Why power stations underinvest in front-end water treatment

By Dan Norman, Sales Manager for Water Process Systems at PureTec Separations Ltd

Most power stations focus their water treatment attention on the back end. Reverse osmosis (RO) membranes, resin beds, condensate polishing. That makes sense. Those are the systems that produce the high-purity water the plant actually needs, and they tend to be where the alarms ring loudest when something goes wrong.

But the front end, the point where raw water first enters the site, is where most of the trouble starts. Poor intake water drives up fouling, lifts chemical consumption, shortens membrane and resin life, and quietly drags on maintenance budgets until something fails.

I have spent 26 years working in soil and water, the last decade-plus selling filtration kit into UK industrial sites. In that time I have lost count of the operators who called me about a back-end problem only to find the answer was sat at the inlet. Below is what a properly designed front end looks like in 2026, and why it is worth your attention this year.

What raw water actually contains

The water arriving at a CCGT (Combined Cycle Gas Turbine) station, a paper mill, a food and beverage plant or a data centre on river abstraction is rarely as clean as the schematic suggests. 

It carries a mix of sand and silt from boreholes, organic fines and algae from rivers and lakes, plastic, leaves and other surface debris, fibrous material that wraps around screens, and invasive species like zebra and quagga mussels (which now reach as far up the Thames as Oxford). Add eels and elvers at unscreened intakes and you have the full picture.

Each of these behaves differently. Sand sinks, fibres wrap around screens, and algae deform and squeeze through anything fine enough to catch them. Mussel veligers slip through anything coarser than 40 microns and colonise heat exchangers downstream. 

I’ve seen plants spend serious money rebuilding heat exchangers after a single season of unscreened veliger ingress, when a properly sized intake screen would have stopped the problem at the source. 

Treating all of it with the same kit, or hoping the downstream filters will catch up, is what creates the operational headaches you actually see.

The back-end takes the hit

When the front end is undersized or missing, the cost shows up later in places that are harder to fix.

Ultrafiltration (UF) and RO membranes foul faster. Backwash frequency climbs, chemical cleaning intervals shorten and resin beds need replacing earlier than they should. Heat exchangers get colonised, and boilers see more carryover. None of these failures shout loudly on day one. They show up as a steady increase in OPEX, more service callouts, and the kind of unplanned downtime that gets explained away as "old kit" when it is actually starvation at the inlet.

The reverse is also true. A properly designed front end takes load off everything that follows. Less backwashing, lower chemical use, longer membrane and resin life, and fewer late-night calls.

There is another wrinkle worth flagging. A lot of UK industrial intake kit is based on 30 year old technology and runs as a single point of failure with no backup. When it goes down, the back end goes down with it. Sites tend to live with this because the kit is old but still working. The day it stops is the day they find out the back end has nothing to fall back on.

What PureTec now offers at the intake

We have spent the last year extending our range upstream. PureTec now covers the full filtration train from 10mm at the intake down to 1 micron, before handing over to the fine filtration and membrane systems we have always offered. 

That includes:

• Self-cleaning intake screens

Hydraulically powered using return feed water, no electrical drive needed. Designed as N+1 so flow holds if one screen fouls. Keeps eels, fish, leaves and large debris out of the pumps. The eels swim away rather than dying against a fixed screen, which has environmental and legal implications for protected populations.

• Hydrocyclones

Simple tube design with one moving valve. Removes up to 97 percent of dense particles (specific gravity above 2.0) such as sand and rust. Ideal for sandy boreholes running at constant flow.

• Media filters

Large vessels filled with a filtration medium, usually sand or glass. They come in all sorts of shapes and sizes, and can have more than one grade of sand or glass in them to aid filtration. Very simple designs and very simple to operate, they do require more water to backwash than screens or disc. 

• Disc filters 

This is the workhorse for surface water. Stacks of polypropylene discs trigger backwash at 0.5 bar differential pressure, one pod at a time, so you keep 100 percent forward flow through the cleaning cycle. 

Typical range is 200 microns down to 20 microns. Air-assisted versions cut backwash water from around 30 litres to 12 litres per cycle. 

I have specified discs and screens across most of the UK industrial water market over the last decade, and the kit only earns its keep if it gets cleaned reliably and the spares arrive on time. 

We have partnered with Azud, a Spanish manufacturer, because of their quality and the security of supply they guarantee. That matters to us, having seen large periods of disruption hit other parts of the market.

• Drum filters

Low-pressure operation, vast surface area, ideal for bulk removal of biological material before final polishing.

