GB2380616A

GB2380616A - A signal combining device

Info

Publication number: GB2380616A
Application number: GB0113239A
Authority: GB
Inventors: Johannes Benedict
Original assignee: Nokia Oyj; Nokia Inc
Current assignee: Nokia Oyj; Nokia Inc
Priority date: 2001-05-31
Filing date: 2001-05-31
Publication date: 2003-04-09
Also published as: GB0113239D0

Abstract

A bias T device and a method for combining a first signal DC with a second signal RF is disclosed. The device is provided with a first signal splitting means 30 having at least two isolated transmission lines 34,40. The first signal splitting means 30 are adapted to receive said first signal DC at one input and said second signal RF at another input. A second signal splitting means 32 having at least two isolated transmission lines 34' 40' is also provided. Said second signal splitting means 32 are coupled to said first signal splitting means 30 such that the respective sets of transmission lines comprise isolated signal routes through said device. An output provides an output signal comprising a combination of said first signal and said second signal. A coupling effect between at least two of said transmission lines is used to combine said first signal with said second signal.

Description

SIGNAL COMBINING DEVICE Field of the Invention This invention relates to signal combining devices and particularly to bias T devices suitable for broadband and high current applications.

Description of Background Art A bias T is a type of signal combining device which can combine a first input signal with a second input signal to provide an output signal comprising a combination of the first and second inputs. In general, the first and second inputs are independent from one another and are not affected in any way by connection to the bias T device or its mode of operation. This type of device may also be referred to as a Bias"Tee"device.

Owing to these properties bias T devices are useful in device characterisation and testing applications. For example, bias T devices may be used to apply direct current (DC) offsets or level shifts in radio frequency (RF), pulsed radio frequency and base band testing of active devices.

Figure 1 shows a known bias T device. The bias T device 10 comprises a first input terminal 12 for receiving a radio frequency (RF) varying signal, a second input terminal 13 for receiving a direct current (DC) signal and an output terminal 14 which comprises a combination of the radio frequency and DC signals. Within the bias T a radio frequency transmission line 15 is connected to the radio

frequency input 12 and is provided with a capacitor 16. A direct current transmission line 17 connected to the direct current input 13 is provided with an inductor 18. The radio frequency and direct current transmission lines 15,17 are connected at a node 20 located on the output side of respective components 16 and 18. The node 20 is connected to the output terminal 14.

In use, the capacitor 16 on the radio frequency transmission line 15 presents a low impedance to radio frequency signals and a high impedance to direct current. The capacitor 16 thus acts as a series blocking capacitor preventing the direct current signal from interfering with the radio frequency signal supplied to input 12. The inductor 18 on the direct current transmission line 17 presents a high impedance to radio frequency signals and a low impedance to direct current. Hence the inductor prevents the radio frequency signals which pass through the capacitor 16 from interfering with the direct current input signal supplied to input terminal 13.

The known bias T device of Figure 1 would typically operate in direct current ranges of up to about half an amp and radio frequency power ranges of up to 40GHz. The current handling and radio frequency power handling of known bias T devices is limited because such devices typically modify signal flow by virtue of transmission line geometries, for example spiral wire structures or through the effects of components connected in series with the transmission lines. A problem with such devices is that high currents and/or high radio frequency power cause heating effects which limit the range of applications of the devices.

Summary of Invention The invention seeks to provide a device and a method by means of which improved signal combining may be provided.

According to an aspect, this is provided by a bias T device for combining a first signal with a second signal, the device comprising: a first signal splitting means having at least two isolated transmission lines, said first signal splitting means being adapted to receive said first signal at one input and said second signal at another input; a second signal splitting means having at least two isolated transmission lines, said second signal splitting means being coupled to said first signal splitting means such that the respective sets of transmission lines comprise isolated signal routes through said device; and an output to provide an output signal comprising a combination of said first signal and said second signal, wherein a coupling effect between at least two of said transmission lines is used to combine said first signal with said second signal.

