|
Arabic
Bengali
Chinese
English
French
German
Hebrew
Hindi
Italian
Japanese
Korean
Malay
Polish
Portuguese
Spanish
Turkish
Ukrainian
Vietnamese|
ENCYCLOPEDIA OF RADIO ELECTRONICS AND ELECTRICAL ENGINEERING Nodes KB transceiver. Encyclopedia of radio electronics and electrical engineering
Encyclopedia of radio electronics and electrical engineering / Civil radio communications In continuation of the publication of KB TRX nodes [1], I offer readers the final version of the main transceiver board. This node does not have any unique solutions, the circuitry is variations on the theme of TRX RA3AO and Ural-84M. The main requirements when choosing a design are repeatability, simplicity while maintaining the maximum achievable characteristics. The element base available today is used. Many decisions can be criticized - the creative process is endless, with constant alterations and improvements it is difficult to see the finished version, but it was necessary to stop and produce printed circuit boards in an industrial way. Initially, the transceiver was conceived for SSB operation as the main mode of radiation. To narrow the bandwidth, a four-crystal erasure filter with band adjustment is introduced. For fans of narrow-band reception, it can be recommended, as is done in branded TRX, to go to additional costs for the manufacture or purchase of high-quality narrow-band quartz filters. As a rule, a home-made ladder filter made of quartz, the most popular among radio amateurs, has insufficient characteristics for high-quality narrow-band reception. For these purposes, you need to make a filter according to the differential-bridge circuit or use very high quality quartz. You can buy a set of branded filters, although they will be comparable in cost to all other costs for the transceiver. The "up conversion" option was not considered due to the lack of a fairly simple and well-established frequency synthesizer circuit. This construction option makes sense in a device with continuous coverage from 1 to 30 MHz, and for operation in nine narrow amateur bands, acceptable selectivity can be provided by a cheaper IF 5 ... 9 MHz. Many people experience problems with carrier suppression of at least 40 dB when shaping the SSB signal directly to the IF. It seems to me that this problem is more contrived than it really is. In almost all cheap branded transceivers, the formation takes place at an IF of 8 ... 9 MHz. I think it is unlikely that anyone will hear an unsuppressed carrier, for example, in the TRX FT840 or TS50. The quality of the SSB signal conditioner assembly depends on the literacy and perseverance of the manufacturer. Excellent performance can be obtained using the simplest modulator on varicaps, as is done in the TRX Ural-84. Just do not need to strive to receive from the modulator levels sufficient to build up the output stage - then it is not possible to suppress the carrier. When working out the main board, elements were used that can be found on almost any radio market. Something special, with gold-plated conclusions, with an VP index was ruled out immediately. For example, the required gain can be obtained from two stages on imported BF980s. But they are not always on sale, so domestic analogues of KP327 are used, although they have worse parameters. The board lacks any irreplaceable parts. The sensitivity from the input of the board, which can be achieved without careful debugging of each stage individually - 0,2 ... 0,3 μV, with the selection of parts and careful tuning - 0,08 ... 0,1 μV. One of the transceivers with such a main board and a synthesizer described in [2] had a sensitivity of 0,4 μV with UHF turned off and two-signal selectivity when two signals were fed with a spacing of 8 kHz, 95 dB. The measurements were taken by UT5TC. These are not limit values, because the transceiver used input band-pass filters on frames with a diameter of 6 mm with fairly high attenuation and conventional high-frequency diodes in the mixer. Although, as experience shows, in transceivers that are designed for normal daily work on the air, you should not chase the dynamic range figures. A value of 80 dB suits most radio amateurs. The use of a super dynamic receiver only makes sense in TRX for head-to-head competition and provided that all competitors are operating on line signals. Problems with interference from the neighbor's transmitter often arise not from the low dynamic range of the receiver, but from the fact that the unfortunate radio amateur, trying to outshout everyone, tunes his transmitter according to the principle - all arrows to the right all the way. According to the observations of US5MIS, which has been turning the knobs of the FT840, Surf and RA3AO for many years, all these techniques sound almost the same to the ear. But when comparative measurements were carried out using the same method, TRX RA3AO responded to a level of 1 V in the adjacent channel, "Surf" - to 0,8 V, and FT840 - to 0,5 V. But the convenience of operation, stability and service took their toll - left FT840. I describe all this not to show how good our