Paul Industries fabricates and installs process piping for Washington plants, including the ones reducing or removing their dependence on a central steam system. Taking steam out of a plant is not a like-for-like swap. The header disappears, the tracing has to come from somewhere else, design conditions change, and a partially decommissioned steam system creates hazards of its own. The piping consequences are where these projects go over budget when nobody has thought them through.
Request a quote or call 201-450-8280
Hot water is a different fluid to distribute
Steam distributes heat by carrying latent heat, which is why a modest steam pipe can deliver a great deal of energy. Hot water carries sensible heat only, so delivering the same duty requires either a much larger flow or a much larger temperature drop across the load, and usually both. That single difference drives most of the piping consequences.
Pipe sizes increase. A plant that replaces steam distribution with hot water distribution will generally need larger lines, and larger lines in existing racks and through existing penetrations is a survey question before it is a design question. It is worth walking the intended route early, because discovering that a line cannot follow the old route is a schedule problem rather than a cost problem.
Pumping appears where it did not exist. Steam moves itself. Hot water does not, so circulation pumps become part of the system, with their own power draw, their own control and their own failure modes. That pumping energy is genuine and it should be in the comparison rather than assumed away, though at Washington’s 6.61 cents per kilowatt-hour it is rarely decisive.
Return matters as much as supply. A steam system has condensate return that plants often neglect. A hot water system has return that it cannot function without, and a system with poor return balance delivers uneven temperature across the plant in a way that is hard to diagnose after commissioning. Balancing valves, and a commissioning exercise that actually balances rather than assuming design flows, is the difference.
Expansion and pressurization. A closed hot water circuit needs expansion capacity and pressure control, and the arrangement is different from anything a steam plant has. It is straightforward, it is inexpensive, and it is regularly left out of early cost estimates.
Tracing without steam
| Option | Suits | Watch for |
|---|---|---|
| Electric trace | Freeze protection and moderate process maintenance | Circuit design, controls, and continuous power draw |
| Hot water trace | Runs near an existing hot water circuit | Lower achievable temperature than steam |
| Point-of-use heating | A single load far from everything else | Often cheaper than extending any distribution |
| Retaining a small steam island | A handful of genuinely high-temperature duties | Standby losses of a boiler serving very little |
The last row is the decision that most often gets made badly. A plant electrifies most of its heat, finds three duties that genuinely need steam, and keeps the boiler running for them. The boiler then idles most of the time, its standby losses are unchanged, and a large share of the savings the project promised never materializes.
The honest options at that point are to find a way to serve those three duties without steam, to accept that the boiler stays and size the project’s savings accordingly, or to replace the large boiler with a much smaller one matched to what remains. The third is frequently the right answer and it is rarely in the original scope.
Decommissioning done properly
Steam lines that are no longer in use are not harmless. They are a real hazard and a real liability, and the way they are dealt with matters.
Isolate physically, not just operationally. A closed valve is an operational isolation and it will eventually be opened by somebody. Blanking, spool removal or physical disconnection is what makes a line genuinely dead.
Drain and leave drained. Water left in a dead line freezes, corrodes and creates water hammer if the line is ever repressurized. Low point drains opened and left open is a cheap permanent solution.
Remove rather than abandon where practical. Abandoned lines accumulate: they confuse the next project, they make drawings unreliable, they collect dust in food environments, and they occupy rack space that the new hot water lines need. Removal costs money once; abandonment costs a little every year forever.
Update the drawings. An as-built that still shows a steam header that has been dead for five years will eventually cause somebody to make a decision based on it.
Mind the insulation. On older plants, insulation on steam systems may contain material requiring specific handling. That assessment belongs at the planning stage, because finding out during demolition stops the job.
Standards and what we build
Process piping follows ASME B31.3 with the fluid service category determined and documented, and the design conditions for a hot water system are meaningfully lower than for steam, which occasionally allows lighter schedules and different material choices than the plant is used to specifying. Where the piping is sanitary we build to ASME BPE with orbital welding to AWS D18.1, weld documentation retained, verified slope and drainability, and passivation to ASTM A967 after cleaning per ASTM A380.
We handle the survey work, the new distribution and return, tracing replacement, the decommissioning and removal of redundant steam systems, and the drawing updates that keep the plant honest afterwards.
Standards referenced: ASME BPE · EIA electricity price data · ASTM A967 · ASTM A380 · ASME B31.3
Frequently asked questions
Do you install process piping in Washington?
