"This paper will introduce the functions, requirements and main characteristics of LNA and PA, and then introduce the typical GaAs and Gan devices and the matters that should be kept in mind when using these devices for design.
The promotion of performance, miniaturization and higher frequency operation is challenging the limitations of two key antenna connection components of wireless system: power amplifier (PA) and low noise amplifier (LNA). Efforts to make 5g a reality and the use of PA and LNA in VSAT terminals, microwave radio links and phased array radar systems have contributed to this transformation.
The requirements for these applications include lower noise (for LNA) and higher energy efficiency (for PA) and operation at higher frequencies up to or above 10 GHz. In order to meet these growing needs, LNA and PA manufacturers are moving from the traditional all silicon process to gallium arsenide (GaAs) for LNA and gallium nitride (GAN) for PA.
This paper will introduce the functions, requirements and main characteristics of LNA and PA, and then introduce the typical GaAs and Gan devices and the matters that should be kept in mind when using these devices for design.
Sensitivity of LNA
The function of LNA is to obtain extremely weak uncertain signals from the antenna, which are usually on the order of microvolts or less than - 100 DBM, and then amplify the signal to a more useful level, usually about 0.5 to 1 V (Fig. 1). Specifically, in a 50 Ω system, 10 μ V is - 87 DBM, 100 μ V equals - 67 DBM.
This gain can be easily realized by using modern electronic technology, but when LNA adds various noises to the weak input signal, the problem will be far from so simple. The amplification advantage of LNA will completely disappear in such noise.
Figure 1: the low noise amplifier (LNA) of the receiving path and the power amplifier (PA) of the transmitting path are connected to the antenna via a duplexer, which separates the two signals and prevents the sensitive LNA input from being overloaded by the relatively strong PA output.
Note that LNA works in a world full of unknowns. As the front end of the transceiver channel, LNA must be able to capture and amplify the low-voltage signal with very low bandwidth consumption and the related random noise caused by the antenna. In signal theory, this situation is called unknown signal / unknown noise problem, which is the most difficult part of all signal processing problems.
The main parameters of LNA are noise figure (NF), gain and linearity. The noise comes from heat source and other noise sources, and the typical value of noise coefficient is 0.5 - 1.5 dB. The typical gain of a single-stage amplifier is between 10 - 20 dB. Some designs use cascaded amplifiers with a higher gain stage after the low gain and low NF stage. This design may achieve a higher NF, but it becomes less important once the initial signal has "increased"（ For details on LNA, noise and RF receivers, refer to the article "low noise amplifier can maximize receiver sensitivity" in techzone.)
Another problem of LNA is nonlinearity, because synthetic harmonic and intermodulation distortion can deteriorate the received signal quality. When the bit error rate (BER) is quite low, signal demodulation and decoding become more difficult. The third-order intermodulation point (IP3) is usually used as the characteristic parameter of linearity to associate the nonlinear product caused by the third-order nonlinear term with the linearly amplified signal; The higher the IP3 value, the better the linearity of the amplifier performance.
Power consumption and energy efficiency are usually not the primary issues in LNA. In essence, most LNAs are devices with relatively low power consumption and current consumption between 10 - 100 mA. They provide voltage gain to the next stage, but do not deliver power to the load. In addition, only one or two LNAs are used in the system (the latter is often used in the multi-functional antenna design of Wi Fi and 5g interfaces), so it is of little significance to save energy through low-power LNA.
In addition to the operating frequency and bandwidth, various LNAs are relatively similar in function. Some LNAs also have gain control function, so they can cope with the wide dynamic range of input signal without overload and saturation. In mobile applications with a wide loss range from base station to mobile phone channel, such a wide variation range of input signal strength is often encountered, even in a single connection cycle.
The routing of the input signal to the LNA and the output signal from it are as important as the specifications of the components themselves. Therefore, designers must use complex modeling and layout tools to achieve the full potential performance of LNA. Due to poor layout or impedance matching, high-quality components may easily deteriorate. Therefore, it is necessary to use the Smith chart provided by the supplier (see "Smith chart: an 'ancient' graphic tool still crucial in RF Design") and a reliable circuit model supporting simulation and analysis software.
