Modern RF amplifiers require both linearity and high efficiency. Linearity requirements are derived from the use of modern modulation methods such as QAM (Quadrature Amplitude Modulation) and OFDM (Orthogonal Frequency Division Multiple Access (Reference 1). These amplifiers also require efficiency to reduce power consumption and reduce heat dissipation. Developers typically mount modern RF amplifier components in the mast. In these "roof" amplifier designs, the enclosure can be fanless and exposed directly to daylight. For every 1W of power consumption, it means less 1W of heat sink cooling. In addition, overdriving the amplifier causes distortion and harmonic spikes, making demodulation impossible. These spikes will fall into the adjacent band, perhaps a band that the mobile phone company does not own. The FCC (Federal Communications Commission) has strict restrictions on this ACLR (adjacent channel leakage ratio).
So, you have two reasons to achieve good linearity: this allows you to precisely modulate the signal so that your signal does not interfere with adjacent signals. Equally important, you get the best power efficiency at the output stage. The problem is that linearity and efficiency are mutually exclusive.
The distortion of the RF amplifier can be viewed in both the frequency and time domains. In the time domain, it is possible to visually see a chamfered or flat-topped sine wave through an RF amplifier, just like an audio signal that is driven too close to the voltage rail (Figure 1). In the frequency domain, the amplifier distortion appears to contain the "edge" of the harmonics, which enters the range of adjacent bands (Figure 2). For any amplifier, the higher the power desired, the more severe the distortion will be. At RF frequencies, there are not only amplitude distortion, but also phase distortion, as well as distortion due to thermal transients and electrical memory effects (Figure 3). Phase distortion occurs in the fast slew rate region, and the RF output lags the input signal, such as when the carrier signal enters the ground, or when a modulation envelope must immediately change to a different level.
In order to load more information within a certain bandwidth, modern modulation techniques rely on an accurately received RF signal envelope. With accurate voltage and phase, you can decode a constellation that represents a point in a digital code. This code produces a stream of digital data that is then further decoded into a baseband voice or data signal.
Older modulation methods are less sensitive to the linearity of the amplifier. Both AM (Amplitude Modulation) radios and analog TV broadcasts use the AM method, which relies on the peak value of the RF signal. Any distortion has the same effect on all peaks and has little effect on the quality of all received signals. The FM (FM) radio and analog TV audio signals use the FM method, which depends on the zero crossing of the waveform. Therefore, any amplitude nonlinearity has no effect. Phase distortion has an effect on zero crossing, but they are uniform and do not affect FM modulation.
There are several techniques for increasing the linearity of an RF amplifier. First, better transistors can be used. Thus, manufacturers will employ GaAs (gallium arsenide) and other III-V semiconductor processes in the production of RF transistors, that is, chemical compounds composed of at least one group III element and at least one group V element. Alternatively, try SiGe (silicon germanium) transistors, perhaps with a CMOS process (Reference 2). Although SiGe is slower than GaAs and has a large noise, it is usually sufficient, especially at frequencies below 3 GHz. Engineers face the pressure of using CMOS in RF amplifiers because of its low cost, but the low operating voltage of CMOS makes it difficult to implement in power amplifiers. CMOS also has a high noise figure. The reduction method is to increase the size of the transistor structure, but this method also increases the stray capacitance and reduces the operating frequency of the product. RFMD and other companies offer CMOS on sapphire, with a dielectric isolation layer underneath all transistors (Reference 3). This approach has a cost advantage and reduces stray capacitance.
The market-driven reality is that engineers can use CMOS to make low-power RF amplifiers for Wi-Fi hotspot applications. Mobile phones require more special processes, such as SOI (insulated silicon), and GaAs will dominate the recent mobile phone base stations.
Once your power amplifier has a well-linear transistor technology, the next step is to focus on the architecture of the amplifier. You can switch from a discontinuously driven architecture (such as Class C) to a more continuous type, such as Class AB. Class C is highly efficient because it uses a transistor to drive a tank circuit that produces an RF sine wave for transmission. Unfortunately, Class C amplifiers do not adapt to modern linear requirements, especially base stations. One way to achieve good linearity is to reduce the drive to the amplifier so that the transistor is not nearly saturated and the output voltage swing is completely within the power rail. Unfortunately, this solution is the least efficient.
To solve this problem, try a Doherty amplifier, which is a composite device that uses a main channel and an auxiliary RF channel to save power when the signal strength is low, and when higher power is required, Still adapt to larger signal swings (Figure 4). The Doherty amplifier architecture works well, but it adds the number and complexity of the ideal simple amplifier stage.
Manual Motor Starter,Motor Starter,KNS12,China Motor Starter
Wenzhou Korlen Electric Appliances Co., Ltd. , https://www.zjaccontactor.com