Filter Run Turbidity Creep After Coagulant Change
A 6 MGD conventional surface-water plant treats a reservoir source that has just turned over, raising raw turbidity from 4 NTU to 22 NTU and dropping raw alkalinity to 38 mg/L as CaCO3. The operator switches from alum to ferric chloride to improve settling, keeping the same 25 mg/L dose and the same target finished turbidity of 0.10 NTU. Within six hours, combined filter effluent rises from 0.06 to 0.28 NTU and stays there. Individual filter turbidities are 0.15, 0.18, 0.22, and 0.31 NTU. Filter run times have shortened from 40 hours to 14 hours. Filter-to-waste is available but currently bypassed. The clearwell is at 71 percent and demand is normal. Jar tests from the prior shift showed good floc at pH 6.8, but the current filter influent pH reads 7.4.
1. What is the most likely cause of the turbidity breakthrough, and what immediate operational action do you take?
Read the worked answer
The numbers point to a coagulation chemistry problem, not a filter media failure. Ferric chloride consumes alkalinity differently than alum and shifts the optimal pH; the jar test at 6.8 does not match the plant at 7.4. At 38 mg/L raw alkalinity, a 25 mg/L ferric dose leaves little buffering margin, so the pH drifted upward and the floc is weak and pin-floccing through the filters.
Immediate actions:
- Put filters on filter-to-waste now to protect the clearwell while you diagnose. Finished water above 0.3 NTU is a compliance risk and the clearwell has enough volume to ride out a short filter-to-waste period.
- Confirm pH and alkalinity at filter influent with a freshly calibrated meter.
- Run a bench jar test at the actual plant pH (7.4) across a dose and pH sweep, including a coagulant aid or acid addition to bring pH back toward 6.8.
- If pH correction is needed, feed acid or adjust coagulant dose to restore the target pH, then watch filter effluent for recovery before returning to service.
Do not simply backwash harder; that treats the symptom and shortens runs further.
2. The 0.31 NTU filter is the worst performer. How do you decide whether to take it out of service versus keep it running?
Read the worked answer
Use the individual filter effluent (IFE) framework and the combined limit as the decision drivers. A single filter at 0.31 NTU is above the 0.3 NTU combined limit and signals a problem, but the immediate question is whether the combined stream can stay compliant.
Decision logic:
- If the other three filters are stable and the combined effluent can be held at or below 0.10–0.15 NTU with the worst filter isolated or on filter-to-waste, keep it online on filter-to-waste and investigate.
- If the worst filter is trending upward and the combined stream is climbing, take it out of service and backwash, then evaluate the backwash (air scour, bed expansion, mudball check).
- Check for media loss, bed disturbance, or a channeled bed on that filter. A filter that will not recover after a proper backwash should be removed from service for media inspection.
Document the IFE readings, the action taken, and the recovery, because consecutive IFE exceedances trigger reporting obligations.