Skip to content

Fleet Maintenance: Engine Hours or Odometer, and Why the Standard Tracks Both

By Saeed Baatiah · Published:

Short answer

Equipment that works while stationary, a generator, a crane, a high-idle truck, consumes its operating life in engine hours, not kilometres. A highway truck does the opposite. SAE J1939 defines a service countdown by both distance and hours because they measure genuinely different kinds of wear, and no single counter fits every asset.

Key takeaways

  • SAE J1939 defines the service-interval countdown twice, once by distance, once by hours, inside the same message.
  • The FMS-Standard used by seven European truck OEMs implements the distance countdown and leaves the hours one unused, by design choice, not oversight.
  • Equipment with high idle time accrues engine hours without matching odometer distance, the two counters diverge for exactly that reason.
  • Generator and equipment manufacturers already schedule service in operating hours, on a "whichever comes first" rule against calendar time.
  • There is no single rule for a fleet, the asset's duty pattern decides which counter should drive its maintenance schedule.

01Two counters, two different kinds of wear

An odometer counts distance, which is a good proxy for wear on components that only wear while the vehicle moves, tyres, brake pads, suspension bushings, wheel bearings. An engine-hours counter counts time the engine has run, which is a better proxy for wear that accumulates whether the vehicle moves or not, oil breakdown, filter loading, bearing wear inside the engine itself, and the cumulative heat cycling that ages seals and hoses. The two counters measure different physical processes, and a vehicle that spends a large share of its engine-running time stationary will wear its engine oil and internals on a schedule the odometer cannot see at all. Scheduling maintenance on the wrong counter does not just risk over- or under-servicing; it risks servicing the wrong component category entirely.

02What the odometer measures, and what it misses

SAE J1939 defines Total Vehicle Distance as SPN 245, a four-byte parameter with a resolution of 0.125 km per bit, and a High Resolution variant as SPN 917 at 5 metres per bit for applications needing finer precision, both transmitted in dedicated PGNs. Both are genuinely useful for anything correlated with road contact and mileage accumulation. What they structurally cannot capture is engine-running time with no distance attached: a generator truck running a power take-off at a job site, a refrigerated unit idling for hours at a loading dock, or a crane truck stationary with the engine powering hydraulics. In every one of those cases the odometer records zero wear while the engine accrues real operating hours, and a service schedule keyed only to distance will silently under-service exactly these vehicles.

03What engine hours measure

SAE J1939-71 defines SPN 247, Engine Total Hours of Operation, as a four-byte parameter with a gain of 0.05 hours per bit, transmitted in PGN 65253, and adopted as mandatory in all factory-fitted FMS gateways since 1 October 2009 under the joint FMS-Standard specification signed by Daimler, Scania, Iveco, MAN, Volvo Trucks, DAF and Renault Trucks, a strong signal that heavy-vehicle manufacturers regard this counter as essential regardless of how a truck is used. A separate parameter, SPN 246, Total Vehicle Hours, tracks accumulated operating time of the vehicle as distinct from the engine specifically. Because engine hours accumulate purely from running time, they capture idling, power-take-off operation and any other stationary engine use that an odometer cannot register at all.

04The idling gap between them

The arithmetic that makes the two counters diverge is idling. A truck that idles heavily accrues engine hours at a rate an odometer never sees, because idling by definition adds zero distance. US DOE and Argonne National Laboratory data on long-haul trucking states this relationship directly: a long-haul truck idles roughly 1,800 hours per year, burning around 1,500 gallons of diesel doing so, and the same source notes that where manufacturer warranties and maintenance intervals apply to hours operated rather than miles travelled, the cost of idling is greater than fuel alone, because that idling time is quietly consuming service-interval budget an odometer-only schedule would never charge against the vehicle. A high-idle duty pattern is precisely the case where tracking only distance produces a maintenance schedule that runs behind the engine's real wear.

