Mixer Technology: The Devise of Increasing Importance
Mixers are three port devices that are used to translate one frequency higher or lower for the purpose of transmitting or processing the signal more effectively. Mixers definitely had a pivotal role in the history of beyond the radio receiver development. With the advent of TV and radio broadcasting, the demand for radio technology grew beyond the ranks of the military and the hobbyists to the all-powerful consumer. As the 21st century is churning, mixer technology is constantly improving to jive with the wireless technologies that are booming. The changes it had undergone with time coincide with manufacturing efficiencies to reduce the cost and improve the performance of various electron devices. In essence, mixer technology will always be essential because it is a crucial tool in the transmission or processing of information.
According Dunleavy & Connick (2004), a mixer is a three port device that is used to translate one frequency higher or lower for the purpose of transmitting or processing the signal more effectively. As in Figure 1, a mixer has the input (Radio Frequency, RF) signal port and output (Intermediate Frequency, IF) signal port, uses a third Local Oscillator (LO) port to drive the mixer.
Figure 1. The Terminal Nonlinear Device of a Mixer.
The importance of mixers comes in systems to enhance the upconversion and downconversion of signals is crucial to radio operation. In some sense, mixers are highly nonlinear components, in which the higher-order terms in a system transfer characteristic are intentionally used to translate between one frequency and another. In another sense, however, the relationship between the input signal and its translated counterpart needs to be quite linear, in which all the usual linear concepts of superposition and matrix algebra apply (Gilmore 2003, p. 433).
Futhermore, Dunleavy & Connick (2004) explained that mixers are used to perform the multiplication function in the time domain. The output of a mixer up to the second order term is:
However the desired signal is either the sum or difference frequencies shown below and the other products are often filtered out:
Some basic parameters used to describe a mixers performance are:
Above the subscript denotes the frequency the power is measured and the superscript represents the port at which the power is measured. In essence, in the field of electrical engineering, mixers are essential for operations such as frequency translation, modulation and detection. As a consequence, they are an important building block for both transmitters and receivers.
The history of mixers has been identified to date back in the late 1800s. French physicist Edouard Branly (1846– 1940) had introduced a coherer. It was based on the properties of small metal particles between two electrodes in an evacuated glass tube. This invention of vacuum tubes was a great step forward toward better transmitters and receivers. In 1904, the British physicist John Ambrose Fleming (1849– 1945) invented the rectifying vacuum tube, the diode. In 1906 the American inventor Lee De Forest (1873– 1961) added a third electrode, called a grid, and thereby invented the triode. The grid controlled the current and made amplification possible. The efficiency of the electron tubes was greatly improved by using concentric cylinders as electrodes. Later, De Forest and the American engineer and inventor Edwin Armstrong (1890– 1954) independently discovered regenerative feedback in 1912. This phenomenon was used to produce a continuous carrier wave, which could be modulated by a voice signal. Armstrong invented also the superheterodyne receiver. These techniques made broadcasting possible. AM stations began broadcasting in 1919 and 1920. Regular TV transmissions started in Germany in 1935. Armstrong’s third great broadcasting invention was FM radio, but FM broadcasting was accepted not until after World War II (Rainen 2003, p. 6-7).
For Long (2003), the history of mixer technology in use today was mathematically known several hundred years ago and reduced to practice by 1935 when the Double Pole, Double Throw (DPDT) vibrator modulator was used in low frequency amplifiers. Mixers were quickly adapted at that time to receivers as well as transmitters for their efficient frequency translation properties. In 1949, this function was performed by the 7360 vacuum tube in radio circuits and six vacuum tubes in rocket telemetry circuits. In 1963, Jones of Burroughs Corp. patented a common mode feedback addition to the transistor version of the circuit to mitigate the poor performance. This circuit in integrated circuit form was market by Sylvania in 1964 as an Engineering Case Library (ECL) exclusive or gate. People today who get their history from journalists instead of reality call this topology the “Gilbert Cell Mixer”.
