Views: 0 Author: J-VALVES Publish Time: 2026-11-13 Origin: Site
Steam humidification is the metered injection of dry saturated steam into an air stream to raise its absolute moisture content - typically to hold occupied spaces, cleanrooms, hospitals, museums, data centers and production halls between 40 and 60 percent relative humidity. The device that meters that steam is a modulating control valve, and the train around it - pressure reducing station, separator, strainer, traps - is exactly the valve infrastructure this site writes about. Yet humidifier steam lines fail more often on valve-station details than on the humidifier itself. This article works through the load arithmetic, the physics of why steam humidification is near-isothermal, the steam-quality requirements, the control valve sizing with a worked Kv example, the absorption-distance rules, and the procurement clauses that decide whether the system holds setpoint through a cold snap.
A boundary note first, because this article sits in a steam cluster that includes two earlier guides: a desuperheater selection guide covers removing superheat from process steam so it stays steam, and a steam trap selection guide covers draining condensate from steam lines. Humidification is the third job: turning the steam into air moisture. The steam leaving a humidifier lance never comes back as steam - only the line condensate is recovered - and that asymmetry shapes every design decision below.
Air holds water as vapor, and the amount it can hold depends sharply on temperature: air at 20 deg C holds about 14.7 g of water per kg at saturation, air at 0 deg C only about 3.8 g/kg. Cold outdoor air heated to room temperature therefore arrives bone dry - outdoor air at 0 deg C and 80 percent RH contains roughly 3 g/kg, but after heating to 21 deg C its relative humidity collapses to about 18 percent. Nothing in the heating process adds water; the fix is to add water vapor deliberately, and steam is the cleanest way to do it.
Steam humidification is classified as an isothermal humidification method - the moisture is added by injecting vapor that was evaporated elsewhere (a boiler or a dedicated steam generator), so the evaporation energy does not come from the air itself. This distinguishes it from adiabatic systems - evaporative pads, spray, ultrasonic and compressed-air atomizers - where water evaporates directly into the air and draws the latent heat from the air, cooling it as it humidifies. Section 3 quantifies that difference, because it decides which system the downstream heating and cooling coils must be sized around.
The single most quoted sizing error in this field is sizing by room volume. Humidification load is set by airflow and by the absolute humidity difference, not by the space being conditioned. The governing relation is:
m_steam (kg/h) = rho_air (kg/m3) x V_air (m3/h) x delta-x (kg water per kg dry air)
with rho about 1.2 kg/m3 at sea level (about 1.07 at 1,000 m altitude). The delta-x is read on a psychrometric chart or calculated between the entering air state and the target state - and the entering state is the mixed air after recirculation, which is why the outdoor-air fraction dominates the winter load. Three worked examples, all from published design references:
Case | Air data | Humidity delta-x | Steam load | Selected capacity |
|---|---|---|---|---|
AHU winter makeup air | 12,000 kg/h dry air, mixture needs +3 g/kg | 3 g/kg | 36 kg/h | 40-42 kg/h (+10-15 percent margin) |
Print house, 40 percent fresh air | 30,000 m3/h total, outdoor 1.9 g/kg at -7 deg C / 75 percent RH, room target 21 deg C / 50 percent RH = 8.7 g/kg, mixed = 0.4 x 1.9 + 0.6 x 8.7 = 6.0 g/kg | 2.7 g/kg | 1.07 x 30,000 x 0.0027 = 87 kg/h | 100 kg/h (about 80-85 percent modulation) |
Imperial check | 100,000 CFM, W1 = 0.0012 to W2 = 0.0078 lb/lb, air density 0.075 lb/ft3 | 0.0066 lb/lb | 49.5 lb/h (22.5 kg/h) | approx. 55 lb/h |
Two discipline points from these examples. First, the margin is 10-15 percent, not 50 percent: an oversized generator cycles too frequently, concentrates minerals faster in the cylinder, and destabilizes part-load control, because most units modulate cleanly only down to about 20 percent of capacity. Second, the load scales with the outdoor-air share. If ventilation later increases - a compliance change, a CO2-driven demand-controlled scheme - the humidifier that was sized on last year's air balance quietly loses its winter headroom. The load calculation should be a documented deliverable in the humidifier submittal, showing air flows, states, and the fresh-air fraction, not a nameplate figure.