Another aspect of the present invention provides a method of combining signals, the method comprising the steps of: receiving a first input signal at an input of a first signal splitting means having at least two electronically isolated transmission lines; receiving a second input signal at another input of said first signal splitting means; splitting at least one of the signals at the first splitting means; passing signals from said first signal splitting means to a second signal splitting means having at least two electronically isolated transmission lines ; and combining said first and second signals by means of a coupling effect between said electrically isolated transmission lines.

More specific forms of the invention are defined by the dependent claims.

Brief Description of Drawings Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings in which: Figure 1 illustrates a known bias T device; Figure 2 illustrates a high current bias T device embodying the present invention; Figure 3a is a schematic plan view of a 900 hybrid circuit such as those used in the embodiment of Figure 2; Figure 3b is an exploded perspective view of a 900 hybrid circuit; Figure 4 illustrates the embodiment of Figure 2 with signal routes illustrated schematically; Figure 5 illustrates S-parameter transmission results for the embodiment of Figure 2; Figure 6 illustrates another high current bias T device embodying the present invention; Figure 7a is a schematic plan view of a 900 hybrid circuit such as those used in the embodiment of Figure 6; Figure 7b illustrates an exploded perspective view of the 900 hybrid circuit; Figure 8 illustrates the embodiment of Figure 6 with signal routes illustrated schematically; and Figure 9 is a flow chart illustrating a mode of operation of bias T devices embodying the present invention.

Detailed Description of Preferred Embodiments

Figure 2 shows a high current bias T device 20 implemented using two back-to-back hybrid circuits. The bias T has a first input port 22 adapted to receive a direct current signal and a second input port 24 adapted to receive a radio frequency signal. The bias T device 20 has two further ports, the first of which 26 provides an output signal combining the signals at the input ports 22 and 24 and the second of which 28 is terminated by a 50 load. This 50 load is optional and is connected to the bias T in order to counter some imperfections of the hybrids as will be explained hereinafter. The bias T device 20 comprises a first 900 hybrid circuit 30 coupled to a second 900 hybrid circuit 32. The input ports 22 and 24 are provided on the first 900 hybrid circuit 30 and the output ports 26 and 28 are provided on the second 900 hybrid circuit 32.

The bias T device of Figure 2 is used here to combine a radio frequency signal supplied at port 24 with a direct current signal supplied at port 22 to provide an output signal at port 26 comprising direct current and radio frequency components without either of the signals at ports 22 and 24 being perturbed in any way.

Figure 3a and 3b illustrate a 900 hybrid circuit of the type used to implement the bias T device of Figure 2. The schematic plan view of Figure 3a shows a transmission line 34 connecting a first input port 36 to a diagonally opposed output port 38. A second transmission line 40 connects a second input port 42 to a second diagonally opposed output port 44. The transmission lines 34,40 may be of any convenient type, for example they may be waveguides or microstrip lines. In this embodiment, the transmission lines are disposed within a block of dielectric material 50.

With reference to Figure 3b, the dielectric block 50 is comprised of an upper substrate 52, a thin isolation layer 54 and a lower substrate 56. A first metalised region 60 disposed on a lower surface of the upper substrate layer 52 defines the transmission line 34. A second metalised region 62 disposed on an upper surface of the lower substrate layer 56 defines the transmission line 40. The transmission lines 34,40 follow diagonal paths on opposed faces of the respective substrate layers 52,56. The isolation layer 54 disposed between the transmission lines 34,40 electronically isolates the transmission lines from each other. The inputs 36 and 42 are also electronically isolated from one another.

The two transmission lines of each hybrid circuit are arranged in such a way that a coupling occurs between the lines for signals within the operational bandwidth of the 90 C hybrid. For signals having frequencies outside the operational bandwidth of the hybrid circuits no coupling occurs (e. g. a DC signal) and isolation between the transmission lines is effective to prevent signals on different lines from mixing.