Over the years several companies have tried to market the Gilbert mixer with little success. This circuit only came into its own when the cellular systems were put in place and the handsets had severe restrictions on size and battery performance. The limitations of this circuit are mitigated by using Field Effect Transistors (FET) which has less Intermodulation Distortion (IMD) that bipolar transistors.
Mixer definitely had a pivotal role in the history of radio receiver development. For as the number of permanent transmitting stations grew exponentially, from a scant 100 stations in the U. S. in 1905 to over 1100 stations only ten years later, the scarcity of spectrum and the accompanying drive to higher and higher frequencies (a drive that continues to this day) stimulated the development of radio receiver architectures that were increasingly sensitive and selective. With the advent of TV and radio broadcasting in the 1920s and 30s, the demand for radio technology grew beyond the ranks of the military and the hobbyists to the all-powerful consumer. The economic incentives for satisfying this demand added fuel to the fire, as evidenced by the rapid pace of technological progress, a renewed interest in “short-wave” radio, and a marked increase in patent litigation. The technologies developed along the way in part to meet the increasing demands of radio reception – such as the vacuum tube (or Audion, as it was
originally called), the piezoelectric resonator, and later on the transistor and the integrated circuit – tell the story of electronics in general, not just of radio (Shaeffer 1999, p. 2).
The most common application of a mixer is the superheterodyne receiver design. Nearly all radio and TV receivers said today are “superhets”. In the superhet, the incoming RF signal (FRF) is converted to an intermediate frequency (IF) by “beating” a local oscillator (LO) signal (FLO) against the RF frequency. That process is called heterodyning, from which the name superheterodyne was derived (Carr, January 1999). A superheterodyne receiver can provide both good selectivity and sensitivity, because the noise bandwidth can be limited to the channel bandwidth without compromising the receiver’s ability to tune across the entire RF band. Its basic components are shown in Figure 2.
Figure 2. The Basic Architecture of the Analog Superheterodyne Receiver (Source: Besser, 1999).
Figure 3 shows the block diagram of a typical superhet receiver. The radio signal at FRF is picked up by antenna, and then (sometimes) amplified in an RF amplifier circuit. This signal is mixed with FLO In the mixer to produce two output signals, in addition to F1 and F2:F1+F2 (sum) and F1-F2 (difference). These frequencies correspond to the results when m = n = 1.
The IF amplifier is used to process the output signal of the mixer. Either the sum or the difference signal can be used for the IF, as they are mirror images of each other. In AM and FM Broadcast Band (BCB) receivers, the difference frequency is typically used. In the United States and Canada, manufacturers usually use 455 kHz as the AM BCB IF, and 10.7 MHz as the FM BCB IF. In car radios, 262.5 kHz is used as the AM BCB IF. European and Japanese radios have sometimes used other frequencies for the IF, but they are close to these values. In recent years, High Frequency (HF) shortwave receiver designs use two conversion steps. The first IF is around 50 MHz, but this frequency is later down converted to a second IF such as 9 MHz, 8.83 MHz, or 10.7 MHz. In Fig. 3 the difference frequency FLO-FRF is shown being selected. The IF amplifier provides most of the signal gain found in the receiver. It also provides most of the selectivity of the receiver. The reasons for doing the frequency conversion is that the IF amplifier operates at one frequency only, and that means it is easier to obtain quality selectivity characteristics (the best filters are single-frequency devices) and high gain without either variation in gain with frequency or spurious oscillations.
Figure 3. The Block Diagram of a Typical Superheterodyne Receiver. Though the difference-signal output of the mixer is used here, either the sum or difference signal output could be used (Source: Carr, January 1999).
On the contrary, diode mixers are important because of their low noise characteristics. While single diode mixers are not normally used at frequencies below 1 GHz, their analysis provides some useful insight into the operation of mixers in general. It should be noted that diode capacitance can have a detrimental effect upon mixer performance and low capacitance devices, such as the Schottky barrier diode, will normally be required for operation at higher frequencies.