For these reasons, almost all high-performance LNA suppliers working in the GHz range will provide evaluation boards or verified printed circuit board layouts, because every aspect of the test setup is crucial, including layout, connectors, grounding, bypass and power supply. Without these resources, designers need to waste time evaluating the performance of components in their applications.
A representative of GaAs based LNA is hmc519lc4tr. This is an 18 to 31 GHz PHEMT (pseudocrystalline high electron mobility transistor) device from analog devices (Fig. 2). This leadless 4 × The 4 mm ceramic surface mount package provides 14 dB small signal gain, 3.5 DB low noise figure and + 23 DBM high IP3. The device draws 75 Ma from a single + 3 V supply.
Figure 2: hmc519lc4tr GaAs LNA provides low noise gain for low-level inputs from 18 to 31 GHz; Most package connections are for power rails, grounded, or not used.
From a simple functional block diagram to multiple external capacitors with different values and types, a design process is required to provide appropriate RF bypass with low parasitic effect on the three power rail feeds, designated as VDD (Fig. 3).
Figure 3: in practical application, hmc519lc4tr LNA requires multiple bypass capacitors with the same rated voltage on its power rail to provide large capacitance for low-frequency filtering and small capacitance for RF bypass, so as to minimize RF parasitic effect.
Based on this enhanced schematic, an evaluation board is generated, detailing the layout and BOM, including the use of non FR4 printed circuit board materials (Figures 4 (a) and 4 (b)).
Figure 4 (a)
Figure 4 (b)
Figure 4: considering the high frequency of these LNA front ends and the low-level signals they must capture, a detailed and tested evaluation design is very important. It includes a schematic diagram (not shown), circuit board layout (a) and BOM, and details of passive components and printed circuit board materials (b).
MACOM maal-011111 is a GaAs LNA for higher frequencies and can support 22 to 38 GHz operation (Figure 5). The device can provide 19 dB small signal gain and 2.5 DB noise figure. The LNA is a single-stage device on the surface, but there are actually three cascaded stages inside. The first stage is optimized for minimum noise and medium gain, and subsequent stages provide additional gain.
Figure 5: for the user, maal-011111 LNA is a single-stage amplifier on the surface, but it uses a series of gain stages internally to maximize the input to output signal path SNR and increase significant gain at the output.
Similar to the LNA of analog devices, maal-011111 requires only one low-voltage power supply, and the size is only 3 × 3 mm, extremely compact. Users can adjust and balance some performance specifications by setting the bias (power) voltage between 3.0 and 3.6 v. The recommended board layout shows the critical PCB copper sheet size required to maintain appropriate impedance matching and ground plane performance (Figure 6).
Figure 6: the proposed layout makes full use of MACOM's maal-011111 to provide input and output impedance matching at the same time. Note that for impedance controlled transmission lines and low impedance ground planes, printed circuit board copper sheets (dimensions in millimeters) are used.
PA drive antenna
Contrary to the difficult signal acquisition challenge of LNA, PA obtains relatively strong signals from the circuit, has high SNR, and must be used to improve signal power. All general coefficients related to the signal are known, such as amplitude, modulation, waveform, duty cycle, etc. This is the known signal / known noise quadrant in the signal processing diagram, which is the easiest to deal with.
The main parameter of PA is the power output at the relevant frequency, and its typical gain is between + 10 and + 30 dB. Energy efficiency is another key parameter second only to gain in PA parameters, but the use of model, modulation, duty cycle, allowable distortion and other aspects of driven signal will complicate any energy efficiency evaluation. The energy efficiency of PA is between 30 and 80%, but this is largely determined by many factors. Linearity is also a key parameter of PA, which is determined by IP3 value as in LNA.
Although many PAS use low-power CMOS Technology (up to about 1 to 5 W), in recent years, other technologies have been mature and widely used, especially when considering energy efficiency as a higher power level with energy efficiency as a key indicator of battery life and heat dissipation. When several watts or higher power is required, PA using gallium nitride (GAN) has better energy efficiency at higher power and frequency (typically 1 GHz). Especially considering energy efficiency and power dissipation, Gan PA is very cost competitive.
