This Annex describes the certification procedure regarding the friction losses of wheel ends for non-driven axle applications. The certification of wheel ends on driven axles is included in the procedure laid down in Annex VII.
Alternatively to the certification of wheel ends, the standard friction losses of wheel ends as set out in point 6 can be applied for the purpose of the determination of vehicle specific CO2 emissions.
For the purposes of this Annex the following definitions shall apply:
‘wheel bearing’ means the bearings that are used to support one wheel end in a vehicle.
‘wheel end’ means the assembly of components that establish the connection between the wheel and the axle, which includes the wheel bearings, seals and lubricants as well as the wheel hub, if available, and all other components relevant to the rotational friction, and may exclude the brake disc and wheel flange.
‘radial load’ means the load applied to the wheel end perpendicular and vertically to the shaft axis.
‘axial load’ means the load applied to the wheel end in the direction of the shaft axis considering the dynamic wheel radius.
‘load line position’ means the position on the wheel end through which the radial load is applied.
‘wheel end manufacturer’ means the legal entity that produces the wheel end.
‘wheel end family’ means a manufacturer's grouping of wheel ends which through their design, as set out in point 2.3, have similar design characteristics and CO2 and fuel consumption properties.
‘customer’ means the legal entity that sells the vehicle or axle in which the wheel end is installed.
‘testing entity’ means the legal entity responsible for testing the wheel end, either the wheel end manufacturer or a third-party.
‘seal’ means the part of the wheel bearing designed to prevent the intrusion of particles or liquids in the wheel bearing, or to prevent lubricant leakage.
‘clearance’ means the total distance through which one bearing ring can be moved relative to the other in the axial direction.
‘preload’ means the negative operating clearance in the wheel bearing.
‘inner ring’ means the ring or rings of the wheel bearings with smaller diameter than the outer ring.
‘outer ring’ means the ring or rings of the wheel bearings with greater diameter than the inner ring.
‘measurement’ means the measurement of friction losses in the wheel end expressed as a friction torque in Nm.
‘bearing rated load’ means the maximum design load as defined in the wheel bearing specifications.
‘pitch diameter’ means the distance in a wheel bearing between the geometrical centre of two rolling elements when the two rolling elements are diametrically opposed.
‘run-in procedure’ means the procedure of conditioning an unused wheel end under load in order to bring it to a state of representative in-use conditions.
The wheel ends used for the verification of friction loss measurements shall be new.
They shall be the same wheel ends as defined by specifications, as intended for series production, and as will be installed in the customer’s applications.
These specifications include, but are not limited to, the dimensions, the materials, the surfaces’ quality and treatments, the numbers of rollers, the seal, the lubricant’s type, quality, and quantity as well as any other characteristic relevant for the friction of the wheel end.
For the purpose of the CO2 certification of a wheel end family, at least four different wheel ends from the family parent shall be tested according to the procedures described in points 3. and 4. using for each the same speed and load target steps.
The following criteria shall be the same to all members of a wheel end family:
rolling elements’ quantity;
rolling elements’ diameter within ± 0,5mm (when measured perpendicular and at the centre of the long axis);
rolling elements’ length within ± 1mm (when measured along the long axis);
pitch diameter within ±1mm;
number of rows;
outer ring contact angle with the rolling elements with ± 1deg;
the lubricant type: oil or grease;
load-line position (in the case the family parent is not tested at the indicated position in Figure 2).
The family parent of a wheel end family shall be the member with the highest friction.
If a family has more than one member, the testing entity shall justify the choice of the family parent based on the component properties.
The bearing rated load for the family shall be the highest bearing rated load of all family members.
For each family member, the testing entity shall provide quantifiable data on:
the seals performance (e.g. friction losses);
the lubrication (oil or grease) performance (e.g. viscosity);
the preload / clearance range (e.g. maximum and minimum).
The approval authority may request the testing entity to provide additional justification, including by means of simulations or calculations, when it considers that the properties listed in the fourth paragraph are sufficient to justify the choice of the family.
The testing entity shall apply a run-in procedure on the wheel ends.
The run-in procedure shall use the same test set up and have the same requirements as for the measurements of friction losses.
The run-in procedure shall comprise of four successive phases.
During the first phase, the wheel end shall be run clockwise at a constant speed of 300 rpm with a radial load applied corresponding to 50 % of the bearing rated load for a duration of 60 ±2 minutes.