The main task of the diodes is to take the RF and LO signals and mix them together to form new signals. The main requirement is that there must be an even number of diodes so that each side can be fed the combined signals evenly. Some studies also say that a single-diode mixer configuration illustrate the basic operation of the three sections that make up a mixer. However, such a configuration is not recommended, since both the losses and the noise are very high, and you will not obtain optimum performance (Laverghetta 2005, p. 122).
For most applications of a mixer, the output is either the sum of the two signals (F1 + F2) or the difference (F1 – F2). If the sum is used, the mixer is called an upconverter. If the difference is used, the mixer is called a downconverter. The choice of an upconverter or downconverter depends simply on the type of filter that is used in the mixer. With the proper filter at the output of the mixer, you have a completed circuit. Thus, the mixer must be a nonlinear device in order to mix the signals together. It is assumed that the signals were the same level and would be amplified equally in the linear case.
For the mixer, the input signals are very different in level. The RF signal usually is a very low level signal. Many times it is the signal that comes in right off an antenna, and the level may be -70 to -80 dBm or even lower. The LO signal, as we have previously stated, is the key signal for making the mixer operate properly. The level of the LO signal is substantially higher than that of the RF signal and may be 0, +5, or as high as +27 dBm. This very high level hits the diodes at the same time the RF signal does and results in the diodes being driven very hard and into the nonlinear region. That is how the mixer creates the nonlinear effect and mixes the two signals. If the LO signal level falls off for any reason, there is a distinct possibility that the mixer will not mix the two signals. In that case, there will be no output from the mixer, since the output filter is designed to pass only the sum or difference frequency of the two inputs. So it can be seen that the level of the LO is crucial to the operation of the mixer (Abrie 1999).
Moreover, certain terms are associated with mixers. One term is conversion loss. Insertion loss describes directional couplers because it is the low loss going from the input to output on the straight-through transmission line and usually is a very low value. The key idea behind insertion loss is that it is a loss from input to output of a device (e. g., the directional
coupler). The conversion loss in a mixer is similar to the insertion loss in a directional coupler in that it is a measure of the performance from the input to the output. For a mixer, it is the difference in signal level (in decibels) between the RF input and the IF output. This loss is a little more difficult to measure than the insertion loss of a directional coupler because the input and the output of a mixer are at different frequencies, whereas the directional coupler insertion loss is at the same-frequency input or output. The best way to look at the conversion loss of a mixer is with a spectrum analyzer, which can distinguish between the two frequencies and give you an absolute power level for each signal. Then it is a matter of taking the difference in decibels and obtaining a figure for conversion loss. Typical values range from 6 to 9 dB. Some mixer assemblies do not have a loss from the RF input to the IF output. These are more appropriately called mixer-amplifier assemblies rather than simply mixers. An amplifier at the IF output raises the level of the output so there is a gain rather than a loss through the entire circuit.
Another term that is important when we discuss mixers is the isolation, particularly the LO-to-IF isolation. This value (in decibels) tells you how well the LO signal, which is considerably higher than any other signal in the mixer, is attenuated so it does not interfere with the desired output. Some mixer circuits have separate filters built in that are designed to greatly attenuate the LO signal at the IF output so that it does not overpower the required IF signal, thus producing a more efficient mixer circuit. The value of RF-to-LO isolation also is important. It is necessary to keep the RF and LO signals apart at the input, and these signals should be combined only at the diodes so they can do the task they are supposed to do. Values for both RF-LO and LO-IF isolation should be in the range of at least 30 dB; 40 to 45 dB would be even better (Coleman, 2004).
The secret to many mixers, and in fact much of the active research in mixers, is in the baluns or combiners that simultaneously impress the strong LO switching waveform across the mixer added to the much smaller RF input signal. The term balun is generally used for the three or four-port device that is configured to linearly sum the incident voltages at the two balun input ports (the LO and RF), rather than for achieving single-ended to differential conversion as is commonly the case in other types of circuits (e. g., push-pull amplifiers). Of course, the same circuit can often be used for either function. The effective voltage applied across the time-varying conductance is then the input signal voltage. For although the mixer model could have three ports, diodes have only two terminals and transistors three, so some means of feeding the device with two signals and for extracting the third needs to be created.
