Harsh-Driving Thresholds: What the Factory Defaults Actually Mean

Short answer

A factory harsh-driving threshold is a starting value the manufacturer picked so the feature does something out of the box. It is not a legal limit and not an industry standard. The same vendor ships different defaults on different models, and every threshold is configurable across a range twenty times wide.

Key takeaways

• The Teltonika FMB120 Green Driving defaults are 2.5 m/s² acceleration, 2.7 braking and 3.4 cornering, each configurable 0.5–10.
• The FMC650 ships 2.2 and 2.5 for the first two instead, the same vendor, different numbers, so neither set is a standard.
• Wialon states outright that no generally recommended threshold standards exist.
• Events are gated by minimum speed (10 km/h, cornering 30 km/h) and minimum duration (500 ms to accept, over 1 s to record).
• A threshold that is harsh for a loaded tanker is normal driving for an empty light van, so one fleet-wide number scores two vehicle types wrongly.

01What a default value is for

When a manufacturer ships a feature that needs a number, it has to put something in the field. The value chosen has one job: make the feature produce plausible output on an unspecified vehicle, so a new user sees it working rather than seeing nothing. It is not derived from a regulation, it is not the outcome of a study on your fleet, and it is not a threshold anyone has certified as the boundary of unsafe driving. Read that way, a default is closer to a placeholder than to a specification. The evidence for this reading is in the documentation itself. The Teltonika FMB120 parameter list gives Green Driving defaults of 2.5 m/s² for acceleration, 2.7 for braking and 3.4 for cornering, and states that each is configurable anywhere from 0.5 to 10. A designer who believed a specific number was correct would not offer a twentyfold range around it.

02The same vendor disagrees with itself

The clearest single argument against treating a default as authoritative is that the same manufacturer publishes different ones. On the FMC650, the documented defaults are 2.2 m/s² for Max Acceleration (parameter ID 11004) and 2.5 m/s² for Max Braking (ID 11005), with cornering at 3.4 m/s² (ID 11006), again each across a 0.5 to 10 range. Set those beside the FMB120 figures of 2.5 and 2.7 and the picture is unambiguous: one vendor, two product families, two answers to the question of what counts as harsh. If a universally correct threshold existed, Teltonika would ship it on both. What the difference actually reflects is the sensing hardware, the firmware family and the vehicle class each model was designed around, none of which is your fleet.

03The platform vendors say so explicitly

This is not a reading imposed from outside; the software side states it plainly. Wialon's guidance on eco-driving control says that no generally recommended standards exist for these values, and then lists what actually moves the reading: engine power, vehicle condition, weight, body load, tyre grip, road surface and weather. Every item on that list varies within a single fleet and several vary within a single shift. A tanker at 40,000 kilograms and the same tanker empty do not decelerate the same way under the same pedal pressure. Wialon also ranks the data approaches by accuracy, device-calculated highest, raw per-axis average, GPS-only lowest, which means the number you should choose depends on how the event is being detected in the first place, not only on the vehicle.

04The gates around the threshold

A threshold never acts alone, and reading it in isolation overstates how sensitive the feature is. On the FMB920 and FMC920 firmware family, acceleration and braking are only evaluated above 10 km/h and cornering above 30 km/h, so low-speed manoeuvring in a yard is excluded before any comparison happens. Duration gates follow: an excursion must last at least 500 milliseconds to be accepted at all, and more than a second before a record is written. Teltonika also latches the scenario, once active it stays active until the measured value falls back below the configured parameter, which is why one continuous hard brake produces one event and not a burst of them. Change the threshold without knowing these gates exist and you will attribute the resulting change in event count to the wrong cause.

05Why one number cannot serve two vehicle types

Consider what the same 2.5 m/s² braking threshold means across a mixed fleet. A light van, unloaded, on good tarmac, can exceed it in ordinary defensive driving at a roundabout, so its driver accumulates events that describe the vehicle, not their behaviour. A fully loaded tanker cannot reach it easily at all, so a genuinely late, aggressive stop stays under the line and never appears. The same number therefore generates false positives at one end of the fleet and false negatives at the other, simultaneously. Since the scores built from these events feed driver rankings and sometimes pay, the consequence is not merely noisy data. It is a ranking that partly measures which vehicle a driver was assigned to that month, and drivers work that out faster than managers do.

