How to convert the tiny sensor output signal into ADC input voltage?

Question: Is there a module that allows me to directly convert the tiny sensor output signal into ADC input voltage?

Answer: Yes, a new generation of instrumentation amplifier series (such as ADI’s integration solution) is expected to accomplish the following tasks in one fell swoop: suppress common-mode signals, amplify differential-mode signals, convert the voltage to the required ADC input voltage, and protect the ADC from Overvoltage influence!

In countless industries, automobiles, instrumentation, and many other applications, a common challenge is how to correctly connect tiny sensor signals to ADCs for digitization and data acquisition. The sensor signal is usually very weak, may be very noisy, and looks like a very high impedance source above a large common mode (CM) voltage. These are undesirable for ADC input.

This article will introduce the latest integrated solutions that can completely solve the problems raised by engineers that are beyond the scope of the current ability. The article will detail the design steps in order to configure a complete sensor interface instrumentation amplifier to drive the ADC input.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 1: The challenge from the sensor to the ADC

What is suitable for the sensor? Why is there a problem?

The short answer to this question is the instrumentation amplifier. The object that the sensor is suitable for connection is the instrumentation amplifier.

The instrumentation amplifier has high precision (low offset) and low noise characteristics, and will not destroy small input signals. Its differential input is suitable for many sensor signals (such as strain gauges, pressure sensors, etc.), and can suppress any existing common mode signals, leaving only the original small voltages that we are interested in, and there will be no unwanted common mode signals . The instrumentation amplifier has a large input impedance and will not load the sensor, ensuring that the fragile signal is not affected by the signal processing.

In addition, instrumentation amplifiers usually use a single external resistor to provide a large gain and selectable gain range, so they are very flexible, allowing the target small signal to adapt to voltages and ADC analog inputs that are much higher than the signal path noise level. Instrumentation amplifiers are designed for precision performance and are internally adjusted to maintain their performance over a wide operating temperature range and are not affected by changes in power supply voltage. Instrumentation amplifiers also have extremely low gain errors, which also help them maintain accuracy and limit measurement or signal errors caused by swing changes

What do you like about ADC input?

Driving the ADC input is not so easy. The switching operation of the internal capacitor (CDAC in Figure 2) of the front end will cause charge injection, which makes it a difficult task to transmit a stable signal with high linearity for ADC quantification. The driver driving the ADC input must be able to handle these huge charge injections and quickly stabilize before the next conversion cycle. In addition, depending on the ADC resolution (number of bits), the noise and distortion of the driver should not be a limiting factor.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 2: ADC input drive is challenging

It is no easy task to meet the above requirements, especially for low-power drivers. In addition, due to the modernization of the semiconductor manufacturing process, the ADC operating power supply voltage is gradually decreasing. One of the undesirable side effects of this trend is that the ADC input becomes more susceptible to input overvoltage and may cause injury or damage. This requires external circuitry to prevent this overvoltage. Not only should this type of external circuit not add any measurable noise to the signal, it should not limit the bandwidth or cause any form of distortion. In addition, it is very hoped that the entire circuit can react quickly and recover quickly from an overvoltage event.

There are also challenges in offsetting the input signal to match the ADC analog input voltage range. Any circuit components added to perform this task must comply with all the constraints listed earlier (ie, low distortion, low noise, adequate bandwidth, etc.).

It would be great if the instrumentation amplifier could directly drive the ADC...

All instrumentation amplifiers on the market have some shortcomings, so more circuit components are needed to complete the path from the physical world (sensor) to the digital world (ADC). Traditionally, instrumentation amplifiers are not the circuit component of choice for driving ADCs (some ADCs are more precise than others). The instrumentation amplifier has done enough, and it seems unfair to hope it can do more!

Overcoming the harmonic distortion (HD) of the ADC driver is a difficult challenge. The following is an expression of the distortion performance that the ADC driver must meet or exceed, which is a function of the ADC resolution: How to convert the tiny sensor output signal into the ADC input voltage?

