A machine with four wheels is a radio problem you can ignore. A machine with 160 wheels is a radio problem you have to design, and it does not announce itself as one — it announces itself as "the display sometimes shows dashes".
This note is about planning that radio link on multi-axle equipment: why a single receiver runs out of reach, where repeaters belong, how many a machine really needs, and how to tell a planned installation from a guessed one.
Why a single receiver is not enough
Our receivers accept up to 256 sensors, which sounds like capacity solves the problem. It does not. Capacity is about how many sensors can be registered; range is about whether the radio link closes. On a long vehicle, the chassis itself is the obstacle: steel structure, axles, hydraulic runs and containers of cargo sit directly between a sensor and the receiver.
At 433.92 MHz, that geometry matters more than the nominal range figure. A datasheet may say 100 metres in open air. Between the front and rear of a 20-axle carrier there is very little open air.
It is worth separating two numbers that are often confused. Receiver capacity is an addressing limit — how many sensor identities the receiver can track and report. Radio range is a physical limit, and it does not improve when you buy a receiver with more capacity. On a long machine, the second number is the one that decides whether the project works, and it is settled by placement rather than by specification.
What the radio link has to survive
Tire sensors are battery powered, and that single fact shapes the whole design. An internal sensor is expected to last five to six years; a replaceable CR1632 at least two. The radio link therefore has to close on a small amount of energy and a low duty cycle, which is exactly why the answer is not "transmit harder".
What the signal has to pass through is not open air but a machine: steel plate, axle housings, hydraulic runs, containers of cargo. At 433.92 MHz the signal behaves much like light, so anything dense in the path costs signal strength. A repeater is useful precisely because it is wired to the machine at DC 10-30 V and can listen continuously with no battery constraint, and because it is sealed to IP68 and can be mounted where the radio path needs it rather than where the cab supply is. Placing one repeater well is far cheaper than making every sensor work harder for five years.
Repeater placement is a geometry exercise
The approach that works is unglamorous: treat the machine as a series of radio zones, and place a repeater wherever one zone's sensors cannot reach the next zone's receiver or repeater.
- Walk the chassis and mark where steel mass sits between the front and the rear.
- Group wheel positions into zones of roughly equal radio distance.
- Place one repeater per zone boundary, in a position with the least metal in the direction data must travel.
- Verify with a link-quality readout, position by position, before the machine leaves the workshop.
- Record the result, including the zone each repeater serves, so a later repair is a lookup rather than an investigation.
On a 160-wheel beam carrier this produced twelve repeaters feeding two receivers — not one repeater per axle, and not a guess. The layout we use for the multi-axle TPMS range follows the same logic, and the delivered configuration is documented on the beam carrier installation page.
How many repeaters: the real answer is per machine
A rule of thumb such as "one repeater every four axles" is not an engineering answer, and it is wrong in both directions on real machines: it over-specifies a compact 20-axle trailer and under-specifies a long carrier whose cargo changes the radio path every trip.
Repeater count follows three things: where the steel mass sits, how the wheel positions group into radio zones, and where a mounting point exists with an unobstructed path in the direction the data has to travel. That is why a competent supplier quotes a repeater count per machine, after looking at the layout, rather than per product.
Our own reference points are specific. A 160-wheel beam carrier was delivered as two receivers, twelve repeaters and 160 sensors, with ten units in service. The largest single machine we have instrumented carries 224 wheels across 28 axles. Neither configuration came from a ratio, and neither used one repeater per axle.
The failure mode that costs the most
A badly planned installation does not usually fail immediately. It fails as intermittent data: a wheel that reports most days, then does not, usually on a machine parked in a different orientation, loaded differently, or with a container that happens to sit over an antenna.
Because it is intermittent, it is easy to misdiagnose. The sensor is swapped, then the receiver, and the problem returns. Once the vehicle is in service, every return visit is expensive and the customer's confidence is already falling.
That is why we commission with a per-position link check and record it. A written record turns "the system is unreliable" into "position 47 needs a repeater moved 600 mm", which is a repair rather than an argument.
Sensor addressing at scale
At 256 positions, sensor identity becomes a process problem as much as a technical one. Three rules keep it manageable:
- Physical positions are numbered once, in a scheme agreed with the customer, and never renumbered.
- Sensor IDs are bound to positions during commissioning and the binding record is handed over with the machine.
- Replacement procedure is written down, including how to rebind after a sensor swap, so a tire fitter does not have to invent it.
We have seen fleets lose more time to unrecorded sensor swaps than to radio problems.
What this means when specifying a system
The specification that works lists wheel positions and their physical grouping, not just a total. It says who will own the sensor-to-position mapping and in what format it will be delivered. And it asks the supplier to justify repeater count from the chassis layout rather than quoting a rule of thumb.
Two smaller items are worth adding while the specification is open. First, the electrical and sealing details, because a receiver on a mining or port machine lives in a hostile place: our hardware runs from DC 10-30 V, is sealed to IP68 and uses 433.92 MHz license-free radio, with pressure accuracy of ±7 kPa and a CAN data cycle of 10 ms by default. Second, the alarm semantics, because a low-pressure flag that must trigger a derate is a different requirement from one that merely lights a lamp.
Those items are the difference between a system that reports every wheel and a system that mostly reports every wheel.
Send us the axle layout and wheel count and we will propose receiver and repeater counts for that machine. See the CAN bus TPMS integration options or send us the machine details.