The claim that steam humidification "adds moisture without changing the air temperature" deserves quantification, because it is not literally true - but it is close enough to change coil selection. Injecting saturated steam at roughly 100 deg C into air at 20 deg C adds enthalpy in two parts: the latent portion (about 2,257 kJ/kg at atmospheric pressure) stays locked in the vapor and does not heat the air, while only the sensible difference between steam at 100 deg C and vapor at room temperature shows up as a temperature rise. Working the enthalpy balance through, injecting 10 g/kg of dry saturated steam raises the dry-bulb temperature by roughly 1.5 K - about 0.15 K per g/kg. On a psychrometric chart the process line therefore tracks the steam's enthalpy line, which is nearly vertical at ordinary air temperatures.
Adiabatic humidification is the mirror image. With no external heat input, the latent heat of evaporation is drawn from the air itself, and the dry-bulb falls by about 2.5 K for every 1 g/kg of moisture evaporated - so a 10 g/kg adiabatic process cools the air roughly 25 K. That cooling is a feature in data centers and it is a liability in winter comfort heating, where the cooled air then needs reheat before it can be delivered.
Attribute | Steam (isothermal) humidifier | Adiabatic humidifier |
|---|---|---|
Air temperature effect | Rises about 0.15 K per g/kg added | Falls about 2.5 K per g/kg added |
Energy source for evaporation | External: boiler fuel or electricity, roughly 0.7 kWh per kg steam | The air's own sensible heat; reheat often needed downstream |
Water in contact with air | None - vapor arrives already evaporated and sterile | Yes - pads, reservoirs and droplets need hygiene management |
Hygiene profile | Steam at over 100 deg C is sterile; no wetted reservoir in the air path | Reservoirs and recirculating water demand biocide/treatment programs |
Visible plume | Micron droplets form and re-evaporate within the absorption distance | Fine mist or fog at discharge, longer visible persistence |
Absorption distance | Typically 0.3-2 m, set by manufacturer charts | Usually shorter for atomizing types, but carryover droplets heavier |
Typical best use | Hospitals, cleanrooms, museums, offices - hygiene-critical, comfort RH | Data centers, printing and textile halls - evaporative cooling wanted |
The isothermal property has one direct control consequence worth stating now: because steam injection barely cools the air, a temperature control loop is essentially blind to humidification. All regulation must come from a humidity sensor, which is why sensor placement (Section 10) matters so much more in steam systems than in evaporative ones.
The humidifier wants dry saturated steam at reduced pressure. Typical practice divides the steam path into two pressure levels: a distribution pressure of 2-4 bar g from the boiler side, reduced at the humidifier to about 0.5-2 bar g at the lance inlet, with some designs injecting at even lower pressure. Two reasons push the injection pressure down. Hotter, higher-pressure steam carries more sensible enthalpy and adds a slightly larger temperature rise to the air; and the dispersion orifices are sized for a target exit condition, so feeding them high-pressure steam overrides the designed droplet behavior.
Dryness is the harder requirement. Any liquid carried into the dispersion tube shows up as "spitting" - droplets blown into the duct that wet duct walls, filters and anything below. The defense is a standard steam-train discipline: correctly sized and sloped steam lines with drains at low points, a line separator with a float trap immediately before the humidifier, and a steam-jacketed dispersion tube that keeps the tube walls hot so steam does not condense against them on its way out of the orifices. When drainage fails, the symptoms are consistent and diagnostic: thread-like erosion at the modulating control valve trim (persistent condensate flashing through the seat), dripping at the lance, and water pooling at the duct bottom. A drip of that kind is a drainage fault signature, not a humidifier fault - the correction belongs in the upstream train.
For electric or electrode steam generators, feedwater quality is part of steam quality: manufacturers typically specify conductivity between about 125 and 1,250 microsiemens/cm for direct units, and demineralized or RO water for resistive units to prevent scale on elements and in dispersion orifices. Scale blocked orifices throttle the true capacity long before the load calculation is ever questioned.