Other important mixer characteristics are port-to-port isolation and image rejection. In the former, it is to guaranteed, for example, that no LO signal appears at the RF or IF ports in order to prevent desensitization of the preamplifier or IF amplifier, or even its radiation through the receiving antenna. In the latter, we ought to prevent noise present at the Image Frequency (IM), from being converted onto the desired IF. This is particularly important in multichannel systems where the undesired IM channel (one which is the mirrored image of the RF across the LO symmetry axis) may have an amplitude even higher than the sought channel. As a matter of fact, the main role of the input coarse channel-selection filter is IM rejection. Because this IM frequency is located only two times the IF apart from the RF input, available IM rejection filter cut-off slopes may dictate the IF frequency value, and so the number of different IF chains in the receiver’s architecture. There are many distinct mixer topologies, which differ from the selected nonlinearity or the number of equal coupled nonlinearities. According to the selected nonlinearity, there are diode mixers, normally using high-speed Schottky diodes, or transistor mixers, either based on FETs or Bipolar Junction Transistors (BJTs). Referring to the number of those nonlinear devices needed, the mixer can be unbalanced (only one device), singly balanced (at least two equal devices), or doubly balanced (at least four equal devices). However, every one of these follows a common design procedure, which will be exemplified for one simple, but also of practical relevance, unbalanced topology: the active FET mixer (Garcia et al., 1999).
For FETs, this model can be used in mixers in both active and passive modes. Active FET mixers are transconductance mixers using the LO signal to vary the transconductance of the transistor. They have the advantage of providing the possibility of conversion gain rather than loss and can also have lower noise figures than passive designs. In FET mixers, and as long as LO pumping does not lead to gate-channel junction conduction or breakdown remains approximately unchanged. So, finding optimum matching conditions is probably not too difficult. On the contrary, when the nonlinearity is a diode or bipolar transistor, changes significantly with LO drive, and so obtaining a conjugate match may require some changes. That is clear with an unbiased pn or Schottky junction that shows an almost pure reactance (due to the junction depletion charge) when LO is too small to induce forward conduction. This inhibits efficient power transfer to the device, and the pumping regime remains practically unchanged. However, when some power begins to be delivered to the junction, its impedance starts to show some resistive component and an adjustment on LO source impedance is required. Then, additional power can be delivered, the input impedance gets progressively more resistive, and a new adjustment is needed. Those steps should then be repeated until a matching condition, under the desired LO drive, is reached.
On the other hand, the Schottky-barrier diode is perhaps the simplest modern solid-state microwave device in existence and the easiest to characterize accurately. The junction current voltage (I-V) and capacitance characteristics can be expressed by simple closed-form equations that are accurate for almost all purposes; there is little need to make a trade-off in the diode model between accuracy and simplicity.
It is also important to note that most diode mixers used at microwave and even millimeter-wave frequencies are balanced. The advantages of balanced mixers over single-diode
mixers are (1) rejection of spurious responses and intermodulation products, (2) inherent LO-to-RF isolation, (3) in some cases, inherent LO-to-IF or RF-to-IF isolation, and (4 ) rejection of AM noise in the LO. The most important disadvantage of balanced mixers is their greater LO power
requirements. Commercially available balanced mixers are frequently small, lightweight, inexpensive, broadband components. In many applications, their good spurious-response properties are essential. Furthermore, in systems where the LO and RF bands overlap, balanced mixers must be used, because it is impossible to separate the LO from the RF by filtering (Maas 2003, p. 339).
