Coagulant Overfeed and Filter Breakthrough After a Storm
A 6.0 MGD conventional surface-water plant treats a river source with alum and polymer, flocculation, sedimentation, and four dual-media filters. After a 2-inch rain event, raw turbidity climbed from 4 NTU to 24 NTU in six hours and raw alkalinity fell from 105 to 42 mg/L as CaCO3. The operator raised the alum dose from 28 to 48 mg/L. Two hours later, settled water turbidity is 4.1 NTU (normal 1.2), Filter 2 and Filter 3 effluent read 0.34 and 0.38 NTU (normal 0.07), finished water pH is 6.3 against a 7.4-7.8 target, and clearwell chlorine residual is steady at 0.9 mg/L. Filter runs have fallen from 60 hours to 14 hours. The plant has one jar tester, 2.0 MG of finished-water storage (about eight hours at 6 MGD), and an idle polymer feed.
1. What is failing in the treatment train, and what are your first-hour actions in priority order?
Read the worked answer
Diagnosis. The alum increase is consuming alkalinity the raw water no longer has. Alum destroys roughly 0.45-0.5 mg/L of alkalinity (as CaCO3) per mg/L of dose, so 48 mg/L of alum takes about 22 mg/L of the 42 mg/L available. The result is a pH crash to 6.3, incomplete alum hydrolysis, soluble aluminum and pin floc that the clarifier cannot settle and the filters cannot hold. Settled water at 4.1 NTU proves the failure is upstream of the filters; the filters are only reporting it.
Priority actions.
- Jar-test the current raw water before changing anything. Do not chase turbidity with more alum.
- Restore alkalinity and pH ahead of coagulation - roughly 12 mg/L of caustic soda (about 0.8 mg NaOH per mg/L of alkalinity as CaCO3, plus what the added alum consumed) to bring coagulated pH back near 7.0, with final pH trim to the 7.4-7.8 target.
- Re-optimise the alum dose downward and start the idle polymer as a coagulant/filter aid.
- Reduce loading: cut plant rate, wash Filters 2 and 3 and return them slowly, and check for media loss, mudballs and a rising turbidity profile.
- Check clarifier blanket level and sludge removal - solids carryover is why runs fell to 14 hours.
- Log every action and reading.
2. Filters 2 and 3 read 0.34 and 0.38 NTU. Is that a violation, and what must you do and record?
Read the worked answer
Not on these two readings alone. A conventional plant must keep individual filter effluent at or below 0.30 NTU in at least 95 percent of the measurements taken each month, and combined filter effluent at or below 0.30 NTU in 95 percent of monthly measurements and never above 1.0 NTU. Two grab samples are a trend; compliance is a monthly calculation based on how many readings were taken. With, say, 40 individual-filter readings in the month, two exceedances still satisfy 95 percent - but you must actually have those readings, so increase IFR sampling to at least once every four hours per filter.
Know the escalation triggers: any individual filter above 1.0 NTU after four hours of operation must be reported to EGLE by the end of the next business day, with a filter self-assessment within 14 days; above 0.5 NTU in two consecutive months requires a self-assessment within 14 days; above 1.0 NTU in three consecutive months, or above 2.0 NTU at any time, requires a comprehensive performance evaluation.
Record the readings with times, the jar-test and dose changes, the corrective actions, and the fact that the excursions coincided with a coagulant upset.
3. Suppose jar tests cannot get settled water below 2 NTU within four hours. How do you decide between continuing production and curtailing, given 2.0 MG of storage?
Read the worked answer
2.0 MG at 6 MGD is eight hours of supply, so this is a managed decision, not an emergency - yet.
Framing the choice. Producing water that is not being filtered properly trades a quantity problem for a public health problem: turbidity is the surrogate for the filtration barrier and for protozoan removal credit. Never let out-of-spec water reach the clearwell without informing EGLE and customers.
Order of options. (1) Hold the plant rate where the filters stay under 0.30 NTU - partial output plus storage usually covers the day. (2) Take the worst filters out, run the rest conservatively, and accept lower production. (3) Ask large users to cut back, and open any interconnection or purchase water - this buys storage. (4) If storage falls toward 50 percent (1.0 MG) with no recovery, call EGLE, plan a precautionary advisory and conservation notice, and document the timeline.
The judgement the exam rewards is acting on the turbidity trend early, communicating before the clearwell is the constraint, and never treating storage as an excuse to keep making non-compliant water.