Paul Industries designs and installs process water, evaporation and water recovery systems for Idaho dairy and food plants. There is a fact about dairy processing that changes how a plant in an arid state should think about water entirely: milk is roughly 87 percent water, and a plant concentrating or drying it is separating out an enormous volume of water that arrived in the tanker. In a state that is third in the nation for milk production and short of water, that stream is an asset rather than an effluent.
Request a quote or call 201-450-8280
The water arrives in the tanker
Idaho regained third place nationally for milk production in 2025, producing roughly 18.26 billion pounds and narrowly passing Texas. A substantial share of that milk is processed in-state into cheese, powder and concentrated products, all of which involve removing water.
The scale of the resulting stream is easy to underestimate until the arithmetic is done. Milk is about 87 percent water. Concentrating or drying it therefore liberates a volume of water that is comparable to, and on a powder plant considerably larger than, the plant’s entire fresh water intake for processing purposes.
In a wet climate that stream is a disposal question. In southern Idaho it is a supply question, and that reframing is what makes recovery worth engineering properly rather than treating as an environmental gesture.
The water separated from milk is generally of good quality. It has been evaporated and recondensed, which removes dissolved solids effectively. What it carries is volatile organic material from the milk, any carryover from imperfect separation, and whatever the condensate system itself contributes. Polishing is therefore about organics and microbiology rather than about minerals, which is the opposite of the usual water treatment problem and leads to a different treatment train.
Where the recovered water can go
| Use | Treatment typically needed | Justification burden |
|---|---|---|
| Cooling tower makeup | Filtration and biocide control | Low; an engineering question |
| Boiler feed | Polishing to boiler water standard | Low; conventional water treatment |
| Exterior and non-product cleanup | Filtration and disinfection | Low to moderate |
| CIP pre-rinse | Disinfection with monitoring | Moderate; documented in the food safety system |
| CIP final rinse and product contact | Full treatment to potable standard, monitored | High; continuous demonstration required |
The pattern worth noticing is that the easy wins are large. Cooling tower makeup and boiler feed together account for a great deal of a dairy plant’s fresh water consumption, and neither raises a food safety question at all. A plant that recovers its evaporator condensate purely for utility duties may displace most of its fresh water demand without ever touching the product-contact question.
Going further, into CIP rinses and product contact, is where the Pasteurized Milk Ordinance requirements for water derived from milk become the governing document. Those requirements are specific, and the important engineering consequence is that the system must be able to demonstrate compliance continuously rather than prove it once. That means monitoring with defined limits and, critically, automatic diversion when a limit is exceeded, with somewhere for diverted water to go.
Building diversion in from the start is inexpensive. Retrofitting it into a system designed on the assumption that the water would always be good is awkward, and a system without it will eventually deliver questionable water somewhere it should not have gone.
| Continuous load | Per year | Over ten years |
|---|---|---|
| 25 kW | $16,841 | $168,410 |
| 50 kW | $33,682 | $336,820 |
| 100 kW | $67,364 | $673,640 |
At 7.69 cents per kilowatt-hour, modestly below the 8.13 cent national average (EIA, 2024), Idaho’s power is not the deciding factor in this decision. Water availability is. In a basin where additional supply is difficult to obtain at any price, a recovery project is about whether the plant can expand at all rather than about payback on a utility bill, and that is a different conversation with a different set of numbers.
The heat comes with the water
Evaporator condensate leaves hot, which means a recovery project is also a heat recovery project whether or not anyone designed it that way.
Capturing that heat before the water is cooled for storage is usually straightforward and is frequently the part of the project with the shortest payback. Preheating incoming process water, feeding a hot water system, or serving CIP heating are all natural destinations. The design point to get right is sequence: take the heat out first, then treat and store the water cool, because warm stored water is a microbiological problem that then has to be managed with chemistry or with UV rather than with temperature.
That sequencing decision is one of the more common errors we see. A plant recovers condensate into a warm tank because that is where the pipe reached, then spends the next several years managing the growth that follows. Cooling before storage, and taking the heat somewhere useful on the way, solves both problems with one piece of equipment.
