Low-cost, high-precision digital control scheme for battery test equipment

Battery testing equipment is an important part of the post-processing system of the lithium-ion battery production line, and it is very important to the quality of the lithium-ion battery. The core function of battery testing equipment is to perform high-precision constant current or constant voltage charging and discharging of lithium-ion batteries. The traditional control method is mainly based on analog control schemes built with discrete devices.

Author: Jared Liu

Battery testing equipment is an important part of the post-processing system of the lithium-ion battery production line, and it is very important to the quality of the lithium-ion battery. The core function of battery testing equipment is to perform high-precision constant current or constant voltage charging and discharging of lithium-ion batteries. The traditional control method is mainly based on analog control schemes built with discrete devices. Compared with the traditional analog control solution, the digital control solution implemented with TI’s C2000™ as the core will become the mainstream development direction of battery test equipment in the future due to its low cost, high precision, more flexibility, and better confidentiality. . In this article, I will introduce in detail how to effectively reduce system costs through TI’s C2000 digital control program, and ensure extremely high current and voltage control accuracy.

1 low cost

The typical structure of TI’s C2000 digital control scheme is shown in Figure 1: The current/voltage amplifier samples the current/voltage of the battery charging and discharging, and the analog signal is converted into a digital signal through the analog-to-digital converter ADC and sent to the C2000. C2000 performs loop calculations based on constant current or constant voltage commands and sampling signals, outputs PWM with a certain duty cycle to adjust the switching of MOSFETs, and finally makes the buck/boost converter perform constant current or constant voltage on the lithium battery according to the command. Discharge.

Low-cost, high-precision digital control scheme for battery test equipment

figure 1

Compared with the analog solution, voltage, current commands and loop control are all generated and completed in C2000, eliminating the need for high-resolution digital-to-analog converter DAC and error amplifier, effectively reducing system costs. TMS320F280049 is a C2000™ 32-bit MCU with 100MHz main frequency and 256KB flash memory. Through high-resolution 16bit PWM, it can control up to 8 independent channels of synchronous buck/boost converters. Using the digital control scheme of TMS320F280049 can save more than 30% of the BOM cost compared with the traditional analog control scheme.

In addition, because lithium-ion batteries are widely used in 3C products, electric vehicles, energy storage, and many other fields, the currents of various types of lithium-ion batteries often vary greatly. This leads to the fact that if the battery test equipment adopts analog control, it is often necessary to select different hardware schemes according to the current size, which increases the development cycle and equipment cost. If the digital control scheme of C2000 is adopted, you can freely switch between low-current or high-current mode without changing the hardware: when the current is low, each of the 8 channels can operate independently; when the current is high, it will be more Two channels are operated in parallel to output a larger current.

Low-cost, high-precision digital control scheme for battery test equipment
figure 2

As shown in Figure 2, when multiple channels are running in parallel, each channel will use the same constant voltage loop, and the constant current loop will be independent. You can achieve a larger output current range by simply connecting the outputs in parallel. Therefore, compared to analog control, the use of C2000’s digital control scheme can adapt to a wider range of test scenarios without changing the hardware, greatly reducing equipment costs.

2 High precision

Through calibration, battery test equipment can often remove most of the initial system errors. The remaining sources of errors that are difficult to be calibrated mainly include: temperature drift of current sense resistor, offset and gain temperature drift of current and voltage sense amplifiers, offset caused by input common-mode voltage changes, non-linearity of ADC, temperature of reference voltage source drift. In this article, the error value is calculated according to the temperature variation range of ±5°C.

Current detection resistance:

The temperature drift of the current sense resistor is an important source of total system error. For CC control, a high-precision current sense resistor with a few milliohms and a low temperature coefficient is required. This article uses a high-precision, current-sensing metal strip SMD power resistor, the resistance value of the detection resistor is 5mΩ, and the temperature drift value is 10 ppm. Then, the error caused by the temperature drift of the current sense resistor is 50 ppm.

