HELLO CHINA

INDUSTRY CLUSTER · Automotive manufacturing

Tianjin Dongli new-energy vehicle power systems

Dongli new-energy power systems share a language of operating conditions rather than isolated peak values. One hardware-software combination must work in cold starts, loaded climbs, after fast charging and through durability. Cell consistency, torque response, insulation, thermal control and calibration generate separate conclusions that return to vehicle cooling, communication, structure and safety interfaces.

Formation and regional foundations

Dongli connects automotive research, testing facilities and component projects with vehicle demand, allowing batteries, motors, controls and powertrain testing to feed one another. Operating problems are decomposed on benches and then confirmed in vehicles. Expansion projects describe planned construction, so present facilities remain separate from future capacity.

A battery, motor, controller and cooling loop become a power system only inside one operating condition. Dongli’s manufacturing, joint laboratories and test facilities can separate cold, grade, fast-charge, durability and electromagnetic problems; when recombined in a vehicle, hardware, software, interfaces and environment must remain the same set.

Core geography: A Dongli network linking batteries, electric drives, controls, calibration and testing across the development zone, New Energy Innovation Port and CATARC base

One anomaly loop through a Dongli power system

From a cold-soaked sample to vehicle closure

Each link carries sample identity, configuration and timebase; positions show test responsibility rather than plant coordinates.

  • Cold-soaked battery — State of charge, temperature and cell spread determine usable power.
  • Synchronous torque bench — Request, current, motor and wheel signals share a timebase.
  • Cooling-branch observation — Flow, local temperatures and derating expose thermal margin.
  • Calibration revision desk — Protection thresholds, filters and recovery bind to a checksum.
  • Vehicle interface sample — Cooling, harness, bus and installation complete vehicle review.

01 · Cold soak exposes battery power margin first

Lower temperature changes resistance, capacity, charge acceptance and auxiliary demand.

Cold start separates usable battery power through changes in cell resistance, available capacity and charge acceptance. Fix cold-soak duration, chamber temperature and initial battery state of charge together. If cabin or battery heating starts early, mark auxiliary consumption on the power trace so preheating benefit is not confused with intrinsic cell capability. Chamber-door time is recorded because brief warming can alter the first discharge.

Resting voltage alone can overstate capability; under load, a weak cell reveals itself through voltage sag and temperature spread. Two modules with similar resting voltage can diverge sharply in loaded voltage drop and temperature rise. A cold-condition conclusion covers the weakest cell, contactor action and the exact BMS power-limit event rather than only pack averages. Alongside pack average, retain the coldest sensor and neighbouring cells.

After a defined soak, record cell voltage, temperature, insulation and available power before start, acceleration and regenerative-braking cycles. For launch and regeneration, record requested, permitted and delivered power with drive mode and thermal state. When low voltage or temperature triggers an exit, retain the cell identifier and recovery behaviour on both sides of the event. External charging or a power cycle during recovery is treated as a separate event.

  • Soak and state of charge are fixed together.
  • Loaded traces identify the weakest cell.
  • Power limiting retains trigger and recovery.

02 · A neat spread after balancing may be temporary

Consistency comes from cycling trends and measurement-chain diagnosis, not one screenshot.

Differences in capacity, resistance and self-discharge grow through cycling, while BMS balancing compensates only within limits. Capacity, direct-current resistance and self-discharge come from different measurement windows and cannot be replaced by one instantaneous voltage spread. Retesting the same module before and after cycling shows whether divergence is stable or grows within a state-of-charge region. Trend plots keep one scale so auto-ranging does not exaggerate small differences.

A closer value after balancing does not remove the cause; connection resistance, measurement error and cell aging require different remedies. A poor sense-line contact can create a false weak-cell signature, while genuine ageing follows the same cell after channels are exchanged. Diagnosis separates the measurement chain from the electrochemical object before any decision to replace a complete pack. After sense repair, a short cycle confirms that the apparent fault has not moved channels.

Select a high-spread module, swap measurement channels, and combine capacity, AC resistance, temperature rise and rest decay to identify the cause. Select the largest outlier for capacity, resistance and rest-decay checks, while measuring busbar drop and temperature-sensor bias. Link the retest to earlier cycling under the module serial number and judge the post-repair trend rather than one reset value. Storage state for a removed cell is logged separately to preserve sample identity.

  • Capacity and resistance use different windows.
  • Sense faults are separated from cell ageing.
  • Retests follow one module serial number.

03 · Wheel response is not motor-bench torque

Control, motor, gearing, half-shafts and tyres jointly shape pedal response.

