Pressure and flow recovery: Science and thermodynamics

Why pressure converts so well, yet gas needs reheating

What this tab covers The thermodynamics: the grade (temperature or pressure) of the energy being recovered, energy versus exergy efficiency, and how harsh the waste stream is on equipment.

Pressure is a high-quality form of energy, so turbines and exchangers can convert most of it to useful work. The catch is that the quantity at most sites is small, and gas expansion needs heat put back in.

Grade of energy

Favourable

The source here is pressure, not temperature, which is why the 100 C rule does not apply. A city gate gas station may drop gas from 70 to 20 bar gauge. Water mains and treatment plants often throttle away hundreds of psi (1 psi is about 0.07 bar) across pressure-reducing valves. At Manitou Springs in Colorado, about 235 of 250 psi was available to a turbine. Seawater reverse osmosis (RO) rejects brine still at near membrane pressure. Compressed-air heat is the one thermal stream here, and it is low grade, with air-cooled outlets only 30 to 40 F above the cooling air.²³⁵⁶

Gas letdown example 70 to 20 barg (50 bar drop)³ Water main example about 235 psi usable of 250 psi inlet⁶ US conduit siting threshold about 18 ft head (about 8 psi) surplus pressure⁵ Compressor exhaust air 30-40 F above cooling air inlet²

Exergy versus energy efficiency

Mixed

Rotary pressure exchangers in RO pass brine pressure straight to incoming seawater at up to 98% efficiency, cutting plant energy use by up to 60%. Gas turboexpanders reach about 85% isentropic efficiency (the share of the ideal expansion work actually captured), or about 70% after generator and electrical losses. Gas cools by about 0.5 C per bar as it expands, so it must be preheated. In one modelled 2 MW retrofit, heating duty rose from 1,300 kW to 3,280 kW. The extra heat almost equals the power output (calculated), so the gain depends on where that heat comes from. Compressor heat recovery is the reverse case. As much as 80 to 93% of compressor electricity becomes heat, and 50 to 90% of that can often be reused, but only as warm air or water.¹²³

PX pressure exchanger efficiency up to 98%¹ Turboexpander efficiency 85% isentropic, about 70% overall³ Extra preheat per 2,000 kW power 1,980 kW (calculated, 3,280 minus 1,300)³ Compressor input turned to heat 80-93%²

Media compatibility and corrosiveness

Mixed

Seawater and brine are highly corrosive, so ERI's PX uses alumina ceramic parts designed for more than 25 years. Natural gas is clean, but the outlet must stay about 25 C above its dew point to stop liquids and hydrates (ice-like solids) forming. Raw water and wastewater carry debris that can damage small turbines. Compressor heat comes from oil coolers, so water circuits need a heat exchanger to keep oil out.¹³

PX ceramic design life more than 25 years¹ Gas outlet margin about 25 C above dew point³

Variants compared¹²³⁴⁵⁶

Variant	Typical size	Conversion efficiency	Main constraint
RO pressure exchanger (PX)	Plants up to megaproject scale	Up to 98%	Tied to new or retrofitted desalination demand
Gas letdown turboexpander	About 0.1-2 MW	About 70% overall	Needs preheat heat source
In-pipe hydro turbine	About 10-70 kW	Not sourced	Small output per site
Pump as turbine	Small sites	Not sourced	Suits steady head and flow
Compressed-air heat recovery	Per compressor, e.g. 100 bhp	50-90% of heat recoverable	Low temperature; needs a heat demand

Sources

Energy Recovery, Inc. Form 10-K for fiscal year 2025 (company filing) Energy Recovery, Inc. (US SEC filing), 2026
Fact Sheet 10: Heat recovery with compressed air systems (updated 2026) Compressed Air Challenge (sourcebook developed with US DOE), 2026
H2 production through gas pressure utilization at natural gas transmission network stations (G. Vallios, METRON SA) H2Tech, 2023
Hybrid turboexpander and fuel cell system for power recovery at natural gas pressure reduction stations (Howard, Oosthuizen and Peppley), Chemical Engineering Transactions 21 AIDIC, 2010
Water and wastewater systems hydropower guide (document 24-50213) Pacific Northwest National Laboratory and Idaho National Laboratory for US DOE Water Power Technologies Office, 2024
Manitou Springs water treatment plant case study (vendor case study) Rentricity, 2022