home-made (or semi-home-made, like Surf) technique is, but to make it clear that the pursuit of dynamic range makes sense up to a certain level and under specific conditions. I think that many happy owners of super-dynamic RA3AOs would be happy to exchange them for the "frail" FT840s in terms of dynamics. I want to touch on another stereotype common among our radio amateurs. This is the belief that the synthesizer is "noisy". After the birth of Kovel synthesizers, none of my transceivers was with a VPA, only and only a synthesizer. Above, I described the sensitivity achievable from the input of the main board when used as VFO synthesizers. What kind of noise can we talk about when neither G4-102A, nor G4-158, nor G4-18 can measure the ultimate sensitivity. I had to make a separate crystal oscillator, power it from batteries, shield it with a double screen, and use an attenuator up to 136 dB to evaluate the sensitivity of the board. Let's move on to the description of the main board itself, which includes: - switchable UHF, reversible mixer, passive diplexer, matching reversible FET stage, main crystal filter (picture 1); - IF array, reference oscillator, detector (picture 2); - ULF and AGC node (picture 3). Let's consider the circuit diagram in detail. High-frequency amplifier (VT5) - with X-type negative feedback circuit [7]. Possible parameters of this type of amplifiers range from:
Simply put, UHF is not overloaded on 40 meters even in the evening when the level of interference is very high. The extreme sensitivity is such that it allows you to hear the noise of the air at 28 MHz, even in rural areas. One of the best transistors for such an amplifier is KT939A. KT606A was included in the board as cheaper and more common. No need to worry too much that UHF worsens the dynamic range of RX (again I'm talking about "dynamics", I'm sinful, I myself was once fond of limiting figures). Firstly, UHF is switchable, you can always turn it off. Secondly, turning it on is usually required only on the quietest bands during low penetration, when all stations are heard at a low level, and it is unlikely that any of the stations will overload this cascade. And thirdly, "the devil is not so terrible as he is painted." Almost all industrial RPUs, for example, R399A, use UHF, and non-switchable ones. The configuration of this cascade depends on the needs of the user. Depending on the type of transistor and its mode, it is possible to provide either the maximum possible sensitivity, or the minimum effect of this stage on the upper limit of the dynamic range. I wrote about the mixer in a previous article [6], its circuitry is borrowed from [4]. The main advantages of this option are reversibility and a sufficiently large dynamic range (Dbl - up to 140 dB) with a low local oscillator level. Of course, in terms of the number of parts, it is more complicated and more expensive than commonly used mixers. But we must not forget that this node determines the quality of the entire receiver, and saving on it is meaningless. The thoroughness of the mixer settings also determines how the receiving part will perceive the air, what can be heard there, and how much "garbage" will be given out for transmission, how complex the band-pass filters will have to be made so that it is possible to work without TV1. Part of the divider (D1) had to be installed directly at the mixer in order to ensure antiphase signals at the input of the arms VT1, VT2 and VT3, VT4. This is the most important requirement on the part of the local oscillator. If you are using a conventional local oscillator, anti-phase signals must be generated in a different way. A variant of the simplest docking with the Kovel synthesizer is also used here. The use of the trigger is also due to the fact that at its output the signal is as close as possible to the meander. When docking with a conventional GPA, you need to use other ESL microcircuits, for example, types LM, TL, etc. The main requirement is that at the input of transistor switches there must be equal in level, but ideally antiphase high-frequency signals. The keys use transistors KT368 and KT363 recommended in [4]. Experiments with other transistors were not carried out. The mixer is operable with various types of diodes. It can be assumed that Schottky diodes will be the best. The transition from KD922 to KD512, KD514 does not cause any noticeable deterioration in the parameters (subject to the selection of diodes). In my opinion, the main advantage of KD922 diodes over all the others is that they are supplied selected and packaged in individual containers (therefore, mixing is excluded). With carefully selected KD503, the mixer works in much the same way as with KD922. The symmetry and workmanship of the T1 transformer is very important. Input resistances from input T1:
This must be taken into account when coordinating with the DFT. You can try different turns ratios to get the input impedance closer to 50 ohms, but it turned out to be easier to change the DFT coupling coils to suit the specific resistance of the main board. To match with subsequent stages, a conventional diplexer is used. On fig. 1 shows the diplexer data for IF=9 MHz. In principle, you can not install this node. A good agreement can be obtained by selecting the VT15 KP903 mode, however, the use of a diplexer allows you to get the highest possible sensitivity, and if you do not completely get rid of the affected points, then significantly reduce their level. The active bidirectional VT15 stage after the mixer should have the lowest possible noise figure, not degrade the dynamic range of the mixer and compensate for the attenuation introduced by the mixer, DFTs and diplexer. The most common and high-quality transistor for this cascade is KP903A. You can use KP307, KP303, KP302 (with the maximum slope value), KP601. After VT15, the signal through the transformer T3 is fed to the quartz filter ZQ1. Resistor R26 is used for matching, it may not be required. This procedure can also be performed using R22. A ladder six-crystal quartz filter was used as ZQ1 (Fig. 4). To narrow the bandwidth in CW mode, additional capacitors are switched on in parallel with the outer resonators using a relay. Such a CW filter, of course, cannot be called high-quality. Narrowband CW fans require the use of a separate crystal filter. Why is a six-crystal filter applied? Usually practiced eight and even ten plates. But do not forget that this filter is also used for transmission, and for acceptable SSB quality, a bandwidth of about 3 kHz is required. But for reception in conditions of overloaded amateur bands, a band of 2,2 ... 2,4 kHz is sufficient. Therefore, a compromise was chosen: a bandwidth of -3 dB - 2,3 ... 2,4 kHz with a smaller squareness. As a result, we have quite high-quality reception and a good transmission signal (which cannot be said about the signals that are formed using eight-crystal filters). Another advantage over the eight-crystal filter is less attenuation in the transparency band. This ensures the achievement of the maximum sensitivity of the entire amplification path.
To increase the attenuation outside the transparency band in the IF path, a cleanup four-crystal filter was used (Fig. 5). The total attenuation of both filters exceeds 100dB. Figures 4, 5 show the averaged data of quartz ladder filters made of plates in housing B1, which are most often encountered. The cleanup filter cuts the noise introduced by the IF path, and due to the applied smooth bandwidth adjustment, it allows you to slightly deviate from interference in the SSB mode. One should not, of course, place high hopes on such a variant of a smooth bandwidth change. Firstly, the narrowing occurs only on one side of the filter slope, and secondly, it is problematic to get more than 40 dB from a four-crystal ZQ. But the complication is so simple and cheap that it makes no sense to refuse such a service, albeit a small one. The filter should be designed for a bandwidth of 2,4 kHz. With a smooth narrowing of the band by varicaps, the upper slope approaches the lower one, depending on the quality factor of the quartz, up to the band of 600 ... 700 Hz. But due to the low squareness of the filter, even with such a bandwidth, it is possible to receive SSB stations. This mode is often used in the ranges of 160, 80 and 40 m. Instead of the indicated varicaps, several KB 119, KB 139 connected in parallel can be used.
The crystal filter ZQ1 is consistent with the IF path (Fig. 2) through the resonant circuit L3 with the coupling coil. If the filter resistance is noticeably different from 300 ohms, the selection of the number of turns of the coupling coil is required. Transistor VT7 turns on during transmission. The second gate controls the output power of the transceiver. The UFC line is assembled on KP327 transistors. Circuitry borrowed from RA3AO. In my opinion, this is one of the best options for building such a path. Here you can use double-gate field-effect transistors and other types. BF980 turned out to be the best. Our industry failed to copy the characteristics of this transistor, KP327 in comparison with the BF980 is worse both in Ksh and in Kus, although Kus of transistors is not of decisive importance. For VT8, you need to choose a transistor with minimal noise. Usually the best specimens come across among the KP327A. VT9, VT10, VT11 can also be replaced by KP350. The advantage of KP327 over KP350 and KP306 is in the best value of Ksh, resistance to static, and "gold diggers" do not react to them in any way, because. transistors do not contain precious metals. To adjust the gain, the property of saturation of the throughput characteristics of field-effect transistors on the first gate at a low voltage on the second one was used [2]. Excessive gain is removed by shunting the IF circuits with resistors R38 and R46. You should not increase the RF levels on the first gates of the transistors so that the instantaneous voltage value does not exceed the opening threshold of the static protection zener diodes (15 V). Otherwise, the zener diodes open and block the operation of the AGC - this applies to the last two cascades of the IF. The detector and reference oscillator, preliminary ULF and AGC are similar [2]. The VT13 transistor (Fig. 3) can be used to turn the AGC circuit on and off and to block the AGC during transmission so that the S-meter readings are not distorted, which in this mode shows the output power of the transmitter. As VT 13, you can use both a field-effect and a bipolar transistor. The bipolar transistor has a lower collector-emitter resistance, so it shunts the AGC circuit better. The AGC rectifier amplifier circuit is similar to [2]. The timing characteristics of the "fast" chain have been changed, the capacitance of C74 had to be increased to 0,047 ... 