Each 900 hybrid circuit 30,32 acts as a three decibel coupler with a transmission line 34 connected between ports 36 and 38 and a second transmission line 40 between ports 42 and 44. In one envisaged use, a direct current provided at the input port 36 will be supplied directly to the output port 38 without any current leaking onto the transmission line 40. However, the hybrid circuit would split a radio frequency into two separate radio frequency output signals. An input signal of X decibels would be split into two equal amplitude output signals at ports 44 and 38 and each having

a power of X-3 decibels. A signal diagonally traversing a 900 hybrid circuit experiences a 90 phase shift relative to a signal following a straight through signal path. Thus the radio frequency output signals at ports 44 and 38 are of equal amplitude and have a relative phase offset of 900.

The above assumptions neglect line losses and imperfections in the hybrid circuit.

Figure 4 shows the bias T device 20 of Figure 2 with signal routes through the device illustrated schematically. The two 900 hybrid circuits are coupled together such that the transmission line 34 of the first hybrid circuit 30 connects to the transmission line 40'of the second hybrid circuit 32. This forms a continuous signal path 72 between ports 22 and 26 of the device. The transmission line 40 of the first hybrid circuit 30 connects to transmission line 34 of the second hybrid circuit 32 forming a continuous signal path 70 between ports 24 and 28 of the device. The continuous paths are electrically isolated from one another by the isolation layers 54 in respective hybrid circuits 30,32. Signals within the operational bandwidth of the 90 C hybrids experience a coupling effect and are transferred between the two transmission paths. Signals outside this operational bandwidth (e. g. a DC signal) do not couple and so cannot overcome the isolation between the transmission paths to transfer from one to the other.

The direct current input at the port 22 of the first hybrid circuit 30 travels along the signal path 72 through the first hybrid circuit 32 and into the second hybrid circuit 32. The DC current signal is output at port 26 without any DC signal components affecting the signal on the signal path 70 or at either of the ports 24 and 28.

A radio frequency signal which is input at port 24 does not follow a single path. The radio frequency signal is split into two separate signals as it passes through the first hybrid circuit 70. The splitting of the radio frequency is caused by a coupling between the transmission line 40 connected to port 24 and the transmission line 34 which is also used for the DC signal. The first RF and second RF2 radio frequency signals resulting from the coupling affect have a phase difference of 900. The signals RFi and RF2 are output from the first hybrid circuit 30 and supplied to the respective input ports of the second hybrid circuit 32. A coupling effect between transmission lines 34'and 40'of the second hybrid circuit 32 means that the radio frequency signal RFi received at the input connected to line 34'is coupled onto the transmission line 40'. At the same time the radio frequency signal RF2 received at the input connected to line 40'is coupled onto transmission line 34'.

Each of the signals RFi and RF2 is split into two further signal components as they traverse the second hybrid circuit 32. Between these signal components an additional phase shift is imparted. As a result, the signal components interfere constructively at the output port 26. The signals interfere destructively at the output port 28 such that they can cancel each other out.

If the phase offset within a hybrid is not exactly 90 C the signals at port 28 will not be exactly combined out-of-phase and hence they won't cancel each other out completely. The 50Q load absorbs the remaining signal preventing them from interfering with signals at port 26.

The result is that the radio frequency signal input at the port 24 appears without loss at the output port 26 where it is combined with the direct current signal also supplied to that port. The line losses and losses caused by imperfections are neglected within this specific description since they are small within the realized device. However, within other embodiments (e. g. using hybrids with larger imperfections) they may not be neglected.

The above described embodiment thus uses a coupling effect between separate transmission lines in 900 hybrid circuits to combine a radio frequency signal with a DC signal. The coupling effect depends mainly on the distance between the two transmission lines and does not depend substantially on the width of the transmission lines. Accordingly, the above embodiment overcomes problems with the prior art in that thick transmission lines can be used in preferred embodiments. Therefore preferred embodiments are not limited to low current ranges or narrow radio frequency operating bandwidths. Therefore, no trade-off is necessary between high DC current handling capability and broad bandwidth of the signal-combining device.

