander slow line. In the future, the samples stored on the storage capacitors, which are used as high-speed Schottky diodes, are sent to the analog-to-digital converter in a converted time scale.

Using a combination of meander and straight lines [3], on the one hand, it is possible to simplify the distribution of the strobe pulses that select the corresponding sections of the signal under study, and on the other, to reduce the effective sampling time, which increases the speed of registration.

The closest technical solution to this proposal is a device for recording pulsed electrical signals [3], containing a meander information slowdown microstrip line (IMSPL) made on a dielectric board, the taps of which are connected via appropriate keys to the corresponding sampling and storage elements connected through series-connected expanders and a multi-input switch with corresponding inputs of a multi-bit ADC, controlling key inputs via communication elements are connected to the taps of the line MPL strobe signal (MPLSS) formed on a dielectric board and extends parallel with the axis IZMPL.

The disadvantage of the device [3] is the presence of a meander line, which limits the bandwidth of the device and the location of the IMSPL and MPLS in the same plane, which increases spurious coupling between the lines, degrades the quality of the signal samples, and necessitates the installation of additional communication elements (for example, capacitors). In addition, there are structural problems of placing a number of elements in the minimum gap between the lines.

The technical result of this invention is to expand the bandwidth, increase the sampling frequency of the device, increase the manufacturability of the design.

The technical result in a device for recording pulsed electrical signals, which is made on a dielectric board with a relative dielectric constant ε1 IMSPL, the taps of which are connected via corresponding keys to the corresponding sampling and storage elements connected through series-connected expanders and a multi-input switch with corresponding inputs of a multi-bit analog-digital the converter, the control inputs of the keys through the communication elements are connected to the taps MPLS, performed on a dielectric board with a relative permittivity ε2, is achieved by the fact that the ISMPL and MPLSS are made on rectilinear or circular MPLs whose dielectric constants are connected by the relation ε1≫ε2, whose motherboards are located in different planes (one above the other), so that the sides boards with solid metallization are connected to each other electrically and mechanically, and the communication elements are made in the form of combined slots in the reverse metallization of each of the boards with a diameter of dsv, and in the MPLS board made directly above the strobe signal line, and in the IMSPL board, slots are made under the corresponding contact pads with a transverse dimension d spaced apart from the IMSPL by a distance ΔY, determined by the installation size of key elements (for example, high-speed mixing diodes).

The essence of the invention is to slow down the propagation of the recorded signal compared to the strobe signal, which is achieved due to the different values of the dielectric constant of the boards, as well as the separation in the space of the propagation lines of the registered and strobe signal, which allows to minimize parasitic inductances. In addition, the proposed design provides the ability to build layers, gives the fundamental possibility of increasing the number of samples and lengthening the time axis.

The structural diagram of the proposed device is presented in figure 1.

Figure 2 (a, b) shows the relative position of the line (for the case of lines of annular shape) and a fragment of a transverse section at the location of one of the ring samples of reverse metallization. The positional designations of the elements in Fig.1,2 match. The designations of sizes with an index of 1 correspond to IMPL, the sizes with an index of 2 correspond to MPLSS.

The device contains (end-to-end numbering in FIG. 1 and FIG. 2 (a, b)): IMSPL 1, made, for example, on a board from a high-frequency (HF) dielectric of the FLAN16 type (ε1 = 16), which slows down the microstrip line 2 of the strobe signal made on a circuit board, for example, of a reinforced foil fluoroplastic having a dielectric constant ε2 = 2.6, keys 3, sampling and storage elements 4, pulse expanders 5, multi-input switch 6, multi-bit ADC 7. The figure also shows elements that are not involved in the claims, for example, agreed upon loads 8, 9 IMLF and MPLSS, made in the form of resistors, input connectors 10, 11 of these lines, spurious inductive elements Lп. The communication elements between MPLSS 2 and contact pads 12 of the key elements are annular slots 14 of boards 15 and 16, made in reverse metallization 13, respectively, IZML 1 and MPLS 2, through which MPLS 2 and pads 12 are formed with an equivalent capacitance of SSB. Keys 3 can be performed on 2 on-board diodes, for example, high-speed Schottky diodes of the type HSCH 5314 manufactured by Hewlett-Packard. The connection point of the diodes on the corresponding contact pad 12 is the control input of the key and through parasitic inductive elements Lp and the equivalent capacitance Csv connection is connected to the slowing MPLS. The output contact pad 17 of each key performs the function of the storage capacitance Sn due to the capacity of the pad on the reverse metallization layer.