Therefore, for 16-bit ENOB, SINAD≥98dB

Instrumentation amplifiers currently on the market are usually not designed to drive ADC inputs. The most common reason for this is that these components lack the linearity required for high-resolution ADCs. Linearity or harmonic distortion (also known as THD, or total harmonic distortion) is the most likely limiting factor, and instrumentation amplifiers cannot directly drive the ADC. When the complex waveform is digitized, once it is interfered by the distortion item, the signal cannot be distinguished from such interference, and the data capture will be destroyed! The driver should also be able to quickly stabilize from the ADC input charge injection transient explained earlier.

The current solution is improved

Now, the new instrumentation amplifier series can not only complete all the things traditionally done by instrumentation amplifiers, but also can directly drive the ADC and protect the ADC input very well! LT6372-1 (supports 0dB to 60dB gain) and LT6372-0.2 (supports –14dB to +46dB gain/attenuation) can assist in completing the task of precision sensor interface and directly drive the ADC input.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 3: Ideal sensor amplifier/ADC driver

Using high-precision, low-noise instrumentation amplifiers such as the LT6372 series to directly drive ADC analog inputs has obvious advantages, without the need to add an amplification or buffer stage. Some of the benefits include: reducing the number of components, reducing power consumption and cost, reducing board area, providing high CMR, excellent DC accuracy, low 1/f noise, and selecting gain through a single component.

Many high-speed operational amplifiers selected as ADC drivers may not have the low 1/f noise characteristics of the LT6372 series because the latter is manufactured using a proprietary process. In addition, additional buffering and gain stages may be needed to amplify tiny sensor signals. When an instrumentation amplifier is used to directly drive the ADC, neither the amplifier stage nor the reference voltage source has comparable additional noise sources or DC offset terms.

LT6372-1 and LT6372-0.2 have extremely high input impedance, can be interfaced with sensors or similar signal input, and provide large gain (LT6372-1) or attenuation (LT6372-0.2) without causing load effects, and at the same time its low distortion And low noise can ensure accurate conversion without degrading performance, and support 16-bit and lower resolution ADCs operating at rates up to 150kSPS. Figure 4 shows the achievable bandwidth of each component at a given gain setting.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 4: Frequency response of LT6372-1 and LT6372-0.2 at various gains

The relationship between LT6372-1 distortion and frequency is shown in Figure 5. It should be ensured that the distortion term does not affect the THD performance of the ADC at the highest target frequency. Take ADCLTC2367-16 as an example, its SINAD specification is 94.7dB. To ensure that the driver is not the main factor, Figure 5 shows that the LT6372-1 is a suitable choice for frequencies less than about 5kHz.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 5: The relationship between LT6372-1THD and frequency

The subtlety of LT6372-1 used in ADC driver

In addition to the aforementioned advantages, the separate reference voltage architecture of the LT6372 series (shown as separate RF1 and RF2 pins in Figure 6) allows the signal to be directly and effectively translated into the ADCFS voltage range without using additional The reference voltage source and other external circuits to achieve the same purpose, thereby reducing cost and complexity. For most ADCs, REF2 (shown here connected to the VOCM DC voltage) will be connected to the ADCVREF voltage, which will ensure that the ADC analog input mid-level is VREF/2.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 6: LT6372 separate reference voltage is used to move the signal to the ADC analog input signal range

The built-in output clamps (CLHI and CLLO) of the LT6372 series ensure that the sensitive inputs of the ADC will not be damaged or possibly damaged by positive or negative transients. This series supports a distortion-free output swing to reach the clamp voltage, and can respond and recover quickly, thereby protecting the ADC and quickly returning to normal operation after a possible transient triggers any clamp.

The analog input of some SARADCs brings a challenging load to the amplifier drive. The amplifier needs to have low noise, fast settling characteristics, and high DC accuracy to keep the disturbance of the interfering signal at an LSB or less. Higher sampling rates and higher-order ADCs place higher demands on amplifiers. Figure 7 shows the input of a typical SARADC.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 7: SARADC input in acquisition/sampling mode

The switch position shown in Figure 7 corresponds to the sampling or acquisition mode, in which the analog input is connected to the sampling capacitor CDAC, and then the conversion starts in the next session.