During the second phase, the wheel end shall be run counterclockwise at a constant speed of 300 rpm with a radial load applied corresponding to 50 % of the bearing rated load for a duration of 60 ±2 minutes.
During the third phase, the wheel end shall be run clockwise at a constant speed of 500 rpm with a radial load applied corresponding to 100 % of the bearing rated load for a duration of 660 ±2 minutes.
During the fourth phase, the wheel end shall be run counterclockwise at a constant speed of 500 rpm with a radial load applied corresponding to 100 % of the bearing rated load for a duration of 660 ±2 minutes.
The run-in procedure shall be documented by the testing entity with regard to run-time, speed, radial load, and bearing temperature, and reported to the approval authority.
The lubricant type, quality and quantity shall be the same as defined by specifications, as intended for series production, and as will be in customer’s applications.
If the wheel end manufacturer is not delivering lubricant with the wheel bearing, the customer shall provide the necessary information on the lubricant that will be used in the final application to allow accurate testing of the wheel end.
If the lubricant is of the oil type, the oil level within the bearing shall be as defined in the axle specifications. In the absence of a specification the maximum geometrically possible oil level of the axle shall be applied.
If the bearing operating clearance/preload can be adjusted, the clearance/preload used for testing the wheel bearing shall be set at the arithmetic mean of the clearance/preload range defined in the specifications, within a tolerance of ±20 μm.
The seals used for testing the wheel end shall be the same as defined by the specifications, as intended for series production, and as will be installed in the customer’s applications.
If the wheel end manufacturer is not delivering seals with the wheel end, the customer shall provide the necessary information on the seals that will be used in the final application to allow accurate testing of the wheel end.
The temperature in the test cell shall be maintained at 25 °C ± 10 °C. The ambient temperature shall be measured within a distance of 1 metre to the wheel bearing’s outer ring and documented in the test report. It shall be a target temperature for the testing entity, of which systematic deviations across tests are not allowed.
The wheel bearing temperature shall be measured on the bore-side of the inner ring located on the inner-side of the vehicle. During measurements, the wheel bearing temperature shall be kept at a maximum of 60 °C. For that purpose, air cooling may be applied in accordance with section 3.3.5.
The test set-up shall be as illustrated in Figure 1.
Torque measuring devices shall be installed in order to measure friction losses in the wheel end, and in such a way that parasitic effects are minimized.
A speed measuring device shall be installed to measure the rotational speed of the wheel end.
A temperature measuring device shall be installed to measure the temperature of the bore-side of the inner ring on the inner-side of the vehicle.
A load measuring device shall be installed to measure the radial load applied on the wheel end.
The test set-up shall consist of an electric machine used to apply a rotational speed to the wheel end, and of a device capable of applying a radial load onto the wheel end.
The wheel end shall be installed such that the outer ring of the wheel bearing is rotating and used for speed input, while the inner ring is not rotating.
Gearings and couplings are allowed between the electric machine and the wheel end, provided that they do not influence the results of the measurements.
The calibration laboratory facilities shall comply with the requirements of either IATF 16949, ISO 9000 series or ISO/IEC 17025. All laboratory reference measurement equipment, used for calibration and/or verification, shall be traceable to national (international) standards.
The measurement accuracies set in points 3.2.3.1 to 3.2.3.4 shall concern the complete measurement chain, including sensors and additional sources of inaccuracy. The specified tolerances for uncertainty shall not be used for systematic deviations when measurement instruments are applied with higher accuracy.
The uncertainty of the torque measurement for the measurement of the wheel end’s friction torque shall not exceed ± 0,2 Nm.
In the case of a higher uncertainty, the measurements shall be calculated as set out in point 3.4.6.
The uncertainty of the load measurement for the measurement of radial load applied on the wheel end shall not exceed ± 1 kN.
If the radial load is applied as a mass, this shall be converted by applying the gravitational constant of 9,81 N/kg.
The uncertainty of the rotational speed measurement for the measurement of the wheel end speed shall not exceed ± 2,5 rpm.
The uncertainty of the temperature measurement for the measurement of the ambient temperature shall not exceed ± 2 °C.
The uncertainty of the temperature measurement for the measurement of wheel bearing temperature shall not exceed ± 2 °C.