Santroll’s disclosed motors and controllers occupy the electric-drive stage; reduction backlash, half-shaft twist and tyre adhesion join the response chain only in the vehicle. A bench result limited to steady-state efficiency cannot answer launch shock or hesitation during regenerative transition. Wheel testing also retains gear-oil temperature and tyre pressure so mechanical-state changes can be excluded.

An abrupt request can cause shock, while a slow one harms hill starts; regeneration must also blend continuously with hydraulic braking. Requested torque, inverter current, motor output and wheel measurement need a common clock; tens of milliseconds of offset can otherwise resemble control delay. Reverse-impact from mechanical lash is diagnosed separately from current-loop oscillation. Two control revisions are compared only after acquisition filtering is fixed.

Place pedal, requested torque, motor output, wheel torque and speed on one timeline to inspect rise, overshoot, oscillation and transitions. Capture complete traces for low-speed grade launch, hard acceleration, pedal release and brake-regeneration blending, then calculate rise time, overshoot and oscillation separately. If tyre slip occurs, the run no longer rates motor control and moves to an adhesion-limited record. Regeneration logs the hydraulic-brake entry point so blending is not mistaken for motor fluctuation.

  • Signals share one timebase.
  • Mechanical lash does not masquerade as control delay.
  • Four torque transitions are recorded separately.

04 · A long grade challenges cooling more than one sprint

Steady heat, local hotspots and post-stop soak form the full boundary.

A sustained loaded climb raises battery current, motor copper loss, inverter loss and cooling demand differently from brief acceleration. A sustained climb accumulates heat in the battery, motor, inverter and coolant loop together, which a short peak-power figure cannot represent. Grade, vehicle mass, wind condition and speed set input power, while ambient temperature changes heat-rejection margin. A change in frontal area or rolling resistance requires a new grade-power baseline.

Normal average coolant temperature can coexist with a hotspot at a winding end or power module, and protective derating changes the power available to the vehicle. Raising pump speed does not necessarily increase useful flow when branch distribution, an air lock, or exchanger condition restricts heat transfer. Analysis aligns local temperatures, branch flow, and derating commands on one timeline and measures them again after bleeding or refill.

Collect coolant inlet and outlet, critical temperatures, pump-valve commands and derating state during a loaded steady climb, then close the heat balance. After the specified loaded climb reaches steady operation, continue recording through post-stop heat soak. When bench and vehicle use different coolant, line resistance or radiator airflow, state the differences explicitly rather than carrying a conclusion across. Heat-soak completion is defined by a temperature turning point, not a fixed ten minutes.

  • Grade, load and ambient set input.
  • Local temperature aligns with derating.
  • Observation continues after stopping.

05 · Unplugging does not end an insulation problem

Thermal expansion, condensation and sensor drift can change post-charge readings.

High-power charging leaves cells, busbars, contactors and cooling components hot, so the period after unplugging is not static. At the end of fast charging, the battery, busbars and contactors remain hot while coolant pressure and condensation conditions are changing. Interpret insulation readings with charge voltage, temperature, humidity and contactor position; one fault code cannot locate leakage. The charger identifier and cable temperature enter the fault record at an alarm.

One pre-charge measurement misses hot expansion, coolant seepage or stress at high-voltage connections. Thermal expansion can disturb a high-voltage connection only near charge end or briefly after unplugging, which a room-temperature static check misses. Conversely, sensor drift can create an apparent fall without physical leakage. Condensation assessment needs dew point and enclosure temperature, not humidity alone.

Fast-charge checks use four points: before charging, at maximum power, immediately after unplugging, and after complete cooldown, each recording insulation, temperature and humidity, contactor state, coolant pressure, and diagnostic codes. After an anomaly, high voltage is confirmed discharged before branches are isolated in stages to see which disconnection restores the reading. Teardown begins only after the branch is located, and residual-voltage waiting is never skipped for speed.

  • Insulation is read with contactor position.
  • Brief anomalies require continuous capture.
  • High-voltage branches are isolated in stages.

06 · Calibration thresholds consume physical margin

Every parameter change shifts performance, protection and driveability.

Hardware limits in batteries, motors and inverters become driving behavior through torque limits, thermal protection, regeneration and degraded modes. Calibration converts hardware temperature, current and voltage limits into torque limits, regeneration and fallback. Huayi’s powertrain-testing setting can compare parameter changes, but one result applies only to the controller, sensors and software revision used. Test service and sample-design responsibility remain separate, with authorised approval for changes.

A higher protection threshold can release more short-term power while consuming temperature and life margin, whereas an overly low value may derate during normal variation and harm driveability. Every threshold states its physical risk, filter time, and ordering relative to neighbouring protections. When filter delay exceeds fault duration, the controller may miss the event entirely, so one static value is insufficient for comparison.

