TI's RES80A-Q1 Brings ±0.1% Ratio Matching to 2000 V EV Dividers

Texas Instruments has put a ±0.1% (maximum) ratio-matching spec on an automotive resistive divider that stands off 2,000 V DC continuously. The RES80A-Q1, qualified to AEC-Q200 temperature grade 1, targets high-voltage sensing in ICE, BEV, and hybrid vehicles — a job that used to belong to scientific and medical instrumentation.

The device is a classic two-resistor divider built in thin-film silicon chromium (SiCr) on TI's analog wafer process. With a nominal high-side input resistance of 50 MΩ, it steps EV-bus potentials down into the full-scale input range of a standard ADC or op amp. Eight ratio options (high-voltage to low-voltage, RHV:RLV) span 145:1 to 1000:1, so designers can match the divider to whatever front end they already have.

What does the drift budget look like?

For dividers, ratio matching matters as much as absolute accuracy, because the ratio is what the measurement actually depends on. TI specifies each divider's measured ratio within ±0.1% of nominal, held across the –40 to +125°C temperature range and through aging with a further ±0.2% maximum cumulative drift. Stack those up and an uncalibrated RES80A-Q1 stays inside a ±0.3% envelope over its lifetime.

That uncalibrated number is the one worth noting. It means designers of EV voltage-monitoring loops can potentially skip a calibration step — or use it only to trim residual error — in systems where the high-voltage rail must be known accurately for safety and state-of-charge decisions.

How does it survive 2,000 V?

The resistors sit under a silicon-dioxide (SiO2) insulative layer that encapsulates them and supports sustained operation at up to 2,000 V DC, with a 4,000 V DC, 60-second HiPOT rating for production testing. The package does its part too: the 8-pin DWV SOIC, 7.5 × 5.85 mm nominal body, maintains at least 8.5 mm of creepage and clearance between high- and low-voltage pins. Those two spacing parameters (creepage is surface distance, clearance is through-air distance) largely determine whether a high-voltage design clears regulatory approval.

The underlying process is qualified through time-dependent dielectric breakdown (TDDB) testing: dedicated test structures are held at constant high AC or DC stress until they fail by electrical short, and lifetimes at working voltage are then extrapolated using the Weibull method. TI's TDDB work showed that unipolar and bipolar biasing conditions produce different operational lifetimes for the process — a detail that matters if your signal is AC-referenced rather than a steady DC bus.

What's not included?

Galvanic (ohmic) isolation isn't part of the device. If the measurement path needs it, you add it externally, and several approaches work. TI backs the part with a 23-page datasheet covering electrical data, thermal behavior, layout, packing, and ordering — which practitioners reading this will recognize as above-average depth for a passive component.

The takeaway: EV high-voltage sensing no longer means cobbling together precision high-value resistors from instrumentation catalogs, but it does raise the question of how many designers will trust a monolithic divider's TDDB-extrapolated lifetime over discrete thick-film stacks in safety-relevant battery-management paths.