Why CASS Reactors Fail Differently Than Continuous Activated Sludge
It is 2 a.m. Your SCADA alarm fires: effluent NH3-N has doubled to 14 mg/L and the next decant starts in 90 minutes. Before you start tweaking aeration timers, you need to understand one thing — a CASS (Cyclic Activated Sludge System) reactor is not a small continuous-flow plant. It is a five-phase sequencing batch reactor variant, and most of the diagnostic habits you learned for conventional activated sludge do not apply here.
CASS operates on a fixed time cycle: fill → react → settle → decant → idle, as described in the CASS reactor configuration patent CN103553280A. There is no return activated sludge (RAS) line, no internal recycle, and no steady-state influent. Everything is controlled by phase timers and level setpoints. That means a "return rate" calculation, F/M ratio, or hydraulic residence time check from a continuous-flow playbook will send you down the wrong path inside the first ten minutes of troubleshooting.
Modern CASS plants in 2026 typically run 3–4 cycles per day, with cycle times between 4 and 8 hours depending on influent strength and temperature. A municipal plant at 15–20 °C usually settles on a 6-hour cycle (1 h fill, 3 h react, 1 h settle, 0.5 h decant, 0.5 h idle), while a high-strength industrial stream may need 8 hours. A second key difference from classic SBR: CASS includes a pre-anaerobic selector zone at the inlet, which gives it bio-P removal capability and changes how filamentous bulking presents. Bulking in CASS is more often tied to selector hydraulics than to dissolved oxygen alone, because the selector is where the floc-formers and filamentous organisms first compete.
The CASS Symptom → Cause → Fix Matrix
This is the table you run to at 2 a.m. Six symptoms, ranked by frequency in municipal and industrial CASS plants, each mapped to the two or three most probable causes and the one parameter check that confirms it. Before you pull anything apart, run the "three-number test" — SVI, SVI₃₀, and zone settling velocity (ZSV) on a 1 L graduated cylinder. It takes ten minutes and will resolve roughly 60% of settle-phase complaints before you touch the PLC.
| Symptom | Probable cause | Confirm with | Immediate fix |
|---|---|---|---|
| High effluent TSS (>30 mg/L) | Filamentous bulking; sludge blanket too high at decant | SVI > 250 mL/g; blanket within 0.3 m of decant weir | Extend settle time 20–30 min; reduce decant volume |
| High effluent NH3-N (>5 mg/L in summer, >10 mg/L in winter) | Aeration failure OR nitrifier washout | DO < 1.5 mg/L = aeration; SRT < 8 d = washout | Restore DO to 2.0 mg/L; or raise SRT by reducing wasting |
| High effluent TP (>2 mg/L) | Selector bypass; excessive NO₃ in return from previous cycle | Ortho-P > 4 mg/L in mixed liquor; ORP < −100 mV not held | Restore anaerobic selector retention; check idle-phase sludge loss |
| Foam carryover (especially brown, stiff, or white billowing) | Surfactant shock; nocardioform bloom; high SRT | Inlet surfactant spike via conductivity tracer; SVI > 200 | Install a rotary mechanical bar screen for CASS headworks if FOG/surfactant is bypassing; dose anti-foam |
| Turbid supernatant during decant (>30 NTU) | Straggler floc; denitrification floaters; weir misalignment | NO₂ > 1 mg/L rising at end of react; weir level ±5 mm | Re-level weir; shorten react phase 15 min before settle |
| Sludge washout (MLSS drops > 15% in 48 h) | Hydraulic overload; excessive decant volume; bulking overflow | Daily flow > 110% of design; decant > 30% of working vol. | Reduce decant; add PLC-controlled chemical dosing for nutrient removal to recover nitrifiers |
On foam specifically, the IJRTE review on MBR/SBR-class reactors found that pre-treatment "significantly reduces" foaming caused by surfactants (IJRTE, 2021, doi:10.35940/ijrte.d6591.1110421). The mechanism is identical in CASS: the more surfactant that reaches the aeration zone, the more stable the foam layer becomes, and the more biomass it strips from the reactor. Headworks screening and equalization are the upstream control points, not the foam-control spray nozzle.