For each critical calibration, retain purpose, basis, revision and affected conditions, then compare boundary cases before and after the change. Choose one boundary case and hold hardware, environment and driver request constant before and after the change, comparing torque, temperature, warnings and recovery. Calibration file, build checksum and test data share one change number, with unverified side effects left as closure conditions. The rollback revision also receives start-up and safety checks.

  • Results bind to a software checksum.
  • Thresholds state physical basis and filtering.
  • Hardware remains fixed across the comparison.

07 · Post-vibration grounding enters EMC testing

Harness, shielding and connections retain configuration before and after durability.

Long vibration, temperature and load cycling can loosen connections or change grounding, producing intermittent electromagnetic-compatibility symptoms. CATARC’s new-energy test centre publicly covers powertrain, charging and electromagnetic-compatibility facilities, allowing post-durability communication or grounding anomalies to be revisited in different environments. Facility scope does not mean one sample has completed every programme. Packaging and connections are photographed before and after laboratory transfer.

Rerouting a harness may improve electromagnetic behaviour while changing abrasion points, bend radius and cooling; added shielding also consumes weight and packaging space. If vibration loosens a ground intermittently, an emissions or immunity failure may appear only in certain conditions. Each change is rechecked in the same pre- and post-durability configuration, with shielding terminated at specified points and temporary diagnostic jumpers removed before a formal conclusion is recorded.

Before durability, establish baselines for contact resistance, insulation, communication errors and emissions and photograph harness retention and shield termination. Repeat the measurements at the same locations afterwards, while every repair and harness adjustment enters configuration history. Reproduce the original position before changing one connection or route at a time; multiple simultaneous replacements destroy attribution. Repair tools, torque and operator are retained with the configuration change.

  • Facility scope is not a complete sample programme.
  • Shield changes receive mechanical review.
  • Only one suspected variable changes at a time.

08 · The final report must fit back into a vehicle

Bench results become usable only after matching vehicle interfaces and configuration.

A power-system release report matches the tested item to the final vehicle configuration item by item, including voltage, harness, cooling, communications, mounting, mass, and software revision. Dongli joint laboratories can separate motor, controller, and thermal questions on benches but cannot assume interface responsibility for a particular vehicle. Before installation, the customer identifies every difference and decides which bench results remain valid and which duties need additional vehicle testing.

After a component passes its own bench, vehicle installation still adds bus congestion, line resistance, mounting stiffness, and structural resonance. If a report retains only peak efficiency and torque without harness, coolant, firmware, and installation revisions, a field difference cannot be assigned to product or interface. Even without an immediate fault, each new boundary becomes a durability watch item followed under a consistent vehicle state.

At closure, map requirement number, test configuration, result, deviation and owner, with remaining vehicle-level interfaces listed explicitly. Close the review requirement by requirement with hardware, software, condition, result and deviation, identifying work completed only at component level. Vehicle confirmations receive owners and due conditions, while equipment in phase-two construction is not listed as a resource already used. Open items stay in the vehicle list rather than disappearing into an overall-pass phrase.

  • Component and vehicle boundaries are listed separately.
  • Installation can create new risks.
  • Equipment under construction is not counted as used.

Sources and reading notes

A revision triggers different retest scope: sensors or harnesses affect measurement and EMC, cooling parts affect heat balance, and control parameters affect torque and protection. Identify the physical chain first, then choose component, powertrain or vehicle testing; uncovered interfaces remain vehicle-confirmation items.

  • National power-system distinctive cluster: 1.
  • Vehicle makers named for joint laboratories: 4.
  • Opening year of the first test-center phase: 2022.
  • Test buildings in the first phase: 20+.

Related companies

These companies have a locally relevant industry role supported by public information. Order is not a ranking, inclusion does not imply authorization, partnership or endorsement, and Hello China does not participate in or guarantee transactions.

A company may request a free correction or removal through the formal channel if its record is inaccurate or it does not wish to be shown. Do not send business licences, authorization letters or other sensitive files through insecure channels.

  • CATARC NEV Test Center — Tianjin planning records identify the project entity and Dongli site, while science-authority material separately records the base opening and test capabilities.
  • Huayi Tianjin Testing — Huayi Technology’s statutory disclosure confirms the Tianjin entity, Dongli address and powertrain-testing business, while local project material supports its development-zone relationship.
  • Tianjin Santroll — Dongli industrial material identifies motors and controls as core power-system stages, while Santroll’s site confirms the Tianjin entity and electric-drive products.
  • Request a company-record correction or removal — Requests are reviewed by a person, handled promptly after verification, free of charge and not tied to commercial cooperation.

Sources

Use the original publishers to check dates, scope and definitions; public material is not a transaction guarantee.