TROUBLESHOOTING
PAC Not Working? Seven Checks Before You Blame the Chemical
Small pin flocs, a hazy supernatant, or a clarifier that was fine last month and is not fine now - these are the most common calls a coagulant supplier receives. In most cases the product is doing what it was made to do and something around it has changed. Work through the seven checks below in order; the first three solve most cases.
Start with the symptom, not the bag
Before touching the dosing pump, write down what you can actually see. The symptom narrows the cause faster than any specification sheet.
| Symptom | Most likely cause | First check |
|---|---|---|
| Pin flocs, hazy supernatant | Underdose, or pH outside the working window | Jar test at three doses; pH before and after |
| Turbidity worse after dosing | Overdose and charge reversal | Halve the dose and repeat |
| Flocs form, then break up in the flocculator | Mixing energy too high | Impeller speed and tip speed |
| Good jar test, poor plant result | Short retention time, short-circuiting, wrong injection point | Dye or tracer test |
| Slow floc only in cold months | Low water temperature | Water temperature; dose adjustment |
| White deposit in lines and strainers | Undissolved solids or hard dilution water | Make-up procedure; strainer inspection |
| High residual aluminium in the outlet | pH drift or overdose | pH profile through the plant |
1. Dose: the jar test still decides
Underdosing leaves charge on the colloids and you get small, weak flocs that do not settle. Overdosing is worse in a subtle way: the particle surface charge reverses, the colloids repel each other again, and turbidity climbs while you keep adding chemical. Operators rarely suspect the overdose, because the instinct is always to add more.
Run three doses - your current one, half of it, and one and a half times. If the half dose gives a clearer supernatant, you were overdosing. If none of the three works, the problem is pH or mixing, not the quantity.
2. pH and alkalinity
PAC works across roughly pH 5.5 - 8.5, wider than alum, but the useful window for charge neutralisation is pH 6.5 - 7.5. Below 5.5 the aluminium stays soluble and never builds a proper floc. Above 8.5 aluminate re-forms and residual aluminium rises.
Alkalinity is the hidden constraint. PAC hydrolysis consumes bicarbonate, and soft water below about 50 mg/L as CaCO3 can lose a full pH unit across the coagulation stage at higher doses. If the pH after coagulant addition is below 6, dose alkalinity first - lime or soda ash - rather than more coagulant.
3. Mixing energy
Coagulation needs violent, brief mixing: 30 - 60 seconds at a velocity gradient around 250 - 800 s-1, enough to disperse the product before it hydrolyses in a pocket of concentrate. Flocculation needs the opposite: 20 - 70 s-1 for 15 - 20 minutes, just enough for particles to collide without tearing the flocs apart.
A floc that forms in the beaker and disappears in the flocculator is usually being sheared. Check impeller tip speed, not only rpm, and check whether the last compartment runs faster than the first.
4. What changed in the raw water
Coagulant demand is a property of the water, not of the plant, and raw water is never stable for long. The usual suspects:
- A storm event - turbidity up, pH down, and a load of fine clay that behaves nothing like the usual solids
- An algal bloom - cells and extracellular organics consume coagulant before they form floc, and the dose can double
- A new industrial discharge into the sewer - surfactants, oils and complexing agents all interfere
- Seasonal turnover in a reservoir - manganese and organics arrive together in autumn
- Water below 10 °C - hydrolysis slows, flocs form later and smaller, and the dose usually needs 10 - 20% more plus longer flocculation time
5. The product itself
Not every failure is external, and there are three product-side causes worth ruling out early. Basicity mismatch comes first: a product with a much higher or lower basicity than the one you qualified will behave differently in the same water. Batch variation comes second - compare the current COA with the previous one rather than with the catalogue. Storage damage comes third: powder that has caked no longer dissolves fully, and the undissolved fraction never reaches the water.
Test one kilogram from the suspect batch against a retained sample of a good batch in the same water. Ten minutes of work settles the question.
6. Dosing equipment and make-up
A correct dose delivered badly produces the same symptoms as a wrong dose. These five checks cover most of it:
- Working solution strength - 5 - 10% as product is the usual band; above 10% the solution becomes viscous, crystallises in lines and doses unevenly
- Dilution water - hard water can precipitate aluminium at the point of dilution, especially with high-basicity product
- Pump calibration - measure the actual delivered volume over ten minutes, do not trust the stroke setting
- Strainers and non-return valves - undissolved solids collect here first, and a partly blocked line delivers a pulsing under-dose
- Crystallisation in winter - outdoor lines on liquid product need lagging or trace heating
7. Sequence and interference
The order of addition matters as much as the dose. PAC goes first, with rapid mixing; PAM goes after, with gentle mixing. If PAM reaches the water before neutralisation is complete, the polymer bridges particles that are still charged and the flocs come out small and fragile.
Watch for oxidants as well. Sodium hypochlorite or permanganate dosed upstream attacks polymer chains and reduces PAM performance, and a heavy oxidant dose can change the organics in the water into a different coagulation problem altogether. Lime belongs before PAC, not after it.
The order to work in
The sequence below takes about half an hour and costs nothing. Changing supplier costs a container of product and two weeks of shipping, and usually solves nothing.
- Look at the floc and the supernatant, and write down which symptom you actually have
- Run a three-point dose test at the current pH
- Measure pH and alkalinity before and after coagulation
- Check rapid-mix time and flocculator tip speed
- Ask what changed upstream in the last two weeks
- Compare the current COA with the previous batch
- Calibrate the dosing pump and inspect the strainer
- Only then ask for a different grade or a different supplier
Frequently asked questions
How do I know whether I am overdosing?
Two signs: the supernatant gets worse, not better, as you add product; and flocs float rather than settle. Confirm with a halving test - if the half dose clears better, the higher dose was the problem.
Can I fix low alkalinity by adding more PAC?
No. More coagulant consumes more alkalinity and pushes the pH down further. Add alkalinity with lime or soda ash ahead of the coagulant, or move to a higher-basicity grade that consumes less alkalinity per milligram of Al2O3.
Why does it work in a beaker and fail in the plant?
Beaker tests use ideal mixing and unlimited time. In the plant the usual culprits are short retention time, short-circuiting in the flocculator, an injection point too close to a pump, or flocs sheared by a transfer pump.
Does freezing damage PAC?
Solid product is not damaged by frost, but caking gets worse afterwards. Liquid product can separate or crystallise; thaw it slowly to 15 - 20 °C, mix, and jar test before returning it to service.