An analysis is made of roughness measurements made at two different samplin
g intervals with a scanning white-light interferometer on 17 plateau-honed
cylinder liners. Plasticity indices are calculated for both sampling interv
als according to two different methods. For both methods the values of plas
ticity index are found to be significantly different for the two sampling i
ntervals, as expected, but there is no correlation between the two sets of
measurements. One plasticity index is inferred to be about three times the
other, in fair agreement with experiment. An error analysis is carried out
which when applied shows that the proportional uncertainties in the two ver
sions of the plasticity index are about the same. A factorial analysis sugg
ests that to achieve a lower value of the plasticity index, the initial hon
ing pressure should be reduced and plateau honing time should be increased.
(C) 2001 Elsevier Science Ltd. All rights reserved.