Why a Two-Stage Pressure Reduction Exists

Why this matters

Two reducing valves in a row is one of those details that gets copied without being understood, and the copying goes both directions. Shops add a second stage to a station that never needed one and then spend years maintaining twice the trim. Shops delete a second stage during a rebuild because "one valve makes the pressure" and hand the customer a station that hunts, screams, and cuts its seat out in a season.

Worse, there are two completely different reasons to put a second valve in a station, and they call for opposite piping. One is solved by valves in series. The other is solved by valves in parallel. A station built with the wrong arrangement still has the original fault, and the fault is now hidden behind hardware that looks like it was addressed.

Steam at station pressures is an invisible, severe scald hazard at the leak point, so nothing in this article is diagnosed by hand or by feel, and no valve body, strainer, or pilot is opened until the section is isolated at both ends, drained through its drip leg, vented, confirmed at zero on the upstream gauge, and locked and tagged under the written energy control procedure required by 29 CFR 1910.147 for stored pressure energy.

The gauge ratio is not the ratio that governs

Take a station reducing a 150 psig header to 5 psig. On the gauges that is a 30 to 1 reduction and it sounds extreme. It is not, because a reducing valve responds to absolute pressure, and absolute pressure is gauge plus atmospheric.

• Inlet: 150 plus 14.7, so 164.7 psia
• Outlet: 5 plus 14.7, so 19.7 psia
• Absolute ratio: 164.7 divided by 19.7, about 8.4 to 1

Now take a station reducing 250 psig to 8 psig. On the gauges that is about 31 to 1, essentially the same headline number.

• Inlet: 264.7 psia
• Outlet: 22.7 psia
• Absolute ratio: about 11.7 to 1

Two stations that look identical on a gauge sit on opposite sides of the design limit. The first is a single-valve job. The second is not. Anyone sizing off the gauge numbers gets both wrong, and gets them wrong in the same direction, which is why the error is invisible until a seat fails.

The two independent limits on a single valve

Pressure ratio. A widely used design limit is about 10 to 1 absolute across one reducing valve, with the valve manufacturer's published rating governing whenever the two disagree. Past that, one seat is dissipating too much energy in one place: noise rises sharply, the trim erodes, and the plug's travel resolution gets so coarse that a small mechanical movement swings the outlet pressure far more than the control can settle.

It is worth naming what this limit is not about. Steam flow through the seat goes critical, meaning it chokes at sonic velocity, once the downstream absolute pressure falls below roughly 58 percent of the upstream absolute pressure. That threshold is about 1.7 to 1, so essentially every steam reducing station in a building is choked, including the ones that need only one valve. Choking is not the thing that separates a one-stage station from a two-stage one, and an article that tells you it is has sent you looking at the wrong number.

Flow turndown. A control valve can only resolve flow down to some fraction of its own capacity. Below that it is effectively slamming between shut and cracked, which reads at the gauge as hunting and reads at the seat as wire drawing. If your maximum load is 6,000 lb/hr and your shoulder-season load is 500 lb/hr, you need 12 to 1 of usable control range out of one valve, and if it also had to be sized for a startup surge, the required range is worse than that.

These two limits are independent. A station can exceed one, the other, both, or neither.

Series solves ratio, parallel solves turndown

Two valves in series split the pressure ratio between two seats. The conventional first cut at the interstage pressure is the geometric mean of the two absolute pressures, because that divides the ratio evenly and gives each valve the same job. Series staging does nothing at all for turndown: both valves still pass the full flow range, so a station that hunts because its load collapses in the summer will still hunt with a second valve in front of it.

Two valves in parallel, sized large and small, split the flow range. The small valve carries the low loads inside its own resolution and the large valve opens on top of it when the load calls for it. Parallel staging does nothing for pressure ratio: both valves still see the same inlet and outlet pressures, so a station that erodes its seat because it drops 12 to 1 in one step erodes both seats instead of one.

State the limit you actually exceeded before choosing the arrangement. That single sentence prevents most of the field errors in this subject.

Case one: 150 psig to 5 psig, steady load

A film evaporator draws a nearly constant 1,800 to 2,000 lb/hr and runs whenever the plant runs.

• Absolute ratio: about 8.4 to 1, under the roughly 10 to 1 limit, so the ratio gate does not trip.
• Turndown: 2,000 divided by 1,800, about 1.1 to 1, nowhere near any valve's resolution limit, so the turndown gate does not trip.
• Outcome: one valve.

