afstudeer poster

1
Seeding of air for PIV-measurements Ronald
Kuunders Faculty of Mechanical Engineering,
Eindhoven University of Technology
Introduction

The flow inside an internal combustion engine is
of high importance for the combustion, and thus
emissions and power. This flow can be measured
or simulated. Particle Image Velocimetry produces
instantaneous velocity fields. This technique is
useful for measuring the turbulence and can be
used to validate numerical simulations. PIV uses
the displacement of particles in the flow. These
particles have to be added and the accuracy of
the PIV-measurement depends largely on the
properties and numbers of these particles.
A rectangle duct flow with an aspect ratio of 21
and a Reynolds number of 5300 was investigated.
The mass flow of air was 8.3 kg/hour. Titanium
oxide particles with a diameter of 1 µm were used
during these experiments. The particle production
of the seeding-device was 4.8106 particles a
second. The occupancy of the titanium oxide in
the ethanol droplets was 1200. The number of
particles can be changed by changing the number
of nebulisers or changing the concentration of
titanium oxide in the ethanol. The measured
velocity-profiles have been compared with the
literature. The primary velocity-profiles
compared very well. The secondary
velocity-profiles differed. The differences were
caused by a different inlet and the fact that the
flow was not developed yet. The instantaneous
velocity-fields were reasonable accurate, the
turbulence could be measured.
In-cylinder flow
The flow inside an engine consists of two
main-flows tumble (left) and swirl (right).
Fig. 1 Flow inside an internal combustion engine
Fig. 3 velocity-profile of half the channel
(left) and an instantaneous velocity map(right).
One cylinder of a 9.2 litre truck engine has to
be investigated in the future. When motored at
1000 rpm the particle-production has to be 8106
s-1 to achieve a resolution of 32 x 32 vectors in
one cylinder. At first a seeding-device,
previously used for LDA-measurements, was used
but it did not produce enough particles. The
seeding-device produces ethanol droplets with
titanium oxide particles in it. The ethanol
evaporates and the particles remain. The particle
production was improved by placing the nebulisers
directly in the airflow.
Conclusions

• The seeding-device produces enough particles for
  a PIV-measurement in a rectangle duct.
• The primary velocity fields compared well with
  the literature.
• The velocity-maps were showing the turbulence.
• The particle-production might be high enough to
  seed one cylinder of a truck engine for a rough
  measurement
• The particle-production can be altered by
  changing the number of nebulisers and/or changing
  the concentration of titanium oxide in ethanol.

flowdirection
Future work
Titaniumoxide in ethanol

• Pulsating flow in seeding-device
• High pressure test with seeding-device

Fig. 2 Cross-section of the seedingdevice
Engineering Thesis Committee Prof. Dr. Ir.R.S.G.
Baert , chairman Dr.ir.R.J.M. Bastiaans,
coach Dr.ir.H.P.van Kemenade Dr.ir.L.M.T.
Somers Dr.ir.C.C.M. Rindt
Faculty of Mechanical Engineering Section
Combustion Engines