GB1593637A

GB1593637A - Pulse-duration modulation circuit

Info

Publication number: GB1593637A
Application number: GB2383978A
Authority: GB
Original assignee: Patelhold Patenverwertungs and Elektro-Holding AG
Current assignee: Patelhold Patenverwertungs and Elektro-Holding AG
Priority date: 1977-07-07
Filing date: 1978-05-30
Publication date: 1981-07-22
Also published as: JPS5418268A; CH622390A5; DE2741038A1; FR2397103B1; DE2741038C2; FR2397103A1

Abstract

For all-band transmitters, modulators with which high amplification efficiency, good modulation linearity and low harmonic or intermodulation distortion can be achieved are required. For this purpose, high pulse repetition frequencies are required. In the proposed PWM circuit, the anode of the modulation tube (PR) is connected via an oscillation coil (SS) to the earthed cathode of the high frequency output tube (HF), and via a free-running diode (FD) to the positive voltage output of the mains rectifier (GR). A low pass filter (TF) is connected in series before the HF output tube. To reduce transmission of low frequency signals via the rectifier (GR) into the alternating voltage mains (N), further filter elements or absorption circuits (FDr1, FDr2, Sk, C) are provided. <IMAGE>

Description

(54) PULSE-DURATION MODULATION CIRCUIT (71) We, PATELHOLD PATENTVER WERTUNGS- UND ELEKTRO HOLDING AG, a Swiss Company of Glarus Switzerland, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed to be particularly described in and by the following statement: This invention relates to a pulse-duration or pulse width modulation circuit.

Figures 1 and 2 show known types of such modulation circuits. In the Figures, N signifies a mains power supply, PA a pulse preparation or pulse forming circuit, PS the pulse control, PR a pulse tube, FD a free-running diode, SS a pulse coil, SR a transmission output tube. In the circuit of Figure 1, all electrodes of the transmission output tube SR must be at high potential. In the circuit of Figure 2 the cathode of tube SR is connected to earth, as a result of which the disturbing effects of the electrode capacities, which can be felt in the first circuit, are avoided. Apart from this, however, the two known circuits have serious disadvantages. The circuit shown in Figure 1 must work with the cathodes of the HF tubes at a mixed low-frequency and direct-voltage potential, which can only be effected with difficulty in short-wave allband transmitters as well as in mediumwave and long-wave transmitters. It is true that this problem is consequently avoided in the circuit shown in Figure 2 but, on the other hand, difficulties arise with resonance in the coupled coil SS and the coupling capacitor. As a result of unwanted resonance, pulse deformation may occur, which immediately leads to distortion problems and losses of efficiency. The pulse repetition frequency must be selected as high as possible, particularly in view of the distortion factor or intermodulation distortion. In this connection, it is worth noting that the circuit illustrated in Figure 1 works with a pulse repetition frequency of 75 kHz, whereas the circuit shown in Figure 2 only works at about 50 kHz. As a theoretical analysis shows, a pulse repetition frequency of 50 kHz is already at the lower limit with regard to the intermodulation products.

In accordance with the invention, there is provided a pulse-duration modulation circuit, comprising a pulse tube having a control grid to which pulse duration signals are applied, said pulse tube having its anode connected to a positive voltage output of a mains power supply by a series circuit of a pulse coil, a high-frequency stage and an inductor of a low-pass filter, the anode of the pulse tube also being connected with the positive voltage output of the mains power supply through a free-running diode, a junction between one end of the pulse coil and the high-frequency stage being connected to earth and the cathode of the pulse tube being connected to the negative voltage output of the mains power supply.

In embodiments to be described herein, a high pulse repetition frequency can be used.

Embodiments of the invention will now be described, by way of examples only, with reference to the accompanying drawings, in which: FIGURE 1 is a diagram of a first known pulse-duration modulation circuit; FIGURE 2 is a diagram of a second known pulse-duration modulation circuit; FIGURE 3 is a diagram of a first embodiment of the invention; FIGURE 4 is a diagram of a second embodiment of the invention FIGURES 5 and 6 are block diagrams of a pulse-preparation circuit used in the embodiments of Figures 3 and 4; and FIGURES 7, 8 and 9 are circuit diagrams of different modifications applicable to the embodiments of Figures 3 and 4.

Each of the Figures 3 and 4 circuits uses a simple pulse coil SS with one end of its single winding at earth potential, which also leads to advantages in the construction and placing of the coil.

