<DOC>
<DOCNO>
315
</DOCNO>
<TITLE>
scale effects at high subsonic and transonic speeds
and methods for fixing transition in model experiments .
</TITLE>
<AUTHOR>
haines,a.b., holder,d.w. and pearcey,h.h.
</AUTHOR>
<BIBLIO>
arc r + m3012, 1954.
</BIBLIO>
<TEXT>
  the major scale effects at high subsonic and transonic
speeds arise from differences between the conditions
under which laminar and turbulent boundary layers separate, and in
how they behave after separation .  for turbulent
boundary layers, these conditions and behaviour do not vary greatly
as the reynolds number is changed and in many
examples, it has been shown that they are similar for the turbulent
layers that occur naturally at high reynolds number
and for boundary layers in which transition to turbulent flow is fixed
artificially .  the scale effects arising in wind-tunnel
tests made at low reynolds number may, therefore, often be
minimised by fixing transition to turbulent flow by
introducing an artificial disturbance such as that produced by
excrescences attached to the surface .  the fact that the
effects of separation are often less severe for laminar layers than
for the turbulent layers that are likely to be encountered
at full scale, makes it all the more important to do this whenever
possible .
  several methods which can be used to fix transition are described,
and the results obtained by using them are compared .
in general, in experiments in two-dimensional flow, good agreement
is found, and explanations can be advanced for
cases in which discrepancies occur .  several uncertainties and
difficulties that arise in fixing transition are discussed
and illustrated by examples .  in particular, special care is needed
in interpreting the results obtained with transition
fixed at very low reynolds numbers (say, less than about r = 1 x 10
based on local chord for wings of about 0.1
thickness chord ratio and possibly higher reynolds numbers for thinner
wings) .
  the difficulties of fixing transition satisfactorily are increased
for three-dimensional wings, particularly if they are
swept-back or highly tapered (i.e., small chord and reynolds
number near the tip) and if the tests cover a large range
of incidence including high incidences for which the flow may
separate from very close to the leading edge .  under
these circumstances, it is frequently necessary to place the
excrescences at different chordwise positions for low and
high angles of incidence, and this is inconvenient in practice .
more research is needed before sound recommendations
can be made as to how and where transition should be fixed
on such models, particularly since in routine testing, it is
often not possible to check the effects of transition-fixing fully .
  in the sections dealing with three-dimensional tests, examples
are given of the spurious results that have been avoided
successfully by fixing transition, of the conditions where even
at low reynolds numbers artificial fixing of transition
may not be necessary to give a turbulent boundary layer ahead
of the shock, and of the conditions under which there
are some doubts whether the methods used for fixing transition
have been satisfactory .
</TEXT>
</DOC>