A humidifier is a control-valve application wrapped in a standard steam distribution system. Component by component:
Component | Function | Sizing / selection note | Failure if wrong |
|---|---|---|---|
Pressure reducing station | Reduces boiler header (6-10 bar g) to the 2-4 bar g humidification distribution header | Parallel 1/3-2/3 station when load turndown exceeds about 10:1; see the PRV station guide | Single oversized PRV throttles violently and wears; low-load pressure creep |
Line separator + float trap | Removes entrained condensate before the humidifier | Rated for the header flow at distribution pressure | Wet steam: spitting, trim erosion, duct wetting |
Y-strainer (100 mesh) | Protects the modulating valve seat and the dispersion orifices | Horizontal installation, blowdown access | Grit embeds in soft seats; orifices plug with scale |
Modulating control valve | Meters steam flow to the lance against the RH signal | Globe valve, equal-percentage trim, pneumatic actuator with positioner; sized by steam Kv formula (Section 6) | Oversized valve hunts at small openings; undersized caps winter capacity |
Safety relief valve | Protects the reduced-pressure lance line if the PRV or control valve fails open | Set below the lowest-rated component downstream, with a stated set-pressure window | Overpressure of lance, dispersion tube and duct-mounted equipment |
Dispersion tube condensate drain + trap | Drains condensate forming inside the lance and jacket | Trap selected for the lance backpressure; see the trap guide | Water accumulates and is blown out as droplets |
Isolation and bypass | Permits maintenance without shutting the AHU | Full-bore isolation both sides of the control valve train | Whole air handler lost to a seat repair |
Nothing here is exotic - it is the same architecture as any small steam conditioning station, which is why it is worth reusing the standard components: see the steam PRV station design guide for the reducing-station layout, the steam trap selection guide for the drainage points, and the strainer product range for the protective filtration. The one component that deserves its own arithmetic is the modulating control valve.
Steam flow through a control valve follows the standard steam sizing relations already derived in the Cv / Kv sizing guide - the humidifier case just fixes the pressure conditions. Take the first worked example: design load 40 kg/h of steam, distribution pressure reduced to 1 bar g at the valve (2 bar a absolute), and the lance discharging into the duct at essentially atmospheric backpressure of 1.1 bar a. The pressure ratio P2/P1 = 0.55 falls below the critical ratio of about 0.58 for saturated steam, so the flow is critical and the sizing relation reduces to:
Kv = ms / (12 x P1), with ms in kg/h and P1 in bar a
Kv = 40 / (12 x 2) = 1.67. The winter minimum load might be 6 kg/h - a night setback in a partially occupied building - needing an effective Kv of 0.25, so the valve must turn down smoothly over roughly a 7:1 range. An equal-percentage trimmed globe valve with a rated Kv of about 2 satisfies this: it operates near 85-90 percent open at design load and still modulates at the minimum. The classic procurement error is choosing the next valve size up "for safety": a DN25 globe with Kv = 8 turns the minimum load into a 3 percent opening, where the valve hunts, the seat wears on a wire-drawing duty, and the RH loop oscillates - the control instability many operators blame on the humidifier is actually a valve-authority fault. The same phenomenon is documented for oversized humidifier generators; the cure is identical in both cases: size for the duty, not for reassurance.
Two further valve details specific to humidification. First, the valve sees wet-steam risk from upstream: persistent condensate reaching the trim erodes it in thread-like grooves, which is why the separator and its trap are specified immediately upstream, not somewhere convenient in the plant room. Second, close-down matters: the leakage class of the selected valve decides whether the humidifier "drips" steam into the duct at zero demand, gradually over-humidifying the space - a leakage-grade requirement (Rate A on low-pressure air is a common specification) belongs in the purchase order rather than being left to the vendor's standard trim.
Steam injected into air that people breathe becomes an air-quality boundary, and boiler steam is not clean. Boiler feedwater carries oxygen scavengers, neutralizing amines and other treatment chemicals, and direct injection aerosolizes them into the occupied space. Hospitals, pharmaceutical cleanrooms, laboratories and museums therefore specify clean steam: steam generated from treated or RO water in a dedicated resistive or steam-to-steam generator, physically segregated from the boiler circuit, so treatment chemicals never contact the humidification vapor. This is one of the strongest procurement arguments for steam over adiabatic systems - the vapor arrives sterile and chemical-free by construction, whereas evaporative systems hold standing water near the air path and demand a hygiene program of dosing, dumping and disinfection.
Holding 40-60 percent RH is itself a hygiene measure: airborne bacteria survival, influenza viability and electrostatic discharge all degrade materially within that band, which is why the same window appears in hospital ward guidance, semiconductor fab specifications and archive conservation standards. The engineering caution cuts the other way too - above about 60 percent RH, mold growth risk and duct condensation rise steeply, so the control band is not a comfort preference but a hard design envelope, and the high-limit sensor (Section 10) is its enforcement mechanism.
