Skip to content

Measurement guide · for test engineers · 4 min read

Distance and speed from wheel encoders

One encoder per wheel of the non-steered axle, averaged, and a stator restraint without play.

  • wheel speed
  • distance
  • encoders
  • vehicle dynamics

The essential

A wheel pulse transducer on the lug nuts gives 50 to 5 000 pulses per revolution, in quadrature with an index. Multiplied by the rolling circumference, the pulses give distance, and their rate gives speed. In a case study on a course of about 1.6 km, one encoder gave a distance error of up to 0.22 %; the average of two encoders, one on each wheel of the non-steered axle, gave 0.0115 %. The remaining error comes from the turns, in which each wheel follows its own path, and from a stator restraint that moves.

01

How accurate is the distance from one encoder?

In a case study on a course of about 1.6 km, the distance from one encoder had an error of up to 0.22 %. In a turn, each wheel follows its own path: one encoder measures the path of its wheel, not that of the vehicle, and the error builds up over a series of turns.

02

Why average two encoders?

Because one encoder measures the path of its own wheel, not the path of the vehicle: in a turn, the outer wheel travels farther and faster than the inner one. The average of one encoder on each wheel of the non-steered axle, left and right, gives the distance and speed at the center of the vehicle.

  • Same course, with a roundabout and a zigzag: two encoders averaged gave an error of 0.0115 %, against up to 0.22 % for one.
  • The non-steered axle keeps the geometry simple; steered wheels add their turning motion, which then has to be compensated in the data.
03

What goes wrong at low speed?

A loose stator restraint adds pulses that the wheel did not make. The stator is the stationary part of the encoder; a rod holds it to the vehicle body. If the rod can move in its holder, the encoder reads that movement as wheel rotation. It matters at near-zero speed and through a change of direction.

  • In a case study, a magnetic restraint without a centering insert gave a slight oscillation on stopping and a backward pulse when the brake was released. With a restraint that keeps the rod centered, neither the data nor the video showed an oscillation.
  • The stator restraint guide shows the two errors that come from how the stator is held, and what removes each.
04

How are the pulses recorded?

As pulses, by any counter input. For recorders that cannot read pulses, the EC-LV conditioner converts them in line to two analog voltages: angular velocity, with ± 10 V DC at full-scale velocity, and angular position, 0 to 10 V DC for 0 to 360°. It is powered from the encoder cable, and the encoder pulses pass through.

To remember

  • 50 to 5 000 pulses per revolution, on the lug nuts of the existing wheel.
  • One encoder: up to 0.22 % on distance. Two encoders averaged: 0.0115 %.
  • One per wheel of the non-steered axle, left and right.
  • A stator restraint without play, above all at low speed and through reversals.
  • Pulses, or two voltages through the EC-LV: speed and angle.

A distance or a speed to measure at the wheel?

Describe the vehicle, the speed range and the recorder. An engineer answers with the encoder, the conditioner and the mounting.

More details to give? Use the full contact form.

Contact us

An application engineer replies within 24 hours on working days. Use the full contact form.