It can meet the design requirements for logarithmic amplifiers for RF power measurement

　　A logarithmic amplifier (commonly called a logarithmic amplifier, sometimes referred to as a logarithmic detector) is an analog component used in RF circuits and electro-optical interfaces. Its transfer function is conceptually simple: the output voltage or current is proportional to the logarithm of the input voltage or current (Figure 1). It achieves this input/output relationship within an input range of 60 to 80 decibels (typically), but in some cases it can be as wide as 120 dB; Some logarithmic amplifiers even reach a dynamic range of 160 dB. Although it is called an "amplifier," it is not an "amplifier" in the conventional sense; It is actually a linear-to-logarithmic converter.

　　Figure 1: Logarithmic amplifier or converter generates output current or voltage signals (Y-axis), proportional to the logarithm or voltage signal of the input current (X-axis); Note that for inputs equal to or below zero, it is undefined, although logarithmic amplifiers have methods that limit this.

　　Given the importance placed on components, deliberately placing nonlinear components in the system may backfire. Electrical contourity and low distortion over a wide range. Logarithmic amplifiers are part of the design, but they are usually not directly in the signal chain. In RF circuits, it is usually part of closed-loop control, which adjusts the gain of the receive channel through automatic gain control (AGC), maintains a near-constant signal level (RSSI) in the channel by measuring input signal strength over a wide range (RSSI—Received Signal Strength Indicator), Figure 2, and controls the transmitted power. In optical circuits, it is used to monitor the current of the laser diode and adjust its changes according to temperature and other operational factors.

　　Figure 2: As shown in the block diagram of an FM receiver, logarithmic amplifiers are typically used in the receiver's AGC feedback loop to maintain signal levels. Although the input signal strength range is wide, it remains within a very narrow range.

　　Although the logarithmic function is not defined for parameters equal to or less than zero, the actual circuit does have non-positive signals. Therefore, logarithmic amplifier designers use various techniques to address this limitation. Logarithmic amplifiers and their applications are divided into three main categories:

　　DC logarithmic amplifiers ("DC" is a bit of a misnomer) are used for slowly changing signals, up to 1 MHz. It is used for optical path power control, as well as medical, chemical, and biological instruments.

　　When a certain type of signal compression is needed, baseband logarithmic amplifiers are used in audio and video circuits, as well as in the IF stage of the receiver signal chain and the signal processing path in ultrasonic circuits. For positive or negative input signals, it has a symmetric output, with the output positive for positive input and negative for negative input.

　　A demodulated logarithmic amplifier compresses and demodulates RF signals, with its output being the logarithm of the rectifier signal envelope. This logarithmic amplifier is used in RF transceiver applications, where the received RF signal strength is used to control the transmitter's output power. The output is based on the absolute value of the input; whether the input is positive or negative, it is positive.

　　[Note: The logarithmic amplifier is different from another nonlinear analog amplifier, the limiting amplifier. This device is sometimes called a clipper amplifier and is linear for most of its range. However, when the input approaches the positive or negative maximum, the amplifier gain begins to decrease and become limited. Therefore, this amplifier "softly limits" and relatively elegantly reaches maximum output, rather than simply saturating the output hard—which causes severe distortion and may take a relatively long time for the amplifier to recover. When the input returns to the normal range, the output also quickly returns to linear mode. ] Logarithmic amplifier design

　　The core of a logarithmic amplifier is based on the well-known logarithmic relationship between current through the diode PN junction and voltage (Figure 3 left), and is used in combination with operational amplifiers in actual circuits (Figure 3 right). Starting from this basic physical principle, logarithmic amplifiers use many topologies and configurations, each offering trade-offs between various performance attributes and priorities required for precision and bandwidth. While internal details may not be directly related to logarithmic amplifier users, they do affect the matching between the logarithmic amplifier and the application. Logarithmic amplifiers that provide high-precision transfer functions—features required for certain applications but not required in others—are commonly referred to as "linear dB" logarithmic amplifiers.

　　Figure 3: The well-known current-voltage relationship of diodes forms the foundation of almost all logarithmic amplifier designs (left); To take advantage of this diode relationship, it is placed in the feedback loop of the basic operational amplifier design (right).