Figure 5 shows transmission S-parameter results obtained using a bias T device according to Figure 2. The hybrid circuits used were rated to pass signals at frequencies in the range of 1.3 to 10GHz. A skilled person will appreciate that these S-parameter results were obtained by connecting a suitable frequency generator to the input port 24 and measuring the response at the output port 26 of the bias T device 20. Referring to Figure 5, the bias T operates reliably in the bandwidth range 1.3 to 10GHz. Satisfactory

results have been obtained using DC input currents of 10 or more amps.

The embodiment described with reference to Figures 2 to 4 uses a DC signal at port 22. However, the device is suitable to combine any signals of which one is outside and the other within the operational bandwidth of the hybrids. That is, the above-described device would combine any signals input at port 22, which may be outside the operational bandwidth of the hybrid circuitry, with any signals input at port 24, which are within the operational bandwidth of the hybrid circuitry. The device in Figures 2 and 4 splits only the signal inserted at port 24 into multiple frequency components with a phase offset. As a result, the device may not require filters or other selective signal blocking means between the two hybrids.

That is, the principle of operation relies on only one input signal component being split into multiple signals while the second signal is directly forwarded from one port to another without being split.

Preferred embodiments such as the one illustrated in Figure 2 can therefore be used to combine two radio frequency signals. For example, this can be achieved by inputting a first radio frequency signal (having a frequency outside the operational bandwidth of the hybrid circuits) to input port 22 and inputting a second radio frequency signal (having a frequency within the operational bandwidth of the hybrid circuits) to input port 24. The signal output at port 26 is then a combination of the first and second RF signals input at ports 22 and 24.

Using different hybrid circuits with larger frequency operating ranges it is possible to provide bias T devices with different operating ranges. For example, other hybrid circuits have operating frequencies in the range 1 to 18GHz or more and so can extend the operating bandwidth of preferred bias T devices up to 18GHz. Other hybrid circuits can extend the operating bandwidth still further.

Appropriately designed transmission lines will permit still higher operating currents. Preferred embodiments thus provide bias T devices which are operable at high currents and over large bandwidth ranges. Such bias T devices are useful for example in the testing of high powered transistors. However, preferred embodiments can also be used in many other applications, such as in amplification applications.

Figure 6 shows another bias T device embodying the present invention. The transmission lines (and inputs) of the hybrid circuits employed in this embodiment are not electrically isolated from one another. The bias T device comprises a first input port 122 for receiving a direct current signal and a second input port 124 for receiving a radio frequency signal. The bias T device 100 has two further ports, the first of which 126 provides an output signal which is a combination of the radio frequency and direct current signals at input ports 124 and 122. The second further port is an output port 128 terminated by an optional 50 electrical load. The bias T device 100 comprises a first 900 hybrid circuit 130 and a second 900 hybrid circuit 132.

In this embodiment, the radio frequency input port 124 is provided on the first hybrid circuit 130 and the direct

current input port 122 is provided on the second hybrid circuit 132. High pass filters 180 and 182 are disposed between the first and second hybrid circuits 130,132, one on each of the signal paths as will be explained herein. The bias T device 100 of Figure 6 can combine a radio frequency signal supplied to the port 124 with a direct current signal supplied to the port 122 in order to provide an output signal comprising both radio frequency and direct current components without affecting either of the input signals in any way.

Figures 7a and 7b illustrate a hybrid circuit of the type described to implement the bias T device 100 of Figure 6.

The transmission lines are arranged to form a quadrilateral shape 134 with spurs 140 connecting to each of the input and output ports. The transmission lines 134, 140 are formed by a metalised region disposed between dielectric layers 152,156. Since there is no isolation between any of the respective transmission lines (or the input and output terminals) the metalised region 160 of the 900 hybrid circuit may be regarded as a continuous transmission line system. Hybrid circuits of the type shown in Figure 7a and 7b can split a radio frequency signal input at port 136 into two radio frequency signals of substantially the same amplitude which are output at ports 144 and 138. The radio frequency signals output at ports 144 and 138 have a relative phase difference of 900 which is imparted to them by the hybrid circuit.