The pulse-preparation PA, the heating and the biasing rectifier for control and screen grid of the pulse tube PR must be connected to their cathode potential, which corresponds to the known prior art. The free-running diode FD may be problematical insofar as when using a heated tube diode, the heating transformer must be at different potential from the heating of the pulse tube PR. This does not lead to any fundamental difficulties or physical problems; ultimately it is a question of transformer insulation and hence a question of price.

The great advantage of the circuit in accordance with the invention consists in that a simple pulse coil SS with single winding (i.e. without a secondary coil) is connected with one end of its winding to earth and at the same time this point is connected to the cathode of the HF stage; this offers the possibility of using the highest possible pulse repetition frequency while at the same time restricting the distortion factor. A relatively high modulation frequency can also be used advantageously.

Because of these characteristics, it is also possible to use the proposed circuit as a modulator for medium and long-wave transmitters.

In Figures 3 and 4, GR signifies the mains rectifier, Mod the modulation signal, TF the low-pass filter, C5 a stray capacity, N the mains, UaREF the anode direct-voltage reference, HF the high-frequency stage, FDr, FDrl, FDr2 filter chokes, and FK the filter capacitor.

One feature of a PDM (pulse-duration modulation) system lies in the fact that an adjustable high-voltage rectifier can be dispensed with. With the pulse system described below, the direct voltage can also be regulated continuously, which is particularly important for short-wave operation.

Another feature is that the cut-out means known as "Crowbar" can be dispensed with.

The anode voltage immediately drops to zero through pulse blocking (blocking the positive trigger pulses). If a flashover occurs in the pulse tube, the high-voltage switch must be triggered through a suitable detector. In this case, the following method can be used to advantage: When the negative edges are lacking at the pulse-tube anode, triggering is effected immediately.

With a construction of the circuit as shown in Figure 3 it is necessary to prevent the low-frequency signals from being transmitted into the mains through the rectifier GR. With the normal filter elements FDr (filter choke) and FK (filter capacitor), the low-frequency signals enter the mains through the capacity between secondary and primary winding of the mains transformer.

A static screen alon, as shown in Figure 4 and designated by StS, does not offer full protection in all circumstances without additional means. Figure 4 provides an improve- ment with a division of the single filter choke FDr of Figure 3 into two component chokes FDrl and FDr2, which are connected to respective poles A,B of the rectifier GR. Considerable transmission losses are achieved as a result. If particularly resistant frequency components appear, capacitors C or series-tuned wave traps SK may be taken to earth from the poles A and B shown in Figure 4, The filter capacitor FK remains the same. The low-pass filter TF, consisting of Cl, L and C2, serves to filter the pulsed signal (other low-pass circuits may, of course, also be used, see C3 in Figures 3 and 4).

The solution shown in Figure 4 is scarcely more expensive than the known solution shown in Figure 2. Instead of a filter choke FDr, two smaller ones are necessary, on the other hand a simple pulse coil SS is sufficient (without secondary coil and therefore involving fewer problems), which is even connected directly to earth, which leads to a saving in space. The free-running diode FD may be a tube diode (heating transformer with high insulation equality and blocking of the low frequency NF flowing back into the mains, see above).

As can be seen from Figure 3, the pulse preparation system PA is at earth potential.

The signals obtained by differentiation of the PDM signal can be transmitted over two sections which are formed, for example by HF sections, by optical fibres or by a pulse transformer with a high insulation quality between primary and secondary windings, the oppositely poled pulse peaks formed at the edges of the PDM signal each being transmitted over one of the said two sections.

The details of the pulse preparation will now be explained with reference to Figures 5 and 6.

The pulse preparation PA, Figures 3 and 4, or 21 in Figures 5 and 6, consists of circuit units known per se, which are illustrated as blocks in Figures 5 and 6. In order to be able to follow better the development of the signals UT22 (negative edge) and UT23 (positive edge) appearing at the outputs T22 and T23, the corresponding voltage-time diagrams for the voltages at the individual connecting lines are included in the drawing.

The sine-wave generator 21.1 delivers at its output a pulse voltage with a repetition frequency which corresponds to the required pulse frequency.

The modulation signal voltage Umod to be amplified is superimposed on a direct voltage U= through a coupling capacitor C.

