Home/Catalog/Filtration units

Filtration units

Reagent-free filtration is the stage where dissolved iron, manganese and hydrogen sulphide are converted into an insoluble form and retained within a bed of granular media. No oxidant is dosed from outside: the oxygen already dissolved in the water does that work, and the filter media acts as the catalyst.

  1. Coarse strainer
  2. Aeration
  3. pH correction
  4. Filtration units
  5. Softening
  6. Reverse osmosis
  7. UV disinfection

Principle

How it works

The unit is a fibreglass vessel with an inner polyethylene lining, filled with granular filter media; beneath it lie a supporting gravel layer and a drainage and distribution system, and a control valve sits on top. In service the water flows downwards, passes through the bed and leaves through the lower drain. There are no moving parts inside the vessel: the only mechanism in the unit is the valve that switches the flows.

The stage has two jobs at once, and that is what sets it apart from an ordinary mechanical filter. While iron is in its dissolved ferrous form there are no particles in the water at all — there are ions, and they pass through any granular bed without resistance. The substance therefore has to be oxidised first and retained second. Catalytic media does both: oxidation takes place on its surface, and the insoluble precipitate that forms stays within the bed.

The oxidant is the oxygen dissolved in the water itself, and the rate of oxidation depends on how much of it there is and on the pH. That sets a direct limit of applicability: where iron is high or oxygen is short — which is exactly what deep borehole water looks like — the dissolved oxygen is not enough, and aeration is installed ahead of the filter. In such a scheme it is not an improvement but a condition of the filter working at all.

The accumulated precipitate is removed by backwashing: the valve sends the flow upwards, the bed expands, the grains are set in motion and what was retained is carried to drain. The backwash flow rate is set by the cross-sectional area of the vessel, not by the capacity of the unit, and is markedly higher than the service flow. This is a requirement on the source rather than on the equipment, and it is the one most often left out of the calculation.

The control valve determines who starts the backwash and when. An automatic valve counts by timer, or by actual throughput if it has a water meter; a manual one is switched by the operator. The valve type has no direct bearing on filtrate quality — it bears on whether the backwash happens on time. A manual version needs no power supply and costs less, but it needs a routine and someone who keeps to it.

Principle

The counterintuitive part

Clear borehole water that turns brown in a bucket within half an hour is a sign not of a poor filter but of oxidation happening downstream of it. While the iron is dissolved there is nothing to retain: filter media works on particles, and the particles do not yet exist. The precipitate forms in the bucket, in the tank or in the pipework — wherever the water has time to take up oxygen. This is why an ordinary mechanical filter does not help here at all, however fine its rating is made.

Iron and manganese need different conditions even though they are removed at the same stage. Iron is oxidised by oxygen at a pH of about 7 already, manganese needs noticeably more, and in practice its removal starts working from a pH of around 8. Hence a common outcome that looks like a fault: the iron has gone, the manganese has stayed. The media is not to blame, an uncorrected pH is, and the cure is alkali dosing ahead of the filter, not a larger unit.

The capacity of the unit is set by the cross-sectional area of the vessel, not by the figure on the valve. The valve caps the flow from above, but the governing quantity is the filtration velocity, that is, flow divided by area. Exceed it and the water passes the bed faster than it can give up its precipitate, and turbidity breaks through into the filtrate. For that reason the «valve capacity» in the data is an upper limit rather than a working point, and a size cannot be chosen on it alone.

Practice

What that gives in practice

The stage closes the most common set of borehole water deviations in one go: iron, manganese, hydrogen sulphide, turbidity and the colour that goes with iron. A separate device for each parameter is not needed — at moderate concentrations a single bed takes all of them.

No reagents are dosed and none are stored on site. What remains among the running costs is backwash water, the electricity for the valve and periodic replacement of the media; there is no salt handling here of the kind ion exchange requires.

