HS CAN Transceiver (ISO 11898-2) — Classic 1 Mbps
← Back to: Automotive Fieldbuses: CAN / LIN / FlexRay
This page shows how to select, protect, verify, and debug a Classic HS CAN transceiver (ISO 11898-2, 1 Mbps) on real vehicle harnesses. The focus is on measurable behavior—termination/reflectance, common-mode headroom, port parasitics, and thermal/fault recovery—so issues become reproducible and fixable.
Scope Guard & Definition (HS CAN Transceiver, ISO 11898-2)
This page is a vertical, engineering-grade guide for Classic High-Speed CAN transceivers (ISO 11898-2) up to 1 Mbps. It focuses on electrical robustness (common-mode tolerance, short-to-battery/ground behavior, ESD/EMC co-design, TxD safety behaviors, and thermal protection), plus a verify → debug loop that prevents “works on bench, fails on harness” surprises.
- Does: converts MCU TXD/RXD logic into a robust differential bus interface (CANH/CANL), enforces port-level safety behaviors, and survives real-harness faults.
- Does not: define arbitration, bit timing math, or higher-rate data-phase constraints (those belong to controller/protocol and sibling pages).
- Where used: body, chassis, and powertrain Classic CAN networks where wiring, ground offsets, ESD/EMI, and fault survival dominate field reliability.
- Dominant/recessive electrical behavior, waveform sanity targets, and termination/stub impact (mechanism + checks).
- Wide common-mode tolerance and ground-offset failure modes (how to detect “CM overrun”).
- Short-to-VBAT/GND survival: current limiting, thermal shutdown/recovery behaviors, and diagnostic observability.
- Port ESD/EMC co-design at the HS CAN connector (TVS/CMC/split termination interactions and placement rules).
- TxD dominant timeout, fail-safe receive behavior, and mode control signals (EN/STB/VIO implications).
- CAN FD transceivers: data-phase constraints, faster timing windows, emission trade-offs — see CAN FD Transceiver.
- SIC / SIC-XL: waveform shaping and symmetry metrics for heavy loads — see SIC / SIC-XL.
- CAN XL PHY: Classic/FD/XL compatibility and higher host interfaces — see CAN XL PHY.
- Selective wake / partial networking (ISO 11898-6): frame-filter tables, false-wake control — see Selective Wake.
- A scope-locked selection logic focused on Classic HS CAN robustness (CM window, short survival, ESD/EMC interaction, thermal behavior).
- A port-level co-design view that ties waveforms ↔ protection parasitics ↔ harness realities.
- A verify/debug path that maps symptoms (bus-off, intermittent errors, post-ESD fragility) to measurable causes.
Where the HS CAN Transceiver Sits (System Position & Topology)
Reliable Classic CAN is a system contract between three layers: the controller (protocol/timing), the transceiver (electrical interface and fault survival), and the harness (topology, ground offsets, and noise return paths). This section builds a root-cause map so later debug steps can assign responsibility quickly.
- Controller side observables: TXD toggles, RXD stability, error counters, bus-off conditions (logic-domain evidence).
- Bus side observables: CANH/CANL waveforms (differential + common-mode), reflections from stubs/termination, symmetry under load.
- Transceiver contribution: output drive behavior, fail-safe receive behavior, dominant timeout, and recovery behavior after faults.
- VIO matters: TXD/RXD thresholds must match the MCU I/O domain; undervoltage behavior can look like “random glitches” if not validated.
- GND is not ideal: ground offsets and return currents shift CANH/CANL common-mode levels; a clean differential trace can still fail if common-mode exceeds tolerance.
- Decoupling is a stability tool: local supply impedance affects edge control and immunity; place high-frequency decoupling close to the transceiver supply pins.
- TVS placement defines where surge current flows: keep it close to the connector so the return path stays out of the transceiver ground.
- CMC reduces common-mode radiation: but asymmetry or extra parasitics can convert differential energy into common-mode noise.
- Split termination damps and centers common-mode: but wrong midpoint network/placement can worsen immunity or distort edges.
Gateway, DoIP, and Ethernet bridging are intentionally excluded here. This page covers the ECU port from logic pins to the harness interface and its electrical robustness behaviors.
PHY essentials: Dominant/Recessive, termination & amplitude
Classic HS CAN robustness starts with a golden waveform model. The minimum debug set is CANH, CANL, Vdiff = CANH−CANL, and Vcm = (CANH+CANL)/2. A link can look “fine” on Vdiff while failing in the field if Vcm is pushed toward the receiver’s tolerance boundary by harness return currents or ground offsets.
- Observe both domains: Vdiff for data integrity and Vcm for receiver headroom (do not rely on Vdiff alone).
- Stable dominant plateau: dominant state reaches a clear level and stays flat without “breathing” under load.
