The short answer

Water hammer kills polyphosphate filters in two different ways, and installers usually check only one of them. The obvious risk is pressure: a surge can crack a housing, blow a thread, or unseat an O-ring. The less obvious risk is mechanical damage to the crystal bed itself. Polyphosphate media is a soluble salt, not a durable filter granule. Repeated shock, vibration, and flow reversal fracture the crystals, create fines, and open channels through the bed. After that, the unit may still hold pressure and still look intact, but its dosing behaviour has changed.

So the practical answer is: before commissioning a polyphosphate filter on any line with solenoid valves, pumps, quick-closing valves, or check valves, measure the actual peak pressure at the filter, identify where the surge is generated, compare the peak against the temperature-derated rating of the housing, verify how the filter is supported, and inspect the crystal bed after any surge event. The rest of this article explains how to do each of those checks.

Why a polyphosphate filter is more vulnerable than it looks

The housing is not a surge vessel

Most polyphosphate units sold into residential, light commercial, and light industrial duty use either a cartridge-style housing or a small media tank. These are designed for steady working pressure, not for transients. Thermoplastic housings in particular lose strength as water temperature rises, so a housing that is comfortable at a given cold-water pressure may have far less margin at 40–60 °C. When a surge arrives, the weakest points are usually the bowl thread, the O-ring seat, and the inlet/outlet boss — not the body wall.

The media is a soluble crystal, not a filter granule

Siliphos-type polyphosphate crystal works by controlled dissolution. Water passes through the bed, a small amount of phosphate enters solution, and the phosphate then interferes with calcium carbonate crystal growth downstream and can form a thin adsorbed film on metal surfaces. That is scale inhibition and, to a limited degree, corrosion inhibition. It is not water softening — hardness ions are not removed — and it is not filtration in the sense of removing suspended solids, and it is certainly not disinfection.

Because the media dissolves rather than traps contaminants, the bed’s physical condition matters as much as its chemical composition. A fractured crystal has more exposed surface area and may release phosphate faster; a channelled bed has less contact time and may release less. Either way, the unit stops behaving the way it was selected to behave.

Name confusion leads to wrong expectations

Siliphos is a brand name for a particular family of polyphosphate crystal. “Phosphate crystal”, “polyphosphate media”, and other trade names used in the market are not automatically interchangeable — composition, dissolution rate, particle size distribution, and approvals can differ. When specifying or replacing media, confirm what the original unit was designed for rather than assuming any phosphate crystal will do.

Where the surge usually comes from

Installers often blame the filter because that is where the leak appears. In practice the source is almost always elsewhere on the line:

• Solenoid valves on RO units, irrigation zones, dosing skids, and appliance feeds. These close in milliseconds.
• Pump start/stop, especially with a non-return valve that slams shut on shutdown.
• Quick-closing ball valves and quarter-turn isolation valves operated by hand during commissioning or servicing.
• Fast-closing check valves, including a check valve installed right at the filter outlet to prevent backflow.
• Pressure-reducing valve chatter or a PRV that fails partially open.
• Thermal expansion in a closed system — a backflow preventer or PRV plus a water heater can raise line pressure with no valve closing at all.
• Entrained air after maintenance. Air pockets collapse and re-expand, amplifying every other surge source.

The important consequence: mitigation placed at the filter does not remove the energy from the line. A hammer arrestor belongs near the valve that causes the surge, sized for the fixture units and pipe size it serves, not screwed onto the filter inlet as an afterthought.

How much surge are you actually dealing with?

For a first estimate, surge pressure is approximately fluid density × pressure-wave speed × the change in flow velocity. Two practical points follow from that:

• Wave speed depends heavily on pipe material. In rigid metal pipe it is roughly 1,000–1,200 m/s; in plastic pipe it is far lower, often 200–500 m/s. This is why the same valve closing on the same flow produces a much gentler surge in PEX or PE than in steel or copper.
• Closure time matters. If a valve closes slowly relative to the round-trip time of the pressure wave along the pipe run, the surge is greatly reduced. This is why a slow-closing valve or a soft-seated solenoid is often the cheapest fix.

