CN0235

The AD7280A is a complete data acquisition system that includes a high voltage input multiplexer, a low voltage input multiplexer, a 12-bit, 1 μs SAR ADC, and on-chip registers for channel sequencing. The HV MUX is used to measure the series connected Li-Ion battery cells as shown in Figure 1. The LV MUX provides single-ended ADC inputs that can be used with external thermistors to measure the temperature of each battery cell; or, if temperature measurements are not required, the auxiliary ADC inputs can be used to convert any other 0 V to 5 V input signal. A precision 2.5 V reference and an on-chip voltage regulator is also included.

The AD8280 is a hardwire-only safety monitor for lithium ion battery stacks. In conjunction with the AD7280A, the AD8280 provides a low cost, redundant, battery backup monitor with adjustable threshold detection and shared or separate alarm outputs. It has a self-test feature, making it suitable for high reliability applications, such as automotive hybrid electric vehicles or higher voltage industrial usage, such as uninterruptible power supplies. Both the AD7280A and the AD8280 obtain power from the battery cells they monitor.

The ADuM5404 includes an integrated dc-to-dc converter, which is used to power the high voltage side of the ADuM1400 and ADuM1401 isolators and provide the VDRIVE supply to the AD7280A SPI interface. These 4-channel, magnetically isolated circuits are a safe, reliable, and easy-to-use alternative to optocouplers.

To optimize the performance of the daisy-chain communication under noisy conditions, for example, when experiencing electromagnetic interference, the daisy-chain signals are shielded on an inner layer of the printed circuit board (PCB). Shielding is provided above and below by a VSS supply plane, which is connected to the VSS pin of the upper device in the chain. Figure 2 shows the top layer of the EVAL-CN0235-SDPZ PCB, which contains the upper shielding for the AD7280A, and Figure 5 shows the bottom layer, which contains the upper shielding for the AD8280. Figure 3 shows the inner layer (layer 2), which contains the shielded daisy-chain signals, and the shielding below is carried out on Layer 3 as shown in Figure 4. Individual 22 pF capacitors are placed on each daisy-chain connection and are terminated to either the VSS pin of the upper device or the VDD pin of the lower device, depending on the direction in which data is flowing on the daisy chain. The PD, CS, SCLK, SDI, and CNVST daisy-chain connections pass data up the chain, and the 22 pF capacitors on these pins are terminated to the VSS of the upper device in the chain.

The SDOlo and ALERTlo daisy-chain connections pass data down the chain, and the 22 pF capacitors on these pins are terminated to the VDD of the lower device in the chain. A direct low impedance trace is used to connect the VDD of the lower device with the VSS of the upper device to hold the two potentials as close as possible together in a noisy environment.

A ground fence at the isolation barrier is used to enclose the low voltage side, which consists of the left hand side of the PCB. This fence consists of a guard ring laced together by vias and connects to the digital ground on all layers throughout the board. Noise on power and ground planes that reach the edge of the circuit board can radiate causing emissions, but with this shielded structure the noise is reflected back.

Input-to-output dipole radiation can also be generated when driving a current source across a gap between ground planes. To help minimize this, a continuous shield is used at the isolation gap whereby the ground planes are extended on all layers throughout the PCB to create a cross-barrier coupling using overlapping shields; and the isolation gap on each layer is kept to a minimum, with a gap of 0.008 inches used on the tested board. For further recommendations to control radiated emissions with isoPower^® devices, such as the ADuM5404 used in this circuit, please refer to Application Note AN-0971.

Test Results

An important measure of the performance of the circuit is the amount of noise in the final output voltage measurement.

Figure 6 shows a histogram of 10,000 measurement samples taken for the VIN3−VIN2 channel. This data was taken with the CN0235 Evaluation Board connected to the EVAL-SDP-CB1Z System Demonstration Platform (SDP-B) evaluation board. Details of the setup are described in the Circuit Evaluation and Test section of this circuit note.

Twelve Li-Ion batteries were connected to the input screw terminals. Note that there are only a small percentage of codes that fall outside the primary bin due to noise. Figure 6 and Figure 7 show 3 LSBs peak-to-peak noise, corresponding to approximately 0.5 LSBs rms.

A complete design support package for this circuit note can be found at www.analog.com/CN0235-DesignSupport.