Optimized Fluidized Sand Biofilters
Optimized Fluidized Sand Biofilters
Steven Summerfelt
Freshwater Institute, Shepherdstown, WV
Michael Timmons
Cornell University, Ithaca, NY
Interface Between
Clear Fluid & Static Bed
Le
L
Water Distribution &
Media Support Mechanism
Bed height
increases according to Ergun’s
equation until bed begins to expand.
remains constant at all water
velocities after the expansion begins. Real
Static Bed
Real
Fluidized Bed
Pressure drop
remains constant for all sand sizes,
1 m of static sand requires about 1 m of
water head to expand.
see Summerfelt and Cleasby (1996)
Superficial velocity
Bed height
sand as a function of water velocity,
using:
water viscosity and density
Real Real
sand size, sphericity Static Bed Fluidized Bed
Pressure drop
void space of the static bed
see Summerfelt and Cleasby (1996)
Superficial velocity
thick biofilms,
velocities require beds
low velocities, thin biofilms,
to be too narrow and tall
biofloc manage- high velocities,
and limits TAN removal capacity
ment required, high loading rate,
without increasing flowrate
low loading rate, high removal rate,
high removal %, deep beds required,
TAN limiting, somewhat self cleaning
shallow beds,
10/2/00
10/9/00
8/14/00
8/21/00
9/18/00
9/25/00
9/4/00
9/11/00
8/7/00
removal efficiency
was > 50%.
Time (days)
10th Annual Recirculating Aquaculture Systems Shortcourse. 13
FSB Performance in Coldwater
FSB first started up on ammonium chloride.
1 .4 0
Note 1. At stocking
-N (m g /L )
1 .2 0
T A N (B io filter In ) the fish density was
T A N (B io filter O u t)
1 .0 0 15 kg/m3 (mean fish
N itrite (B io filter O u t)
0 .8 0
weight = 150 g).
T A N & N2O
0 .6 0
Note 2. Last
0 .4 0 measured fish
0 .2 0
density was 33.5
kg/m3 (mean fish
0 .0 0
36951 36966 36981 36996 37011 37026 37041 weight = 320 grams).
T im e (da y s )
shear
Biofilm Development in
Fine Sand Biofilters
Shear forces tear biofilm
shear
shear
shear
Filter
inlet Filter
outlet
Lower
Scoured-sand 0.031 - 0.048
bed
10th Annual Recirculating Aquaculture Systems Shortcourse. 27
Flow Distribution Mechanisms
Flow distribution methods vary, but are all important!
1-2 cm
Injection Probes
Influent
Manifold
Effluent Weir
Influent
Manifold
Effluent Weir
fluidized-
sand biofilter
(Swanson-type design)
10th Annual Recirculating Aquaculture Systems Shortcourse. 32
Distribution Through False Floor
Formerly Penobscot Smolt Hatchery (Franklin, ME)
Currently Center for Cooperative Aquaculture Research
(Swanson-type design)
ball
ball
valve
valve
slotted inlet
sand P
4.0
2.0
0.0
water lift
0.0 5.0 10.0 15.0 20.0 25.0 30.0
HLR, gpm/ft2
(Freshwater Institute data)
10th Annual Recirculating Aquaculture Systems Shortcourse. 49
Cyclo Biofilter™ Advantage
Cyclo Bio requires less pressure to operate.
0.1-0.3 bar (2-4 psig) less pressure was required to
operate a cyclo bio compared to a modified-pipe
manifold FSB.
assuming a similar fluidized-sand biofilter height.
cyclo bio’s reduce P of piping and inlet orifice
To stripping
column
9 ft dia x 20 ft
strippers cyclo biofiler
fan fan
LHOs UV channel
LHO
sump
side-wall
150 m3 culture tank
drain
10th Annual Recirculating Aquaculture Systems Shortcourse. 54
Cyclo Bio™ at WV Aqua
Three 9 ft dia Cyclo Bio’s installed at char farm
ball
ball
valve
valve
*as published in the 1998 AWWA Sourcebook and 1996 AWWA Buyers Guide
Expansion, %
Expansion, %
120 120.0
100 100.0
80 80.0
60 60.0
40 40.0
20 20.0
0 0.0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30
HLR, gpm/ft 2
HLR, gpm/ft2
US Silica sand Parry Company sand
D10 = 0.275 mm D10 = 0.23 mm
10th Annual Recirculating Aquaculture Systems Shortcourse. 71
Purchasing Filter Sand
Freshwater Institute recently purchased filter
sands from:
US Silica Company (Berkeley Springs, WV)
D10 = 0.275 mm, UC =1.7
$1300 for 15 tons delivered in 100 lb bags on pallets
The Parry Company (Richmond Dale, OH)
D10 = 0.23 mm, UC =1.5
$1800 for 15 tons delivered by pneumatic truck