- Quiet recessive baseline: recessive state avoids spikes that resemble short dominant pulses (noise-to-threshold risk).
- Controlled edges: rise/fall transitions are monotonic enough to avoid multi-threshold crossings (ringing can create double crossings).
- Symmetry sanity: CANH and CANL move in opposite directions with comparable magnitude (large asymmetry hints at parasitic mismatch).
- Headroom check: Vcm stays inside the receiver tolerance window with margin during worst-case operating conditions.
- Symptom: strong overshoot + long ringing after edges, errors increase with harness length.
- Quick check: power-off measure bus resistance at the connector (compare to expected network equivalent).
- First fix: validate termination placement at both ends; limit stubs before changing transceiver settings.
- Symptom: under-damped ringing (double crossings) or over-damped slow edges (threshold dwell).
- Quick check: compare rise/fall time vs. harness load; look for multi-crossing signatures.
- First fix: restore nominal termination network; then re-evaluate EMI controls (CMC / slew settings).
- Symptom: errors appear only with certain nodes connected; ringing frequency changes by configuration.
- Quick check: isolate branches and compare waveforms; correlate errors with harness configuration changes.
- First fix: shorten stubs / move nodes; avoid “fixing reflections” by adding large capacitors (can reduce margin).
- Fast edges: better timing margin but higher EMI and stronger reflection visibility on real harnesses.
- Slow edges: EMI-friendly but can increase “threshold dwell” under heavy bus capacitance, making noise-induced errors more likely.
- Capacitance stacking: harness capacitance + node input capacitance + protection parasitics can deform edges and reduce Vcm headroom.
- Allowed statement (scope-guard): physical waveform distortion can cause controller sampling mistakes; bit timing parameter calculation belongs to controller-focused pages.
Common-mode & ground offset (why “Vdiff looks OK” can still fail)
In Classic CAN field failures, many dropouts are not caused by insufficient differential signal. They occur when Vcm is pushed outside the receiver’s tolerance window by ground offsets, return currents, and transient events. Treat common-mode as a first-class design metric: it links harness physics, protection return paths, and ECU grounding into one measurable margin.
- Errors appear only under specific operating conditions (motor start/stop, charging, high load, cold crank).
- Bus-off events correlate with external equipment plug/unplug or cable routing changes.
- Post-ESD systems become “more fragile” even when Vdiff still resembles the bench waveform.
- Different ECUs show different error sensitivity on the same harness (ground reference dependency).
- Offset: ECU-to-ECU ground potential difference shifts both CANH and CANL together.
- Ripple: inverter/motor noise couples into harness and reference planes as common-mode oscillation.
- Transient: ESD/surge return paths create short ground “steps” that momentarily exceed receiver tolerance.
- Result: Vdiff can remain recognizable while the receiver loses headroom due to Vcm overrun.
- Capture CANH and CANL (single-ended to local ECU ground), then inspect Vcm drift and spikes.
- Log error counters and correlate with operating conditions (motor events, load steps, connector activity).
- Compare Vcm at different nodes to detect ground reference divergence along the harness.
- After ESD events, re-check leakage/asymmetry signatures in protection components (TVS/connector) as part of the CM margin audit.
- Constrain return paths: keep surge current away from transceiver ground (TVS near connector; short return loop).
- Ground strategy: reduce shared high-current return impedance; avoid long ground loops that inject common-mode.
- Harness discipline: routing and shielding/return planning to minimize common-mode coupling in noisy zones.
- Isolation decision rule: if ground offsets and Vcm excursions remain unavoidable, perform an isolated CAN evaluation → Isolated CAN / CAN FD Transceiver.
Isolation device ratings, CMTI, creepage/clearance, and isolated power design are intentionally not expanded here and belong to the isolated CAN subpage.
Robustness: short-to-VBAT/GND, reverse & load transients
Robustness must be evaluated as observable behavior modes under faults: how the device limits stress, what it does to the bus waveform during the event, how it recovers, and what diagnostics are available to prove the root cause. This section converts “short-to-VBAT/GND, tolerance, thermal protection” into verifiable items.
Focus is limited to the transceiver’s fault behavior on the HS CAN port and its immediate supply/IO influence paths. Detailed TVS/surge component selection methodology and vehicle-level EMC standards are linked out to the EMC/Protection page.
- Protection target: output stage survival and bus integrity during a hard pull-up fault.
- Trigger condition: CANH or CANL forced toward VBAT for an extended duration or repeated events.
- Device behavior: current limiting / foldback to cap dissipation; may transition to thermal protection if power persists.
- Recovery: automatic retry after cool-down or latch-off until EN/STB/power-cycle (device-dependent).
- Diagnostics: fault indicator pin / status flag (if present), RXD behavior change, controller error counters.
- Pass criteria (placeholder): no damage after X events; recovers within Y; error rate stabilizes within Z.
