SOME ISSUES CONCERNING THE
AUTOMATIC MEASURING OF
CONCRETE WORKABILITY DURING
MIXING TIME
Bogdan CAZACLIU – IFSTTAR, [email protected]
Aggregates & Materials Processing Laboratory
Basic principles of in-mixer sensors
• Power curve stabilization level WATTMETERbefore stabilization MIXOMETER
• Drag force under mixing VISCOPROBE, MFMmixer stopped …
• Composition water content Moister metersair entrained …
• Other methods
Power curve (or torque)
Ordinary concrete SCC …
tf < th = ts
mixing time
mixing power
high w/p, low sp/c
th < tf
exp. decay → t *
low w/p, high sp/c
mixing power
mixing time
ts
overmixing
Cazacliu B., Mixing self compacting concrete: mixers, mixing methods, mixing time, RILEM Publications PRO90 2013
WATTMETER
Hatt 1921Tests of Concrete Mixer
Level ofstabilized power
WATTMETER
STRONG
• Maintenance- No sensor in the mixer
• Cheap• Well established
• OK for VIBRATED concrete
WEAK
• RheologyPViscosity 2 parameters 1 measure
P = PWall Slip 2 new parameters
PCrashing depends on wearing
• For SCC- No stabilization- Small sensitivity
• For DRY concrete
MIXOMETER
MIXOMETER – mixing mechanisms m
ixin
g p
ow
er
(kW
)
0
1
2
3
0 2 4 6 8 10
time (minutes)
8
dry granules
granular suspension
wet granules
dry powder
powder agglomerates
1st
mechanism2nd
mechanism3rd
mechanism
4th
mechanism
Cazacliu B., In-mixer measurements to describe the mixture kinetics during concrete mixing, Chemical Engineering Research and Development, Vol.86, 1423-1433, 2008
Wet powder Uniform Concrete
MIXOMETER – mixing mechanisms
Power measurement for batchesof a same mix-design
RMC Plant
0%
1%
2%
3%
4%
08/04 11/04 02/05 05/05 09/05 12/05 03/06 07/06
10/20
4/10
Aggregates moisture evolution
Measured by dryingRMC Plant
0 20 40 60 80 100 120 140
0
20
40
60
80 Puissance de malaxage
temps de l’opération (s)
∆ ≈ ± 10 l/m3
moisture mixing power
operation time
same concrete during 2 months
Ngoc-Dong Lê, Amélioration de la régularitédu béton en production, PhD 2007
Estimates:
- The time needed tofinish the mixing
- The deviation inwater content(or consistency)
PatentSkako / IFSTTAR
MIXING TIME (s)
1st batch/4:
Water deviation:
Next batch:
POWER (kW)
Mixing index
MIXOMETER
WATTMETER
STRONG
• Maintenance- No sensor in the mixer
• Cheap• Well established
• OK for VIBRATED concrete
WEAK
• RheologyPViscosity 2 parameters 1 measure
P = PWall Slip 2 new parameters
PCrashing depends on wearing
• For SCC- No stabilization- Small sensitivity
• For DRY concreteMIXOMETER
DRAG
FORC
E
Drag force
LICHTENBERG 1926 - method of and
apparatus for determining the consistency
of concrete
strain gauge
in concrete immerged sphere
VISCOPROBE
3
5
7
9
11
13
-15 0 15 30 45 60 75 90
Viscoprobe average signal
WATER
introduction
3
4
5
6
7
8
-15 0 15 30 45 60 75 90
wattmeter signal
WATER
introduction
Around
nomina
l
-10
-8
+0
+2
+4
+6.5
+9
+10
The viscoprobe average force curves follow the wattmeter curves,with a better differentiation
VISCOPROBE
Cazacliu B., Brunquet E., On-line mixing end-point detection by power measurement, Rilem Proceedings pro054, 2007
3
5
7
9
11
13
-15 0 15 30 45 60 75 90
Viscoprobe average signal
3
4
5
6
7
8
-15 0 15 30 45 60 75 90
wattmeter signal
Upper
+9
+10
+20
+24
Much better with upper water proportioning
Good repeatability – see curves +9 and + 10
VISCOPROBE
3
4
5
6
7
8
-15 0 15 30 45 60 75 90
wattmeter signal
3
5
7
9
11
13
-15 0 15 30 45 60 75 90
Viscoprobe average signal
For low water proportioning(-16l/m3 ie slump flow 49cm ) the viscoprobe average force show a minimum when wattmeter shows a second peak!
Minimum of force for the viscoprobe
Maximum of power !