Steam leaves the dispersion tube as vapor, but meeting cooler air condenses part of it into micron-sized droplets - the white plume - which re-evaporate further downstream. The absorption distance is the straight duct length over which that mixing completes; the usually shorter non-wetting distance is the point beyond which surfaces stay dry. No general formula exists; manufacturers publish selection charts and software, and the distance depends on the RH after humidification (the closer to saturation, the longer), air velocity, air temperature and the number of dispersion tubes. Practical ranges run from about 0.3-0.5 m for multi-tube manifolds at low RH gains to 1.5-2 m or more for single tubes, cold air or high demand.
Two rules make the manufacturer's chart usable in layout. First, nothing that can be damaged by wetting sits inside the absorption zone: no filters, no fans, no sensors, no elbows whose inner surfaces collect droplets. Manufacturer practice converts the chart distance into a base distance Bn and multiplies it by the next downstream component - the multipliers below are typical published values. Second, a widespread misconception needs killing: increasing air velocity does not shorten the absorption distance. For the same steam flow, higher velocity carries the plume further while it mixes; at reduced velocity the same steam enters less air, pushes it toward saturation, and the plume lengthens. The correct lever for a tight duct is more injection points - a multi-tube distributor splitting the same flow - not a fan speed change.
First component downstream of injection | Multiplier on base distance Bn |
|---|---|
Supply fan | x 1 |
Coarse / panel filter | x 2 |
Bag or high-efficiency filter | x 3 |
High-limit humidity sensor (minimum) | x 5 |
A duct-side RH cap completes the layout: the mixed condition right after injection should stay below roughly 90 percent RH, above which internal condensation, standing water, insulation degradation and mold risk climb regardless of the absorption distance. The same worked logic shows why: 20 kg/h injected into a 1.0 x 0.6 m duct at 22 deg C and 40 percent RH gives about 52 percent RH after injection at 4 m/s, but about 63 percent at 2 m/s and 86 percent at 1 m/s - the plume in that last case is nearly filling the duct. The high-limit sensor that cuts steam demand is the enforcement device for this cap.
The room-side limit is set by the coldest interior surface, not by the humidifier. Air condenses on any surface below its dew point, and the dew point climbs fast with RH. A worked example shows how binding this is: a 20 deg C room with a poorly insulated window (U-value about 2.8 W/m2K), outdoor design temperature -10 deg C, and an inner-surface film resistance of about 0.13 m2K/W puts the inner glass surface at roughly 20 - 30 x 2.8 x 0.13 = 9.1 deg C. The dew point of 20 deg C air at 50 percent RH is about 9.3 deg C - so 50 percent RH is already borderline condensation on that glazing, and the practical maximum for such rooms is nearer 45 percent. Upgrading to U = 1.4 W/m2K double glazing lifts the inner surface to about 14.5 deg C, permitting roughly 65 percent RH before condensation.
This is why humidification capacity and building envelope quality are coupled decisions: the humidifier can only serve the setpoint the coldest surface permits. In winter commissions, rooms that "cannot reach setpoint" are sometimes correctly refusing to - the high-limit or the condensation risk is doing its job, and the fault is a cold surface, not a valve. Retrofit priorities (glazing, thermal bridges at sills and slab edges) should be reviewed alongside any humidification upgrade in a cold climate.
The recommended architecture is standard and worth restating because failures cluster at the deviations. The control (demand) RH sensor sits in the occupied space or in the return duct - never inside the absorption zone, where it reads unmixed steam and drives the loop unstable. The high-limit sensor sits at least 5 x Bn downstream of the injection point and over-rides the demand signal, cutting steam when duct RH approaches the cap. The modulating control valve takes the loop output through a positioner; on/off humidification is reserved for small units, because cycling a steam line on a thermostat band produces the swing between wet and over-dry conditions that both the duct and the occupants notice.
Control tuning interacts with valve sizing through the authority chain established in Section 6: an oversized control valve hunts, an oversized generator cycles, and a sensor inside the plume oscillates the whole loop. When an RH loop misbehaves, diagnose in that order before touching the PID settings - in field experience the loop hardware, not the tuning, is the cause in most commissioning disputes. A useful acceptance test is a step response: command a 10 percent RH setpoint step and verify the room tracks within the specified band (commonly plus or minus 5 percent RH) without sustained oscillation of the valve position.