The pipe sizing still moves, and that is where this station usually goes wrong. Saturated steam at 150 psig has a specific volume of roughly 2.75 cubic feet per pound from the steam tables. At 2,000 lb/hr that is about 92 cubic feet per minute, and a 2 in schedule 40 pipe carries it at roughly 3,900 feet per minute, comfortably inside the 4,000 to 6,000 feet per minute band commonly used as a ceiling for saturated steam mains.

At 5 psig the same steam occupies roughly 20.5 cubic feet per pound, about seven and a half times the volume. The same 2,000 lb/hr is now about 683 cubic feet per minute, and a 5 in schedule 40 pipe carries it at roughly 4,900 feet per minute. Two inches in, five inches out, one valve. A station piped 2 in on both sides would run the outlet at velocities where wet steam scours elbows, and no amount of staging fixes that, because the problem is the pipe.

Case two: 250 psig to 8 psig, seasonal load

A campus heating main, 6,000 lb/hr at design and about 500 lb/hr in the shoulder season.

• Absolute ratio: about 11.7 to 1, over the limit, so the ratio gate trips. Series staging is required.
• Turndown: 6,000 divided by 500, 12 to 1, past what one valve resolves well. The turndown gate trips too. Parallel staging is also required.
• Outcome: this station needs both, and that is the point of stating the gates separately. Two valves in series, and the second stage built as a large and small pair in parallel.

Interstage pressure, first cut: the geometric mean of 264.7 psia and 22.7 psia is about 77.5 psia, call it 60 psig. That gives the first stage a ratio of about 3.5 to 1 and the second stage about 3.3 to 1. Both are deep inside the limit, which is what you want, because the margin is what absorbs a header that runs high in the winter.

Pipe sizing across the station, holding roughly 5,000 feet per minute at every point:

Two and a half inches in, eight inches out, for the exact same pounds of steam. The cross-sectional area ratio between inlet and outlet is about 10 to 1, and the specific volume ratio is about 10 to 1, which is not a coincidence: at equal velocity, area has to track volume exactly. That identity is the fastest way to sanity check a station drawing without a sizing chart in front of you.

The interstage pressure is also doing a second job worth naming. It gives the station a place to put a safety relief valve on the intermediate section, so the first stage failing wide open does not present 250 psig to the second stage's body and to whatever sits between them. The relief article in this group covers how that valve gets sized and set, and both interstage and final relief valves get sized on the same basis.

Note that the interstage number is a first cut, not an answer. If the plant also draws 60 psig steam for something else, that draw sets the interstage pressure and the split stops being even. If the second stage's own manufacturer publishes a maximum inlet for the trim you selected, that publication governs over the geometric mean.

What the second stage costs, honestly

Two valves in series is two sets of trim, two pilots on pilot-operated types, two strainers if you do it properly, an additional relief valve on the intermediate section, and a longer station that has to be laid out with straight runs for two sensing lines rather than one. On a maintenance plan that is roughly double the reducing-station labor hours per year.

That cost is the reason not to add a stage on instinct. It is also the reason not to delete one on instinct, because the alternative is a valve that eats its own seat, and a station that fails wide open puts full header temperature on downstream equipment.

Reading an existing station

You can usually tell what a station was built to solve without any drawing.

Look at the piping first, from a safe distance and without touching a hot line. Two valves with a pipe between them that changes size, each with its own sensing line and its own gauge, is a series station. The middle section normally carries its own relief valve. Two valves teed off the same inlet header into the same outlet header, one visibly larger than the other, is a parallel station, and both sense the same downstream pressure with the small valve set slightly higher so it takes the low loads first.

Then check whether the arrangement matches the fault. Convert the two gauge pressures to absolute and take the ratio. If a series station's ratio is well under 10 to 1 and the plant runs a wide seasonal load swing, somebody solved the wrong problem, and the hunting complaint that brought you out will not be fixed by rebuilding either valve.

References

• Valve manufacturer sizing and installation literature, which is the governing authority for maximum inlet pressure, published rangeability, and the pressure ratio a specific trim will accept
• Saturated steam tables from any engineering handbook, for the specific volume and saturation temperature at every pressure used above
• ASME B31.1 Power Piping, in the edition adopted by the authority having jurisdiction, for pressure piping design in the station
• 29 CFR 1910.147, the OSHA general industry energy control standard covering isolation of stored pressure energy before any station component is opened
• See related: What a Pressure Reducing Station Is Managing; What a Safety Relief Valve on a Steam System Protects