The booster based on the tube GU-74B

This lamp is selected on reasons of deriving of output power up to 1000W at small overall dimensions. The lamp is included under the schema with earthed grids, its(her) high steepness therefore will not be utillized. It is connected that the booster was engineered under the transceiver with a transistor output stage by power 100...120W. In sample alternative of actuation with the communal cathode the condition AB will be utillized, the output power compounds 600W. This condition can be recommended, if the output power of the transceiver compounds 25...30W. The coefficient of performance of such booster is low-level (standing current - up to 300mA), that is connected to supply of a linear condition of beefing-up. Lacks of this alternative, apart from a low coefficient of performance - necessity hardly to stabilize stresses screen and controlling of grids. If the transceiver with output power not less than 70 will be utillized... 80W, it is expedient to include a lamp under the schema with communal grids. In such schema the distortings, introduced by a stage, depend on a choice of an operating point on the performance of a lamp a little. It allows to diminish a standing current at the same level of distortings. The lamp thus works almost in a condition B, the heating of an anode in spaces is considerably moderated, and it is possible to receive large powers at smaller heating of the shell of a lamp. Add here absence of necessity of stabilizing of stresses of controlling and screen grids - advantage available. Alone lack of such booster a small control ratio (10...15Db) and accordingly large, rather than in boosters with the communal cathode, excitation power, but almost all this power is not lost anywhere, and acts(arrives) in an offloading. The principal diagram of the booster is rotined in a fig. 1.

The basic characteristics of the booster:

  • Input resistance depending on an operational frequency - 47... 140Om;
  • Equivalent resistance anode P - outline - 2000Om;
  • Stress of an anode - 2200... 2500V;
  • Coefficient of performance on different ranges - 62 %... 70 %;
  • Power forfeited in an offloading - up to 1000W;
  • The excitation power - no more 100W (at the exact coordination the transceiver - booster - is customary 60... 80W).

The standing current of a lamp is very small, it(he) is gentle is inflected depending on an anode stress and copy of a lamp, usually does not exceed 30... 50mA. In the experimental purposes the output power 1200W (Ua = 2200V, Ia = about 1,0A) was obtained. Certainly, it is a short-lived peaked condition. After 5 years of exploitation practically by daily operation in an ether on 1... 2 hours of noticeable loss of issue of a lamp were not supervised. The major advantage of such booster, on my view, is encompass byed to his(its) simplicity. Istochnik of power supply (fig.2) up to the minimum is simplified. The anode stress and stress of heat is required only. For power supply of an automation the rectified stress intended for heats of a lamp will be utillized.


The booster is executed in a metallical box by the size 450x380x230mm, disjointed cross-cut baffle plate on two compartments(baies). In one compartment(bay) the power transformer with a high-voltage rectifier is laied out. In the same compartment(bay) on the front desk are disposed gaging Instruments and toggles of a web, vane and instruments. In the second compartment(bay), disjointed dielectric baffle plate, are disposed a lamp, anode choke and parts of a P-outline. The variable capacitors and disk toggle fix to the front desk. The frameless reels of a P-outline are retained by buses on the disk toggle, capacitors and anode choke. The compartment(bay) with a lamp from above is closed down by a dielectric cover. The vane is disposed on a trailing wall in a compartment(bay) of a lamp, whence it(he) pumps out air. Air acts(arrives) through openings in a base cap vice-versa lamps, drives through the desk and edges(rib) of an anode of a lamp. The lamp is established(installed) upright in a tube manufactured of a thermally sound tissue on epoxy, routing air through edges(rib) of a heat sink. As all grids of a lamp immediately are coupled to the body of the booster, there is a feasibility of the self-made desk, which one can serve, apart from an anode, padding heat removal. The lamp is crimped around of the deduction of a screen grid by a powerful clamp from bronze, or copper, and the clamp densely is fastened to the chassis. Through this clamp is padding on the chassis is drawn off the heat. On the deductions of a lamp the ceramic desk from old lamps, for example 12V1L or 2V27L is mounted. A lamp it is possible to lay out as upright, and is horizontal. Certainly, the applying and "firm" desk for GU-74B is not eliminated. It is very important to provide effective heat removal from a lamp. The edges(rib) in an anode are disposed rather Densely, therefore it is required to provide sufficient air pressure "to drag" through an anode not less than 30 cubic meters at one o'clock. The vanes such as "snail" are most effective for these purposes. But they rather seldom meet, and is difficult(complex) to select eligible alternative on power and sizes. With success it is possible to apply flat vanes from old electronic computers. To orient follows on alternatives with a supply voltage 220V and, it is desirable, metallical fans. In the depicted booster the flat vane of the French production with plastic fans by output of 60 cubic meters at one o'clock is applied. In a nominal condition it(he) will be utillized at a undervoltage of power supply - 160... 170V. In conditions, when the booster will very long be utillized on transmission(transfer), on the vane gives compound stress - 220V. The flat vanes badly work on air delivery, it is more rational to utillize them on "deflation". Certainly, it not the best alternative, since ardent air drives through fans and heats vane. Therefore it is better to utillize completely metallical vane, which one design for such difficult conditions of operation. Though it is possible to utillize anyone with some reserve on output, in the depicted booster five years the vane, former before in exploitation, moreover and with plastic fans works.