- Protection target: output stage and substrate clamps under a hard pull-down fault.
- Trigger condition: CANH or CANL forced to GND (hard short or intermittent contact).
- Device behavior: controlled drive / current limiting; thermal shutdown if dissipation continues.
- Recovery: auto-retry can appear as periodic reconnect attempts; latch strategy improves controllability.
- Diagnostics: fault output (if available) + bus waveform signature + error counters correlation to the fault window.
- Pass criteria (placeholder): no permanent dominant; no permanent bus lock; recover behavior matches spec.
- Protection target: prevent one ECU from holding the network dominant due to a TXD fault.
- Trigger condition: TXD asserted dominant too long (stuck low / firmware fault / VIO glitch).
- Device behavior: dominant timeout forces release to recessive (device-dependent).
- Recovery: may require TXD deassert + mode toggle; confirm whether auto-retry is present.
- Diagnostics: timeout flag (if present) + bus waveform showing dominant release + controller counters.
- Pass criteria (placeholder): no network lock under TXD fault; deterministic release within X.
- Protection target: avoid undefined IO states and bus corruption during supply excursions.
- Trigger condition: VBAT/VCC step up/down, VIO droop, or mode pin glitches during transients.
- Device behavior: may enter undervoltage-safe state; output drive can reduce, affecting Vdiff/Vcm margin.
- Recovery: verify reset/mode sequencing and whether bus returns cleanly without repeated fault loops.
- Diagnostics: correlate supply rails (VBAT/VCC/VIO) with bus waveform + error counter timelines.
- Pass criteria (placeholder): no latch in undefined state; recovery to normal comm within X after event.
For TVS and surge array selection details, coupling models, and test-standard-driven design flow, see EMC / Protection & Co-Design.
ESD/EMC port co-design (HS CAN port level only)
HS CAN EMC is often won or lost at the connector. The minimum port-level closure is: TVS (transient energy handling), CMC (common-mode control), and split termination (damping + common-mode centering). Each element improves emissions/immunity while introducing parasitics that can distort edges or convert differential energy into common-mode noise if mismatched.
Coverage is limited to HS CAN port interactions, placement order, and mismatch side effects. EMC standards, vehicle-level mitigation, and full compliance workflows are linked out to the EMC/Protection page.
- Benefit: reduces common-mode radiation and improves immunity against coupled noise.
- Trade-off: mismatch or extra parasitics can convert differential energy into common-mode noise.
- Placement rule: keep symmetric routing; place near connector side of the port chain when used.
- Failure signature: CANH/CANL edge asymmetry; Vcm ripple increases while Vdiff still looks plausible.
- Benefit: improves damping and helps stabilize common-mode reference at the bus ends.
- Trade-off: incorrect midpoint RC or poor return path can reduce immunity and distort transitions.
- Placement rule: keep midpoint return clean; avoid long loops that inject transient ground steps.
- Failure signature: recessive baseline “floats”; Vcm centering drifts under load or after ESD.
- Benefit: clamps ESD/surge energy and defines where transient current returns.
- Trade-off: capacitance mismatch (Cdiff) and leakage drift can unbalance the pair and increase common-mode conversion.
- Placement rule: place closest to connector with a short, controlled return path.
- Failure signature: “passes ESD once” but becomes fragile later; CANH/CANL asymmetry grows over time.
- TVS pair mismatch converts differential energy into common-mode noise (Vcm grows).
- TVS placed far from connector forces transient current through sensitive ground references.
- CMC routing not symmetric (extra vias/length) creates imbalance and edge asymmetry.
- Split termination midpoint return uses a long loop, reducing immunity and worsening post-ESD fragility.
- Over-aggressive “extra capacitance for stability” slows edges and increases threshold dwell under harness load.
- Connector shell/chassis return not controlled; transient return path becomes unpredictable.
- After EMC tests, only waveforms are checked; component drift/leakage and solder damage are not audited.
For coupling models, component selection depth, and standards-driven compliance flow, see EMC / Protection & Co-Design.
Digital behaviors: TXD dominant timeout, fail-safe, standby/normal, thermal
Datasheet terms such as TXD dominant timeout, fail-safe receive, and thermal shutdown must be treated as system behaviors: when they trigger, how they look on the bus and in counters, how to confirm with minimum measurements, and how to recover deterministically.
Coverage is limited to HS CAN transceiver behaviors and mode control pins (EN/STB) at the port level. Selective wake / partial networking filtering logic and tables are not expanded here; see ISO 11898-6 Selective Wake / PN.
Trigger → Observable symptoms → Quick check → Fix. Observable symptoms should include at least one of: bus waveform signature (CANH/CANL/Vdiff/Vcm), controller error counters, or a transceiver fault/status indication.
- Bus appears held dominant; other nodes fail to arbitrate/transmit.
- When timeout engages, the bus is forced back to recessive (release signature).
- Controller error counters rise rapidly; bus-off may occur if the fault persists.
- Capture TXD + CANH/CANL on the same timeline.
- Correlate the event with error counters and bus-off timestamps.
- If available, log the transceiver fault/status indication.
- Enforce watchdog reset policy and TXD control sanity checks in the MCU.
- Ensure EN/STB and VIO sequencing avoids undefined TXD states during transients.
- Define a deterministic recovery action (mode toggle or reset) consistent with the device’s retry/latch behavior.
- RXD remains stable and predictable rather than toggling with noise.
- False frames/false interrupts are reduced compared to non-fail-safe behaviors.
- Disconnect the bus branch and observe RXD stability (no random toggles).
- Log false interrupt count / invalid frame count during noise-heavy operating modes (placeholder fields).
- Audit connector contact quality and port component drift after ESD/EMC stress (TVS leakage/mismatch, CMC solder, midpoint return).
- Harden MCU input handling (interrupt policy, error filtering) without relying on PN filtering tables.
- Iq shifts between standby and normal in a repeatable way (device-dependent).
- Wake events can be attributed to pin/bus/timer sources if logging is designed (placeholders).
- Mode sequencing mistakes appear as intermittent wake, missed wake, or undefined RXD behavior.
- Record EN/STB, VIO, and RXD during sleep/wake transitions.
- Log wake-source fields (bus/local/timed) and verify that false-wake rate stays bounded (placeholders).
- Define a deterministic mode sequencing rule (VIO stable before enabling normal communication).
- Implement wake attribution logging to separate bus activity from local-pin disturbances.
- Keep ISO 11898-6 PN filtering tables and configuration in the dedicated PN page.
- Intermittent dropouts that recur with a roughly thermal time constant (“every few minutes”).
- Error counters spike near the dropout; bus returns when the device cools.
- If auto-retry exists, periodic reconnect attempts may be visible as repeating waveform patterns.
- Record board/hotspot temperature vs time and align with error counter timestamps.
- Capture CANH/CANL (or Vdiff) before/after the dropout to see drive-state changes.
- Check for retry storms that amplify dissipation (software/hardware interaction).
- Eliminate fault loops (continuous dominant attempts, uncontrolled retries) that raise dissipation.
- Audit port components and harness faults that cause persistent stress (short/leakage/contact intermittency).
- Define a controlled fallback mode on overtemp (report + backoff + standby transition when appropriate).
- Use a watchdog reset policy that prevents indefinite TXD dominant assertion and repeated fault retry storms.
- Log fault evidence as a minimum: mode pins (EN/STB), VIO/VBAT/VCC rails, error counters, and a temperature proxy.
- Implement deterministic recovery actions (mode toggle/reset) aligned with the transceiver’s retry/latch strategy.
- Keep wake-source attribution fields (bus/local/timed) for serviceability; separate false wake from real bus activity.
- After ESD/EMC stress, audit port component drift (TVS leakage/mismatch, CMC solder, midpoint return integrity).
Engineering checklist (Design → Bring-up → Production)
This checklist is a reusable “project bible” asset. Each item is phrased as a doable action with a pass criteria placeholder to enable consistent reviews, bring-up, manufacturing tests, and field diagnostics.
- Define VBAT/VCC decoupling placement to minimize loop area near the transceiver. Pass (placeholder): rail droop/spike within X under Y event.
- Verify VIO compatibility and undervoltage behavior (RXD and bus release under brownout). Pass (placeholder): no undefined mode; release within X.
- Keep CANH/CANL routing symmetric (length/vias/components) to minimize mismatch conversion. Pass (placeholder): edge asymmetry ≤ X; Vcm ripple ≤ Y.
- Place TVS closest to the connector with a short, controlled return path. Pass (placeholder): transient return loop length ≤ X; no post-ESD fragility.
- Use CMC and split termination only with a clean return and symmetric placement. Pass (placeholder): no added bit errors across harness variants.
- Plan return paths so high current transients do not share sensitive transceiver reference ground. Pass (placeholder): Vcm stays in window with margin ≥ X.
- Control connector shell / chassis reference strategy to avoid unpredictable transient returns. Pass (placeholder): ESD event does not shift baseline/offset beyond X.
- Wire fault/status indication (if available) into the MCU for field attribution. Pass (placeholder): fault capture latency ≤ X.
- Define “black box” fields: error counters, bus utilization, temperature, supply minima, wake-source. Pass (placeholder): all fields present for any bus-off event.
- Validate recessive baseline and leakage indicators. Pass (placeholder): recessive stable; leakage within X.
- Measure standby/normal Iq and confirm mode pins behave deterministically. Pass (placeholder): Iq ≤ X in standby; transitions repeatable.
- Verify RXD stability under bus open/floating (fail-safe behavior). Pass (placeholder): no random RXD toggles.
- Capture CANH/CANL + Vdiff/Vcm to validate edge symmetry and ringing. Pass (placeholder): overshoot ≤ X; ring settles ≤ Y.
- Check common-mode headroom under real harness loads and noise sources. Pass (placeholder): Vcm margin ≥ X under Y condition.
- Validate short/open behavior modes and confirm timeout and thermal behavior are observable and recoverable. Pass (placeholder): no damage; recovery within X; no permanent dominant.
- Correlate fault events with diagnostics: fault/status, error counters, temperature, supply minima. Pass (placeholder): attribution complete for every event.
- Lock measurement points and reference ground across stations/fixtures. Pass (placeholder): station-to-station delta ≤ X.
- Use a consistent harness substitute (capacitance/length class placeholder) and keep golden references. Pass (placeholder): golden baseline drift ≤ X over Y.
- Bus utilization + per-window error counters (type + time bucket placeholders).
- Temperature proxy + supply minima/maxima (VBAT/VCC/VIO placeholders).
- Wake-source attribution (bus/local/timed) + mode transitions (standby↔normal).
- Pass (placeholder): any bus-off report includes all fields for root-cause attribution.
Vehicle-level EMC standards and full compliance workflows are covered in EMC / Protection & Co-Design. Selective wake / PN details are covered in ISO 11898-6 Selective Wake / PN.
Verification & debug playbook (measure → judge → reproduce)
This section provides the mainline debug path that should be completed before FAQs. The workflow is evidence-driven: classify the failure, capture the minimum signal set, reproduce with controlled variables, then apply fixes that stay within the HS CAN port scope.
This playbook focuses on HS CAN transceiver + port-level causes. Detailed controller bit-timing calculation is not covered here. If waveform evidence suggests sampling/segment mismatch, return to the controller page: CAN Controller / Bridge.
- Short harness looks OK; long harness fails.
- Stub/topology sensitive; ringing and threshold crossings appear.
- Error bursts correlate with harness variants.
- Bus-off only in specific operating modes (motor, high current, hot plug).
- After ESD, the link becomes more fragile under the same harness.
- Vcm headroom collapses in the problematic condition.
- Changing TVS/CMC/split termination changes stability strongly.
- “Low-C TVS” still makes it worse due to mismatch conversion.
- CANH/CANL asymmetry increases common-mode radiation and sensitivity.
- Dropouts recur periodically (“every few minutes”) and recover after cooling.
- Error spikes align with temperature peaks and retry bursts.
- Heating/cooling shifts the failure boundary consistently.
- Capture CANH, CANL and derive Vdiff + Vcm (math channel is sufficient).
- Record a “golden” waveform and compare only in the failing condition.
- Pass criteria (placeholder): Vcm stays in window with margin ≥ X; ringing settles within Y.
- Check equivalent termination (power-off), verify endpoint placement, and stub lengths.
- Document harness class (length/capacitance placeholder) for reproducibility.
- Pass criteria (placeholder): equivalent termination within X±Y; key stub ≤ X.
- Log temperature proxy near the transceiver and supply current/power during the failure.
- Align temperature with dropout timestamps and retry bursts.
- Pass criteria (placeholder): temperature reaches stable plateau; no periodic dropout under steady state.
- Record controller error counters per time window and bus-off timestamps.
- Log mode pins (EN/STB), VIO/VBAT minima, and fault/status indication (if available).
- Pass criteria (placeholder): counters ≤ Y within X minutes; bus-off rate ≤ Z per hour.
Use a repeatable harness class (capacitance/length placeholder) to separate topology-sensitive reflection issues from common-mode issues.
Apply controlled short/open events using safe, repeatable fixtures. Confirm protection behavior and deterministic recovery without locking the network.
Use controlled heating/cooling to test whether dropouts follow thermal time constants and whether recovery is stable under steady conditions.
Compare the same board before/after ESD/EMC stress; focus on port-component drift (TVS leakage/mismatch, CMC solder integrity, connector contact).
- Next measurement: Vdiff + ringing; endpoint termination sanity (Pack A + B).
- Decision: ringing/settling differs strongly across harness classes → reflection/termination bucket.
- Fix: endpoint placement, stub control, termination correctness; re-check symmetry around port network.
- Pass (placeholder): error counters ≤ Y within X minutes for harness class N.
- Next measurement: Vcm headroom vs time; ground reference integrity (Pack A + D).
- Decision: Vcm approaches/exceeds window in failing condition → common-mode bucket.
- Fix: return-path planning, port ground strategy, reduce common-mode injection; consider isolation if rules indicate (link out).
- Pass (placeholder): Vcm margin ≥ X under worst-case event; bus-off = 0.
- Next measurement: symmetry (CANH vs CANL edge) and Vcm ripple; compare before/after swap (Pack A).
- Decision: mismatch converts differential edges into common-mode noise → protection parasitics bucket.
- Fix: matched arrays, symmetric placement, shortest returns, avoid unequal capacitance; re-baseline waveform.
- Pass (placeholder): stability unchanged across approved port BOM variants.
- Next measurement: temperature + supply current vs time; align to error spikes (Pack C + D).
- Decision: failure aligns with thermal time constant → thermal bucket.
- Fix: eliminate retry storms and persistent stress; validate recovery behavior; improve dissipation paths.
- Pass (placeholder): dropout rate ≤ X in Y hours at steady temperature.
Applications (Classic HS CAN transceiver placement)
These application cards list typical in-vehicle placements for Classic HS CAN (ISO 11898-2). The focus stays on port-level priorities and logging hooks, without expanding into other buses or Ethernet/DoIP gateway details.
DoIP, Automotive Ethernet, and gateway architecture are not expanded here. Port-level protection and logging hooks are described; deeper gateway/Ethernet topics are linked out to the appropriate domains.
- Common-mode headroom under high-current events and noisy grounds.
- Short-to robustness and deterministic recovery behaviors.
- ESD/EMI robustness with stable port-component interactions.
- Ground offset and surge return paths collapse Vcm margin.
- Post-ESD drift in TVS/CMC/connector increases fragility.
- Thermal periodic dropouts caused by persistent stress and retry storms.
- Clear TXD timeout and thermal behavior (retry/latch semantics documented).
- Strong common-mode tolerance with predictable fail-safe behavior.
- Diagnostics hooks usable for field attribution (fault/status placeholders).
- Standby current and deterministic EN/STB transitions.
- Wake-source attribution (bus/local/timed placeholders).
- Stable RXD (fail-safe) to reduce false triggers without PN tables.
- Mode pin glitches under VIO transients cause false wake or missed wake.
- Connector intermittency causes RXD toggling and phantom events.
- Port-component mismatch increases susceptibility to EMI spikes.
- Well-defined standby behavior and robust RXD stability under open/floating.
- Clear wake signaling and the ability to log wake attribution reliably.
- Port network tolerance to matched TVS/CMC implementations.
- Port protection co-design that stays stable across harness and fixture variants.
- Consistent debug logs for service attribution (counters, temperature, rails, mode history).
- Deterministic behavior under faults without network lock.
- Fixture/harness substitutes differ across stations, creating false “regressions”.
- Protection BOM changes break symmetry and increase common-mode conversion.
- Missing black-box fields prevents root-cause attribution after bus-off.
- Usable fault/status hooks and predictable timeout/thermal behavior.
- Robustness across port-component implementations (matched arrays, symmetric layout rules).
- Compatibility with consistent manufacturing evidence capture (station baselines).
Controller timing details: CAN Controller / Bridge. Vehicle-level EMC workflows: EMC / Protection & Co-Design.
H2-11 · IC Selection Logic (HS CAN Transceiver, ISO 11898-2, Classic 1 Mbps)
Selection is treated as a five-step decision tree. The goal is to pick a transceiver type that matches real harness risks: short faults, common-mode headroom, port protection parasitics, diagnostics, and standby strategy.
Scope guard: no CAN FD fast-phase timing metrics (loop delay symmetry / sample-point optimization) and no ISO 11898-6 frame-filter tables. Use internal links to the CAN FD page and the Selective Wake / Partial Networking page when needed.
Five-step decision tree
- Robustness gate: short-to-VBAT / short-to-GND requirement (duration, repetition, recovery behavior, and whether the event is diagnosable).
- Common-mode gate: extreme ground offset / noisy return paths reduce CM headroom. If CM margin collapses in real vehicle events, an isolated CAN option becomes a primary candidate (details belong to the Isolated CAN page).
- Port BOM compatibility: TVS / CMC / split termination parasitics can convert differential energy into common-mode noise or distort edges. Pick a transceiver that tolerates mismatch and preserves symmetry.
- Diagnostics & safety hooks: fault indication / status visibility / predictable fail-safe behavior. Prefer parts that turn “mystery dropouts” into traceable events.
- Power & mode strategy: standby IQ targets and wake attribution (bus vs local). Keep this at mode-control level; do not expand into ISO 11898-6 filtering tables here.
Output of the tree is a recommended type: Robust / CM-headroom (isolation candidate) / Port-parasitic-tolerant / Diagnostics-enhanced / Low-standby. Vendor part numbers below are examples and must be verified for AEC-Q100 grade, temperature range, package, and ordering suffix.
Field glossary (selection-critical, 2–3 lines each)
Example material numbers (HS CAN, ISO 11898-2; Classic 1 Mbps capable)
The list is organized by the decision-tree outputs. Each item is a concrete MPN example; confirm suffix details (package, temperature grade, and ordering code) before locking a BOM.
- Texas Instruments TCAN1051H-Q1 (HS CAN transceiver family; fault-protected behavior emphasis)
- Texas Instruments TCAN1042-Q1 / TCAN1042V-Q1 (HS CAN physical layer compliant; classic CAN capable; ordering variants exist)
- Infineon TLE6250 (HS CAN transceiver; short-circuit proof and overtemperature protection are core design points)
- Microchip MCP2561 (1 Mb/s operation; ISO 11898-2/-5 physical layer class)
- onsemi NCV7342 (HS low-power CAN transceiver family; standby and wake use-cases)
- NXP TJA1042 / TJA1042T (HS CAN transceiver with low-power mode variants in the family)
- NXP TJA1051 / TJA1051T,
112 (HS CAN transceiver family; verify the exact variant and features per project needs) - Microchip MCP2561-E/
MF (ordering-code example; verify package/suffix) - STMicroelectronics L9616 (HS CAN transceiver class; confirm project-specific protection goals at the system level)
- Infineon TLE6250
GV33 (example ordering code; confirm voltage/I/O variant)XUMA1 - onsemi NCV7342
MW3 (example ordering code; confirm package and screening level)R2G
- Confirm AEC-Q100 grade and temperature range for the ECU location.
- Confirm I/O logic level expectations (VIO support, RXD level) and reset behavior.
- Confirm fault behavior model: current limit / thermal / auto-retry vs latch-off; and whether it is diagnosable.
- Validate port BOM synergy with the intended TVS/CMC/split termination on the real harness.
- Keep a link-out for isolation, partial networking, and CAN FD timing topics (avoid expanding scope here).
Recommended topics you might also need
Request a Quote
H2-12 · FAQs (HS CAN Transceiver, ISO 11898-2, Classic 1 Mbps)
These FAQs close long-tail debug questions without expanding the main body. Each answer is intentionally “data-shaped”: Likely cause / Quick check / Fix / Pass criteria (threshold placeholders).
Scope guard: No CAN FD fast-phase timing math and no ISO 11898-6 PN filter tables here. If timing math is suspected, route to the CAN controller timing page.
Short harness is stable, but vehicle harness causes bus-off: check reflections or common-mode first?
Likely cause: [Reflection or CM] harness capacitance/stubs create ringing, or Vcm headroom collapses during real load/ground events.
Quick check: capture CANH,CANL→Vdiff/Vcm on vehicle harness; measure termination Z and stub map; align with error counters/time.
Fix: correct end termination and stub length first; if Vcm is near limits, fix return paths/grounding (then consider isolation per CM rule).
Pass criteria: Vcm stays within window with ≥ X margin; bus-off = 0 over Y minutes; errors ≤ Z/10 min.
ESD test passes, but the bus becomes “more fragile” afterward: TVS leakage/mismatch or connector damage?
Likely cause: [Port parasitics] TVS leakage shift or Cdiff/mismatch increases CM conversion; connector contact damage adds intermittent impedance.
Quick check: A/B compare pre/post-ESD on the same harness: Vcm headroom, symmetry (CANH vs CANL), and intermittent dropouts under vibration.
Fix: replace TVS/connector as controlled swap; restore symmetry and shorten ESD return loop (TVS near connector, clean return path).
Pass criteria: post-ESD errors do not increase by more than X% over Y hours; Vcm margin ≥ Z under worst-case events.
Errors happen only at cold/hot temperature: thermal shutdown or common-mode window shrink?
Likely cause: [Thermal or CM] overtemp cycling under load, or reduced CM margin due to temperature-dependent ground/return shifts and leakage.
Quick check: log temperature vs error counters; capture Vcm at cold/hot during the same harness event; look for periodic drop/recover signature.
Fix: if periodic, reduce sustained drive/short-fault stress and improve heat removal; if CM-limited, fix return path and ground offset sources.
Pass criteria: no periodic dropouts over X thermal cycles; bus-off = 0; Vcm stays within limits with ≥ Y margin at both extremes.
Bit errors rise after swapping TVS vendor: what is the fastest Cdiff / asymmetry sanity check?
Likely cause: [Port parasitics] CANH/CANL capacitance mismatch (Cdiff) converts differential energy into CM noise and shifts thresholds on real harness edges.
Quick check: compare CANH vs CANL edge shape and Vcm swing before/after; inspect placement symmetry and return path; swap back as A/B control.
Fix: select matched low-C arrays with tighter symmetry; enforce symmetric routing/via count; keep TVS at connector with short return.
Pass criteria: Vcm increase ≤ X under the same stimulus; error rate returns to baseline ≤ Y/10 min.
The bus is “held dominant” intermittently: did TxD dominant timeout trigger, and how to confirm fast?
Likely cause: [Digital behavior] TXD stuck-low or control-domain fault forces dominant; dominant timeout releases the bus or flags a fault depending on design.
Quick check: capture TXD + CANH/CANL on one timeline; correlate “dominant plateau → release” with error counters and any fault indication.
Fix: harden MCU reset/watchdog strategy; validate VIO behavior during brownout; ensure mode-control (EN/STB) can recover to a known state.
Pass criteria: dominant-hold events = 0 over X hours; any forced-dominant event is bounded to ≤ Y ms and is diagnosable.
After a short-to-VBAT event, communication recovers but errors increase: thermal stress or drift in CMC/termination?
Likely cause: [Thermal + port] repeated protection cycling heats the port, and stressed components (CMC/termination/connector/TVS) shift symmetry or damping.
Quick check: compare waveforms pre/post-fault (ringing + Vcm); inspect temperature rise and recovery pattern; swap suspect port BOM items as A/B.
Fix: verify short-fault behavior model (limit/thermal/recovery); reinforce port BOM robustness and placement; re-validate damping/termination on harness.
Pass criteria: error rate returns to ≤ X per 10 min after recovery; no parameter drift beyond Y across Z fault cycles.
Errors occur only during a specific event (motor start/charging): capture common-mode swing or ground return path first?
Likely cause: [Common-mode] high-current events shift ground reference and inject return-path noise, pushing CANH/CANL CM levels toward limits.
Quick check: capture Vcm during the exact event; log error counters vs event timestamp; verify return path continuity (connector shield/chassis reference).
Fix: shorten and control the surge/return loop; keep CAN reference clean; reduce CM coupling through symmetric layout and appropriate port network placement.
Pass criteria: Vcm remains within limits with ≥ X margin during the event; event-triggered errors ≤ Y per Z events.
RXD jitters intermittently while differential looks normal: fail-safe, threshold, or common-mode coupling?
Likely cause: [RX path + CM] RXD instability from CM excursions, input threshold sensitivity, or missing fail-safe stability under open/noisy conditions.
Quick check: observe RXD + Vcm concurrently; test controlled “bus open/idle” state; check if RXD stays stable when CAN lines are disconnected.
Fix: stabilize CM headroom and return paths; ensure correct mode-control and input domain stability; verify port BOM symmetry to reduce CM injection.
Pass criteria: RXD toggles ≤ X false edges per Y minutes in idle/open tests; bus-off = 0 in target events.
Plugging/unplugging a specific ECU makes the bus unstable: intermittent return path or stub too long?
Likely cause: [Topology/return] connector return discontinuity changes CM reference, or the node introduces a long stub/cap load that increases ringing.
Quick check: compare termination Z and waveform with ECU present vs absent; measure stub length and node input capacitance proxy; inspect shield/chassis bond.
Fix: shorten/relocate stub, restore proper termination, and repair return/bond continuity; validate Vcm headroom after connector changes.
Pass criteria: waveform ringing decays within X ns (placeholder) and errors stay ≤ Y/10 min with ECU hot-plug scenarios disabled/enabled per procedure.
Production tests pass, but road tests fail: which missing “black box” log fields are most fatal?
Likely cause: [Observability gap] lab does not reproduce harness CM events/thermal drift; missing time-aligned logs prevents root-cause attribution.
Quick check: confirm logging includes error counters by type + timestamps, temperature, VBAT/VIO minima, mode transitions, and wake attribution.
Fix: add the minimum field set and align it with waveform captures during worst-case vehicle events; update test fixtures to mimic harness RC.
Pass criteria: every dropout is attributable to one bucket with evidence within X minutes; unknown-cause incidents = 0 across Y road hours.
Termination “looks correct,” but ringing persists: what can split termination / midpoint capacitor placement break?
Likely cause: [Damping + CM path] midpoint network creates unintended CM injection or poor damping if midpoint cap/return path is long or asymmetric.
Quick check: compare ringing with midpoint capacitor installed vs removed (controlled test); measure Vcm change; verify symmetric routing to the midpoint.
Fix: move midpoint capacitor closer to the intended reference with a short return; enforce symmetry; keep the port network ordered correctly (connector→TVS→CMC→termination).
Pass criteria: ringing overshoot/undershoot decreases by ≥ X% and settles within Y ns; Vcm swing reduces to ≤ Z.
Intermittent drop every few minutes: how to correlate thermal shutdown with communication load?
Likely cause: [Thermal cycling] sustained load or fault-retry storms heat the transceiver until thermal-off, then recovery repeats with a consistent time constant.
Quick check: log temperature + supply current + error counters; check if dropout period matches thermal rise/fall; reduce bus load to see if period stretches.
Fix: reduce sustained dominant stress and retry storms; improve heat spreading/airflow; verify recovery mode and prevent repeated fault cycling at the system level.
Pass criteria: no periodic dropouts over X minutes at max load; temperature stays below a project-defined limit with ≥ Y margin; errors ≤ Z/hour.
