Post-Storm Alum Overdose at a 4.0 MGD Conventional Plant
A 4.0 MGD conventional plant treats a reservoir that just took 2.8 in of rain in six hours. Raw turbidity climbed from 6 to 38 NTU and raw alkalinity fell from 68 to 42 mg/L as CaCO3 (pH 7.4). Acting on last week's jar test, the operator raised the alum dose from 35 to 55 mg/L. Within four hours, settled turbidity rose from 1.5 to 6 NTU, the floc turned light and pin-point, finished water pH fell to 6.2, and filter effluent turbidity reached 0.42 NTU. Flow this shift is 3.5 MGD and the clearwell pH is now below the 7.0 corrosion-control target. The jar test that produced the 55 mg/L figure was run on raw water collected two days before the storm. The plant has dry alum, hydrated lime, and an anionic polymer on hand.
1. Using the chemistry and the numbers provided, explain why raising the alum dose made coagulation worse rather than better.
Read the worked answer
Two mechanisms are working at once.
The dose was chosen from stale data. The 55 mg/L optimum came from pre-storm water at 68 mg/L alkalinity and 6 NTU. The storm changed the particle character (smaller, more organic, higher surface area) and cut alkalinity by 26 mg/L. An optimum dose is a property of a specific water, not a fixed number.
Alkalinity consumption drove the pH down. Alum destroys roughly 0.45 mg/L of alkalinity as CaCO3 per mg/L of alum:
- Alkalinity consumed = 55 x 0.45 = 24.8 mg/L
- Alkalinity remaining = 42 - 24.8 = ~17 mg/L
That is the drop from pH 7.4 to 6.2. With little alkalinity left, alum hydrolysis cannot go to completion; you get soluble aluminum, weak pin-point floc, and poor sweep.
Overdosing can also restabilize. Excess positive coagulant species can reverse the particle charge, so collisions stop producing floc.
The evidence points away from "not enough coagulant": settled turbidity tripled, floc got smaller, and pH fell. Those are overdose/alkalinity symptoms, not underdose symptoms.
2. State the corrective actions you would take during this shift, in order, and name one action you would specifically avoid.
Read the worked answer
Sequence for this shift:
- Stop the bleeding first. Drop the alum back toward 35 mg/L or to a verified jar-test value; do not keep climbing.
- Run a fresh jar test on today's raw water, testing a range (20, 25, 30, 35, 40, 45 mg/L) plus a parallel series with lime or caustic added to restore 20-25 mg/L of alkalinity and pH 6.8-7.2. You are testing two variables: coagulant dose and pH correction.
- Confirm the feeder is actually delivering what the dial says. A calibration cylinder drawdown test over 2 minutes catches a worn diaphragm or a plugged rotameter that would otherwise make the dose change meaningless.
- Restore alkalinity/pH ahead of rapid mix (lime or caustic), then re-check settled turbidity before touching filtration.
- Use polymer as a filter aid if settled turbidity is still marginal while the chemistry settles out.
- Consider PACl as a lower-alkalinity-demand alternative if this source stays low-alkalinity.
What to avoid: do not solve a turbidity problem by adding more coagulant blindly, and do not chase filter effluent turbidity with polymer while the settled water is 6 NTU. Polymer hides a settling failure; it does not fix it.
3. The state wants at least 25 mg/L of residual alkalinity and pH 7.0 or higher after coagulation. At 3.5 MGD, how many pounds per day of hydrated lime (Ca(OH)2) would raise alkalinity by 12 mg/L as CaCO3, and where does it go?
Read the worked answer
Step 1 - convert lime to alkalinity.
Ca(OH)2 has a molecular weight of 74; the CaCO3 equivalent is 50 per equivalent, and lime supplies two equivalents:
- 1 mg/L Ca(OH)2 = 100/74 = 1.35 mg/L as CaCO3
Step 2 - dose needed.
- 12 mg/L as CaCO3 ÷ 1.35 = 8.9 mg/L of Ca(OH)2
Step 3 - convert to pounds per day.
- lb/day = mg/L x MGD x 8.34
- lb/day = 8.9 x 3.5 x 8.34 = ~260 lb/day of pure Ca(OH)2
Step 4 - correct for product purity. Hydrated lime is typically 90-95% Ca(OH)2:
- 260 ÷ 0.92 = ~283 lb/day of commercial product
Where it goes: feed as a slurry into rapid mix ahead of the alum, so the alkalinity is available when the coagulant hydrolyzes. Verify with a jar test before committing the dose, because overshooting raises pH, reduces alum effectiveness, and can cause CaCO3 scaling in the clearwell and on filters. Confirm the existing lime feeder and slurry system can hold 280 lb/day continuously; lime settle-out in the day tank is a common failure.