The unit does not change the salinity of the water. Downstream it stays as hard and as mineralised as it arrived: filtration solves the problem of turbidity and oxidisable impurities, but not the problem of salt composition.

The stage protects everything installed after it. Ion exchange resin, membrane elements and a UV chamber fail from iron and suspended solids sooner than they wear out, and the cost of restoring them bears no comparison to the cost of filter media.

The unit is assembled from standard parts — vessel, media, Runxin valve — and supplied in a version matched to a particular flow and connection. We do not manufacture it: our part of the work is the analysis, the calculation of filtration velocity and bed volume, the choice of vessel and valve, supply, commissioning and subsequent service.

Limits

When you need it

  • The borehole water analysis exceeds the limit for iron, and the water goes cloudy or brown on standing.
  • Manganese is present: the water leaves a dark film on sanitary ware and black marks on laundry.
  • There is a smell of hydrogen sulphide and the source is a groundwater intake.
  • Turbidity and suspended solids have to be removed permanently, rather than by a cartridge changed once a month.
  • The stage is needed as pretreatment ahead of softening, reverse osmosis or UV.
  • Reagent handling is unwelcome on site: there is no room for the vessels and no staff to work with solutions.

When it will not help

  • If the problem is hardness and scale. The bed does not hold calcium and magnesium at all; that calls for ion exchange, and substituting one for the other leaves the scale in place with a perfectly sound filter installed.
  • If salinity, nitrates or sulphates are above the limit. Filtration does not reduce those parameters: reverse osmosis is needed, or ion exchange on a suitable resin.
  • If the iron content is high and there is no aeration in the scheme. The oxygen dissolved in the water will not be enough for oxidation, the media will stop coping, and backwashes will become daily without any improvement in the result.
  • If the source cannot supply the backwash flow. A borehole that yields exactly the design consumption cannot backwash the filter, and the bed will blind within the first few months.
  • If the colour comes from dissolved organics — the typical case for surface and bog water intakes. Catalytic media does not remove it; adsorption or coagulation does.

Practice

Typical cases

Feed
Private house borehole: iron 3.2 mg/l, manganese 0.3 mg/l, hydrogen sulphide noticeable by smell, pH 7.2, demand 1.5 m³/h.
Task
Bring the water within drinking limits for iron, manganese and smell without reagent handling in the plant room.
Scheme
Strainer → pressure aeration → reagent-free filtration on catalytic media → UV disinfection.
Result
Iron and hydrogen sulphide are removed at a single stage. At a pH of 7.2 manganese is only partly removed, and if the analysis after commissioning shows a breakthrough, alkali dosing is added ahead of the filter — the point is provided for in advance, since cutting into finished pipework costs more.
Feed
Mains water at a small production site: iron 0.6 mg/l after old networks, variable turbidity, 6 m³/h on one shift.
Task
Obtain water of constant clarity for a process line and protect the equipment behind it.
Scheme
Reagent-free filtration with a meter-controlled automatic valve → softening → process line.
Result
A metered valve was chosen because the flow is uneven: time-based counting on a single-shift regime would backwash the filter for nothing at weekends and too late on peak days. The backwash runs outside working hours.
Feed
A 20-bed guest house on a borehole: iron 5 mg/l, virtually no oxygen in the water, peak draw-off morning and evening.
Task
Supply water without interruptions at peak draw-off with a limited borehole yield.
Scheme
Aeration column → reagent-free filtration → storage tank → booster set → UV disinfection.
Result
The tank is there not as a reserve but as a decoupler: it lets the filter run at a steady flow against an uneven draw-off and frees the borehole yield for backwashing during the night.

The cases are typical examples, not site reports.

Scheme

Place in the scheme

Reagent-free filtration is rarely a stage on its own. A coarse strainer goes ahead of it — it takes out the sand and scale that damage the control valve — and the rest of the pretreatment is set by the analysis: aeration where oxygen is short or iron is high, pH correction where manganese is present.

