OpenLegislation

Division 3.3.9 Measurement uncertainty

United Kingdom

Original text. Later amendments are not shown.

The part of the calculated total uncertainty UT,loss exceeding 5 % of Tloss or 1 Nm (ΔUT,loss ), whichever value of ΔUT,loss is smaller, shall be added to Tloss for the reported torque loss Tloss,rep . If UT,loss is smaller than 5 % of Tloss or 1 Nm, then Tloss,rep  = Tloss .

Tloss,rep = Tloss + MAX (0, ΔUT,loss )

ΔUT,loss = MIN ((UT,loss – 5 % * Tloss ), (UT,loss – 1 Nm))

For each measurement set, the total uncertainty UT,loss of the torque loss shall be calculated based on the following parameters:

  1. (1)

    Temperature effect

  2. (2)

    Parasitic loads

  3. (3)

    Calibration error (incl. sensitivity tolerance, linearity, hysteresis and repeatability)

The total uncertainty of the torque loss (UT,loss ) is based on the uncertainties of the sensors at 95 % confidence level. The calculation shall be done as the square root of the sum of squares (‘Gaussian law of error propagation’).

  • UT,loss=UT,in2+UT,outigear2
  • UT,in∕out=2×u2TKC+u2TK0+u2cal+u2para
  • uTKC=13×wtkcKref×ΔK×Tc
  • uTK0=13×wtk0Kref×ΔK×Tn
  • uCal=1×Wcalkcal×Tn
  • upara=13×wpara×Tn
  • wpara = senspara * ipara

where:

Tloss
  • Measured torque loss (uncorrected) [Nm]

  • Tloss,rep
  • Reported torque loss (after uncertainty correction) [Nm]

  • UT,loss
  • Total expanded uncertainty of torque loss measurement at 95 % confidence level [Nm]

  • uT,in/out
  • Uncertainty of input/output torque loss measurement separately for input and output torque sensor[Nm]

  • igear
  • Gear ratio [-]

  • uTKC
  • Uncertainty by temperature influence on current torque signal [Nm]

  • wtkc
  • Temperature influence on current torque signal per Kref, declared by sensor manufacturer [%]

  • uTK0
  • Uncertainty by temperature influence on zero torque signal (related to nominal torque) [Nm]

  • wtk0
  • Temperature influence on zero torque signal per Kref (related to nominal torque), declared by sensor manufacturer [%]

  • Kref
  • Reference temperature span for uTKC and uTK0, wtk0 and wtkc, declared by sensor manufacturer [K]

  • ΔK
  • Difference in sensor temperature between calibration and measurement [K]. If the sensor temperature cannot be measured, a default value of ΔK = 15 K shall be used.

  • Tc
  • Current/measured torque value at torque sensor [Nm]

  • Tn
  • Nominal torque value of torque sensor [Nm]

  • ucal
  • Uncertainty by torque sensor calibration [Nm]

  • Wcal
  • Relative calibration uncertainty (related to nominal torque) [%]

  • kcal
  • calibration advancement factor (if declared by sensor manufacturer, otherwise = 1)

  • upara
  • Uncertainty by parasitic loads [Nm]

  • wpara
  • senspara * ipara

    Relative influence of forces and bending torques caused by misalignment [%]

  • senspara
  • Maximum influence of parasitic loads for specific torque sensor declared by sensor manufacturer [%]; if no specific value for parasitic loads is declared by the sensor manufacturer, the value shall be set to 1,0 %

  • ipara
  • Maximum influence of parasitic loads for specific torque sensor depending on test setup (A/B/C, as defined below).

  • A) 10 % in case of bearings isolating the parasitic forces in front of and behind the sensor and a flexible coupling (or cardan shaft) installed functionally next to the sensor (downstream or upstream); furthermore, these bearings can be integrated in a driving/braking machine (e.g. electric machine) and/or in the transmission as long as the forces in the machine and/or transmission are isolated from the sensor. See figure 3.

    Figure 3

    Test setup A for Option 3

  • B) 50 % in case of bearings isolating the parasitic forces in front of and behind the sensor and no flexible coupling installed functionally next to the sensor; furthermore, these bearings can be integrated in a driving/braking machine (e.g. electric machine) and/or in the transmission as long as the forces in the machine and/or transmission are isolated from the sensor. See figure 4.

    Figure 4

    Test setup B for Option 3

  • C) 100 % for other setups