05Which assets should be scheduled on hours

Generator and heavy-equipment manufacturers have already answered this for their own products, and their schedules are informative for any high-idle fleet vehicle. Cummins's Onan generator operator manuals schedule service items at 8, 50, 100, 250 and 500 operating hours, explicitly stating the interval is "at the time period indicated or after the number of operating hours indicated, whichever comes first", and shorten the oil-change interval to every 50 hours in dusty or dirty conditions, relevant to Saudi site work. Caterpillar generator sets carry a dedicated hours-run meter whose stated purpose is assisting maintenance, with service scheduled every twelve months or 500 hours, and battery checks tightened to every 200 hours in high-ambient conditions. Any fleet asset that resembles these duty patterns, construction equipment, cranes, refrigerated trucks with long dock time, standby generators on wheels, should follow the same logic and be scheduled on hours, not distance.

06Which assets should stay on distance

The reverse case is just as real. A line-haul truck that spends nearly all of its engine-running time actually moving accrues engine hours and distance in a roughly fixed ratio, which means an hours-based schedule and a distance-based one converge to similar service points, and distance is usually the easier, more auditable figure to report to a driver or a workshop, since it is visible on the dashboard and does not require CAN access to confirm. The FMS-Standard interface used across seven major European truck brands reflects this priority directly: its Service Information message, PGN 65216, implements the distance-based service countdown, SPN 914, in active use, while its data bytes carrying the hours-based countdown, SPN 916, are explicitly marked "Not used for FMS-Standard" in the interface specification. That is the standard itself choosing distance as the practical default for road-haul trucking, while still reserving the hours parameter in the protocol for the vehicles that need it.

07Why fleets get this wrong

The owner's view is direct on this point, and it explains the recurring mistake: there is no single rule that applies across an entire fleet, and fleets that adopt one anyway are the ones that get it wrong. A fleet operations team often standardises on distance because it is simpler to report and easier for drivers to self-check, then applies that same rule to a generator truck or a crane vehicle that spends much of its working life stationary with the engine running. The equipment then goes into service late, wear accumulates past the interval the manufacturer actually intended, and the eventual failure looks unpredictable when it was, in fact, exactly what the mismatched counter should have predicted. The fix is not choosing the "right" counter once, it is auditing each asset's duty pattern and assigning the counter that matches it.

08Where the counter comes from matters too

How a telematics device produces its odometer figure affects whether it can be trusted for scheduling at all. Teltonika devices support both a CAN-derived odometer, read directly from the vehicle bus where available, and a virtual odometer computed second by second from GNSS positions when no CAN source is present, with the calculation source explicitly selectable and a starting total that can be seeded to match the vehicle's existing dashboard reading. On trucks equipped with the FMS interface, engine hours can similarly be read directly from CAN rather than derived. This matters because the two calculation methods are not interchangeable in accuracy, and a maintenance schedule built on a virtual, GNSS-derived odometer for a vehicle that does have CAN access is using a weaker source than the one already available.

09Counter resets, replacements and continuity

A maintenance schedule is only as good as the continuity of the counter it reads, and both counters can lose that continuity in ordinary fleet operations. An engine replacement resets SPN 247 to whatever the new engine control unit reports, disconnecting it from the vehicle's prior service history unless the fleet records the transition explicitly. A telematics device swap can reset a GNSS-derived virtual odometer to zero unless its starting value is deliberately seeded from the previous reading, which Teltonika's configuration explicitly supports for exactly this reason. Wialon's own service-interval implementation tracks mileage, engine hours and calendar time as three genuinely independent terms specifically so that a gap in one does not silently corrupt a schedule that depends on all three, a design choice worth matching in any fleet's own record-keeping.

10Building a dual-trigger schedule

The practical answer, matching what generator manufacturers and Wialon already implement, is a dual-trigger schedule per asset: set both a distance interval and an hours interval, and service whichever threshold the asset crosses first. For a highway truck the distance trigger will almost always fire first and the hours trigger sits quietly as a backstop for an unusually idle month. For a generator truck or crane the hours trigger will fire first every time, and the distance trigger becomes close to irrelevant. The value of setting both is that the schedule self-corrects if an asset's duty pattern changes, a highway truck reassigned to yard work will start tripping its hours trigger even though its distance stays low, catching the shift in wear pattern without anyone having to notice the reassignment and manually change the schedule.