　　For RF applications, continuous compression logarithmic amplifiers use multi-stage amplification and progressive limiting to form segmental logarithmic approximations. They include a rectifier (detector), each with 5 to 10 low-gain stages (each 8 dB to 12 dB), whose outputs are summed to generate a filter voltage, which is an average decibel-level power measurement above 100 dB. Other RF applications use exponential gain designs, with a narrower gain range (about 60 dB) but higher accuracy; It typically includes a detector whose filtered output makes the detector a square law device, with the output being the power equivalent (RMS) value of the applied signal.

　　Logarithmic amplifiers for optical applications are usually in the "DC stage" because they measure the relatively slow changes in current related to optical power to control the current in the laser diode or the gain of the optical mode amplifier. They may need to complete this work in the range of about a few pin-ampere to a few milliamps, totaling ninety years (span 10 9:1).

　　Logarithmic amplifier specifications

　　The physical implementation of a logarithmic amplifier can be an integrated circuit (IC) or a module composed of a single chip and discrete components. The IC version is smaller, cheaper, has lower power consumption, and offers other advantages, while also delivering excellent performance. They are usually the first choice. When a single IC process technology or individual IC cannot fully meet all necessary application parameters (such as noise, bandwidth, or temperature range), hybrid structures are used.

　　Logarithmic amplifiers have specifications similar to traditional non-logarithmic amplifiers, as well as some unique specifications due to the nature of the device. Additionally, different vendors may have legally different definitions for certain parameters, so it is crucial to check detailed information and test conditions in the datasheet. Top factors include:

　　- Dynamic range over decades: usually measured in dB, with most cases ranging from 60 dB to 120 dB (or higher). In all cases, a wide range may not be necessary, and implementing it could reduce trade-offs in other key specifications.

　　- Bandwidth: For today's RF applications, this is typically a single-digit GHz range, but some advanced devices can reach tens of GHz.

　　- Accuracy: Perfectly conforms to the linear/logarithmic transfer function. It usually ranges between 0.1% and 1%, but can also vary depending on its position within the measurement input range.

　　- Sensitivity: the lowest signal value that a logarithmic amplifier can process; Typically, it ranges from 1 nA or 1 μV, but can be lower; It is usually specified in dBm, typically 50Ω.

　　- Offset: The output of the logarithmic amplifier when the input is at its minimum (not 0, since log 0 is not defined).

　　- Fixed or adjustable references: Some logarithmic amplifiers have fixed scaling factors, such as 0.25 V/ten times (or 10 mA/ten times); Other references allow users to provide to determine the proportional factor. The scale factor can be adjusted relative to dB or decimal, for example, 20 mV/dB or 400 mV/decade.

　　- Unipolar and bipolar input and output: The logarithm of negative numbers is undefined, but many real-world signals are bipolar signals with negative values; To overcome this limitation, baseband and demodulation logarithmic amplifiers use offset, squared, or other techniques to allow inputs below 0 V.

　　The two most challenging issues with logarithmic amplifiers are noise and temperature coefficients. Because logarithmic amplifiers have been in use for decades, they can handle signals in the μV, nV, and even pV range (or μA, nA, or pA). However, if the signal level is very low, the internal noise of the logarithmic amplifier may exceed the signal. For many RF applications, fortunately, as long as the noise spectral density is low enough (usually on the order of nV/√Hz), low noise is not as important as range and bandwidth.

　　Tempco offers the most challenging parameters for logarithmic amplifier designers and users. Because the core of a logarithmic amplifier behaves based on the semiconductor junction, it inevitably changes with temperature. Logarithmic amplifier designers use various design techniques to cancel, compensate, trim, or minimize the temperature coefficient, but it remains a factor affecting overall performance. Like many analog components, logarithmic amplifiers offer detailed specifications suitable for standard commercial, industrial, and even military temperature ranges.

　　The example of a logarithmic amplifier shows the specification range

　　Many analog and mixed-signal IC suppliers offer logarithmic amplifiers. Manufacturers typically provide an overview of error consistency curves, as well as detailed curves showing specific frequencies at each frequency, as well as consistency at low, nominal, and high temperatures.