The resulting voltages Umod + U= is applied to the one input (illustrated as the lower input in Figure 5) of the threshold-value switch 21.3. The threshold-value switch is constructed in the form of a binary component, that is to say the said lower input carries the reference potential UmOd + U=, and whenever the voltage at the other (upper) input exceeds the reference potential, the signal, the signal L appears at the output, that is to say the one binary state, while otherwise the threshold-value switch delivers a zero output signal. Thus, with a fixed threshold-value (voltage at the upper input) rectangular pulses of different duration appear at the output of the thresholdvalue switch 21.3.

These rectangular pulses are -now supplied to the differentiating stage 21.4 (Figure 5,6). From the rising edge of the rectangular pulse, this differentiating stage forms a positive needle pulse which then reaches the control input of the pulse tube through a correspondingly poled diode and through the control line T23. With the falling edge of the rectangular pulse, the differentiating stage 21.4 delivers a negative needle pulse which then reaches a reversing network (normal transistor stage with a common-emitter connection) through a second diode which this time has different polarity. From the output of the this reversing network, the control line T22 leads to the control input of the pulse tube.

Now, in a transmitter, the aim is to adjust a certain output power. If the transmitter power is to be altered, then with the present high-power transmitter, only the direct voltage U= (Figures 5 and 6) need be altered accordingly, because this is a linear function of the direct-voltage component UO of the modulation signal. It is advisable to effect the adjustment of UO with a control loop because this method is the most accurate.

For this purpose, a small extension is necessary at the control device 21, as shown in Figure 5.

By means of an auxiliary direct voltage +UB and a potentiometer Pot, the desired value for the direct-voltage component UO is applied to the negating input of the com parison circuit 21.7. Since a few kV are not being used here, a scale factor k must be taken into consideration; thus Uosoll.k is obtained as a "fictitious" desired value.

Reduced by the same factor k, the direct voltage component UO reaches the non inverted input of the comparison circuit 21.7. The low-pass filter 21.6 (Figure 5) is interposed in the negative feedback branch GK in order to filter out and weaken this direct-voltage component. An operational amplifier wired as a differential amplifier may appropriately be used, in known manner, as a comparison circuit. As is known, this combines the voltages appearing at its two inputs in the form: U= = (Uo.k) - (Uoll.k) or U (UO UON",) k In order to illustrate the regulation let it be assumed that a reduction in the directvoltage component UO (and hence a reduction in power) is being effected by means of the potentiometer Pot (U01I can also be present as a control value (manipulated variable) by the transmitter control).

If UO " K is kept constant, then, as a result of the effect of the control loop, assurance is provided that UO also remains constant. In order that the degree of modulation which has once been set may be retained with an altered direct-voltage component U,, the modulation signal UmOd is conveyed through a multiplication network 21.5 (Figure 5). This alters its amplification in proportion to the direct voltage U=m appearing at its control input.

In order to avoid reactions from the output of the multiplication network 21.5 on its control input, the setting member 21.8 is interposed in the control line. This contains a low-pass filter to ensure that only the direct voltage U= is effective. In addition, the low-pass filter is followed by a potentiometer to adjust the said factor m, with which the required degree of modulation m* can then be adjusted.

PDM (pulse-duration modulated) pulses are again produced in the PS unit (pulse control) Figure 3 and control the control grid of the pulse tube PR.

In pulse-duration modulation circuits, it is often desired to modulate the screen grid of the HF output tube also. According to a known method, a small separate modulation transformer is used, but this is somewhat expensive. Another known method uses a series resistor in the screen-grid circuit; in this case, however, the distortion factor may be increased.

Circuit modifications described below enable the disadvantages of the known methods to be avoided and a simple and cheap screen-grid joint modulation to be realised. According to a first method (Figure 7), a capacitive voltage divider may be installed at the output of the pulse filter, where the capacitor C2 is. In this case, for example C2 5.9 nF and the screen-grid capacity Cg2 60 nF. This can be realised with the circuit of Figure 7. A choke must be installed for the supply of the direct voltage.

The input Pu is connected directly to the terminals of the capacitor C1 in Figures 3, 4.

According to a second method, the mod ulation signal is produced by means of a ta; at the pulse coil SS (Figures 8a and b) ane with a following pulse coil SS' and low-pass filter TP' together with a screen-grid rectifier GR'. Here, however, the internal resistance of the circuit must be very lowfrequency signals, because of the high screen-grid decoupling capacity Cg2, and this may involve difficulties in some circumstances.

A third method consists in the modulation of the screen grid through a separate tube or power-transistor amplifier M (Figure 9).