Figure 8 shows the bias T device 100 of Figure 6 with signal paths through the device illustrated schematically. The two 900 hybrid circuits 130,132 are arranged as in a branch line coupler with two high pass filters 180, 182 coupling them

together. The high pass filters 180,182 are tuned so as to prevent the direct current passing while allowing the radio frequency signal to pass. The direct current signal input at port 122 of the second hybrid circuit 132 is blocked by the high pass filters 180,182. The direct current signal is output at port 126 of the second hybrid circuit 132 without passing into the first hybrid circuit 130.

The radio frequency signal input at port 124 of the first hybrid circuit 130 is split into two separate signals RFi and RF2 as it traverses the first hybrid circuit 130. The second radio frequency signals RFi and RF2 output from the first hybrid circuit 130 have a relative phase offset of 900. The radio frequency signals RFl and RF2 pass through the respective high pass filters 180,182 and are input to the respective input ports of the second hybrid circuit 132. The radio frequency signals RFi and RF2 are split into further radio frequency components while they traverse the second hybrid circuit 132. The output signals derived from radio frequency signal RFi have a 900 relative phase offset after they have traversed the hybrid circuit 132. The output signals derived from radio frequency signal RF2 have a similar 900 phase offset after they have passed through the hybrid circuit 132. Accordingly, the signals derived from the signals RFi and RF2 interfere constructively at the output port 126 and destructively at the output port 128. The result is that the radio frequency signal input at port 124 appears substantially without loss at the port 126 where it combines with the direct current signal also supplied to that port from port 122.

In the example described with reference to Figures 6 to 8, a device is used to combine a radio frequency signal input at

port 124 with a direct current signal input at port 122.

The signal input at port 122 need not be a direct current signal. For example, the signal input at port 122 may be an alternating signal having a frequency anywhere in the range between zero and the cut-off frequency of the high pass filters 180,182. In practice, the high pass filter can be selected to have a lower cut-off frequency below that of the signal input at port 124. In this way, the device of Figures 6 to 8 can be used to combine a first alternating signal input at point 124 with a second signal having a frequency anywhere between zero up to the frequency of the first alternating signal. It will be apparent that the signal input at port 122 can be inside or outside the operational bandwidth of the hybrid 132.

Both embodiments described herein use hybrid circuits to recombine components of at least one of the input signals.

Reference is now made to Figure 9 illustrating the operation in accordance with an embodiment.

At step 901, a combining device receives first and second input signals. In the examples described herein the first input signal is a direct current signal and the second input signal is a radio frequency signal.

At step 902, a first signal splitting means is supplied with the second input signal which it splits into first and second signal components. The first signal splitting means also imparts a relative phase shift to signals which pass through it such that the first and second signal components have a predetermined phase offset.

At step 903, a second signal splitting means is arranged to split each of the first and second signal components into at least two further signal components. The second signal splitting means also imparts a further relative phase shift to signals which pass through it such that the further signal components have a further predetermined phase offset.

At step 904, a representation of the second input signal is provided at an output where interference effects between the further signal components arise.

Finally, at step 905 the device outputs a combination of the first input signal and the representation of the second input. Owing to the nature of the interference effects which lead to the representation of the second input signal the signals are combined at the output substantially without any losses.

The above describes the operation in accordance with Figures 2 to 5 in which only one of the signals is split. In the embodiment as described with reference to Figures 6 to 8 both of the incoming signals are split. In the latter embodiment the signal inserted at port 122 does not necessarily have to be outside the operational bandwidth of the hybrid circuit 132. If the signal is outside the operational bandwidth the signal splitting does not take place and this signal is simply output at port 126 while the other signal is processed as described above. If the signal is inside the operational bandwidth the signal inserted at port 122 is split into two components with a 900 offset.

These components are output at the filters 180 and 182. These two filters are preferably selected such that they reflect the signal inserted at port 122. These signal

components are then re-reflected into the hybrid 132 and each signal component is split into two further components with a 900 offset. The created signals combine constructively at port 126 and destructively at port 122.

It will be apparent that the embodiment of Figures 6 to 8 affords similar advantages to the embodiment of Figures 2 to 4.