• Band filters

Where backwash water has value, band filters recover it instead of sending it to drain. 

They also capture solids that often have scrap value (metal particles from milling are the classic example), turning what would normally be waste and a disposal cost into a small revenue stream. 

This category doesn’t come up in every conversation, but where it fits the economics are clean.

• Wedge wire and screen filters

Worth knowing about, with one real caveat. Long thin particles can pass at angles up to ten times the nominal rating, so they are not always the right answer for fibrous loads.

Picking one over the others is the wrong question. The job is to match the technology to the particle, the flow profile and the duty downstream.

Find out more about full coarse filtration here.

Why feed analysis is where this starts

Before any of the above gets specified, you need data on what you are actually treating.

That means Total Suspended Solids (TSS) in mg/L from a proper lab analysis (not turbidity, the two are different). It also means Particle Size Distribution (PSD), which tells you what is actually getting through your existing kit and what is not. A wider water quality classification helps too, because one "very poor" reading flips the whole rating.

Without this, the specification is a guess. A 50 micron filter sounds reassuring until the PSD shows 99.4 percent of your particles already pass through it. At which point you are spending capital to capture less than one percent of the load, and leaving the real problem untouched. 

I have sat across the table from operators who bought what looked like a sensible system on paper and could not understand why the back end was still struggling. The data on the inlet is almost always where that conversation ends.

There is a similar trap with the filter ratings themselves. A "10 micron filter" on a data sheet might only remove 60 percent of particles at that size. The rest depends on whether the manufacturer is quoting a nominal or absolute rating, and on the test method (most use Arizona Test Dust to a standard, but the percentage removal varies a lot between manufacturers). 

Read the fine print before you buy. The number on the front of the box is the headline. The spec lives further down.

Feed analysis is cheap. The capital decision it informs is not.

Not just power stations

This is a power station blog, but the same logic plays out in other sectors.

In paper mills, river water heading into aging clarifiers loaded at 60 ppm with spikes above 100 ppm is a single point of failure waiting to happen. Pre-filtering before the UF cuts dirt load, backwash water, chemical use and OPEX in one move.

In food and beverage, intake quality changes seasonally and can move product quality with it. Front-end protection stabilises the rest of the train.

In data centres, cooling water increasingly comes from non-mains sources to manage demand and licensing pressure. Coarse and fine filtration at the inlet is how you keep heat exchangers clean and avoid the spiral of chemical dosing that follows.

Each of these operators tends to discover the front end after a back-end failure. The ones who get to it first save themselves the cost of finding out the hard way.

What to do this quarter

If you operate a site that draws from a river, borehole, lake or surface source, and your last serious intake review was more than five years ago, this is the year to revisit it.

Start with a proper feed analysis. TSS and PSD, taken across more than one sampling day to catch variation. Then compare what is arriving at your inlet against what your downstream kit was designed to handle. Most of the gaps live there. Once you have that data, deciding between an intake screen, a hydrocyclone, a disc filter stack or a multi-stage train becomes a much shorter conversation.

One more option to flag: intake screens, disc filter batteries and UF can be containerised. That gives you a way to bridge a decommission-and-rebuild gap without taking the plant offline, and it is a useful route if the site needs treatment capacity faster than a permanent install can deliver. This is something we can help with too.

We are offering free feed water analysis and intake assessments to power, paper, food and beverage and data centre operators throughout 2026. No obligation. 

The point is to know what you are actually treating before you spend on anything else.

If you would like a second pair of eyes on your intake, contact us today. I will scope a screening plan for your site and we can take it from there.

About the author

Dan Norman is Sales Manager for Water Process Systems at PureTec Separations Ltd. 

He brings 26 years of experience in the soil and water industry, with a Postgraduate Diploma in Soil and Water Engineering from Silsoe College (Cranfield University) and a BEng from the University of Warwick. He is also a Certified Irrigation Designer with the Irrigation Association of Australia.

His career has covered municipal water supply, industrial filtration, irrigation system design, oil and gas, and the wider water transfer market in the UK and internationally. Before joining PureTec, Dan spent six years as Technical Sales Manager (UK and Ireland) for a leading global filtration manufacturer, where he specified, installed and serviced intake screens, disc filters, hydrocyclones and self-cleaning systems across the industries covered in this article.

You can reach Dan on 01531 248 372 or email him here.

PureTec Separations Ltd designs, builds and maintains advanced water treatment systems for utilities, power, food and beverage, pharmaceutical, data centre and industrial clients across the UK.