Current detection amplifier:

In order to reduce the temperature rise and power loss caused by high current, the resistance of the current sense resistor is generally small, so the input differential signal of the current sense amplifier generally does not exceed tens of millivolts, and the instrument amplifier is often selected for signal conditioning. The error of the instrument amplifier mainly comes from the following two aspects: when the ambient temperature changes, the offset voltage and gain drift; when the battery voltage changes, the offset voltage caused by the change of the input common-mode voltage. Therefore, when choosing an instrumentation amplifier, you should mainly focus on parameters such as offset voltage drift, gain drift, and CMRR. Table 1 shows the key parameters of TI’s main instrumentation amplifiers used in battery test equipment:

Table 1

Specifications

INA821

INA828

INA819

INA188

Vos max (µV)

35

50

35

55

Drift (Max) (µV/C)

0.4

0.5

0.4

0.2

Gain Error (% Max)

0.15

0.15

0.15

0.5

Gain drift (ppm/°C) (G=1)

5

5

5

5

CMRR (Max Gain) (Min) (dB)

140

140

140

118

GBW (MHz) (G=1)

4.7

2

2

0.6

As a high-precision, low-drift instrumentation amplifier, the INA821 has a maximum offset voltage drift of 0.4 µV/°C, and a ±5°C temperature offset will produce a 2 µV offset voltage, which is a 40ppm full-scale error; the gain drift is 5 ppm/°C, then ±5°C temperature offset will produce 25ppm error; the common-mode voltage rejection ratio is 140dB, then when the input common-mode voltage range changes from 0 to 5V, an offset voltage of 0.5µV will be generated. Under 10A charging current, the voltage signal of the full-scale sampling resistor is 50mV, that is, the input common-mode voltage change brings 10ppm full-scale error.

Voltage detection amplifier:

The error source of the voltage detection amplifier also mainly comes from the drift of the offset voltage and gain, and the offset voltage caused by the change of the input common-mode voltage. Therefore, when choosing an instrumentation amplifier, you should also focus on parameters such as offset voltage drift, gain drift, and CMRR.

TLV07 is a cost-sensitive, low-noise, rail-to-rail output, precision operational amplifier. The offset voltage drift is typically 0.9 µV/°C. A temperature drift of ±5°C will produce an offset voltage of 4.5 µV, which is 1ppm full-scale error; gain drift is mainly affected by the drift error of the input resistance and the feedback resistance, here is 5 ppm/°C, then ±5°C temperature offset will produce 25ppm error. The minimum value of the common-mode voltage rejection ratio is 104dB, so when the input common-mode voltage ranges from 0 to 5V, an offset voltage of 31.5µV will be generated, which is a 6ppm full-scale error.

Analog-to-digital converter and reference voltage source:

The error of the analog-to-digital converter ADC is mainly caused by the non-linearity and the drift of the reference voltage source. ADS131M08 is a 24-bit, 32kSPS, 8-channel synchronous sampling delta-sigma high-precision ADC. Because ADS131M08 is a differential input, it can effectively reduce errors caused by crosstalk between channels. It can be found from the data sheet that the nonlinearity INL of ADS131M08 is only 7.5ppm full-scale error. If the internal reference voltage source is used and the maximum temperature drift is 20 ppm/°C, then the ±5°C temperature offset will produce a 100 ppm error. If the external reference voltage source REF2025 is used, and the maximum temperature drift is only 8 ppm/°C, the ±5°C temperature offset error will be reduced to 40 ppm.

Error summary:

According to the above analysis, the total error value caused by each error source can be calculated, and the total system error of the battery test equipment is shown in Table 2 under constant current and constant voltage control. It can be seen that using the C2000 digital control scheme, the current and voltage error ranges are within 10,000, achieving extremely high control accuracy.

Table 2

Current error

Voltage error

Source of error

Full scale error

Source of error

Full scale error

Shunt resistance temperature drift

50 ppm

Shunt resistance temperature drift

50 ppm

INA821 misalignment temperature drift

40 ppm

TLV07 offset temperature drift

1 ppm

INA821 gain temperature drift

25 ppm

TLV07 gain temperature drift

25 ppm

INA821 CMRR

10 ppm

TLV07 CMRR

6 ppm

ADS131M08 non-linearity

7.5 ppm

ADS131M08 non-linearity

7.5 ppm

REF2025 voltage temperature drift

40 ppm

REF2025 voltage temperature drift

40 ppm

Total error

0.017%

Total error

0.013%

In summary, the use of TI’s C2000 digital control program in battery test equipment can ensure extremely high current and voltage control accuracy while reducing system costs, which is very suitable for applications in various battery test programs.

The Links:   LQ190E1LX78 AA084SB01 POWER IGBT

Related Posts