We design and install evaporator condensate collection, heat recovery, polishing and treatment trains, storage and distribution, monitoring and diversion, and the interfaces to utility and process water systems. Product-contact piping is built to ASME BPE and 3-A sanitary conventions with orbital welding to AWS D18.1, weld documentation retained, verified slope and drainability, and passivation to ASTM A967 after cleaning per ASTM A380.
Standards referenced: EIA electricity price data · FDA Pasteurized Milk Ordinance · ASME BPE · ASTM A967 · ASTM A380 · 3-A Sanitary Standards
Frequently asked questions
Do you build process water and recovery systems for Idaho dairy plants?
Yes, across Boise, Twin Falls, Jerome, Idaho Falls, Burley and statewide: evaporator condensate collection, heat recovery, polishing and treatment, storage and distribution, monitoring and automatic diversion, and the interfaces to utility and process water systems.
How much water comes out of milk?
A great deal. Milk is roughly 87 percent water by mass, so a plant concentrating or drying it liberates a volume comparable to, and on a powder plant considerably larger than, its entire fresh water intake for processing. In an arid state that is a supply question rather than a disposal one.
What quality is the recovered water?
Generally good. It has been evaporated and recondensed, which removes dissolved solids effectively. What it carries is volatile organic material from the milk, carryover from imperfect separation, and whatever the condensate system contributes, so polishing targets organics and microbiology rather than minerals.
Where is the easiest place to use it?
Cooling tower makeup and boiler feed. Together they account for a great deal of a dairy plant’s fresh water consumption and neither raises a food safety question, so a plant can displace most of its fresh water demand without ever reaching the product-contact discussion.
Can it be used for CIP or product contact?
Yes, with full treatment and continuous demonstration of compliance. The Pasteurized Milk Ordinance requirements for water derived from milk govern, and the engineering consequence is monitoring with defined limits plus automatic diversion when a limit is exceeded, and somewhere for diverted water to go.
Why design diversion in from the start?
Because retrofitting it is awkward and expensive, and a system built assuming the water will always be good has no safe response when it is not. Automatic diversion is inexpensive at design stage and it is the difference between a monitored system and a monitored system that can act.
Should we recover heat as well?
Yes, and it frequently has the shortest payback in the project. Condensate leaves hot, and preheating incoming process water, feeding a hot water system or serving CIP heating are all natural destinations for it.
What sequencing mistake should we avoid?
Storing the water warm. Plants recover condensate into a warm tank because that is where the pipe reached, then spend years managing microbiological growth. Take the heat out first, send it somewhere useful, and store the water cool. One piece of equipment solves both problems.
Does Idaho’s power cost drive this decision?
Not really. At 7.69 cents per kilowatt-hour, modestly below the 8.13 cent national average (EIA, 2024), energy is not the deciding factor. Water availability is. Where additional supply is difficult to obtain at any price, recovery determines whether a plant can expand at all.
How do I get a quote for an Idaho water recovery project?
Use the form on this page or call 201-450-8280. Useful inputs are milk volume processed and the products made, current fresh water consumption and its cost or availability constraint, where condensate currently goes, and which uses you would want recovered water to serve.
What treatment does condensate of whey water need before reuse?
Evaporator condensate is low in dissolved solids but carries volatile organics and can support microbial growth, so it is typically treated with reverse osmosis or activated carbon and disinfection before reuse, depending on the destination. Treatment is matched to where the water will go.
Which uses are the first to switch to recovered water in a dairy?
Cooling tower make-up, boiler feed after softening, floor and yard washing and the first rinse of CIP, all of which accept water below potable quality and use large volumes. The first uses are chosen for their tolerance and their volume, not for their proximity to the recovery point.
What are the microbial risks in recovered dairy water and how are they controlled?
Condensate is warm and nutrient-bearing, so it supports bacterial growth in storage and distribution; it is controlled by treatment such as membrane filtration or UV, by keeping storage and lines sanitised, and by monitoring. Recovered water is treated as a process water with its own hygiene programme.
What conditions apply to reclaimed water under the milk ordinance?
Reclaimed water may be used for defined purposes with treatment and monitoring that keep it within the ordinance's quality requirements, with product contact and final rinse uses subject to the strictest conditions and documented approval. The ordinance permits the practice within a framework rather than forbidding it.
How is heat recovered alongside water?