　　For example, ADI's AD8318 is a demodulation logarithmic amplifier that uses progressive compression technology on a cascaded amplifier chain, with each stage equipped with a detector unit (Figure 4). It provides accurate logarithmic consistency for signals from 1 MHz to 6 GHz and offers useful operations at 8 GHz. The input range is typically 60 dB (input impedance is 50Ω), with an error less than ±1 dB (Figure 5), and temperature stability of ±0.5 dB. 4 mm × 4 mm, 16-pin devices have a rated temperature range of -40°C to +85°C, powered by a single 5 V power supply.

　　Figure 4: ADI's AD8318 logarithmic amplifier uses cascaded amplifier chains and progressive compression technology, providing precise log-consistency for signals from 1 MHz to 6 GHz, and operating at 8 GHz frequency.

　　Figure 5: One of many detailed performance charts provided by the supplier for logarithmic amplifiers, comparing the AD8318 output voltage VOUT (almost a straight downline line) and log-consistency (the "swing" line). The 8 GHz input amplitude also shows +25°C (black), -40°C (blue), and +85°C (red) performance.

　　Lingliert offers the LT5537, a wide dynamic range RF/IF detector operating in the frequency range of 10 MHz to 1 GHz (Figure 6). At 200 MHz, its dynamic range is 90 dB, with ±3 dB nonlinear (50Ω input), as shown in Figure 7. The geophone output voltage slope is 20 mV/dB (nominal value), temperature coefficient is 0.01 dB/°C, and C is 200 MHz (typical value). Sensitivity is also measured at 200 MHz, at least -76 dBm. It uses a single power supply ranging from 2.7 V to 5.25 V, with 8-pin packaging in 3 mm × 2 mm packages.

　　Figure 6: LT5537 provides a log-linear relationship between input and output; The input signal is amplified by a series of limiting amplifier stages; A series of detector units rectifies the signal and generates an output current linearly related to the input power.

　　Figure 7: This is a broad overview of the relationship between output voltage, linearity error, and input power. At 200 MHz and three temperatures, Lingliert's LT5537 is supplemented with a wealth of more detailed performance diagrams.

　　The third example is MAX4003 from Maxim Integrated. Their MAX4003 low-power logarithmic amplifier is designed to detect the power levels of RF power amplifiers (PAs) operating in the frequency range from 100 MHz to 2500 MHz (Figure 8). This logarithmic amplifier has a typical dynamic range of 45 dB, suitable for wireless applications, including cellular PA control, transmitter signal strength control for wireless terminal devices, and other transmitter power measurements.

　　Figure 8: Maxim's MAX4003 logarithmic amplifier is a low-power component ranging from 100 MHz to 2500 MHz, with a range of 45 dB; It includes four 10 dB amplifier/limiter stages, each with a 10 dB small-signal gain; The output of each amplifier/limiter stage is applied to the full-wave rectifier, and the detector stage is also located before the first stage, with a total of five detectors.

　　This voltage measurement device is suitable for a typical signal range of -58 dBV to -13 dBV, using various small packages including 8-ball chip stage, μMAX, and thin QFN packages. Vendors provide advanced overview consistency charts for different frequencies (Figure 9), as well as more detailed consistency diagrams for each referenced frequency, including temperature and even package type. The device requires 5.9 mA (3.0 V power supply), and only 13μA when the device is off. It achieves temperature stability across the entire operating temperature range, from -40°C to +85°C.

　　Figure 9: Packaging also affects performance. As shown in the Maxim MAX4003 datasheet, VOUT and logarithmic consistency compared to input power at 2.5 GHz use an 8-pin μMAX package (left) and its 8-ball solder ball-level UCSP Shangtai package (right).

　　Summary

　　Although they have more complex and finer specifications than traditional linear amplifiers, logarithmic amplifiers play a key role in RF and optical systems. Logarithmic amplifiers with GHz-range response manage the front-end gain and transmitted power of the receiver, while low-frequency logarithmic amplifiers measure the current through the laser diode in the fiber optic link.

　　There are many ways to build logarithmic amplifiers, most of which are based on the diode's unique logarithmic voltage/current transfer function. However, practical complete logarithmic amplifiers are far more complex than standalone diodes and must be adapted and balanced with specifications for dynamic range, bandwidth, temperature drift, noise, and other performance parameters. Today's IC-type logarithmic amplifiers deliver excellent performance in compact, low-power, and low-cost packages. Only in fairly specialized cases will hybrid multi-chip logarithmic amplifiers be increasingly needed.