Home / Fleet data logger

Fleet data loggerRecord what changed. Do not pay for what did not.

Recording without a data bill

The naive logger streams every frame. On a few hundred machines that is a large monthly volume of values that did not change since the last message — the same pressure, the same temperature, the same state. The platform learns nothing and the SIM bill knows it.

The recording rule

On change: a value crossing a threshold or moving is written and sent.
On a heartbeat: unchanged machines still report "alive and stable" at the interval the customer chooses.
On recovery: whatever was buffered while offline goes out in order.

Stable machines cost almost nothing. Unstable machines generate exactly the messages the customer needs, which is the point of monitoring in the first place.

Positioning as an event, not a track

GPS and BeiDou dual-mode with geofence: entering or leaving a defined area becomes an event in the same record as the bus data. For a fleet that means fewer coordinates and more useful exceptions — the dashboard shows what changed, not a wall of dots.

Site realities

Power: DC 10–30 V from the machine, sleep and wake cycles configured.
Mounting: a retrofit decision — where the unit sits, how the harness runs, who signs off the installation.
Recovery: hardware watchdog, so a locked unit does not need a site visit.
Coverage: 4G CAT.1 with 2G/3G fallback for remote yards.

Sizing conversation

How many machines, how many points per machine, which changes count as events, and whether location is needed everywhere or only on some machines. Those four answers set the configuration; send them and the reply is in writing.

DTU overview · how the data is delivered.

Data sheet, interface list and a payload example on request — contact us (ralph@gz-can.com · +86 130 7692 2278).

Logging is not the same as streaming

A data logger records; a gateway transmits. Most fleets need both, and the difference matters commercially because they fail in different ways. A logger that streams everything bills you for unchanged data and fills a database with noise; a gateway that only forwards live values loses the history you need when something goes wrong at 3 a.m. in a yard with no coverage.

Our unit does the logging locally and uploads on a trigger, so the record is complete and the data bill is not. Change-triggered upload is the single feature that decides whether a logging project is affordable at fleet scale.

What gets recorded

Three questions that decide a logging project

Where it sits next to your TPMS

Fleets that already run tyre-pressure monitoring often want the pressure history in the same place as the rest of the machine data. That is a payload question, not a second platform: the tyre data and the CAN values can leave the machine together, in one record, if both sources are specified at once.

Questions about logging

Logger or gateway - do we need both?

Usually yes, in one unit. The logger keeps the detail; the gateway moves the summary and the events. Specifying only one of the two is how projects end up with either an unreachable history or a data bill for values that never changed.

How long can it record without coverage?

That is a buffer-size decision set by how long your longest gap is likely to be and how much detail you need to keep. Tell us the machine's typical operating pattern and the interval you want and it becomes a specification item rather than a guess.

Does logging tyre pressure count as a separate system?

No. Pressure and temperature are values like any other - if they are specified at the same time as the CAN values, they leave the machine in the same record instead of a second platform.

Question	Why it changes the design
How long must history be kept?	Determines buffer size and whether the unit uploads everything or uploads summaries and keeps detail locally.
Who reads the history?	If it is a maintenance engineer chasing one fault, you want long local detail. If it is a dashboard, you want aggregates and fewer uploads.
What happens with no coverage?	Buffering is not optional on mobile machines. A unit that drops data when coverage drops is a unit that will be blamed for missing data.