11What neither counter tells you

Both counters are proxies, and neither captures load, terrain or operator behaviour, all of which affect real wear independent of hours or distance. An engine running at high load on a steep grade for one hour ages differently from an hour spent idling in a yard, even though both counters record identical values. A truck running Riyadh's summer heat accumulates thermal cycling stress an hours counter cannot distinguish from a mild-climate hour, and a driver who habitually over-revs before the engine has warmed produces wear no counter registers at all, the counter climbs at the same rate whether the engine was given thirty seconds to circulate oil first or was driven hard from a cold start. This is not a reason to distrust hours or distance as scheduling triggers, they remain the best broadly available proxies, but it is a reason to treat a schedule built on either as a floor, not a ceiling, and to bring in oil analysis or condition-based indicators for assets where load and terrain vary widely.

12Auditing the schedule against actual failures

A dual-trigger schedule is a starting hypothesis, not a permanent answer, and the way to correct it is to compare which counter actually predicted each unplanned failure over a review period, typically a year. If a category of asset consistently fails between its distance-triggered services, the hours interval for that category is probably too loose. If a generator-type vehicle consistently over-services on the hours trigger with no matching wear found, the interval can likely be relaxed. This audit only works if both counters are actually recorded reliably per asset, which returns to the earlier point about data continuity: a schedule cannot be corrected by evidence that was never captured, so the counter-reliability work described earlier is a precondition for this step, not a separate concern.

13How ENTEK.AI positions the two counters

On ENTEK.AI, the Construction solution shows real running hours for each machine, taken from the ignition signal, which is the counter that matters for equipment that works while standing still. The Fleet solution does not offer a maintenance schedule built on these counters today, so the dual-trigger approach in this article is a method to apply in whatever maintenance system you use. We do not have a published figure for how much unplanned downtime a dual-trigger schedule prevents across our own deployments, and we would rather state that plainly than present an estimate as a measurement. What we can state is the method: the counter a maintenance trigger should use is decided by the asset's duty pattern, not by which number is easier to report, and a fleet mixing highway trucks with stationary-running equipment needs both counters tracked, not one chosen for the whole fleet.

Which counter should drive the schedule, by asset type
Asset typePrimary counterWhy
Highway line-haul truckDistance (odometer)Engine hours and distance accrue in a roughly fixed ratio
Generator truck / mobile power unitEngine hoursRuns stationary; odometer sees none of the wear
Crane / construction equipmentEngine hoursEngine powers hydraulics while stationary for long periods
Refrigerated truck with long dock timeEngine hours (dual with distance)High idle for cooling; distance under-represents wear
Mixed-duty delivery truckBoth, dual-triggerDuty pattern varies by route and season

Figures and sources

Questions

Why does SAE J1939 track both engine hours and distance for maintenance?

Because the standard defines the service-interval countdown twice, SPN 914 by distance and SPN 916 by hours, in the same PGN 65216 message, treating them as genuinely different measures of wear. The FMS-Standard used by seven European OEMs implements the distance one for road trucks and marks the hours one unused for that use case, by design.

How do I know which counter to use for a specific vehicle?

Look at how much of its engine-running time is spent stationary. A highway truck should be scheduled on distance. A generator truck, crane or vehicle with long idle or power-take-off time should be scheduled on engine hours. Mixed-duty vehicles need both, with service triggered by whichever threshold is crossed first.

Does idling really matter enough to change a maintenance schedule?

Yes, at scale. US data shows a long-haul truck can idle around 1,800 hours a year, and DOE notes that where warranties and maintenance intervals are set in hours rather than miles, idling cost is greater than fuel alone, because it consumes service-interval budget an odometer-only schedule never charges against the vehicle.

What happens to the engine-hours counter after an engine replacement?

It resets to whatever the new engine control unit reports, disconnecting from the vehicle's prior service history unless the fleet records the transition explicitly. The same continuity risk applies to a GNSS-derived virtual odometer after a device swap, unless its starting value is deliberately seeded from the previous reading.

Sources

Related insights