Yes, across Seattle, Tacoma, Spokane, Yakima, the Tri-Cities and statewide: new hot water distribution and return, sanitary and high-purity piping, tracing replacement, steam system decommissioning and removal, pressure testing and the documentation package.
Why does replacing steam with hot water need bigger pipes?
Because steam carries latent heat and hot water carries only sensible heat. Delivering the same duty needs a much larger flow, a much larger temperature drop across the load, or both. That single difference drives most of the piping consequences, starting with line size.
What is most often missed in the cost estimate?
Circulation pumping, expansion and pressurization for the closed circuit, and the survey question of whether larger lines actually fit the existing routes and penetrations. The last of those is a schedule problem rather than a cost problem, which makes walking the route early worth the time.
Why does return balance matter?
Because a hot water system cannot function without return, and poor balance delivers uneven temperature across the plant in a way that is difficult to diagnose after commissioning. Balancing valves plus a commissioning exercise that actually balances, rather than assuming design flows, is what prevents it.
How do we replace steam tracing?
Electric trace for freeze protection and moderate process maintenance, hot water trace where a run passes near an existing circuit, or point-of-use heating for an isolated load, which is frequently cheaper than extending any distribution to reach it.
What if a few duties genuinely need steam?
Then the boiler stays, and this is where these projects most often disappoint. A large boiler idling to serve three small duties keeps its standby losses, so much of the promised saving never appears. Replacing it with a much smaller boiler matched to what remains is often the right answer and is rarely in the original scope.
Can we just leave the old steam lines in place?
You can, and it costs a little every year forever. Abandoned lines confuse the next project, make drawings unreliable, collect dust in food environments and occupy rack space the new lines need. Removal costs money once. Where lines do stay, they must be physically isolated and drained, not merely valved off.
Why is a closed valve not enough?
Because it is an operational isolation and somebody will eventually open it. Blanking, spool removal or physical disconnection is what makes a line genuinely dead. Water left in a dead line also freezes, corrodes and causes water hammer if it is ever repressurized, so low point drains should be opened and left open.
Does anything need checking before demolition?
Insulation on older steam systems may contain material requiring specific handling, and that assessment belongs at planning stage. Finding out during demolition stops the job, and it converts a scheduled shutdown into an unscheduled one.
How do I get a quote for a Washington piping project?
Use the form on this page or call 201-450-8280. Useful inputs are the heat loads and their temperatures, the existing steam system layout and age, which duties genuinely require steam, available rack and penetration routes, and whether decommissioning and removal are in scope.
What loop temperature should a hot water system run at?
As low as the highest-temperature process load allows, because heat pump efficiency and distribution losses both improve at lower temperature, while staying above the range where Legionella grows in stored water. Most process heating loads are satisfied well below the temperature a boiler would have supplied.
How is flow controlled in a hot water distribution loop?
With variable speed pumps and pressure-independent control valves at the loads, so that each load takes the flow it needs and the pumps deliver only that, rather than constant-speed pumping against balancing valves. Variable flow is where the pumping energy saving is made.
How is a hot water loop pressurised?
With an expansion tank and a pressurisation unit sized for the loop's volume and temperature range, with relief protection, so that the water expands and contracts without venting or cavitation. The expansion provision is often undersized on a loop that replaced steam.
Which heat exchanger types serve hot water process loads?
Plate exchangers for most duties because of their close approach temperature, which matters when the supply is cooler than steam was, and shell and tube where fouling or pressure demands it. The exchanger's approach temperature sets how low the loop can run.
What happens to the condensate system when steam is removed?
Traps, condensate receivers and return pumps are decommissioned with the steam lines, drained and removed or blanked, because a condensate system connected to nothing accumulates water and corrodes. The condensate side is half the steam system and is retired with it.
Is an asbestos survey needed before removing old steam lines?
Yes on any Washington plant built before the 1980s, because steam line insulation and gaskets of that era commonly contain asbestos, and removal is regulated work. The survey precedes the demolition plan.
What electrical capacity does electrified heating need?
Heat pumps draw a fraction of the heat they deliver, while electric boilers draw the full load, so the service upgrade depends on which technology serves each duty and on the plant's existing capacity. The electrical study is part of the conversion design.
What utility incentives support steam decommissioning in Washington?