Evaporator condensate and pasteuriser cooling water are warm, and recovering that heat into CIP water or boiler feed reduces energy alongside the water saving. Designing the two together avoids recovering water cold and reheating it.
What about whey processing water?
Whey concentration by membranes and evaporation removes water that can be recovered, and whey permeate and its water streams are part of the plant's water balance. Whey processing is where much of the recoverable water is.
How is the water balance built for a dairy plant?
By metering every intake, every recoverable stream and every use, then matching recovered streams to uses by quality and timing. The balance shows how far recovery can go and what treatment each step needs.
What storage is needed for recovered water?
Enough to buffer the mismatch between when water is recovered and when it is used, with the storage designed to be cleaned and to prevent growth, because warm recovered water grows organisms quickly. Storage is part of the treatment design.
How does Idaho's water rights situation affect dairies?
Idaho water is allocated under prior appropriation, and groundwater in the Snake River Plain is managed under conjunctive administration. A plant's supply is limited by its rights, which gives recovered water real value.
What about wastewater discharge for dairies?
Dairy effluent is high in organic load and is treated on site or discharged to municipal treatment under surcharge, and water recovery reduces both volume and load. Land application under permit is used in some rural locations.
How is recovered water monitored?
With online conductivity, turbidity and disinfectant residual where applicable, and with microbiological sampling on the schedule the plant's programme sets. Monitoring is what makes reuse defensible to an inspector.
Does potato processing in Idaho recover water similarly?
Potato plants use large volumes for washing, blanching and cooling and can recover and reuse within the process under FDA rules. The engineering is similar; the regulatory framework is different from dairy.
How is water recovery retrofitted into a dairy that is running?
By installing the collection and treatment alongside the existing systems, connecting uses one at a time with a bypass to fresh water, and validating each connection before the next. The plant never depends on the new system until it has proven itself on a use that can tolerate a failure.
What is the biggest sequencing mistake in dairy water recovery?
Building the treatment before deciding the destinations. Water treated to a quality nobody needs, or not treated to the quality the best use needs, is the usual result.
How quickly does a dairy water recovery project pay back?
It depends on water and wastewater costs and on the energy recovered with it, but plants with municipal supply and surcharge typically see returns within a few years, and plants constrained by water rights see value beyond cost.
Whey, and the stream that changed from cost to product
Any conversation about water in an Idaho dairy plant runs into whey, because cheese production generates it in enormous volume and because what a plant decides to do with it determines a large share of its water and energy engineering.
Whey is what remains after the curd is separated, and it is mostly water with a modest solids content, so on a volume basis a cheese plant produces far more whey than cheese. Historically that was a disposal problem. It is now, for most operations of any scale, a product line: whey protein concentrates and isolates, lactose, and permeate streams each have markets, and the processing to reach them is largely a matter of separating water from solids in stages.
The engineering that follows has three parts and they interact.
Membrane separation. Ultrafiltration concentrates protein while passing lactose and minerals; nanofiltration and reverse osmosis concentrate further and remove water. These are electrically driven pressure processes rather than thermal ones, so they are comparatively efficient at removing bulk water, and they are sensitive to fouling in a way that makes cleaning regime and flux monitoring part of normal operation rather than maintenance.
Evaporation and drying. Membranes take a stream so far and thermal processes finish the job. The design rule worth stating is to remove as much water as possible with membranes before handing the stream to an evaporator, because pressure-driven water removal costs considerably less than boiling water off. Plants that evaporate a stream a membrane could have concentrated first are paying a permanent premium.
The permeate is water. The streams passing through those membranes are largely water with a low solids content, and like evaporator condensate they are a recoverable resource rather than an effluent. In an arid basin, a plant that treats both its evaporator condensate and its membrane permeate as water supply rather than as wastewater changes its water balance substantially.
The practical advice we give is to map every stream leaving the process by volume and by how hard it would be to make useful, before designing any treatment. Plants frequently discover that the largest recoverable stream is not the one they were focused on, and that the sequencing of membranes and evaporation, decided years ago for reasons nobody now remembers, is costing more than the recovery project would save.
Short of water at an Idaho dairy plant?
Tell us your milk volume and where condensate currently goes. On a powder or concentrate plant that stream is often larger than the supply you are trying to secure. Call 201-450-8280 or use the form below.