Steam humidification buys its isothermal character with external energy: evaporating 1 kg of water as steam requires roughly 2,600 kJ, about 0.7 kWh of electricity in a resistive generator, or roughly 0.85-0.9 kWh of fuel through a boiler running at 75-85 percent efficiency. For the 87 kg/h print-house example that is on the order of 60-70 kW of continuous winter duty at full load - real money, and the reason load arithmetic honesty (Section 2) pays for itself twice, once in equipment cost and once in energy cost. Adiabatic systems avoid this external energy but pay for it in reheat: the 2.5 K per g/kg cooling must be reversed before the air can be delivered to an occupied space in winter, so the apparent saving shrinks wherever the cooled air must be reheated for comfort.
Two smaller energy items belong in the operating budget. Electrode and resistive generators blow down 5-15 percent of capacity to control mineral concentration - water, heat and treatment chemicals leave with it, and the blowdown policy is a genuine efficiency lever. And uninsulated steam manifolds and dispersion tubes reject sensible heat into the air stream and generate condensate continuously; published testing of insulated dispersion assemblies reports reductions in unwanted heat gain and condensate generation in the 65-75 percent range, which is why insulated or jacketed dispersion hardware is specified rather than bare tubes.
Symptom | Most likely root cause | Verification | Correction |
|---|---|---|---|
Spitting / wet discharge at the lance | Wet steam: separator ineffective, drains blocked, trap failed open or closed | Check trap temperature cycling; open drain leg at low points | Restore drainage; recheck line slope and trap selection |
Thread-like erosion grooves in control valve trim | Persistent condensate flashing through the seat | Inspect trim; check separator condensate rate | Fix upstream drainage; consider hardened trim |
RH hunting, valve position oscillating | Oversized control valve (poor authority) or sensor inside absorption zone | Compare measured opening with Kv calculation; move sensor | Resize valve by steam formula; relocate sensor to room/return |
Corrosion and mold at duct bottom | Absorption distance too short, or duct RH above the 90 percent cap | Check layout against manufacturer chart and multipliers | More injection points or longer straight duct; enforce high-limit |
Room RH never reaches setpoint in cold weather | Load underestimated (outdoor-air share grew) or generator undersized | Recompute load with current ventilation rates | Recalculate and resize; verify fresh-air fraction |
Scale blocking dispersion orifices | Untreated or hard water in the generator; blowdown neglected | Inspect orifices; measure feedwater conductivity | Water treatment policy; RO/resistive unit; blowdown schedule |
High-limit sensor trips repeatedly | Sensor inside absorption zone, or genuine RH cap exceeded | Measure distance against 5 x Bn; log duct RH at injection | Relocate sensor or reduce demand; recheck absorption layout |
# | Specification clause | Why it is in the PO |
|---|---|---|
1 | Steam quality stated as dry saturated, with distribution pressure band and lance inlet pressure band | Wet or over-pressure steam defeats the dispersion design before the humidifier is ever blamed |
2 | Generator type matched to hygiene requirement: direct electrode/resistive, or clean steam (steam-to-steam / segregated) for occupied healthcare, pharma and archive spaces | Boiler treatment chemicals must never reach the air stream |
3 | Feedwater specification and conductivity band; blowdown policy stated | Scale in orifices is a slow capacity collapse; blowdown is 5-15 percent of capacity costed |
4 | Documented load calculation with air flows, psychrometric states and fresh-air fraction | Room-volume sizing and missing outdoor-air share are the two classic selection errors |
5 | Absorption distance verified against manufacturer charts for the actual duct, velocity and next component multiplier | Filters, fans and sensors inside the wetting zone are permanent moisture failures |
6 | Control valve sized by steam Kv formula with authority check; equal-percentage trim; positioner specified | Oversized valves hunt and wire-draw; undersized valves cap winter capacity |
7 | Valve leakage class specified (low-pressure air test, e.g. Rate A) | Zero-demand leakage over-humidifies the space overnight |
8 | Full valve train: separator with float trap, 100-mesh strainer, safety relief on the reduced-pressure line, drainable dispersion tube with trap | Each item has a documented failure signature; deleting any of them relocates the failure into the duct |
9 | Sensor architecture: control RH sensor in room or return duct; high-limit sensor at 5 x Bn minimum | Sensor placement is the most common control-loop fault in this application |
10 | Hygienic materials and maintainability: corrosion-resistant dispersion tubes, drainable and cleanable, line condensate recovered to a feedwater preheater where practical | Condensate heat recovery shortens generator startup and recovers part of the energy bill |
Steam humidification rewards exactly the discipline this site's steam series has built up: size the load honestly, keep the steam dry through a proper train of separators, traps and strainers, size the modulating valve for its actual duty, and respect the absorption-distance geometry of the duct. A humidifier specified that way holds 40-60 percent RH through the coldest week of the year; one specified by nameplate alone spends the winter hunting, spitting and growing mold in the duct corner nobody inspects. The other guides in the series are collected in the technical article library.