The following signals shall be recorded for the purpose of the calculation of the friction torque losses:
Input rotational speed [rpm]
Wheel end friction torque [Nm]
Applied radial load [kN]
Bearing temperature [°C]
Ambient temperature [°C]
The following minimum sampling frequencies of the sensors shall be applied:
Friction torque: 300 Hz
Rotational speed: 100 Hz
Temperatures: 10 Hz
Load: 10 Hz
The raw data of the friction torque shall be filtered by a suitable low-pass filter such as a Butterworth 2nd order filter with a cut-off frequency of 0,1 Hz. Filtering of the other signals may be applied in agreement with the approval authority. Any aliasing effect shall be avoided.
The raw data shall not be reported.
To determine the torque loss map for a wheel end, the grid points of the friction torque loss map data shall be measured as specified in point 3.4.
The measurement of a grid point may only be repeated if there is a technical justified reason to do so such as the failure of a measurement sensor. This repetition shall be recorded in the test report. The total testing of one wheel end sample, from initiating the run-in until concluding the last grid point, shall be concluded within a maximum of 55 hours, otherwise the test of the sample will be void.
The friction loss map shall be measured with radial loads corresponding to 25 %, 50 %, and 100 % of the bearing rated load.
The target loads shall be reported by the testing entity together with the actual measured load.
The radial load shall be applied onto the wheel end at its centre, so that the load line position is at the centre of the wheel bearing within ± 0,5 mm. The centre of the wheel bearing is determined as the middle of the outside positions of the inner WB rings (see Figure 2).
At the request of the manufacturer and with the approval of the approval authority the load line position may be chosen outside the centre of the bearing. In this case the manufacturer has to provide evidence that this load line position corresponds to the application of the wheel end.
For the purpose of these the measurements set out under this point, no axial load shall be applied onto the wheel ends.
The wheel end shall be tested at 250 and 500 rpm. All rotational speed points shall be measured in a clockwise and counterclockwise direction in accordance with the testing sequence specified in point 3.4.1. The results may be reported as the average measured values of the clockwise and counterclockwise direction.
The wheel end may be air cooled by a fan using ambient air at ambient temperature as defined in point 3.1.1. Other external cooling or heating shall not be allowed. In the case that air cooling is used, the same cooling condition shall be applied for all tested wheel ends at all grid points.
The testing sequence to be applied depends on the measurement configuration of the test set-up.
In the case that the measurement configuration is such that the radial load and the friction torque are both determined individually by dedicated torque measurement device, the wheel end testing shall follow Testing sequence A as described in point 3.4.1.1.
In the case that the measurement configuration is such that the radial load and the friction torque are determined simultaneously by the same torque measurement device, the wheel end testing shall follow Testing sequence B as described in point 3.4.1.2.
If, based upon the functional descriptions referred to in the second and third paragraphs, the testing entity cannot judge which test sequence shall be used, Testing sequence A shall be applied.
The friction measurements of the grid points shall start at the highest radial load downwards to the lowest radial load, while at each load step first the highest and then the lowest rotational speed shall be tested. Once the grid point at the lowest load and lowest rotational speed has been measured, the rotational direction on the wheel end is reversed and the previously described sequence is repeated.
The testing sequence is shown schematically in Figure 3.
The friction measurements of the grid points shall start at the highest radial load and the highest rotational speed. Then the rotational direction is reversed and then the same load/speed point is measured. While keeping the same load, the rotational direction is again reversed and the friction is measured at the lower rotational speed. This load/speed point is also measured in both rotational directions. The previously described sequence is repeated for the 50 % and 25 % radial load settings.
The testing sequence is shown schematically in Figure 4.
For each grid point the testing entity shall allow for a stabilisation period of 117±2 minutes before starting the measurement. In addition, the following stabilisation periods shall be applied:
For Testing sequence A:
Before the first grid point and before the seventh grid point (after the rotational direction has been reversed) the stabilisation period shall be extended by an additional 60±2 minutes. The stabilisation times are indicated in Figure 3.
For Testing sequence B:
Before the first grid point the stabilisation period shall be extended by an additional 60±2 minutes. Before the fifth and the nineth grid point the stabilisation period shall be extended by an additional 30±2 minutes. The stabilisation times are indicated in Figure 4.
The friction for each single grid point shall be measured during the last 180 seconds of the corresponding constant speed phase. In the case that the stabilisation criterion as described in section 3.4.3 is not fulfilled during the last 180 seconds of the grid point, the measurement may be taken from the first earlier uninterrupted segment of 180 seconds where the stabilisation criterion was fulfilled.