Low
-16
-11
VISCOPROBE
Particle Image Velocimetry Nuclear Magnetic Resonance
What is the structure of the flow
Philippe Poullain, Étude comparative de l’écoulement d’un fluide viscoplastique dans une maquette de malaxeur pour bétons : PIV, IRM et simulation numérique, PhD 2003
Experimental
3 drag force probes
- Viscoprobe (planetary movement)- MFMi (circular movement)- MFMe (circular movement)
blades
wW
W
W
W
MFMi
MFMe
Viscoprobe
scraper
5 levels of mixing speed W:( with constant w/W )
100% (nominal), 75%, 50%, 25%, 10%
What is the structure of the flow
Ngoc-Dong Lê, Amélioration de la régularité du béton en production, Degree awarded May 2007
Rheometry on samplesfor each tested concrete
(9 SCC mixture tested)blades
wW
W
W
W
MFMi
MFMe
Viscoprobe
scraper
for a known rheology the drag force is a measure
of the concrete velocity around the probe
Drag force = F (rheology, velocity)
Experimental
What is the structure of the flow
code FloMix (IFSTTAR) :
- velocity field determined by
FME fictitious domain methods- 2D, non-slip boundary conditions- several Bi and Re numbers tested
Velocity intensity scaled 0 to 0.5 of scraper velocity
Bi = 0, 32, 64 and 100; Re = 1
Bi = τo / µΩ
Re = ρΩR² / µ
Numerical simulations
What is the structure of the flow
addition of two flows + mixing star flowinteraction flow
low Re / Bi high Re / Bi
Non-inertialreference frame !
structure of the flow
0
50
100
150
200
250
300
350
400
450
0 10 20 30 40 50 60 70 80
Force at v = 0 by Viscoprobe, Fo (N)
Yield stres
s by rheo
metry
, t o
(Pa)
BAP1
BAP2
BAP3
BAP4
BAP5
validity limit
wall slippingin-mixer lower density …
c* = F0/t0RVisco²
Concrete rheology - Drag force
0
10
20
30
40
50
60
70
80
90
0 10 20 30 40 50 60 70 80 90 100
Viscoprobe slope, p (N.s/m)
Visco
sity by rheo
meter, µ
(Pa.s)
l* = p/µRVisco
wall slippingin-mixer lower density …
validity limit
Concrete rheology
0
20
40
60
80
100
120
140
160
180
200
0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2 2,1
Vitesse (m/s)
Force de trainée (N)Ftrainée
rhéomètre
100%75%50%
25%10%
Rheometer fluidity curve
In-mixer fluidity curve
velocity (m/s)
Drag force (N) Vibrated concrete
WATTMETER
STRONG
• Maintenance- No sensor in the mixer
• Cheap• Well established
• OK for VIBRATED concrete
WEAK
• RheologyPViscosity 2 parameters 1 measure
P = PWall Slip 2 new parameters
PCrashing depends on wearing
• For SCC- No stabilization- Small sensitivity
• For DRY concrete
MIXOMETER DR
AG FORC
E
COMPO
SITION
Moisture content
Cazacliu B., In-mixer measurements to describe the mixture kinetics during concrete mixing, Chemical Engineering Research and Development, Vol.86, 1423-1433, 2008
13
23
33
mixing power [kW]
70
74
78
0 30 60 90 120 150
time (s)
local moisture [%]
3
7
11
drag force [N]
wet paste fluiddry
W/P 0.309
13
23
33 mixing power [kW]
70
74
78
0 30 60 90 120
time (s)
local moisture [%]
3
7
11
drag force [N]
wet paste fluiddry
W/P 0.326
∆∆∆∆ ≈ 2.5 L/m3
Moisture content
Moisture content
Composition ≠ Workability
Microstructure + Temperature
Basic principles of in-mixer sensors
• Power curve stabilization level WATTMETERbefore stabilization MIXOMETER
• Drag force under mixing VISCOPROBE, MFMmixer stopped …
• Composition water content Moister metersair entrained …
• Other methods
Other methods
Clagett 1938 - Method of determining the
consistency of concrete mixtures
CONCLUSION
COSTS RHEO
WATTMETER
MIXOMETER
MOISTURE METER
VISCOPROBE
WATTMETER
MIXOMETER
MOISTURE METER
VISCOPROBE
CONCLUSION
Slump (cm): 0 … 10 … 20 …
MOISTURE METER
VISCOPROBE
WATTMETER
MIXOMETER
OUINON NON
OUINON
OUINON
OUI NON
THANK YOU !
Acknowledgement to SKAKO Concrete