Downstream of the filter come the stages it protects: softening, reverse osmosis, UV. The order here cannot be rearranged — ion exchange resin and membrane elements take the iron and suspended solids on themselves and fail early, and a UV lamp loses its effect to turbidity, which absorbs the radiation.

The backwash line is treated as part of the scheme, not as an auxiliary drain. Its flow is calculated, pipework of a matching diameter is provided and a discharge point is chosen; backwash water carries the retained precipitate and is not returned to a recirculating system without settling.

On sites with continuous draw-off a single unit stops the supply while it backwashes. This is decoupled either by a storage tank downstream or by two vessels working alternately; the choice between them is a matter of flow, space and how acceptable an interruption is, and it is settled at the scheme stage.

Scheme

Automation and control

  • Counting the interval between backwashes. By timer where the flow is steady and predictable; by water meter where it is not. The metered version costs more, but on sites with a variable load it pays for itself in backwash water that would otherwise be spent for nothing.
  • A ban on backwashing during draw-off hours. A backwash takes from ten minutes to half an hour, and the supply stops for that time; an automatic valve is given a night window.
  • Monitoring the pressure drop across the unit. A rising drop at an unchanged flow means the bed is blinding and is grounds for an out-of-turn backwash long before turbidity appears in the filtrate.
  • An interlock on no water at the inlet. Backwashing an empty vessel does not restore the bed and resets the counter, which means the next backwash will be doubly overdue.

Operation

Running it

Energy

The unit consumes little of its own: electricity goes only to the control valve drive, which runs for minutes a day, and a manual version needs no power at all. The significant running item here is a different one — backwash water. Depending on the size and the frequency of backwashes it amounts to anything from a few per cent to a tenth of the volume passed, and on sites with a limited source yield it is this consumption that decides the scheme.

Consumables

  • Filter media: catalytic beds last several years, and the life is set by grain attrition and loss of activity rather than by the calendar; the time to replace shows in backwashes growing more frequent at unchanged water quality
  • The supporting gravel layer — topped up when the media is replaced, with no separate life of its own
  • Seals and drive parts of the control valve: a repair kit covering several years of service
  • Reagent for pH correction, where the scheme includes a dosing point for manganese removal
  • Water analyses upstream and downstream of the stage: without them the backwash routine and the media replacement date stop resting on data

What goes to drain

What is removed leaves with the backwash water: iron and manganese hydroxide sludge, retained suspended solids and some attrited media. Backwash water is turbid, coloured and discharged in a surge — over a short time a volume many times the service flow goes to drain. Where the discharge goes to a septic tank or a drainage field this is accounted for separately: a surge stirs up the sludge in the septic tank and carries it further down the scheme. The sludge is not toxic, but its volume and turbidity call either for a discharge point of its own or for a receiving chamber with settling.

Sizing

Sizing

What we account for

  • Feed water analysis: iron, manganese, hydrogen sulphide, turbidity, oxidisability, pH, dissolved oxygen content, hardness
  • Design and peak flow of the site, and the pattern of draw-off through the day
  • Source yield and inlet pressure: backwashing needs a flow above the service rate, and it is the source that has to provide it
  • Whether an interruption of supply during backwash is acceptable: it decides whether a tank or a second vessel is needed
  • Backwash discharge conditions: where the water goes and whether the receiver takes a surge
  • The site: headroom for the vessel and media, access for media replacement, temperature regime
  • What comes after the filter: the requirements of softening, membranes or UV for their feed water decide how far the water has to be cleaned here