Implementations of the invention should not be limited to the configurations of the described embodiments.

Specifically, the described embodiments are examples of configurations which may be used to implement preferred methods and are not intended to define the only apparatus features/method steps which can be used.

Claims

CLAIMS: 1. A bias T device for combining a first signal with a second signal, the device comprising: a first signal splitting means having at least two isolated transmission lines, said first signal splitting means being adapted to receive said first signal at one input and said second signal at another input; a second signal splitting means having at least two isolated transmission lines, said second signal splitting means being coupled to said first signal splitting means such that the respective sets of transmission lines comprise isolated signal routes through said device; and an output to provide an output signal comprising a combination of said first signal and said second signal, wherein a coupling effect between at least two of said transmission lines is used to combine said first signal with said second signal.
2. A device according to claim 1, wherein at least one of said first and second signal splitting means comprises a hybrid circuit.
3. A device according to claim 1 or 2, wherein each said first and second signal splitting means comprises a 900 hybrid circuit.
4. A device according to claim 1,2 or 3, wherein components of said second input signal are transferred between said first and second signal paths by means of said coupling effect.

<Desc/Clms Page number 18>
5. A device according to any preceding claim, wherein said first signal has a frequency that is outside the operational bandwidth of said signal splitting means and said second signal has a frequency that is within the operational bandwidth of said signal splitting means, whereby only the second signal is split by the splitting means.
6. A device according to any of claims 1 to 4, wherein said first and second signals have frequencies that are within the operational bandwidth of said signal splitting means.
7. A device according to any preceding claim, wherein said first signal splitting means is arranged to split said second input signal into first and second components, said first and second components having a predetermined phase offset.
8. A device according to claim 7, wherein said second signal splitting means arranged to receive said first and second signal components and to split each of said first and second signal components into at least two further signal components, said second signal splitting means being adopted to import a relative phase shift such that said further signal components have a further predetermined phase offset.
9. A device according claim 8, wherein said further signal components interfere to provide a representation of said second input signal substantially with losses.
10. A device according to any preceding claim comprising a second output terminated with a predetermined electrical load.

<Desc/Clms Page number 19>
11. A device according to claim 9, wherein two of said further signal components interfere constructively at said first output port and two of said further signal components interfere destructively at said second output port.
12. A device according to any preceding claim, wherein a signal path of the device comprises a waveguide.
13. A device according to any preceding claim, wherein a signal path of the device comprises a microstrip transmission line.
14. A method of combining signals, comprising the steps of: receiving a first input signal at an input of a first signal splitting means having at least two electronically isolated transmission lines; receiving a second input signal at another input of said first signal splitting means; splitting at least one of the signals at the first splitting means; passing signals from said first signal splitting means to a second signal splitting means having at least two electronically isolated transmission lines; and combining said first and second signals by means of a coupling effect between said electrically isolated transmission lines.
15. A method as claimed in claim 14, wherein the transmission lines of said first and second signal splitting means are coupled such that respective sets of transmission lines in the hybrid circuits comprise at least two isolated signal routes.

<Desc/Clms Page number 20>
16. A method as claimed in claim 14 or 15, wherein said first signal splitting means splits said second input signal into first and second signal components.
17. A method as claimed in claim 16, comprising imparting a relative phase shift such that said first and second signal components have a predetermined phase offset.
18. A method as claimed in claim 16 or 17, wherein said second signal splitting means splits each of said first and second signal components into at least two further signal components.
19. A method as claimed in claims 17 and 18, wherein the second slitting means imparts a further relative phase shift such that said further signal components have a predetermined phase offset with respect to one another.
20. A method as claimed in any of claim 14 to 19, wherein high pass filtering in penetrable to said first input signal is applied to at least one of the signals between said first and second signal splitting means.
21. A method as claimed in claim 18 or any claim dependent on claim 18, wherein a representation of said second input signal is provided from interference effects between said further signal components.
22. A method as claimed in any of claims 14 to 21, wherein both input signals are split by the first splitting means.