3D-Particle Tracking Velocimetry an overview by Beat L ü thi

Some applied flows one would like to know more about from Matthias Machacek (2003) PhD Thesis ETH applied flow in wind tunnel: car

Some applied flows one would like to know more about from Matthias Machacek (2003) PhD Thesis ETH more abstract but still applied flow: delta wing

Some applied flows one would like to know more about G. Wilkesanders, Ch. Skallerud, Univ. of Colorado at Boulder Visualized in rheoscopic fluid made of fish scales in water. The wake of a bluff body generated a von Karman vortex street

Main idea of 3D-PTV How to measure a flow field? How to get 3D information? How to get Lagrangian information? How to not disturb the flow? 3D-PTV

Main idea of 3D-PTV 3D-PTV = image based three dimensional Lagrangian flow measurement technique (Particle Tracking Velocimetry) flow + CCD cameras + computers = 3D-PTV

Main idea of 3D-PTV from Matthias Machacek (2003) PhD Thesis ETH smoke streaks yield ’only’ quantitative information

Main idea of 3D-PTV from Matthias Machacek (2003) PhD Thesis ETH 3D-PTV yields quantitative, Lagrangian flow trajectories

Main idea of 3D-PTV from Matthias Machacek (2003) PhD Thesis ETH … zooming in more: flow details behind delta wing

Main idea of 3D-PTV from Heinrich Stüer (1999) PhD Thesis ETH more ’fundamental’ flow: backward facing step

Main idea of 3D-PTV from Berg (2006) PhD Thesis Risø ’ fully fundamental’: isotropic turbulence

Main idea of 3D-PTV to follow a 3D (!) particle position as opposed to 2D PIV! started 1983 ……

some 3D-PTV groups USA, Cornell Reynolds number Denmark, Risø particle dispersion Switzerland, ETH velocity derivatives and many more groups: Eindhoven, Tel Aviv, G ö ttingen,

list of technical aspects flow tracers illumination cameras observation volume camera callibration particle detection from 2D to 3D positions particle tracking

flow tracers

illumination

cameras

observation volume

camera callibration

particle detection

from 2D to 3D positions

particle tracking

flow tracers high tech, accurate, expensive: Idea: Søren Ott & Jakob Mann, Risø, Denmark fly ash  sieving   50-60 µm low tech, accurate, cheap:

illumination LED array, TU/e Lorenzo del Castello, Herman Clercx trend towards smarter solutions

fast digital cameras pixel: 500x500 frame rate: 50Hz pixel: 1000x1000 frame rate: 5000Hz or pixel: 250x250 frame rate: 80’000Hz data storage is main bottelneck

particle detection and position typically the ’image situation’ is far from ideal

camera callibration teach the cameras with know grid points problem: how to have space filling target? solution in part: callibration on flow tracers

teach the cameras with know grid points

problem: how to have space filling target?

solution in part: callibration on flow tracers

from 2D to 3D position callibration and 2D position accuracy, seeding density, etc. Camera 4 Camera 3 Camera 1 Camera 2 x y z r(x,y,z,t)

tracking through consequtive images tracking criteria: particle must not travel further than their typical spacing codes available at www.3dptv.schtuff.com

overcome seeding density bottelneck? K. Hoyer, M. Holzner, ETH Scanning PTV idea: scan flow with thick laser sheet to get more particles

many dependencies, many choices… field of view depth of view optical working distance camera pixel resolution camera recording rate illumination flow speed flow scales one would like to resolve particle diameter number of tracer particles trackability

final output is the start for analysis if all goes well, one can finally start ’learning’ about the flow

velocity derivatives differentiate convoluted velocity field to get velocity derivatives challenge to get HIGH SEEDING DENSITY B. Lüthi ETH, Søren Ott Risø

applications flows with: turbulence dilute polymers 2-phases convection agregation reactions mean shear entrainment rotation bio-medical conditions emphasis on: velocity velocity gradient acceleration dispersion multi particle Lagrangian aspects (mean) Eulerian flow field challenge: reach higher Reynolds numbers at full spacial resolution

flows with:

turbulence

dilute polymers

2-phases

convection

agregation

reactions

mean shear

entrainment

rotation

bio-medical conditions

emphasis on:

velocity

velocity gradient

acceleration

dispersion

multi particle

Lagrangian aspects

(mean) Eulerian flow field

Some selected outcome