What a wrong choice costs

  • Sizing on flow without regard to filtration velocity. A vessel chosen on the valve's data sheet figure turns out too small in cross-section, and turbidity breaks through into the filtrate with the unit formally in order.
  • No aeration where iron is high. The dissolved oxygen is not enough, oxidation is incomplete, and part of the iron passes straight through the filter, settling in the tank and the pipework instead.
  • Backwash flow left out of the calculation. A source chosen to match consumption exactly leaves nothing to backwash the bed with; it blinds, the pressure drop rises, and the unit loses capacity within the first months.
  • Expecting manganese removal without pH correction. At a pH of about 7 manganese is only partly removed, and the analysis after commissioning does not match the expectation — with not a single part of the scheme faulty.
  • A manual valve where there is nobody to look after the unit. Missed backwashes damage the bed irreversibly: a blinded layer is not fully restored even by several backwashes in a row.
  • Backwash discharge into a septic tank without a receiving chamber. The surge stirs up the sludge and carries it into the drainage field, and the consequences show not at the filter but in the drainage system.

Sizing

What we need for a calculation

  • Water analysis report: iron, manganese, hydrogen sulphide, turbidity, colour, oxidisability, pH, hardness, dissolved oxygen
  • Flow: daily average, design hourly and peak, together with the draw-off pattern
  • Source: borehole or mains, yield, pressure and its seasonal variation
  • Whether an interruption of supply during backwash is acceptable and whether the site has a storage tank
  • Where the backwash water goes and what surge flow the receiver will take
  • The site: headroom, floor area for the equipment, temperature regime, power supply
  • What is planned after the filter — softening, reverse osmosis, disinfection

Questions

Questions

Why is the water clear at the tap and rusty in the storage tank?

Because iron in borehole water is dissolved, not suspended. In that state it passes any granular bed and any cartridge: the particles that could be retained do not yet exist. The precipitate forms later, when the water takes up oxygen — in the tank, in the bucket, in the pipework. The purpose of the stage is precisely to make the oxidation happen inside the filter rather than after it, and that is achieved with catalytic media, with aeration ahead of it where required.

What difference does an automatic valve make against a manual one?

None to the quality of the filtrate, as long as the backwash happens on time. The difference is in who is responsible for that. An automatic valve counts by itself, by timer or by water meter, and does not depend on anyone being present; a manual one costs less and works without power, but a missed backwash damages the bed irreversibly — a blinded layer is not restored to its original state even by several backwashes in a row. A manual version makes sense where the flow is predictable and servicing follows a routine.

How often is the filter media replaced?

Catalytic beds last years, and the life is decided not by the calendar but by the working conditions: grain attrition during backwashes, the actual iron and manganese load, how complete the backwashes are. The approach of the limit shows in advance — backwashes become more frequent, the pressure drop rises, and filtrate quality at the same regime starts to drift. Replacement is planned on those signs rather than waited for through worse water at the tap.

Does this stage remove hardness?

No. Calcium and magnesium pass the bed unchanged, and the scale after filtration is the same as before. Hardness is removed by ion exchange or by reverse osmosis, and that is a separate stage installed after the filter rather than instead of it. Substituting one for the other is a common reason for water that has become clear while the kettle still furs up.

How much water does a backwash use?

The backwash flow is set by the cross-sectional area of the vessel and comes to several times the service flow of the unit; a backwash lasts from ten minutes to half an hour. Against the volume passed that works out at anything from a few per cent to a tenth — the figure depends on the size and on how often the bed has to be washed with the water in question. The requirement falls on the source: the borehole has to deliver that flow, otherwise the backwash will be incomplete.

Can aeration be left out?

It can, if the oxygen dissolved in the water is enough to oxidise all the iron and manganese and their content is moderate. That is settled by an analysis, not by a general rule: deep borehole water usually has almost no oxygen at all. Where there is not enough, the catalytic media works below its capability, some of the iron passes through and precipitates beyond the filter, and backwashes grow more frequent with no improvement in the result.

Next

Start a project

Let's talk about your project

Tell us about your site — our engineer will prepare a process design, a specification and a quotation within one business day.

✓ Free calculation✓ Site visit

By submitting the form you agree to the processing of your personal data. Privacy Policy