Washington's utilities and state programmes offer incentives for industrial electrification and efficiency, with the amount and eligibility depending on the utility and the technology, and the application is made before the work starts. The incentive is checked at design, when it can still shape the project.
How is a steam-to-hot-water conversion phased?
Load by load, with the hot water system commissioned to serve each load before its steam supply is removed, so the plant never loses heat to a process. The boiler runs until the last load is converted.
What heat source serves the hot water loop?
Heat pumps, electric boilers or heat recovery, depending on the temperature needed and the heat sources available. Washington's grid makes electric sources attractive, and heat pumps suit the moderate temperatures most loads need.
What materials are used for hot water distribution?
Carbon steel or stainless depending on water treatment and the process, with insulation sized for the lower temperature and larger surface area. Hot water loops need water treatment to prevent corrosion and scaling.
How does the sequencing decision save money?
By converting the loads that free the most steam capacity first, so that the boiler can be downsized or retired earlier, and by grouping loads that share a hot water loop. The order of conversion decides how long two systems run in parallel.
What about clean steam for sanitary processes?
Clean steam for sterilisation stays as steam, from a dedicated generator, and is not converted to hot water. The conversion applies to heating duties, not to sterilising ones.
How is the converted system commissioned?
By balancing flows, verifying delivered temperature at each load, confirming pump performance and checking the isolation of every removed steam line. The balance report is the acceptance record.
What is the commonest steam decommissioning mistake?
Estimating the conversion from the supply line alone and discovering the return, pumps and tracing halfway through, or leaving old steam lines connected by closed valves. Both are avoided by a full scope at the start.
Phasing it without stopping the plant
The question that determines whether a heat conversion project happens at all is rarely whether it pays. It is whether it can be executed on a plant that has to keep producing. A proposal requiring a six-week shutdown will not be approved regardless of its payback, so the phasing is part of the engineering rather than a scheduling detail added afterwards.
The approach that works runs the two systems in parallel for a period and moves loads across one at a time.
Build the new distribution while the old one runs. Hot water supply and return installed alongside the existing steam header, commissioned and proven on a test load before anything of consequence is connected to it. This is the expensive phase and it is also the one with no production risk, which makes it the right place to be thorough.
Convert the easiest load first, deliberately. Not the largest and not the most valuable, but the one whose failure would be least disruptive. A single non-critical duty proves the distribution, the controls, the balance and the commissioning approach at small scale. Whatever is wrong with the design will show up here, where it can be fixed cheaply.
Then convert by shutdown window. Most plants already have scheduled downtime, whether seasonal, for annual maintenance, or simply a weekend. Sequencing conversions into windows that already exist avoids creating new ones. The corollary is that the project runs over a longer elapsed period than a single-shutdown approach, which has to be accepted at the outset rather than discovered when someone asks why it is not finished.
Keep the steam available until the new load is proven in production. Not just commissioned, but proven across a full production cycle including the conditions that only appear occasionally. The temptation to disconnect the old supply as soon as the new one works is strong and it removes the fallback exactly when it is most likely to be needed.
Decommission only when nothing is left on it. The steam system comes out at the end, as its own piece of work, once the last load has moved and been proven. Trying to remove sections as they are freed up leaves the remaining system in odd configurations, with dead legs, unbalanced condensate return and traps serving nothing, which is how a system that worked fine becomes unreliable during the very period it is still needed.
The sequencing decision that saves the most money
There is one ordering choice worth making explicitly, because it changes the capital requirement substantially.
Before converting any load, reduce it. Every duty being moved should first be examined for whether it needs to be as large as it is. Wash temperatures set higher than validation requires, tanks heated continuously that could be heated on demand, lines traced that no longer carry anything, vessels with insulation that was removed during a maintenance job in a previous decade. These are cheap to fix and each one directly reduces the size and cost of the equipment that will replace them.
A plant that converts its existing loads faithfully buys equipment sized for its historical inefficiency and then owns that oversizing for twenty years. A plant that spends three months reducing loads first buys smaller equipment, spends less capital, and runs closer to design point where efficiency is best.
It is unglamorous work and it competes badly for attention against the interesting new technology. It is also, consistently, where the largest share of the return comes from.
Taking steam out of a Washington plant?
Tell us which duties genuinely need steam and which are there by habit. That list decides whether the boiler can actually go. Call 201-450-8280 or use the form below.