Before the start of this phase, switch S2 has discharged the CDAC voltage to 0V or other bias point, such as FS/2. At the beginning of the sampling period, S1 is closed and S2 is open. The voltage difference between VSH and the analog input causes a transient current to flow so that the CDAC can be charged to the analog input voltage. For higher sampling rate ADCs, this current may be as high as 50mA. The capacitor CEXT helps to alleviate the step change of the amplifier output voltage caused by the current step, but the amplifier will still be interfered by it and needs to be established in time before the end of the acquisition period. The resistor REXT separates the driver from CEXT and reduces its impact on stability when driving large capacitors. Regarding the choice of REXT and CEXT values, it is necessary to weigh the greater isolation caused by this current injection and the settling time performance degradation caused by the low-pass filter formed in this way. This filter also helps to reduce out-of-band noise and improve SNR, but this is not its main function.

ADC front-end RC component value design

There are many factors to consider when choosing the values of REXT and CEXT. The following is a summary of the factors that affect the ADC dynamic response measured by FFT or other methods:

CEXT: Acts as a charge bucket for input charge recoil, minimizing the voltage step, thereby improving the settling time.

Too big: it may affect the stability of the amplifier and may reduce the LPF roll-off frequency too low to allow the signal to pass.

Too small: The charge recoil of the ADC input is too large to be established in time.

REXT: Provide isolation between the amplifier output and CEXT to ensure stability.

Too big: The settling time constant may be too long. When the ADC input non-linear impedance is taken into account, it may also cause THD to increase by 1. May increase the IR drop error.

Too small: Due to CEXT, the amplifier may become unstable or its forward path establishment may be affected.

The following are some design steps for designing REXT and CEXT values. Take LT2367-16ADC as an example, which is driven by LT6372-1, the maximum input frequency fIN is 2kHz, and the sampling rate is 150kSPS:

Choose a large enough CEXT to act as a charge bucket and greatly reduce the charge backlash:

How to convert the tiny sensor output signal into ADC input voltage?

among them:

CDAC: ADC input capacitance = 45pF (LTC2367-16)

→CEXT=10nF (selected value)

Use the following formula to calculate the ADC input voltage step VSTEP:

How to convert the tiny sensor output signal into ADC input voltage?

among them:

VREF=5V (LTC2367-16)

CDAC: ADC input capacitance = 45pF (LTC2367-16)

CEXT=10nF (before)

→VSTEP=22mV (calculated value)

Note: This VSTEP function assumes that the CDAC is discharged to ground at the end of each sampling period, as does the LTC2367-16. The VSTEP formula in Reference 1 uses different assumptions because it is specific to the ADC architecture and the CDAC voltage remains constant for each sample.

Assuming that the step input is established exponentially, calculate how many inputs REXT×CEXT time constant NTC is required to establish:

How to convert the tiny sensor output signal into ADC input voltage?

among them:

VSTEP: ADC input voltage step calculated before

VHALF_LSB: LSB/2, the unit is volts. For 5VFS and 16-bit, it is 38µV (=5V/217)

→NTC=6.4 time constant

Calculate the time constant τ:

How to convert the tiny sensor output signal into ADC input voltage?

among them:

tACQ: ADC acquisition time; tACQ=tCYC–tHOLD

Assuming the sampling rate is 150kSPS:

tCYC=6.67μs (=1/150kHz)

tHOLD=0.54μs (LTC2367-16)

Therefore: tACQ=6.13μs

→τ≤0.96µs

In the case where τ and CEXT are known, REXT can be calculated

How to convert the tiny sensor output signal into ADC input voltage?

→REXT≤96Ω

Now that we have the external RC value, the selected ADC can be properly established. If the calculated REXT is too high, you can increase CEXT and recalculate REXT to decrease its value, and vice versa. Figure 8 shows the selected value of CEXT and the corresponding REXT value to simplify the calculation task under the working conditions of this example.

How to convert the tiny sensor output signal into ADC input voltage?

Figure 8: ADC correctly establishes the corresponding external input RC relationship.

Use the previous steps to find the appropriate starting values for REXT and CEXT. Benchmarks and evaluations should be performed, and these values should be optimized as needed, while keeping in mind the impact of such changes on performance.

to sum up

This article introduces a new series of instrumentation amplifiers that can assist in connecting sensors and data acquisition components. The article discusses the characteristics of these components in detail, and illustrates how to design ADC front-end components through a practical example to ensure that the combination of driver and ADC can achieve the expected resolution.