06Calibrate before you compare

There is a mechanical prerequisite that no threshold value can substitute for. Teltonika documents an accelerometer auto-calibration procedure that must run before advanced detection is meaningful: the vehicle stopped on a straight road, then driven above 30 km/h for five seconds, so the device learns which of its axes points forward. Skip it and the device may read a portion of forward acceleration as cornering, or vertical road shock as braking. Wialon's guidance points the same way by noting that the vertical Z axis is excluded from the calculation entirely, with X carrying acceleration and braking and Y carrying turns. Tuning a number on top of an uncalibrated sensor is tuning the wrong layer, and it produces a threshold that only works for the specific mounting error it was fitted to.

07A method for setting your own numbers

Treat the default as a first hypothesis and test it. Group the fleet by vehicle class and typical load, because those are the variables Wialon names as decisive. Run each group at the shipped default for a defined baseline period with the events recorded but not scored, so nobody is penalised by a value you have not yet validated. Then read the distribution rather than the count: if almost every trip in a group produces events, the threshold is below normal operation for that vehicle and needs raising; if a group produces almost none while incidents still occur, it needs lowering. Adjust one parameter at a time, keep a dated log of the value and the reason, and only start scoring once a group's event rate is stable and the events you sample look like real driving faults.

Figures and sources

• Teltonika Green Driving harsh-event thresholds on the FMB120, documented numeric defaults: acceleration 2.5, braking 2.7, cornering 3.4 m/s²; each configurable 0.5–10 (Source: Teltonika Telematics Wiki, FMB120 Parameter list, as of 2026-08-31)
• FMC650 default Max Acceleration threshold (parameter ID 11004), a different number from the same vendor: 2.2 m/s² (range 0.5–10) (Source: Teltonika Wiki, FMC650 Parameter list, as of 2026-08-31)
• FMC650 default Max Braking threshold (parameter ID 11005): 2.5 m/s² (range 0.5–10) (Source: Teltonika Wiki, FMC650 Parameter list, as of 2026-08-31)
• FMC650 default Max Cornering threshold (parameter ID 11006): 3.4 m/s² (range 0.5–10) (Source: Teltonika Wiki, FMC650 Parameter list, as of 2026-08-31)
• Minimum speed at which acceleration and braking events are evaluated: 10 km/h (cornering: 30 km/h) (Source: Teltonika Telematics Wiki, Simple Eco Driving (FMB920/FMC920 firmware family), as of 2026-08-31)
• Minimum duration for a harsh event to be accepted, and to be recorded: 500 ms to accept; >1 s to write a record (Source: Teltonika Telematics Wiki, Simple Eco Driving (FMB920/FMC920 firmware family), as of 2026-08-31)
• Green Driving latching behaviour, device-side hysteresis so one excursion yields one event: scenario stays active until the measured value decreases below the set parameter (Source: Teltonika Telematics Wiki, FMC650 Features settings, as of 2026-08-31)
• Wialon's stated position on published threshold standards, and the variables it names as decisive: no generally recommended standards exist; engine power, vehicle condition, weight, body load, tyre grip, road surface, weather (Source: Wialon Help Center, Eco driving control (recommended criteria values), as of 2026-08-31)
• Accelerometer auto-calibration procedure required before advanced detection runs: stopped on a straight road, then >30 km/h for 5 s (Source: Teltonika Wiki, FMC650 Features settings (Auto calibration), as of 2026-08-31)

Questions

Sources

• Teltonika Telematics Wiki, FMB120 Parameter list, the Green Driving defaults and their ranges
• Teltonika Wiki, FMC650 Parameter list, a different default set from the same vendor
• Wialon Help Center, Eco driving control: no generally recommended standards exist

Related insights

• Driver Behaviour Scoring: What a Harsh-Event Score Can and Cannot Tell You
• False Alarms Are a Two-Layer Problem: The Debounce Settings Already in Your Devices
• Fleet Maintenance: Engine Hours or Odometer, and Why the Standard Tracks Both