0,1 μF. The K174UN14 microcircuit was used as the terminal ULF, in a typical inclusion, the bandwidth from above is determined by the C69, R80 chain; the gain can be adjusted by resistor R81. The ULF output can be loaded on a speaker or through a divider R84, R85 on headphones. Details Coils L1...L6 are wound on frames with a diameter of 5 mm, with a tuning core SCR-1. L3 ... L6 contain 25 ... 30 turns of PEVO wire, 2. LCB - 3...4 turns at the "cold" end of L3. L9, L10 - chokes with an inductance of 50 ... 100 μH. L11 - inductor 0...30 µH. Transformers T1 ... TZ are wound with PEVO wire, 16 on K 10x6x3 rings made of ferrite 1000 nn. T1 contains 10 turns of twisting into three wires, T3 - 9 turns of twisting into two wires, T2 is wound with a twist of three wires: winding I - 3 turns, II - 10 turns, III - 10 turns. Yielding to the desire to ensure the "single-board" of the entire design of the transceiver, we decided to separate the reference local oscillator on the main board. This, of course, complicated the situation with the "affected points". Some of them could be avoided altogether if the reference local oscillator were made in a separate shielded compartment. With a successful IF, the number of points does not exceed 3 ... 5 for all nine ranges. It is possible to get rid of them almost completely if you tinker with additional groundings of the microcircuit power bus and metallization around this node. Location of parts on the board (Fig. 7) The board setup is typical, it has been repeatedly described in amateur radio literature. The values of the elements R1 and C1 depend on which node is used as a local oscillator. If this is a Kovel synthesizer, R1=470...680m, C can have a value from 68 pF to 10 nF. The quality of matching is noticeable by ear by the minimum number of "noise points" from the synthesizer. Elements LI, L2, C7, C9 are tuned to resonance at the IF frequency. Resistor R19 can have a rating of 50 ... 200 ohms. The quality of the matching of this node determines the overall decrease in the level of "lesions" and a slight increase in sensitivity. ZQ1 matching is achieved by resistors R22, R26, Kf and the selection of the number of turns LC8. The cleaning filter ZQ2 is matched with resistors R52 and. R54. The overall gain of the IF path can be selected using R28, R38, R46. Resistors R39, R47, R53, R60 affect Kus and determine the quality of the AGC cascading. On the manufacture of transformers. Ferrites with a permeability of 400 ... 2000 were tested, the diameter of the rings was 7 ... 12 mm, twisting of wires and without twisting. Conclusion - everything works. The main requirements are the accuracy of manufacturing, the absence of a winding short circuit to ferrite and the obligatory symmetry of the arms. Diodes in the mixer should be selected at least according to the open junction resistance and capacitance. Transistors VT1, VT2; VT3, VT4 must be selected as identical complementary pairs. In the VT5 emitter, the R and C values in the chain are not indicated. They depend on the type of transistor. For KT606 R - within 68 ... 120 Ohms, and C should be adjusted to the maximum gain at 28 MHz (usually 1nF). Using R29, you can select the current through the transistor, for example, according to maximum sensitivity. KP327 transistors are soldered from the bottom of the board. On top of the board, from the side of the installation of parts, foil is left, the holes are countersunk. The coils are covered with screens. For purchase of printed circuit boards or customized assemblies, please contact the author, frequency - 3,700 after 23.00 MSK. Literature: 1. Radio amateur. - 1995. No. 11,12.
Author: A. Tarasov (UT2FW), Ukraine, Odessa region, Reni; Publication: N. Bolshakov, rf.atnn.ru
The world's tallest astronomical observatory opened
04.05.2024 Controlling objects using air currents
04.05.2024 Purebred dogs get sick no more often than purebred dogs
03.05.2024
▪ Panoramic attachment for digital camera
▪ section of the site Consumer Electronics. Selection of articles ▪ Article Law - I am dumb before the law. Popular expression ▪ article How do insects breathe? Detailed answer ▪ article by Arnebius coloring. Legends, cultivation, methods of application ▪ article Bouncing coin. Focus secret
Home page | Library | Articles | Website map | Site Reviews
www.diagram.com.ua |
See other articles Section 
Latest news of science and technology, new electronics:
All languages of this page