MF4 troubleshooting
A relevant vehicle event is buried in a long or numerous MF4 recordings, making manual inspection slow and uncertain. This guide helps establish whether the limitation is signal selection, decoding, timestamps, recording coverage, or measurement volume.
Start the diagnostic workflowObservable symptoms
The MF4 measurement opens and contains data, but the relevant behavior is not obvious in plots or signal views. This differs from an unreadable measurement, where the file or its metadata cannot be loaded.
The event appears irregularly or at different positions in multiple MF4 measurements, so searching by elapsed recording time does not provide a reliable match.
Signals selected from the measurement change near the reported behavior, but their timing or values do not consistently correspond to the event. This differs from a confirmed signal correlation supported by repeated recordings.
Long or numerous MF4 recordings make repeated navigation, plotting, and comparison slow enough that potentially relevant sections are skipped.
Possible causes
The observed vehicle behavior may be represented by a signal that was not included in the initial inspection. Without a useful signal correlation, the event remains difficult to distinguish from normal measurement activity.
If the DBC defines different messages, signals, scaling, or metadata from those represented in the measurement, decoded values can appear plausible while failing to describe the reported event.
The MF4 measurement may begin after the relevant behavior, end before it, or omit the signals needed to identify it. A problem description alone cannot establish that the event was recorded.
Signals from the same measurement may not be compared at the correct points in time, making a real relationship appear absent or shifting the apparent event location.
Large measurement files or numerous recordings increase the cost of repeated plotting and comparison, allowing a relevant interval to be overlooked even when the required data is present.
Ordered investigation
Use the problem description to record the reported behavior, reproduction context, expected behavior, and any known timing or signal clues. This separates an evidence gap from a failure to describe the event precisely.
Open representative MF4/MDF measurements and inspect their metadata, timestamp ranges, and available signals. This rules out missing coverage or an unsuitable recording before deeper signal analysis.
Compare candidate CAN signals with the supplied DBC and verify that the selected messages, signal names, scaling, and metadata correspond to the recorded data. This narrows the search when decoded values are misleading.
Plot and compare candidate signals around intervals associated with the reported behavior, using repeated measurements where available. Consistent timing relationships support a signal correlation; inconsistent relationships rule out those candidates but do not identify the root cause.
When manual navigation is too slow, parse the MF4/MDF measurements and apply the same signal-selection and comparison logic across recordings. This reduces search effort and produces comparable evidence for anomaly investigation.
If no candidate interval can be supported, record which signals, time ranges, and recordings were checked and compare that result with the problem description. This identifies whether additional measurement data or domain information is required.
Verification gates
Load each representative MF4/MDF measurement and read its metadata, timestamp range, and available signals.
The measurement loads, exposes usable metadata, and contains timestamped signals covering the relevant recording period.
A load or coverage deviation points to an incomplete, unsuitable, or otherwise unusable measurement; inspect the available MF4/MDF data before searching for the event.
Compare the DBC message and signal definitions with the candidate data selected from the MF4 measurement, including signal identity and scaling.
The DBC definitions correspond to the recorded data and produce values that are consistent with the problem description.
A mismatch can make the event appear absent or incorrectly located; obtain the applicable DBC or reassess the candidate signals.
Plot candidate signals around the intervals identified from the problem description and compare their timing with the observed behavior.
At least one candidate signal shows a repeatable, time-consistent relationship in measurements where the behavior is present.
No relationship rules out those candidates as sufficient evidence; inspect other available signals, recording coverage, or the problem description.
Apply the same signal selection and comparison to more than one MF4 measurement containing comparable context.
The candidate relationship remains interpretable despite different recording lengths or event positions.
An inconsistent result may indicate intermittent vehicle behavior, insufficient context, differing signal definitions, or missing evidence.
Run the repeatable Python analysis over the selected MF4/MDF measurements and compare the resulting candidate intervals with manual plots.
The workflow returns stable candidate intervals that can be reviewed in the measurement viewer or analysis output.
Differences between automated and manual results require inspection of parsing, signal selection, timestamp handling, or DBC interpretation.
From evidence to action
When the event is present but represented by different signals or decoded incorrectly, update the selected signal set and use the matching DBC definitions. Re-run the comparison across representative MF4 measurements.
When the required evidence is absent from the recordings, request measurements that cover the reported behavior and include the signals needed for signal correlation. The problem description should accompany the additional data.
When file size or recording volume is the limiting factor, implement a Python analysis script that parses the MF4/MDF measurements, evaluates the selected signals, and records candidate intervals for review.
When the available MF4 data, DBC, and problem description remain inconsistent, document the checked evidence and involve the measurement owner, supplier, or OEM team responsible for the recording and signal definitions.
FAQ
Diagnostic support
If MF4 event searches remain difficult to reproduce, a focused engineering analysis or Python analysis script can turn the investigation into a repeatable workflow.