  • Dr1 - a choke on a ferrite rod (collar)((ring)) by permeability 1000... 2000, rod - dia 8... 12mm. The dia 1,5 is reeled simultaneously by two wires in enamel insulation... 1,8mm. An amount of coils - 10... 15.
  • Dr2 - a cathodic choke. A hull - dia 20... 25mm, dielectric handset. Spooling - coil to To coil, wire in enamel insulation a dia 0,52... 0,62mm, 100 coils.
  • Dr3 - an anode choke. A ceramic handset a dia 35mm. Short-range to an anode the coils are reeled With a pitch 1mm - 20 coils, then 170 coils - a coil to a coil. A wire - in enamel insulation a dia 0,52... 0,62mm.
  • Dr4 - a high frequency choke of any type which is holding out a stress 300V. Coefficient of self-induction -2,5... 3mGn.
  • L1 - the frameless reel P - outline. A copper handset a dia 6mm, 9 coils, diameter of a tube expander - 47mm. Removals from 4, 5, 6, 8 coils.
  • L2 - the frameless reel, wire - in enamel insulation a dia 2,4mm, diameter, tube expander - 47mm, 14 coils, removal from 6 coils.
  • L3 - the reel - range 1,8MHz. A ceramic, ribbed hull a dia 60mm. A wire in enamel insulation a dia 1,5mm, 17 coils.

The reels L1 and L2 are reeled with a pitch, equal half of diameter of an applied wire.

  • VD1 - the diode the deflocculant D817A.... G. A stress of stabilizing 100V
  • VD2 - the high frequency silicon diode.
  • VD3, VD4 - any silicon diodes.
  • VT1, VT2 - KT815 or any silicon trss which are holding out operating current K1 and K2.
  • C1 - ceramic, U > 300V.
  • C2, C3, C4 - bypass capacitors, U > 300V.
  • C13, C14, C15, C16 - bypass capacitors, U > 50V.
  • C5 - 1500... 2200pF. High-quality ceramic, U > 5kV. Skipped power > 1,5kW.
  • C6 - 2200... 10000pF. A block high-voltage capacitor, U > 4kV.
  • C7 - the series capacitor, is hooked up by contacts disposed on C8 for 3,5 and 1,8MHz. High-quality ceramic, U > 5kV, on a reactive power > 10kW.
  • C8 - 10-260pF. A gap between laminas > 2,5mm.
  • C9 - a gap between laminas > 0,3mm.
  • C10 - ceramic or mica. U > 1000V, on a reactive power > 10kW.
  • C11 - ceramic, U > 500V.
  • PA1 - the gaging head on 1mA.
  • K1 - the relay REN-33 or any high frequency relay which is capable to skip power up to 100W and a current of contacts > 1,5 A.
  • K2 - the relay REW-15, high frequency relay. A stress on contacts - up to 300V, current - up to 5A.