On another side, Garcia (1994) explained that a singly balanced mixer can also be realized by replacing the 180 degree hybrid by a quadrature hybrid. The individual mixers are the same as those used with the 180 degree hybrid, and, as before, they are connected to mutually isolated ports. The main differences in operating characteristics between the 180 degree and quadrature hybrid mixers are in the port’s Voltage Standing Wave Ratio (VSWRs), isolation, and spurious response properties. In the 180 degree mixer, the input VSWR at the LO and RF ports is dominated by the VSWRs of the individual mixers, and the RF/ LO isolation is dominated by the isolation of the hybrid. The quadrature hybrid, however, operates in a very different manner. LO power reflected from the individual mixers does not return to the LO port, but instead exits the RF port; similarly, reflected RF power exits the LO port. The LO/ RF and RF/ LO isolation is therefore equal to the input return loss of the individual mixers at the LO and RF frequencies, respectively; the port isolation of the quadrature hybrid mixer depends primarily on the input VSWRs of the two individual mixers, not on the isolation of the hybrid itself. As long as the LO and RF source VSWRs are good, the mixer’s LO and RF input VSWRs are also good. However, if the RF port termination has a poor VSWR at the LO frequency, the circuit’s balance can be upset and the LO pumping of the individual mixers becomes unequal; similarly, a poor LO port termination at the RF frequency can upset RF balance.
While there are many different types of singly balanced mixers that have been developed, all are fundamentally realizations of either the 180-degree or quadrature structures and nothing else exists. An example is the crossbar mixer, which is in fact a type 180-degree hybrid mixer. In the crossbar mixer two diodes are connected in series across the RF waveguide, and the LO is coupled to the diodes via a metallic strip (the crossbar) that acts as a coupling probe in the LO waveguide. The probe is also used for the IF output. The orientation of the probe and the RF and LO waveguides is such that the probe does not couple the LO and RF waveguides.
Singly balanced mixers have many of the desirable properties of balanced mixers, yet can be treated in many ways like single-diode mixers. It is practical for singly balanced mixers to have matching circuits and dc bias, giving them good conversion efficiency, flat bandwidth, and low VSWR. The structures used for doubly balanced mixers do not allow for practical matching circuits and dc bias. Accordingly, it can be more difficult to optimize a doubly balanced mixer.
Doubly balanced mixers are used primarily in applications where their superior spurious-response properties are essential, and those applications comprise most types of modern microwave systems.
Currently, modern communication systems have stringent dynamic range requirements and mixers are an important building block in transceiver design. This is because the receiver’s dynamic range is often limited by the first down-conversion mixer. The design of mixers forces many compromises between conversion gain, LO power, linearity, noise figure, port-to-port isolation, voltage supply, and current consumption. The most fundamental choice in FET mixer design is the choice of active or passive mixer. Active FET mixers achieve conversion
gain and require lower LO power than their passive counterparts. Passive FET mixers (FETs operating in the linear region) typically demonstrate conversion loss and excellent interrnodulation performance at the expense of LO power. A reduced LO drive is a significant advantage in low voltage / low power Integrated Circuit (IC) design because large LO drives are difficult to generate in a low voltage environment and result in an increase in power dissipation. This also dictates increased LO-RF and LO-IF isolation in order to maintain the same refection as would be obtained with a lower LO drive. A low supply voltage is desired for hand-held wireless applications to reduce the weight from the number of stacked battery cells and for the corresponding reduction in power dissipation in the digital circuitry. The primary advantage of a passive mixer is increased dynamic range at the expense of LO drive.
In electrical engineering, the concept of mixers was essential to forward the function of nearly all modern radio receivers. This is because they use mixers to translate signals from one frequency to another. Although transistor radio devices are connected in old integrated circuit designs, Schottky diodes are the most popular mixer technologies used in microwave and wireless hybrid mixers. Presently, Datavideo SE-800 digital mixer now allows us to switch between up to four analog or digital video sources, and record the mixed output to a digital camera or recorder, or output a stable signal for live broadcast (Frank, September 2003). Recently, using the semiconductor process technology, a high-performance integrated CMOS direct-downconversion mixer was developed. This is capable of an input second-order-intercept point (IIP2) of +78 dBm in Universal Mobile Telecommunications Systems (UMTS) applications. The mixer was fabricated with an 0.18-[micro]m CMOS process with outstanding linearity and noise performance (Microwaves & RF, July 2006). This development would definitely increase the performance parameter of mixers in emerging wireless and next-generation cellular communications systems. Along with the increasing needs for high data rate, low power, low cost, and portability, the next-generation wireless communication standards, with the use of mixers, are required to operate reliably in different environments, such as macro-, micro-, and pico-cellular, as well as urban, suburban, and rural, both indoors and outdoors. Thus, they need to be bandwidth-efficient and capable of being deployed in diverse environments.