IMSPL and MPLSS can have the same or different value of wave impedance. The exact value of the strip width W of each of the lines is determined based on a given wave impedance Z ₀ and the thickness of the board Н: W = f (H, Z ₀ , ε, t), where t is the thickness of the metallization layer, ε is the dielectric constant according to the above, for example , in [5] graphs.

In figure 2 (a, b) the following parameters and dimensions are indicated:

ε1 and ε2 are the relative permittivities of the IMSPL and MPLSS;

w1 and w2 are the strip widths of the IMSPL and MPLSS boards;

t1 and t2 are the thickness of the corresponding strips;

h1 and h2 are the dielectric thickness of the IZPL and MPLSS boards;

dsv is the diameter of the annular slots in the reverse metallization of the boards;

d - the size of the pads (12) of key elements (3);

ΔY is the size of the gap for installing key elements (3);

ΔL is the linear distance between sampling points according to the Sample.

The sampling and storage elements 4 consist of a series-connected storage capacitor Sn and an isolation resistor Rp, the connection point of which is connected to the key; the second end of the resistor Rp through the corresponding extender 5 and a common multi-input switch 6 is connected to a multi-bit ADC 7, which digitizes the time-extended equivalent of each sample. As storage capacitors of the sampling and storage elements, in addition to the capacity of the site, chip storage capacitors of small capacity can be used. As extenders, charge amplifiers can be used on standard op amps, for example, on quadruple op amps AD 824 from Analog Devices.

As multi-bit ADCs, standard high-speed ADCs of suitable capacity type AD 90 ** of the same company can be used.

The proposed device operates as follows. The recorded pulsed electrical signal is fed to the information retarding microstrip line 1 and distributed along it. At that moment, when the signal front has almost reached the end of line 1, a strobe signal delayed relative to the signal being detected is received at the input of the microstrip line 2 of the strobe signal and propagates, “catching up” the front of the recorded signal. The strobe signal has the form of a triangular pulse with a width at half maximum of about 100 ps. Passing through the communication elements 14 to the keys 3, the strobe signal opens them for a short time and makes it possible for each sample of the registered signal through the public keys 3 to charge the storage capacitor Sn of the sampling and storage element 4. These pulse samples are through decoupling resistors Rp of the sampling and storage elements 4 fed to the inputs of expanders 5, after which the sample signals expanded to units of microseconds, the amplitude of which is proportional to the amplitude of the samples from the input signal, through the N-channel switch 6 are fed to the inputs of the ADC 7, from the output of which the digitized samples are sent to the PC RAM for processing. Thus, all samples of a pulsed electrical signal are recorded, which can be restored by appropriate reading and visualized if necessary.

Consider in more detail the operation of the device. The strobe signal, catching up at the end of the line the front of the signal under investigation, will take N samples from it with an effective sampling step: ΔTeff = ΔT ₁ -ΔT ₂ , where

ΔT ₁ is the transit time of the ΔL portion of the recorded signal;

ΔT ₂ is the transit time of the ΔL section by a strobe signal.

Let us calculate the main parameters of a system of two MPLs (MPLSS and IZML) made on high-frequency dielectrics with different permittivities ε1 and ε2. Let ΔL be the linear distance between adjacent terminals for reading the signal. Then the effective sampling step Teff, calculated as the difference in the transit times of the signals ΔL along two parallel lines, will be:

In formulas (1, 2), ε1 and ε2 are the effective relative permittivities calculated, for example, by the formulas from [5].

As a rule, in the design specification, the Teff and the duration of the digital sweep are usually set to Tcr, so the distance ΔL is determined from formula (1). In this case, the total length of the signal lines and strobe signal L, in contrast to a system with a meander line, is the same and equal to:

где w - ширина проводника и L - длина линии выражены в одних единицах длины, Z - волновое сопротивление МПЛ, Ом. С ростом потерь, при увеличении длины линии снижается полоса пропускания линии исследуемого сигнала, ограничивая временное разрешение, что характерно для прототипа. Из формулы (3), подставляя в нее Ап=3 дБ (затухание на частоте, равной граничной Fгр), и длины ИЗМПЛ для предлагаемого устройства и прототипа, получаем:

At the same time, for a system with a meander, the meander line length of the signal will be Nz times greater than the length of the strobe signal line, and the coefficient Nz is equal to the deceleration coefficient of the IMSPL: Nz = Tsig / Tstr, where Tsig and Tstr are the propagation time along the signal and strobe lines between sampling points. This, in turn, will lead to an increase in attenuation losses in the conductors An by a factor of Nz, which are estimated for the strip line using the formula given in [5]:

where w is the width of the conductor and L is the length of the line are expressed in units of length, Z is the wave resistance of the MPL, Ohm. With increasing losses, with increasing line length, the bandwidth of the line of the signal under study decreases, limiting the time resolution, which is typical for the prototype. From formula (3), substituting An = 3 dB (attenuation at a frequency equal to the boundary Fgr) and the IMSPL length for the proposed device and prototype, we obtain:

Thus, the proposed device achieves an increase in the bandwidth of at least Nz ² times, provided that the main factor in the losses in the SIPL is attenuation loss. Spatial separation in different layers of the signal and strobe lines removes the limitations on the gap between MPLS and ISPL inherent in the prototype. This, in turn, makes it possible to reduce to a possible minimum (tenths of a nH) the values of parasitic inductances in the paths of a signal sample from IMSPL to expanders. In this case, it becomes possible to increase the sampling frequency to at least 10 GHz (by reducing the distance between the sampling points of the recorded signal) and thereby register pulsed signals with short edges (of the order of 100 ps).

The manufacture of a microstrip signal propagation line is much simpler than the manufacture of a meander line, which improves the manufacturability of the device.

In addition, the proposed design has the following advantages:

- The proposed device can be performed on the ring configuration of the IMSPL and MPLSS, which reduces the dimensions of the device and eliminates the need for rare microwave dielectrics with a high dielectric constant. For example, as a material for IMSPL, you can use a serial microwave insulator FLAN-16 (ε = 16), and as a material for MPLSS a microwave insulator FAF (ε = 2.6).

- Patented communication design of MPLSS and ISPL allows minimizing the influence of lines on each other, and therefore, improving the quality of the read signal, reducing the distortion of the strobe signal and the signal under investigation during propagation.

- The design of the proposed device allows the buildup of layers to expand the time axis by increasing the number of points of the read signal (while maintaining a high sampling frequency). Duration of the time axis: Tvr = Tcr · k, where k is the number of layers, each of which has a digital scan duration of Tcr.

Thus, the proposed device, having the advantages of a prototype, has compared with it the advantages of expanding the bandwidth by at least 2 times, increasing the sampling frequency, improving the manufacturability of the device.

LITERATURE

1. RF patent No. 1347858, 1985. "A method for discrete registration of the shape of a single electric pulse and a device for its implementation."

2. RF patent No. 1826743, 1991, "Oscillographic analog-to-digital recorder of single electrical pulses."

3. T.E. Makevan, J. Kilkenny 32-GHz single pulse recorder R&D Magazin, oct. 1993 - Prototype.

4. Thomas E. McEwan HIGH SPEED TRANSIENT SAMPLER, US Patent No. 5471162, MKI H 03 K 5/125, NCI 327/92.

5. V.I. Volman Reference for the calculation and design of microwave strip devices, M .: Radio and communications, 1982

Claims (1)

A device for recording pulsed electrical signals, comprising an information slowdown microstrip line (IMSPL) made on a dielectric board with a relative dielectric constant ε1, the taps of which are connected via appropriate keys to the corresponding sampling and storage elements connected through series-connected expanders and a multi-input switch with the corresponding inputs of a multi-bit ADCs, control inputs of keys through communication elements connected to the taps of microbands of a strobe signal line (MPLS) made on a dielectric board with a relative permittivity ε2, characterized in that IMSPL and MPLS are made on microstrip lines (MPL), the dielectric constants of which are connected by the ratio ε1≫ε2, the boards are located one above the other so that the sides of the boards with continuous metallization are electrically and mechanically connected to each other, and the communication elements are made in the form of combined slots in the continuous metallization of each of the boards, and in the MPLSS board, the slots are made sredstvenno over the line the strobe signal, and a board slot located under IZMPL respective row of contact pads spaced a distance from IZMPL ΔY, defined only mounting dimensions keys.

Images