The LTC68xx family uses a laboratory grade Zener reference, a technology ADI has perfected over 30 years. Figure 6 shows the drift over temperature of the battery measurement IC error for five typical units. The drift in the full automotive range of –40°C to +125°C is less than 1 mV.
Figure 7 shows a comparison of the long-term drift for a bandgap voltage reference IC and a buried Zener voltage reference IC. The initial measurements are calibrated for 0 mV of error. Ten years of measurement drift is predicted from drift after 3000 h at 30°C. The picture clearly shows a much better stability of the Zener reference over time, at least 5× better than bandgap reference. Similar tests for humidity and PCB assembly stress show the superior performance of the buried Zener over the band gap voltage reference.
Another limiting factor for accuracy is noise. A car battery is a very harsh environment for electronics due to the electromagnetic interference generated by the electric motor, the power inverter, the dc-to-dc converters, and other high current switching systems in an EV/HEV. The BMS should provide a high level of noise rejection in order to maintain accuracy. Filtering is the classical method used to reduce unwanted noise, but there is a trade-off between noise reduction and speed of conversion. Due to the high number of cell voltages to be converted and transmitted, the conversion time can’t be too slow. SAR converters might be the preferred choice, but in a multiplexed system, speed is limited by the settling time of the multiplexed signal. In this case, sigma-delta (Σ-Δ) converters can be a valid alternative.