The idea that we need a filter downstream of the polyphosphate injection has it backwards. Polyphosphate is a sequestrant — it dissolves completely and its whole job is to keep dissolved minerals from ever becoming particles. Here's the chemistry, plus why our injection order rules out the one theoretical exception.
Narrated overview of this module. The article below has the full procedure and the numbers.
Polyphosphates (hexametaphosphate and the blended “poly” products we feed) are fully water-soluble. Injecting the solution adds dissolved ions to the water — nothing solid. No solid forms, no floc, no cloudiness, no “sediment” materializes from the injection itself. A chemical can only load a filter if it creates particles, and this one is engineered to do the opposite.
Polyphosphate is a sequestrant. Its chain structure wraps around dissolved calcium, iron, and manganese and holds them in solution — so they can't crystallize into scale, and dissolved iron can't grow into visible red-water particles. That's precisely why we dose it: it's threshold scale inhibition and iron/manganese sequestration in one feed.
Dissolved Ca²⁺ and Fe²⁺ are free to find each other, crystallize, and grow into scale & rust particles over time.
Each metal ion is wrapped by the polyphosphate chain — sequestered, invisible, and still 100% dissolved.
The “filter after chemical injection” rule is real — for the family of chemicals whose purpose is to turn dissolved stuff into solids you can catch. Polyphosphate is in the other family. Knowing which family a chemical belongs to tells you whether a filter follows it:
The one legitimate way pH adjustment can create particles is a localized hot spot: right at an injection point, before mixing, a small zone of very high pH can push CaCO₃ out of hard water. That's an injection-point mixing problem — and it's exactly what our sequence eliminates:
Notice the order isn't an accident: the caustic sets pH first and blends completely, then the poly arrives to lock up whatever calcium and iron the water is carrying. The two feeds are complementary — the second one actively protects against the only failure mode of the first.
There are real things to watch with polyphosphate — they're just operations items, not sediment items:
Fully dissolve the product before it goes to the pump. Undissolved chemical slugs are a feed-prep problem — the fix is mixing and makeup practice, not a downstream filter.
In very long residence times or hot water, polyphosphate slowly reverts to orthophosphate and loses sequestering power. Watch dead legs and storage turnover — still nothing a filter would fix.
Poly holds dissolved iron. If the raw water already carries oxidized red-water particles, those existed before the chemical — that's a source-water filtration question, decided on its own merits.
Sequestration is stoichiometric-ish: enough poly for the Fe/Mn/hardness present. Underdosing lets breakthrough happen slowly in the distribution system, not at the injection point.
Filters follow chemicals that make particles. Polyphosphate is the chemical we inject so particles never form.