Care must be taken, however, to ensure that there are no differences in phase between the anode-modulation and the screen-grid modulation voltages. In order to avoid such differences, two modifications are indicated in Figure 9, according to which the modulated signal is supplied through capacitive or ohmic voltage dividers from the anode modulator to the screen-grid modulator.

If the anode voltage is continuously adjustable, then the screen-grid rectifier GR' may, in some circumstances, likewise be continuously adjustable, being controlled through a regulating input Reg. This Reg signal controls the anode voltage at the same time (UO Figures 5; U= Figure 6).

WHAT WE CLAIM IS: 1. A pulse-duration modulation circuit, comprising a pulse tube having a control grid to which pulse duration signals are applied, said pulse tube having its anode connected to a positive voltage output of a mains power supply by a series circuit of a pulse coil, a high-frequency stage and an inductor of a low-pass filter, the anode of the pulse tube also being connected with the positive voltage output of mains power supply through a free-running diode, a junction between one end of the pulse coil and the high-frequency stage being connected to earth and the cathode of the pulse tube being connected to the negative voltage output of the mains power supply.

2. A modulation circuit as claimed in claim 1, in which the mains power supply comprises a mains rectifier and two component filter chokes which chokes are connected between respective said outputs of the mains power supply and respective outputs of the mains rectifier, with a filter capacitor connected between said outputs of the main power supply.

3. A modulation circuit as claimed in claim 2, further comprising series-tuned wave traps connected between said outputs of the rectifier and earth.

4. A modulation circuit as claimed in claim 2, further comprising capacitors connected between said outputs of the rectifier and earth.

5. A modulation circuit as claimed in any one of claims 1 to 4 arranged so that both an anode and a screen grid of the high-frequency stage are modulated.

6. A modulation circuit as claimed in claim 5, comprising a capacitive voltage divider connected at the output of the low-pass filter, its tap being connected to the screen grid of the high-frequency output tube.

7. A modulation circuit as claimed in Claim 5, in which, for providing the modulation signal for the screen grid of the highfrequency output tube, there are provided a tap on the pulse coil, followed by a further coil, a low-pass filter and a screen-grid rectifier.

8. A modulation circuit as claimed in Claim 5, in which, for modulation of said screen grid, there is provided a separate tube or power-transistor amplifier.

9. A modulation circuit as claimed in Claim 8, in which the modulated signal is supplied from the anode modulator to the screen-grid modulator through capacitive voltage dividers.

10. A modulation circuit as claimed in Claim 8, in which the modulated signal is supplied from the anode modulator to the screen-grid modulator through ohmic voltage dividers.

11. A modulation circuit substantially as herein described with reference to Figures 3,5 and 6 or 4,5 and 6 optionally modified according to Figure 7,8 or 9 of the accompanying drawings.

A.A. THORNTON & CO.,

**WARNING** end of DESC field may overlap start of CLMS **.

Claims

**WARNING** start of CLMS field may overlap end of DESC **. ulation signal is produced by means of a ta; at the pulse coil SS (Figures 8a and b) ane with a following pulse coil SS' and low-pass filter TP' together with a screen-grid rectifier GR'. Here, however, the internal resistance of the circuit must be very lowfrequency signals, because of the high screen-grid decoupling capacity Cg2, and this may involve difficulties in some circumstances. A third method consists in the modulation of the screen grid through a separate tube or power-transistor amplifier M (Figure 9). Care must be taken, however, to ensure that there are no differences in phase between the anode-modulation and the screen-grid modulation voltages. In order to avoid such differences, two modifications are indicated in Figure 9, according to which the modulated signal is supplied through capacitive or ohmic voltage dividers from the anode modulator to the screen-grid modulator. If the anode voltage is continuously adjustable, then the screen-grid rectifier GR' may, in some circumstances, likewise be continuously adjustable, being controlled through a regulating input Reg. This Reg signal controls the anode voltage at the same time (UO Figures 5; U= Figure 6). WHAT WE CLAIM IS:

1. A pulse-duration modulation circuit, comprising a pulse tube having a control grid to which pulse duration signals are applied, said pulse tube having its anode connected to a positive voltage output of a mains power supply by a series circuit of a pulse coil, a high-frequency stage and an inductor of a low-pass filter, the anode of the pulse tube also being connected with the positive voltage output of mains power supply through a free-running diode, a junction between one end of the pulse coil and the high-frequency stage being connected to earth and the cathode of the pulse tube being connected to the negative voltage output of the mains power supply.

A.A. THORNTON & CO.,