Cycle-Phase Failures: Where Modern CASS Plants Actually Break
Operators describe problems by clock phase — "the settle failed" or "the decant was dirty" — because that is what the HMI shows. The matrix below maps each of the five phases to its typical 2026-era failure mode and the diagnostic that catches it.
| Phase | Typical failure | Root cause | First diagnostic |
|---|---|---|---|
| Fill | Short-circuiting; surfactant slug | Inlet too close to decant zone; shock load from upstream | Conductivity tracer (NaCl pulse) at inlet vs. decanter |
| React | DO never reaches setpoint; phase ends early | Blower VFD fault; probe membrane fouling; PLC scan drift | Pull PLC event log, not HMI countdown |
| Settle | Blanket rise / bridging | Denitrification gas lifting floc; sludge blanket too deep | Check NO₂ > 1 mg/L at end of react; SVI > 200 |
| Decant | Turbid supernatant; floating scum entering decanter | Weir misalignment > ±5 mm; scum baffle missing; over-decant | Manual tape measurement at four points around the weir |
| Idle | Near-zero idle time | Hydraulic overload (influent outpacing cycle capacity) | 7-day influent flow trend vs. design |
The settle-phase bridging failure deserves a second look because it is the most expensive. When the reactor runs out of carbonaceous BOD near the end of react, any residual NO₂ converts to N₂ and the rising gas bubbles attach to floc, lifting the blanket. The blanket then enters the decanter and the effluent TSS spikes. The fix is rarely more aeration — it is a shorter react phase or a small anoxic selector volume upstream of the aeration zone so NO₃ finishes before settle begins.
Operating Setpoints That Prevent 80% of CASS Upsets
Print this table and tape it to the panel. Every row is a band, not a single number, because influent strength and temperature shift the optimum. These are the 2026 working ranges used across municipal and food/beverage/slaughterhouse CASS plants.
| Parameter | Municipal band | Industrial band | Out-of-band action |
|---|---|---|---|
| MLSS | 2,500–4,000 mg/L | 3,500–5,500 mg/L | > 5,500 → expect settle failure; check wasting rate |
| SVI | < 150 healthy; 150–250 warning; > 250 active bulking | Same thresholds | > 300 mL/g for 3 cycles → specialist call |
| DO (aerobic react) | 1.5–2.5 mg/L | 2.0–3.0 mg/L | < 1.0 mg/L → nitrification at risk |
| DO (SND target) | 0.3–0.8 mg/L | 0.5–1.0 mg/L | Only valid if simultaneous nitrification/denitrification is designed in |
| SRT | 10–20 days (summer); 20–25 days (winter) | 15–25 days | < 8 days at 15 °C → nitrifier washout |
| Decant volume | ≤ 30% of working volume per cycle | ≤ 25% for higher MLSS | > 35% → blanket disturbance likely |
| F/M (per cycle) | 0.05–0.15 kg BOD/kg MLSS·d | 0.10–0.20 | Outside band → check fill volume vs. MLSS |
On aeration energy specifically, the 2026 best practice is to run DO at the low end of the band (1.5 mg/L) during carbonaceous removal and ramp to 2.0–2.5 mg/L only in the last 60–90 minutes of react to finish nitrification. Plants that hold DO at 2.5 mg/L through the entire react phase typically spend 25–35% more on aeration kWh with no measurable effluent improvement — see the SBR aeration energy optimization for 2026 engineering guide for the calculation method.
2026 Automation & PLC Faults: The Hidden Majority of CASS Problems
Most CASS plants that go off-spec in 2026 do not have a biological problem first — they have a sensor or PLC problem. Four patterns dominate field service calls and they are all visible in the PLC event log before they are visible in the effluent.
DO probe membrane fouling is the single most common sensor fault. Cleaning cadence should be weekly for municipal influent and 2–3 times per week for industrial (food, textile, leachate). A membrane that has drifted will read high, so the PLC throttles the blower back, and the biology slowly loses nitrification capacity. The polarization voltage check (probe out of liquid, mV reading within manufacturer spec, usually ±10 mV) catches a failing membrane before the effluent does.