In the case that the test set-up is equipped with a support of the wheel end by means of a support bearing, which is required to be rotated in both directions during the measurement of each grid point, the friction shall be measured during the last 180 seconds of the clockwise rotation of the support bearing and during the last 180 seconds of the counterclockwise rotation of the support bearing.
The stabilisation criterion shall be met when the standard deviation of friction torque during measurement does not exceed 15 % of the mean value or 0,4 Nm, whichever value is the highest.
For every individual sample, all recorded values for each grid point shall be averaged to an arithmetic mean over the measurement duration. Next, these arithmetic mean values of the same grid point shall be averaged over all samples to one arithmetic mean value per grid point.
For each grid point:
The wheel end speed value before averaging shall not deviate from the set value by more than ± 5 rpm;
The radial load value before averaging shall not deviate from the set value by more than ± 2 kN;
No systematic deviation from the set values is allowed.
If the above specified criteria are not met, the measurement of the respective grid point is void. In this case, the measurement for the entire affected speed and load step shall be repeated, and the reason for voiding the grid point shall be recorded in the test report. After passing the repeated measurement, the data shall be consolidated.
In the case that the uncertainties on the measured friction torque are below the limit set in point 3.2.3.1, the reported friction torque loss shall be regarded as equal to the measured friction torque losses.
In the case of higher uncertainties, the part of the uncertainty exceeding the limit shall be added to the measured friction torque losses.
The final wheel end friction torque loss at a given speed and load shall thus be calculated as follows:
TreportedTmeasured + max0, Ut – Ulimit
Where:
Treported is the calculated friction torque loss at a given speed and load reported for the CO2 certification of wheel ends [Nm];
Tmeasured is the measured friction torque loss according to section 3.4.4 at a given speed and load [Nm];
Ut is the absolute value of the torque uncertainty (>0), expressed in Nm;
Ulimit is 0,2 Nm.
For the calculation of the final friction value for the wheel end, the grid points of the reported torque loss map shall first be averaged for all the wheel end samples in accordance with section 0, corrected in accordance with section 3.4.6, if applicable, and then weighted according to Table 1 for non-driven axle wheel end applications.
250 rpm | 500 rpm
25 % load | 0,4 % | 2,4 %
50 % load | 7,9 % | 35,3 %
100 % load | 9,5 % | 44,5 %
The wheel end manufacturer may declare the weighted average friction as calculated in section 3.5 as the certified value for the wheel end family. Alternatively, the wheel end manufacturer has the option to declare any higher friction value. The declared friction value shall be rounded to 1 place to the right of the decimal point.
Every wheel end certified in accordance with this Annex shall be so manufactured as to conform, with regard to the description as given in the certification form and its annexes, to the approved type. The conformity of the certified CO2 emissions and fuel consumption related properties procedures shall comply with those set out in Article 31 of Regulation (EU) 2018/858.
Conformity of the certified CO2 emissions and fuel consumption related properties shall be checked on the basis of the description in the certificate set out in Appendix 1 and the specific conditions laid down in this point.
The wheel end manufacturer shall test, at least every second year from the date of the family parent’s certification, the number of wheel end families shown in Table 2. The number of wheel end families to be tested depends on the production volumes of the year previous to the year when the conformity of production testing is due.
At least two wheel ends of the same family member shall be tested.
Production number | Number of wheel end families to be tested
0 – 100000 | 2
100001 – 150000 | 3
150001 – 250000 | 4
250001 and more | 5
For conformity of the certified CO2 emissions and fuel consumption related properties testing, the wheel end manufacturer shall apply the same procedure as described in point 3, including the run-in procedure and validation criteria.
A conformity of the certified CO2 emissions and fuel consumption related properties test is passed when the weighted average friction value from the conformity testing is lower or equal to the declared friction value for wheel end family, with an allowed tolerance margin of +10 %.
If the production conformity testing is not passed, three additional wheel ends shall be tested using the same procedure. The recorded values of all tested ends, including the three additional wheel ends, shall be averaged for each grid point to an arithmetic mean. If the conformity of production test is again not passed, the provisions set out in Article 23 shall apply.
If a family member proves to have higher friction than the family parent, the family member shall be reclassified into another wheel end family, and require a new certification.
The standard friction loss for non-driven axle applications shall be 4,8 Nm.