Actuation voltage of all relays - 12... 14V.

  • S1 - the high frequency toggle which is holding out a stress up to 3kV and a current up to 7A.

In quality C8 any variable capacitor with a gap between laminas not less 2mm and powerful springing contacts of the body of rotary laminas will approach. It is necessary to pay attention to a zero capacitance, she(it) should not be more 10pF. To bridge over range 1,8MHz, the maximal capacitance should be not less 500pF. For range 3,5MHz there is enough 300pF. The capacitor from old Soviet radio station RSB-5 is applied. For an abatement of a zero capacitance the side metallical walls are carved. The mobile contact, which one is on this capacitor, utilised for hooking up of the padding capacitor C7 for ranges 1,8... 3,5MHz. If it is supposed to work only on an offloading with resistance not above 50-75Om, in quality C9 it is possible to apply a variable capacitor from old lamprooms of recipients with a gap not less than 0,3 mms. For range 1,8MHz a total capacitance C10 and C9 - not less 3000pF. At a choice S1 it is necessary to pay attention to power and reliability of contacts. The contacts should be deleted from earthed structural elements for minimization of introduced capacitance. If not it is possible to provide a communal zero capacitance P - outline less 35pF, C8 it is possible to include not in a beginning L1, and having receded 0,5...1 coil from an anode. The anode choke Dr3 should be reeled up(wound on) on a hull with good dielectric rigidity (ceramics, teflon). It(he) can be executed segmented, i.e. through some of coils to do(make) gaps 2... 3mm, that also moderates total capacity and odds of stray resonances.

Automation and portrayal.

The contacts K1.2 at transferring to transmission(transfer) close the stabilizing diode VD1, peeling a barrier voltage. For control of the relay the transistor clues on VT1 and VT2 utilised. It is made that it was possible to utillize a stress +12V TX from the transceiver for commutation of the booster. At actuation of the relay K1 the dwell (R4, C12) is applied. For what it is made? In quality K2 the powerful relay with a large gap between contacts is applied, the response time of such relay is much more, rather than response time of the small relay K1. Therefore at the moment of transferring from reception on transmission(transfer) there can be a unstable state of the booster, when K1 already has hooked up the transceiver to the booster, and the antenna at the expense of an inertance K2 yet is not hooked up. Now it is uneasy to guess, that will originate in the booster, if in the transceiver there is no such delay, i.e. at the moment of transferring to transmission(transfer) at once on an exit TRCVR there is a cue. At its best, will burn slightly permanently contacts K2. It is much more than the inferior alternatives - down to punching of a lamp. If you are assured of the transceiver and precisely know, that the high frequency cue on ANT a plug-and-socket occurs timelagged (sufficient for actuation K2 in the booster), can simplify an automation of the booster, for example include in bridge both relays and to utillize one clue.

The schema of a power supply unit is rotined in a fig. 2.

In the booster one transformer T1 by dimensional power 1600W is applied. The stress of heat will follow-up be utillized for power supply of the relay of an automation. It would be desirable to tell about this winding apart. It is not necessary to build up her(it) with a reserve, i.e. to utillize with certainly large The dia of a wire, rather than it is required for a lamp (3,6A). In a time of engagement, when the GUI filament of a lamp still cold also has minimum resistance, in a circuit(network) the very large current flows past, which one can her(it) shatter. The greater reserve there is a winding, the the greater breaking down receives a GUI filament at each actuation of the booster. Therefore at account of diameter of a wire of a filament winding it is not necessary her(it) to count for a current more 4A. The anode rectifier of features has no. It(he) is collected on a voltage doubling circuit. Secondary The winding should provide 800... 900V at a current up to 2A. A maximal reverse voltage Than line-up of diodes the peak current - not less 3A should be not lower 3000V. To each diode the resistor for a random distribution of a reverse voltage is by-pass. The bypass capacitors are applied for removal of so-called "flat noise", which one sometimes arises, and for "occluding" and random distribution of instantaneous spikes of an overstrain of a web. These spikes can arise owing to transient phenomenons, which one last the 100-th fractions of a second, but it is enough of it for a breakdown of diodes. At usage of the transformer on P - figurative Ferri lactas, it is necessary to abide some rules of spooling - network and secondary winding divide equally and lay out on both rods. Network S1 the toggle of revolutions of the vane S2 - on a current not less 0,5A should be counted for a current more 12A. In the rectifier of an automation VD1... VD4 it is possible to utillize any silicon diodes on a stress 25V and current 1A.