Thus, we realize mixer technology is constantly improving to jive with the emerging technologies that are sprouting nowadays. The changes it had undergone with time coincide with manufacturing efficiencies to reduce the cost and improve the performance of various electron devices. In essence, mixer technology will always be essential because it is a crucial tool in the transmission or processing of information. To continuously improve the integration level and novel features in existing transceiver subsystems and to focus and explore research and visionary ideas and applications for mixer technologies will be very helpful in bringing new possibilities for the future generation.
CMOS direct-downconversion mixer hits +78 dBm IIP2 for UMTS.(downconversion mixer capable of an input second-order-intercept point )(Brief article).” Microwaves & RF 45.7 (July 2006): 60.
Franks, D. Eric. “Living live: Datavideo SE-800 digital video mixer.(Test Bench)(Product/Service Evaluation).” Videomaker 18.3 (Sept 2003): 28(2).
Abrie, P. L. (1999). Design of RF and Microwave Amplifiers and Oscillators, Norwood, MA:Artech House.
Besser, L. (2003). Practical RF Circuit Design for Modern Wireless Systems : Passive Circuits and Systems, Volume I. Norwood, MA: Artech House, Incorporated, 2003. p 85.
Carr, J.J. (1999, January). Working With Mixers. Electronics Now, 70(1).
Carvalho, N. B., and Pedro, J.C. (2002). A Comprehensive Explanation of Distortion Sideband Asymmetries. IEEE Transactions on Microwave Theory and Tech., 50(9): 2090– 2101.
Coleman, C. (2004). Introduction to Radio Frequency Engineering. West Nyack, NY: Cambridge University Press.
Dunleavy, L. and Connick, R. (2004). On-Wafer Mixer Measurement up to 40 GHz. University of South Florida College of Engineering, Poster No. 15. Retrieved 12 September 2006 at http://www.eng.usf.edu/~schlaf/REU/Symposium/Spring2004/ConnickS2004.pdf#search=%22mixer%20is%20a%20%22three%20port%20device%22%22
Garcia, J. A., et al. (1998). Characterizing the Gate-to-Source Nonlinear Capacitor Role on GaAs FET IMD Performance. IEEE Transactions on Microwave Theory and Tech., 46(12): 2344– 2355.
Gilmore, Rowan. (2003). Practical RF Circuit Design for Modern Wireless Systems: Active Circuits and Systems, Volume II. Norwood, MA: Artech House, Incorporated.
Gonzalez, G. (1997). Microwave Transistor Amplifiers— Analysis and Design, Second Edition, London: Prentice-Hall.
Laverghetta, T. (2005). Microwaves and Wireless Simplified (Second Edition). Norwood, MA: Artech House, Incorporated.
Maas, S. (2003). Nonlinear Microwave and RF Circuits (Second Edition). Norwood, MA, USA: Artech House, Incorporated
Norwood, MA: Artech House, Incorporated.
Pedro, J. (2003). Intermodulation Distortion in Microwave and Wireless Circuits. Norwood, MA: Artech House, Incorporated.
Raisanen, A. (2003). Radio Engineering for Wireless Communication and Sensor Applications.
Shaeffer, D.K. (1999). The Design and Implementation of Low-Power CMOS Radio Receivers. Hingham, MA: Kluwer Academic Publishers.
Sterling, C.H. (2002). Stay Tuned: A Concise History of American Broadcasting. Mahwah, NJ: Lawrence Erlbaum Associates, Incorporated.