The MBR Book: Principles and Applications of Membrane

Table 3.7 O&M data
Design 137
Kubota Mitsubishi Rayon Zenon Norit X-Flow
Membrane aeration
capacity (Nm
90–180 75–120 100 (cycled) 140
3 /h)
Cycle (min) 8 on/2 relaxa 20 h on/4 h off c bflshf 20 h on/4 h off h
Net flux (LMH)
● Normal 8.3–12.5 5–8 20e 15–20, 37i ● Peak 32.5–42 20.3–30.6 35e 50
Biological aeration
capacity (Nm
160 160 100 140
3 /h)
F/M ratio 0.04–0.18 0.02–0.14 0.04–0.18 0.04–0.12
HRT (h) 10.2–15.4 15–22 7.6–12.3 15.2
SRT (day) 27–70 31–87 26–51 42–66
MLSS (g/L) 10.5–12 8.9–11.6 10.4–11.2
Chemical cleaning NaOCl, 0.5% NaOCl, 0.5% NaOCl, 1%, NaOCl, 0.5%,
reagents Oxalic acid, 1% followed by followed by followed by
acid 0.3% citric acid Ultrasili Derived data
SAD m a (Nm 3 /(h m 2 )) 0.75 0.28–0.38 0.54 g 0.33–0.6 h
SAD p (m 3 air/m 3 60–90 normal; 48–56 normal 27 30–40 normal
permeate) 18–23 peak 12–14 peak 15 12–16 peak
Mean permeability, 200–250 w/o r a 200 normal d 200–250 250 normal
LMH/bar 500–800 w r a 140–150 peak d 320–350 75–200 peak
350 peak w r a after clean
Permeability decline 1.5 b 0.39 b 20 b
�K�t, LMH/(bar h)
aRelaxation introduced mid-way through Phase I; permeability data refers to without (w/o.) and with (w.)
relaxation.
bRefers to peak flux operation: for the Zenon membrane this was 60 LMH.
c “Night-time” relaxation introduced during Phase III, along with backflushing at 20 LMH.
dAssumed to be with relaxation.
eRefers to 500 c module.
fAuthors state “ratio of net to gross flux was 83–85%”; bflsh � backflushed.
gIntermittent operation.
hNight-time relaxation introduced during Phase II.
iWith weekly maintenance clean.
jCombination of sulphuric and phosphoric acid.
0.075–0.11 for the Zenon. Ferric dosing was applied at different concentrations for
the different technologies.
The protocol adopted was for operation under flow conditions set by the flow to the
sewage treatment works: a fixed proportion of the flow was directed to the MBR pilot
plants. This led to rather conservative average fluxes. Operating (or gross) fluxes
ranged between 10 and 20 LMH for most of the time, leading to even lower net fluxes
of 8–15 LMH. As a result, high permeabilities were generally recorded, particularly
for the Kubota membrane and especially following the introduction of cleaning
strategies such as relaxation. Commensurately high SAD numbers (SAD p � 60–90, m 3
air/m 3 permeate in the case of Kubota) resulted. Peak flow tests conducted on each of
the technologies always resulted in a significant drop in permeability.