How to Read a Steam Table Without Memorising It

Why this matters

Techs avoid steam tables because they look like a page of undifferentiated numbers, so they guess instead, and guessing on a steam system produces coils that never make setpoint and return lines sized for a fluid that is not in them. The table is not something to memorise. It is four questions with four columns, plus two arithmetic shortcuts that get you within about one percent without opening anything. What a plant actually needs is a card with its own four or five operating pressures on it, taped inside the boiler room door, and this article builds one.

The safety note that belongs here and not later

Everything below is desk work until you go and read a pressure. When you do: read gauges from beside the pipe rather than in front of any drain, cap, plug or gauge fitting, because a corroded gauge siphon or a plugged pigtail can let go under pressure at the face. If you need a surface temperature to confirm a table value, use a contact probe rated above the expected reading on a cleaned spot, keep the lead and your hand off the pipe, and do it on a section you have not just disturbed. Do not confirm a saturation temperature by touch at any pressure.

The four questions and the column that answers each

A saturation table has one row per pressure and the same handful of columns on every one. You will use four.

A fifth column, total heat of saturated vapour (hg), is just hf plus hfg and answers "how much did the boiler have to add", once you subtract the enthalpy of the feedwater it started with.

Two setup facts that cause more trouble than the columns do. The table is in absolute pressure and your gauge is in gauge pressure. Add local atmospheric pressure, near enough 14.7 psi at sea level, to a gauge reading before you look it up. The enthalpy datum is liquid water at 32 F, so hf is not heat above room temperature and it is not heat above your feedwater temperature. When you want to know what returning condensate saves the boiler, you subtract the feedwater's own hf, you do not use hf as the saving.

Read the direction, not the value

Walk two rows apart and note which way each column moves as pressure rises. This is the part worth actually knowing, because it is what lets you sanity-check a number someone hands you.

• Tsat rises, steeply at the bottom of the range and much more slowly at the top.
• hf rises, because hotter liquid holds more sensible heat.
• hfg falls. This is the one that surprises people. Higher pressure steam carries less latent heat per pound, so a given load needs slightly more mass flow at high pressure than at low.
• hg rises very slowly across the whole range a building steam system uses, by only a few percent from atmospheric up to 100 psig.
• vg falls hard. This is the reason high-pressure distribution uses small pipe.

If someone tells you a higher header pressure will cut the pounds per hour, the hfg column says otherwise, and the vg column says what they were actually thinking of, which is pipe size.

The two shortcuts that replace the table for most field work

Shortcut one: hf in Btu per pound is roughly saturation temperature minus 32. That works because liquid water's specific heat is close to 1 Btu per pound per degree F over the range building services use, and it holds to about one percent up to a few hundred psi. It drifts light as you go higher, because water's specific heat climbs with temperature: at roughly 338 F it reads about 1 percent under the table, and by 400 F closer to 2 percent under. Below about 250 F it is essentially exact.

Shortcut two: hfg is hg minus hf, and hg you can carry in your head as a short list - about 1,150 Btu/lb at 0 psig, about 1,165 at 15 psig, about 1,190 at 100 psig. Subtract the hf you just estimated and you have latent heat.

Those two together give you Tsat, hf and hfg from a pressure and nothing else. Specific volume does not have a shortcut worth the risk: vg times absolute pressure is roughly 400 in these units for low-pressure work and it is within a few percent below about 15 psig, but by 100 psig that estimate runs roughly a tenth low. Size pipe off the table, not off the estimate.

The artifact: a plant card, filled in

Here is the card for a real arrangement - a boiler holding a 100 psig header, a pressure reducing valve feeding a 15 psig building main, a flash vessel at 5 psig, and a vented receiver at 0 psig, with make-up water arriving at 60 F.

Enthalpies are Btu per pound, specific volume is cubic feet per pound, and all of them are rounded from published tables. Check the two shortcuts against the card before you trust them anywhere else: 338 minus 32 is 306 against a tabulated 309, about 1 percent light exactly as described; 250 minus 32 is 218 against 219; 212 minus 32 is 180 exactly. And 1,190 minus 309 is 881, which is the hfg column.

Now answer three plant questions off the card and nothing else.

How much steam does a 1,000,000 Btu/hr load on the building main take? The coil condenses steam and drains at saturation, so it uses hfg at 15 psig: 1,000,000 divided by 946 is about 1,060 lb/hr.

How much of the boiler's work is sitting in the condensate leaving that coil? The boiler added hg at 100 psig minus the make-up enthalpy, 1,190 minus 28, or 1,162 Btu/lb. The condensate leaving the coil holds hf at 15 psig minus that same make-up baseline, 219 minus 28, or 191 Btu/lb. That is 191 divided by 1,162, about 16 percent of everything the boiler put in, walking out of the building every time that condensate goes to a drain instead of back.

How much of it flashes when it hits the receiver? Condensate at 15 psig dropping to 0 psig cannot stay liquid at 219 Btu/lb, because saturated liquid at 0 psig only holds 180. The surplus, 219 minus 180, is 39 Btu/lb, and it boils off part of the flow at 970 Btu/lb: 39 divided by 970 is about 4 percent by mass. Do the same drop from the 100 psig header instead and the surplus is 309 minus 180, which is 129, so 129 divided by 970 is about 13 percent. The higher the pressure you are draining from, the more of the flow arrives as vapour, and the two flash calculations differ by more than a factor of three across this one plant.

Notice that the second and third answers used the same column for two different purposes and never touched hfg at the source pressure. That is the whole skill.

What would change these answers

Superheat takes you off this table entirely. A saturation table only describes a fluid at its boiling point. If steam is superheated, its temperature is above Tsat and the pressure no longer tells you the temperature, so you need a superheat table and a second measurement. Most building service steam is saturated or close to it, but steam downstream of a pressure reducing valve picks up a small amount of superheat as it expands, and steam from a firetube boiler carries a little moisture instead, which pushes the useful hfg the other way.

Altitude moves the whole table under you. Atmospheric pressure at 5,000 feet is roughly 12.2 psia rather than 14.7, so a gauge reading 0 psig there sits on a row nearer 203 F than 212 F, and every gauge-to-absolute conversion on the card changes. Write the local atmospheric pressure at the top of the card so nobody converts with the sea-level number out of habit.

Wet steam changes what a pound delivers. Steam carrying 3 percent moisture delivers roughly 3 percent less latent heat per pound than the hfg column says, because that fraction of the mass is already liquid. If a plant's steam quality is poor, the card is still right and your steam is not.

Verifying a value you interpolated

You will need a pressure that is not a row. Linear interpolation between two rows is fine, but know which way it lies. The saturation curve is concave, so a straight line drawn between two rows sits below the true curve everywhere in between: interpolate Tsat halfway between 0 and 15 psig and you get about 231 F against a true value near 233 F.

That is a fixed systematic bias with a known sign, not a random spread. It does not average out over several interpolations, and it does not shrink because you did it carefully. It gets worse the further apart your two rows are and the lower on the curve you are working, so on low-pressure work take rows close together or read a finer table. Where you are comparing two interpolated values from the same region, the bias largely cancels in the difference, which is why an interpolated flash fraction is more trustworthy than an interpolated absolute temperature.

References

• ASME Steam Tables (IAPWS industrial formulation) as the source of the tabulated saturation properties used on the card above
• 29 CFR 1910.133 for eye and face protection when reading or replacing pressure instruments on a live system
• See related: Why Pressure Sets Temperature and What That Decides; What Latent Heat Buys You in a Distribution System; Why Condensate Is the Expensive Half of a Steam System