Cree / wolfspeed cghv14800f (1200 to 1400 MHz, 800 w devices) are some of the latest GaN based PA representatives. This combination of energy efficiency, gain and bandwidth of HEMT PA optimizes the pulse L-band radar amplifier, enabling designers to find many applications in air flow control (ATC), weather, anti missile and target tracking systems. 50 V power supply is used to provide typical energy conversion efficiency of 50% and higher, and 10 × 20 mm ceramic package with metal flange for cooling (Figure 7).
Figure 7: cghv14800f 1200 to 1400 MHz, 800 W, Gan PA 10 with metal flange × The 20 mm ceramic package must meet difficult RF and heat dissipation requirements at the same time. For mechanical and thermal integrity reasons, pay attention to tightening (not welding) the package to the printed circuit board when installing the flange（ Image source: Cree / wolfspeed)
Cghv14800f is powered by 50 V power supply, which usually provides 14 dB power gain and energy conversion efficiency > 65%. Like the LNA, it is important to evaluate the circuit and reference design (Figure 8).
Figure 8: in addition to the device itself, very few components are required for the demonstration circuit provided for cghv14800f PA, but the physical layout and heat dissipation considerations are key; Considering the installation integrity and thermal objectives, the PA is fixed to the plate with screws and nuts (not visible at the bottom) through the encapsulated flange.
Equally important among many specification tables and performance curves is the power dissipation derating curve (Figure 9). This curve shows the relationship between the available power output rating and the housing temperature, indicating that the maximum allowable power is a constant 115 ° C and then linearly reduced to the maximum rating of 150 ° C.
Figure 9: due to its role in transmitting power, the PA derating curve is required to show the designer that the allowable output power decreases with the increase of housing temperature. Here, the rated power drops rapidly after 115 ⁰ C.
MACOM also provides GaN based PAS, such as npt1007 Gan transistors (FIG. 10). Its DC to 1200 MHz frequency span is suitable for broadband and narrowband RF applications. The device typically operates on a single power supply between 14 and 28 V and provides a small signal gain of 18 dB at 900 MHz. The design is designed to withstand 10:1 SWR (standing wave ratio) mismatch without device degradation.
Figure 10: MACOM's npt1007 Gan PA spans the DC to 1200 MHz range for broadband and narrowband RF applications. Designers get additional support through various load tension diagrams（ Image source: MACOM)
In addition to the diagrams showing the performance basis at 500, 900 and 1200 MHz, npt1007 also supports various "load stretching" diagrams to help circuit and system designers who strive to ensure stable products (Figure 11). The load tension test is completed using paired signal sources and signal analyzers (spectrum analyzer, power meter or vector receiver).
This test requires seeing the impedance change of the equipment under test (DUT) to evaluate the performance of the PA (including factors such as output power, gain and energy efficiency), because all relevant component values may change due to temperature changes or changes in the tolerance zone around its nominal value.
Figure 11: the load tension diagram of npt1007 PA exceeds the minimum / maximum / typical specification standard table to show PA performance when its load impedance deviates from its nominal value (this occurs in actual use due to initial production tolerance and thermal drift)（ Image source: MACOM)
No matter which PA process is used, the output impedance of the device must be fully characterized by the supplier, so that the designer can correctly match the device with the antenna, achieve maximum power transmission and maintain SWR consistency as much as possible. The matching circuit is mainly composed of capacitors and inductors, and can be realized as discrete devices, or manufactured as a part of printed circuit board or even product packaging. The design must also maintain the PA power level. Again, the use of Smith chart and other tools is the key to understand and carry out necessary impedance matching.
In view of the small chip size and high power level of PA, packaging is a key problem for PA. As mentioned earlier, many PAS dissipate heat through wide heat dissipation package leads, flange support and heat sink under the package as a path to the copper sheet of the printed circuit board. At higher power levels (approximately higher than 5 to 10 W), the PA can have a copper cap so that the radiator can be mounted on top, and fans or other advanced cooling technology may be required.
The power rating and small size associated with Gan PA mean that it is essential to model the thermal environment. Of course, it is not enough to keep the PA itself within the allowable conditions or junction temperature range. The heat dissipated from the PA shall not cause problems to the circuit and other parts of the system. Consideration must be given to the treatment and resolution of the entire heat
Our other product:
