# BMS communication: CAN and CAN-FD records

> Channel data shows what the tester did. BMS frames show what the pack thought. DBC mapping, DTCs, HVIL and a shared time base.

Canonical: https://batteryxlab.com/notes/bms-communication-in-testing/
Language: en

- [Home](https://batteryxlab.com/)
- Resources
- [Notes](https://batteryxlab.com/notes/)
- BMS communication: CAN and CAN-FD records

Measurement

# BMS communication: CAN and CAN-FD records

Channel data shows what the tester did. BMS frames show what the pack thought. DBC mapping, DTCs, HVIL and a shared time base.

28 Nov 2026 · 8 min read · [All notes](https://batteryxlab.com/notes/)

A pack test can look clean at the terminals and still be wrong. The tester records the current it delivered and the voltage it saw, but the pack's battery management system (BMS) may have derated the current, opened a contactor, flagged a cell overvoltage, or detected an isolation fault between two samples. None of those events necessarily appears in the channel data. If the test plan only keeps channel data, the reason for a stopped run or a shortened discharge is missing from the record.

CAN and CAN-FD links close that gap, but only if the messages are decoded, time-aligned and stored with the same care as the electrical channels. This note covers what the bus can add, what it cannot replace, and how to handle DTCs and the high-voltage interlock loop (HVIL) without turning them into decoration.

## What the bus carries that the channel cannot

The module and pack systems are published with CAN, SMBus and CAN-FD interfaces and DBC import, and the software page describes SMBus, CAN and CAN-FD with DBC mapping. The applications page is explicit about the intended result: cell voltages, temperatures and state of charge sit on the same time base as pack current and voltage. That is the value of the link. It turns a pack-level electrical record into a record of what the BMS believed while the electrical stimulus was applied.

Datasheet fact: on the BTS-6008A-5V300A-H and BTS-6001A-5V2000A-H, a CAN box is an optional add-on with DBC configuration, and an auxiliary unit adds eight voltage and eight temperature channels. When an AUX channel is connected, the published recording frequency changes from 100 Hz to 10 Hz. That is a real trade: adding auxiliary signals changes the time resolution of the main record. The BMS link and the AUX unit are not free additions to the data path.

Classical CAN carries up to eight data bytes per frame; CAN-FD raises the payload to 64 bytes and separates the arbitration bit rate from a faster data phase. The larger payload is useful for messages that pack many cell voltages into one frame, but it does not make the BMS sample faster. The message cycle time is set by the BMS supplier, not by the bus capability. A 100 ms BMS message is still a 100 ms message on a 5 Mbit/s data phase.

## Why “channel data only” is not enough

The tester is a current source. The BMS is a supervisor. When the two disagree, the pack usually wins: the BMS opens the contactors, requests a lower current, or disconnects a module. The channel record then shows a current that fell to zero or a voltage that collapsed, but it does not show which cell or which limit caused the action. Without the BMS frame, the event looks like a tester fault or a cell failure.

A second gap is the internal state that never reaches the terminals. Pack voltage is a sum; a single cell can be at the upper limit while the sum is still inside the tester's window. Temperature is local; the hottest cell may be far from the pack sensor the tester uses. State of charge and state of health are estimates that the BMS maintains, and they are part of the test condition even when they are not part of the electrical measurement.

A third gap is timing. The tester may command a step at t = 0, while the BMS publishes its limit at the next 100 ms cycle. If the test profile ignores that limit, the tester can request a current the pack will not accept. The result is a protection event that looks like a test failure. The fix is not a faster bus; it is a control strategy that reads the BMS limit and applies the lower of the two.

## Reading the bus correctly

A DBC file defines message identifiers, signal positions, bit lengths, byte order, scaling, offset, units and multiplexing. Two mistakes are common. The first is using a DBC that does not match the pack's software version: the identifiers may still decode, but a signal can move or change scale. The second is treating a decoded value as a measurement without checking its resolution and validity bits. A cell voltage signal with 10 mV resolution cannot support a 1 mV comparison, and a signal marked invalid may still carry a plausible number.

Record the DBC file version or a hash with the test. Record the bus speed, the termination, the message cycle times and the number of frames received versus expected. If a frame is lost, the decoded channel has a gap; if the gap is filled by interpolation, the record no longer shows what the BMS sent. Keep the raw frames alongside the decoded signals, at least for the tests that matter. Raw frames are the evidence; decoded channels are the convenience.

Time alignment deserves its own decision. The tester timestamps the electrical sample when it is recorded. The BMS timestamps the message when it is sent, or not at all. The receive timestamp includes bus arbitration, transmission and driver latency. For most pack tests, a millisecond-level alignment is enough; for a protection event, the order of the contactor command and the current fall matters. Log the receive time and, where the BMS provides it, the BMS time, and state which one the analysis uses.

## DTCs: capture before you clear

A diagnostic trouble code is the BMS's own record that a diagnostic test failed. The code is not the same as the live value: a cell voltage can return to normal while the DTC remains confirmed. Read the DTC status and the freeze-frame data before the test starts, after the test ends, and after any protection event. If the test software clears codes automatically, the evidence disappears.

Engineering judgement: do not clear DTCs as part of test setup unless the test plan explicitly says so and the previous state has been archived. A pre-test scan establishes the baseline; a post-test scan shows what the test caused. If a code is already present, record it and decide whether the test is valid before applying current. A pack with a stored isolation fault is not a clean test article, even if the fault is not currently active.

The same discipline applies to the BMS's reported limits. Log the allowed charge and discharge current, the voltage window and the temperature limits at the start of each step. If the BMS changes a limit mid-test, the change should be visible as a step in the decoded record, not inferred from a current that unexpectedly stopped rising.

## HVIL: log the loop, not the inference

The high-voltage interlock loop is a low-voltage continuity loop routed through the high-voltage connectors and covers. If the loop opens, the BMS is expected to open the contactors. The tester can infer an HVIL event from the loss of pack voltage, but that inference is ambiguous: a contactor command, a fuse, a BMS fault or a loose connector can produce the same electrical signature.

Log the HVIL state as a digital channel, together with the contactor state and the BMS's reported fault. Verify the loop before the test and after the fixture is closed. Do not bypass an HVIL circuit to “finish the run”; if the loop cannot be closed, the test article is not ready. The module and pack page lists contactors, insulation monitoring and emergency stop as part of the safety configuration, and HVIL belongs to that same chain.

An HVIL event during a test should stop the current and record the sequence: HVIL state, contactor command, contactor state, pack voltage, current and the BMS fault. That sequence is what lets the team distinguish a connector problem from a BMS decision. Without it, the event becomes a story told from memory.

## A minimum BMS record

- DBC file name, version or hash; bus speed; termination; message cycle times.

- Raw frames and decoded channels; received-versus-expected frame counts; gap markers.

- BMS-reported pack voltage, current, SOC, cell voltage minimum and maximum, temperature minimum and maximum, and the active limits.

- DTC status and freeze-frame data before and after the test; the time of any code transition.

- HVIL state, contactor state and any isolation or insulation-monitoring flag.

- The time base used for alignment, and the estimated latency between BMS transmission and tester receipt.

This list is a starting point, not a universal schema. The pack supplier's communication specification defines the messages that exist; the test plan defines which ones are safety-relevant. What matters is that the record can answer “what did the BMS see, what did it command, and when” without going back to the supplier.

## Takeaways

- Channel data records the stimulus; BMS data records the pack's interpretation. Keep both on one time base, and keep the raw frames as well as the decoded signals.

- The CAN box and AUX unit are optional additions with a cost in time resolution: on the BTS-6008A and BTS-6001A, AUX recording is published at 10 Hz versus 100 Hz without it.

- Capture DTCs and freeze frames before clearing anything, and log HVIL and contactor state as signals rather than inferring them from pack voltage.

## Request a configuration review

Tell us the channel count, current range and protocol. We will check the measurement class against the note and return a configuration, not a generic datasheet.

[Request a configuration review](https://batteryxlab.com/contact/?note=bms-communication-in-testing)

## References

- [BTS-6008A-5V300A-H](https://batteryxlab.com/products/bts-6008a-5v300a-h/) — [datasheet (PDF)](https://batteryxlab.com/assets/docs/BTS-6008A-5V300A-H.pdf)

- [BTS-6001A-5V2000A-H](https://batteryxlab.com/products/bts-6001a-5v2000a-h/) — [datasheet (PDF)](https://batteryxlab.com/assets/docs/BTS-6001A-5V2000A-H.pdf)

- [BTS-4008 / 6000A / CA series](https://batteryxlab.com/series/bts-4008/)

- [EV battery testing](https://batteryxlab.com/industries/ev-battery-testing/)

- [BESS testing](https://batteryxlab.com/industries/bess-testing/)

- Peer-reviewed use of this instrument class: [Alkaline-based aqueous sodium-ion batteries for large-scale energy storage](https://doi.org/10.1038/s41467-024-44855-6) — Wu (2024), Nature Communications. DOI: 10.1038/s41467-024-44855-6

- [Module & pack test systems](https://batteryxlab.com/module-pack/)

- [BMS integration in BTS 9.1](https://batteryxlab.com/software/)

- [EV and grid applications](https://batteryxlab.com/applications/)

[← All technical notes](https://batteryxlab.com/notes/)

## Contact

- Email: sales@batteryxlab.com
- Phone: +852 4436 9298
- WhatsApp: https://wa.me/85244369298
- Request a quote: https://batteryxlab.com/contact/
- Full site index for agents: https://batteryxlab.com/llms.txt