The instrument PA1 in a power source (fig.2) meters Ua and Ia - depending on a position of the toggle S3. For a graduation serve accordingly R3 and R4.

PA1 In the booster (fig.1) serves for measuring a high frequency stress on an amplifier output. The capacitor C11 serves for leveling amplitude frequency response curve of a meter. At the help K3 advance the observations PA1, having weighted the booster on an equivalent load.


The condition of a lamp at feeding demanded stresses is erected(mounted) automatically. The set-up is minimum and is reduced to guard rope of removals from reels L1 and L2 (fig.1), that it is best to make through a meter of frequency response curves. For want of instruments P - an outline adjust, achieving maximal output power, having weighted the booster to equivalent resistance 50... 75Om. At exact set-up the fall of a plate current in a resonance does not exceed 15... 20 % From value at not tuned P - outline.

Joint with the transceiver

In many publications about boosters with feeding of stimulation in the cathode is spoken about an input resistance of such system(device) about 50Om. Unfortunately, it not absolutely so. The input resistance oscillates depending on an operational frequency. It is necessary to allow for it at a choice of the schema of the coordination of an output stage of the transceiver with an amplifier input. Differently, as it frequently happens, on some ranges the booster will not return that power, which one it(he) provides on other.

The input resistance of the booster compounds: on 1,9MHz - 98Om; on 3,5MHz - 77Om; on 7MHz - 128Om; on 14MHz - 102Om; on 21MHz - 54Om; on 28MHz - 88Om. This problem does not arise, if in the transceiver there is an aerial(antenna) tuner, or the output stage has manual setting P - outline on each range. Then it is possible optimally to agree an amplifier input an exit of the transceiver.

In a case, when in the transceiver there are no sets up of an output stage on resistance of an offloadings which are distinct from 50...75Om, this routine should be realized in the revertive order, i.e. an input resistance of the booster to reduce to 50... 75Om with the help of the elementary balancing unit. This system(device) can be placed(installed) in the booster or to execute it(him) in the separate body and to utillize both with other boosters, and with antennas, the resistance which one is distinct from 50...75Om. In our alternative of a request to such system(device) minimum - it should skip power up to 100W and to reduc an input resistance of the booster (47...140Om) to standard - 50.. 75Om. In a fig.3 one of alternatives of such system(device) is reduced. It is customary P - an outline. As coefficient of self-induction it is possible to utillize a variometer or bleeding coil, the adjustable capacitors can be substituted with constants, previously having selected them on each range. The reel has 17 coils, wire - dia 1,0... 1,2mm, frameless, is reeled on a tube expander a dia 15...17mm. She(it) is minimized by a collar(ring) on diameter of the disk toggle, that the removals had minimum length. From one end of the reel the removals are made from each coil, last - arbitrary through 2... 3 coils. The disk toggle - on 11 positions. It is enough of this pitch of set-up for the optimal coordination.

For long and fault-free operations the new lamp is necessary for exposing to aging. Specially it concerns lamps, which one anywhere did not work, but have lain some years. Than the lamp - that longer her(it) is higher it is necessary to train. In home conditions the lamp is held out alive by(with) heats 10... 20 hours. If the lamp has more 10 years - some day. Then it is necessary to give an anode stress and again to maintain even 6... 10 hours. If there is a possibility, in the beginning it is better to give a undervoltage. The dusting thus is required small, let lamp will get warm. After that it is better to work in an ether some days at under power. And only after that it is possible to output the booster on a full condition. It is not necessary to overlook(forget) before disconnect of the booster to produce slightly to work to the vane, that temperature of the shell of a lamp was declined.