As a rough rule of thumb, abruptly stopping 1 m/s of flow in rigid metal pipe can generate a transient on the order of 10 bar above the static line pressure. That is enough to explain a cracked bowl on a system whose gauge reads a comfortable 4 bar all day. Do not rely on a rule of thumb for a final decision, though — measure it.

Measurement is straightforward: fit a peak-holding gauge or a fast-logging pressure transducer immediately upstream of the filter, then operate the suspect valves and pumps through their normal cycles several times. Static pressure readings taken with everything closed tell you almost nothing about what the housing experiences.

Installer checklist

Fixing the cause, not just the symptom

In order of effectiveness and cost, the usual measures are:

1. Slow the closure. Replace hand-operated ball valves used as routine shutoffs with slow-closing types, or specify soft-seated solenoids with adjustable closing time.
2. Install a correctly sized arrestor at the source. Pre-charged bladder arrestors are sized by fixture units and pipe size, and they must be charged to the correct pre-charge pressure to work.
3. Control closed-system pressure. Verify the PRV setting and confirm thermal expansion is being absorbed, not stored in the pipework.
4. Remove air. Purge thoroughly and fit vents at high points on long runs.
5. Support the installation properly. A hanging filter bowl on an unsupported pipe run will eventually fail at the threads regardless of surge.
6. Only then consider a stronger housing. Upgrading to a higher-rated body helps, but if the transient still exceeds the rating, you have moved the failure point downstream rather than removing it.

What to confirm before you buy or replace media

Once the mechanical issue is understood, the purchasing decision usually comes down to a few items that only the supplier can confirm for the exact product:

• Whether the media is genuine Siliphos-type crystal or a different polyphosphate chemistry, and what that means for dissolution behaviour in your water.
• The housing’s pressure rating and its temperature derating curve.
• Whether the exact product, in the exact application, holds the documentation required for potable water in your destination market. Food-grade status alone does not establish drinking-water suitability.
• The correct media form for your unit — cartridge, granular fill, or bulk crystal — and whether the two are interchangeable.
• Service life and replacement criteria expressed in terms of your flow rate and water chemistry, not a generic figure.

Be cautious with any supplier who quotes a service life or a dosage without knowing your hardness, flow, temperature, and consumption pattern. Those inputs change the answer substantially.

FAQs

Can water hammer really crack a polyphosphate filter housing?

Yes. Repeated moderate surges cause fatigue at the bowl thread and O-ring seat even when no single event exceeds the burst rating. A single large transient — typically from a fast-closing solenoid or a pump trip — can split a housing outright. Peak measurement is the only reliable way to know which situation you have.

Does water hammer change how much phosphate dissolves?

It can, but not predictably. Fractured media exposes more surface area, which tends to increase release; channelling reduces contact time, which tends to decrease it. The net effect depends on the bed and the flow pattern, so inspection rather than assumption is the right approach.

Will a water hammer arrestor protect the filter?

Only if it is correctly sized and installed near the source of the surge. An undersized or wrongly charged arrestor, or one fitted at the filter rather than at the offending valve, provides little protection.

Is a polyphosphate filter the same as a water softener?

No. It does not remove calcium or magnesium. It inhibits scale formation and can provide limited corrosion inhibition by altering how scale crystals form and by adsorbing onto metal surfaces. If you need hardness removal, you need ion exchange or another softening process.

Does the polyphosphate filter itself cause the water hammer?

Rarely. The media and housing add a small pressure drop but do not generate surges. The exception is when an installer fits a fast-closing check valve immediately at the outlet — that component, not the filter, becomes the hammer source.

Talk to us about your specific installation

If you are specifying or troubleshooting a polyphosphate filter on a line with known surge problems, ENVO CHEMICAL can discuss the media and housing options that fit your situation. To give you a useful answer rather than a generic quotation, please include: the application and equipment type, source-water hardness, pH and temperature, design flow rate and pipe material, measured static and peak pressures at the filter location, the housing model currently in use, required media form, and the quantity and destination market you are buying for. With those details, we can confirm what is technically appropriate and what still needs to be verified for your application.