The problem the configuration solves
Trackers rarely report a real charging status. In production data, most devices reported an unknown status despite battery levels of 95 to 100%. That is why the system cannot rely on this field and infers charging on the server side, based on the trend of the battery level.
The inference rule is simple: when the battery rises, the device is charging; when it does not change, the state stays as before; when it falls, the device has been removed from the charger. The condition, however, is that the tracker is located at a multi-charger location, and that is decided by the map configuration.
Stage 1: inventory the charging points
Before configuration, list all the places where devices are charged: racks in the tool store, cabinets in the foremen's office, stations at the gate. For each, record the number of sockets and the responsible person. This list is the basis for the later location names.
Pay attention to informal places. Devices are often charged wherever it is convenient, for example in the break room. If such a place is not marked in the system, the inference will not work for devices put down there.
Test of three battery scenarios at a charger (example)
- Reading 1
- Reading 2
Stage 2: name the charger locations unambiguously
The location name should clearly indicate that it is a multi-charger, and be consistent in all places. Example convention: the word charger plus the hall name and the station number. A fixed pattern makes the server rule easier and helps the administrator read the pool view.
Avoid names that could denote both a charger and another place, for example just the name of a room. A name without a word indicating charging means the server will not recognize the location, and the pool will not show the marker.
- A common naming pattern for all chargers.
- The name contains a word indicating charging, plus the hall and station designation.
- No ambiguous names that match other places.
Stage 3: assign a transmitter to the charger location
A tracker is recognized at the charger on the basis of the transmitter it hears. Make sure that each charger has a transmitter assigned to its location, and that its range does not cover other places where devices are not charged. In a building, one transmitter per 100 to 200 m2 is usually assumed, so in tight rooms it is worth checking the separation of locations in practice.
After the assignment, test it on a sample: place a device on the charger and observe whether, after a few heartbeat cycles, the server indicates the correct location.
Stage 4: check that the marker appears in the pool
The calculated charging state appears in the tracker pool view in the delivery form as a marker, without an extra database query in the heartbeat path. After configuration, check the view: devices on a charger should carry the marker, and devices put down elsewhere should not.
Test three scenarios: a device on a charger with a rising battery, a device removed from the charger, and a device with an unchanged battery. Each should give a result consistent with the rule.
Rollout pitfalls
The first pitfall is a location name that does not match the recognition rule: the inference then does not work, and the view shows no error, only a missing marker. The second is a charger in an area where the transmitter is too far away, so the server assigns the device to another location. The third is expecting the marker to appear after the first reading: the inference needs a comparison of at least two consecutive battery readings.
The fourth pitfall concerns team communication: the people issuing devices must know that the marker is an inference from the trend, not a reading from the device, so that they do not treat it as a guarantee of charging.
- The location name does not match the charger recognition rule.
- The transmitter is too far from the charger, or its range covers other places.
- Expecting a result after a single battery reading.
- No information for the team that the marker is an inference from the trend.
Video
Frequently asked questions
Why does the name of the charger location matter?
The server infers charging only for a device in a multi-charger location. A badly named location means no marker, with no error message.
How many readings does the server need to recognize charging?
At least two consecutive battery readings, because the inference is based on comparing the trend: rising, unchanged or falling.
Does the marker in the pool guarantee that the battery is full?
No. The marker means an inference from the trend, and the charge level should be read from the battery value.
Are you configuring multi-chargers?
We will help you plan location names and transmitter placement at the chargers, and then show a test of the three scenarios.
Let's talk