Why is steam humidification called isothermal when the steam is far hotter than the air?
Because almost none of the steam's heat ends up raising the air temperature. The dominant part of steam's enthalpy is the latent portion - about 2,257 kJ/kg at atmospheric pressure - which stays locked in the vapor phase and does not transfer to the air. Only the small sensible difference between steam at around 100 deg C and vapor at room temperature warms the air, and working the enthalpy balance through, injecting 10 g/kg of dry saturated steam raises the dry-bulb by roughly 1.5 K - about 0.15 K per g/kg. On a psychrometric chart the process line therefore follows the steam enthalpy line, nearly vertical at ordinary air temperatures. Adiabatic systems are the mirror image: evaporation draws heat from the air, cooling it about 2.5 K per g/kg added, which is why steam humidification can serve comfort RH in winter without reheating the supply air.
Can boiler steam be injected directly into the air stream?
Not for occupied or hygiene-critical spaces. Boiler feedwater carries oxygen scavengers, neutralizing amines and other treatment chemicals, and direct injection aerosolizes them into the air people breathe. Hospitals, pharmaceutical cleanrooms, laboratories and museums therefore specify clean steam: vapor raised from treated or RO water in a dedicated resistive or steam-to-steam generator, physically segregated from the boiler circuit, so treatment chemicals never contact the humidification vapor. For non-occupied industrial spaces, direct boiler steam is accepted practice - confirm the chemicals list with the boiler operator - and the steam itself is sterile either way, having passed through temperatures above 100 deg C.
What absorption distance does a steam humidifier need?
There is no universal formula - manufacturers publish selection charts and software, and the distance depends mainly on the RH after humidification (the closer to saturation, the longer), air velocity, air temperature and the number of dispersion tubes. Practical ranges run from about 0.3-0.5 m for multi-tube manifolds with small humidity gains to 1.5-2 m or more for a single tube, cold air or high demand. The manufacturer's base distance is then multiplied by the first downstream component: about x 1 for a supply fan, x 2 for a coarse filter, x 3 for a bag or HEPA filter, and the high-limit humidity sensor must sit at least 5 times the base distance away. One caution: increasing air velocity does not shorten the distance - for the same steam flow, higher velocity carries the plume further; the correct lever for a tight duct is more injection points, not fan speed.
How is the humidifier steam load calculated?
Multiply the air mass flow by the absolute humidity difference: m_steam (kg/h) = air density (about 1.2 kg/m3 at sea level) x airflow (m3/h) x delta-x (kg water per kg dry air), with delta-x read between the entering mixed-air state and the target state on a psychrometric chart. Worked example: 12,000 kg/h of dry air needing +3 g/kg requires 36 kg/h, and with a 10-15 percent margin selects a 40-42 kg/h generator. Two disciplines matter more than the formula: size by airflow and absolute humidity - never by room volume, the most common error in the field - and base delta-x on the mixed air with its outdoor-air fraction, because outdoor air in winter is the dominant load and its share changes with ventilation upgrades.
What pressure should humidification steam be supplied at?
Two levels, both lower than the boiler header. Distribution typically runs at 2-4 bar g from the reducing station, and the humidifier lance inlet at about 0.5-2 bar g, with some designs injecting even lower. Low injection pressure keeps the added sensible temperature rise small, preserves the designed behavior of the dispersion orifices, and leaves the modulating control valve working against a near-critical pressure ratio, which makes the steam sizing relation linear and the control smooth. The reduced-pressure line also needs its own safety relief valve, set below the lowest-rated component downstream, in case the reducing station or control valve fails open.