Decant level sensor drift — ultrasonic or pressure — causes over- or under-decanting. The 2-point calibration is simple: pull a manual tape at two known volumes (typically 20% and 80% of working volume) and adjust the sensor offset to match within ±5 mm. Do this quarterly, not annually.
PLC cycle-time creep is the silent killer. A 30-minute react phase that "feels" right on the HMI may actually be 22 minutes because the legacy program has scan-time drift, watchdog resets, or unhandled interrupts. Pull the event log, not the HMI countdown — the event log shows the actual phase start and end timestamps.
Remote monitoring in 2026 now catches most of the above before the operator does. Edge gateways flag DO probe drift (rising slope with falling influent load), blower current anomalies, and level-sensor out-of-band events as discrete alarms, turning the plant from "reactive at 2 a.m." into "set and watch." If your plant does not have this layer, it is the single highest-ROI upgrade available in 2026.
When to Bring In a Specialist vs. Fix It Yourself
The boundary between operator-level fixes and supplier intervention is sharper than most troubleshooting guides admit. Use this rule of thumb on the next upset.
Fix it yourself if: it is a single-cycle upset, a single parameter is drifting (one probe, one timer, one phase), or the foam clears within 24 hours after a hydraulic event. These are normal operational disturbances and the symptom→fix matrix above will resolve most of them.
Call a specialist if: SVI stays > 300 mL/g for three or more cycles, the nitrifier population does not recover within 5–7 days of an SRT extension, the decanter has a structural or mechanical issue (weir frame, traveling bridge, scum baffle), or the PLC event log shows a pattern of phase-skips that the operator cannot reconcile with the HMI.
Before you make the call, collect and export: a 7-day trend of DO, MLSS, SVI, effluent NH3-N, and the full PLC event log as a CSV. A specialist who receives this on the first call will resolve the issue in one site visit instead of three. For overloaded CASS reactors where the tank cannot be expanded, a 2026 retrofit option is the MABR retrofit option for overloaded CASS reactors, which adds aerobic and anoxic biofilm surface area inside the existing aeration zone without increasing tank volume. If the carryover is at the clarifier downstream rather than the CASS decanter, the clarifier scum and FOG diagnostic guide covers the parallel failure mode.
Frequently Asked Questions
What is the most common cause of high effluent NH3-N in a CASS reactor?
Aeration failure, not nitrifier toxicity. Check DO first — if it is below 1.5 mg/L in the aerobic react phase, the nitrifier population is being oxygen-starved rather than poisoned. Raising DO to 2.0 mg/L typically recovers nitrification within 2–3 cycles if the SRT is still above 10 days at 15 °C.
How do I tell filamentous bulking from a settle-phase blanket rise in CASS?
Run the three-number test. SVI > 250 mL/g with a clear supernatant at the top of the settle column = filamentous bulking. SVI < 150 mL/g with the blanket rising 30 minutes into settle and NO₂ > 1 mg/L = denitrification floaters, not bulking. The fix is opposite for each: bulking needs selector or chlorination control, while floaters need a shorter react phase or a small anoxic buffer.
What SVI value should trigger wasting changes in a CASS reactor?
SVI < 150 mL/g is healthy and wasting should hold at design rate. SVI in the 150–250 mL/g range is an early warning — increase wasting by 10–15% to drop MLSS back into the 3,000–3,500 mg/L band. SVI > 250 mL/g is active bulking and requires selector hydraulics review, not just more wasting.
How often should DO probes be calibrated in a CASS reactor?
Weekly cleaning and air-calibration for municipal influent, 2–3 times per week for industrial streams (food, textile, landfill leachate). The polarization voltage check should be monthly. A probe that drifts high will cause the PLC to throttle the blower back, slowly starving the nitrifiers before the effluent NH3-N moves.
Can a CASS reactor be retrofitted with MABR to increase capacity?
Yes. In 2026, membrane-aerated biofilm reactor modules are installed inside the existing aeration zone to add nitrification and partial denitrification capacity without expanding the tank volume. The retrofit typically delivers 30–50% capacity uplift for the same footprint, with the tradeoff being higher biofilm module cost and a different PLC control philosophy around air cycling.
