What to Record About a Steam System So the Next Visit Is Shorter

Why this matters

Most steam service records are unusable a year later, and it is not because they are thin. It is because of what steam is. In a steam system a temperature only means something against the pressure it was taken at, and a pressure only means something against the place it was read. Record either one alone and you have written down a number nobody can convert back into a condition, including yourself.

The practical cost is that every visit starts from zero. A shop that has been to a building four times has four visits' worth of readings and no trend, so the fifth visit re-diagnoses from scratch, which is how the same trap gets replaced twice in two years and the actual fault survives both repairs.

A steam reading is recorded as a pair, at a named location, under a named condition. Everything below is that one rule applied to the fields that matter.

The hazards these instructions create

Fitting a gauge where one does not exist is a scheduled job with isolation, not part of a walk: isolate the supply and the return, lock and tag it as control of hazardous energy under 29 CFR 1910.147 in general industry or under the construction energy control requirements of 29 CFR Part 1926 on a site, drain to a point piped away from any walkway, confirm zero on a gauge you have just watched read something else, and let the metal fall below 120 F verified with a non-contact reading taken at arm's length. A note in a logbook saying a line was isolated is not an isolation; the standard requires the physical device and the tag, and the record is evidence of it rather than a substitute for it.

Measuring a lift with a tape puts you next to hot pipe on a ladder. Work from a stable platform rather than off the piping, never use a pipe or a hanger as a handhold or a ladder rest, and keep the tape and your forearms off bare hot metal, which is at steam temperature and burns on contact.

Seating a contact probe on already-bare metal calls for gloves rated for that surface temperature and eye and face protection under your PPE hazard assessment per 29 CFR 1910.132. Do not brush, cut or lift pipe insulation to expose metal for a reading: thermal system insulation on steam and condensate piping is presumed asbestos-containing material under 29 CFR 1926.1101 in construction and 29 CFR 1910.1001 in general industry until sampled, and that is an inhalation route that gloves do not address.

Several of the readings below are taken at a cold start, which is when a main with standing condensate is most likely to hammer. Take first-call readings from a position that is not beside a main with any hammer history, and treat a main that hammers as a fault to be fixed rather than a condition to stand near and observe.

What gets recorded once, and verified rather than copied

These do not change between visits, they are wrong on drawings more often than anything else in a mechanical room, and every one of them is needed to interpret a reading taken years later. Five entries, per system:

The trap register. A permanent legible tag number on every trap, fixed to a bracket or pipe support rather than to anything that gets replaced with the trap, plus the body type, the orifice designation read off the nameplate, and what it serves. Where a nameplate is unreadable, record that it is unreadable instead of guessing, because a guessed orifice propagates into every later estimate.

The lift, measured with a tape. The vertical distance from each trap outlet up to the return main it discharges into. This is the single number most often absent from a building's records and it is load-bearing for every drainage question that will ever be asked about that equipment.

Where the return main pressure is read, by which gauge, and whether that return is nominally vented or pressurised.

The operating pressure control settings, cut-in and cut-out, and the relief valve setting, taken from the nameplate and the start-up sheet.

Which equipment has a vent and which has a vacuum breaker, whether either is piped away, and where the discharge lands. A device that is not fitted is recorded as not fitted, because "no entry" and "none present" read identically a year later and mean opposite things.

What gets recorded every visit, and the four fields each reading needs

Every reading carries all four. Drop any one and the reading is not comparable to the same reading taken next year.

Location, named specifically enough that the next person stands where you stood. "Coil return header" is a location; "at the unit" is not.

Condition. Load, time, running or first call, outdoor temperature. A steam system behaves differently at a cold start than at ten in the morning, and two readings taken at different conditions are two different measurements, not a trend.

The pressure at that location, gauge identified. On any equipment that shuts off, this needs a compound gauge that reads vacuum as well as pressure; an ordinary pressure gauge sits at zero for both a healthy shut-down space and one pulling a deep vacuum, so it cannot record the difference.

The paired reading, with the instrument named. Metal temperature, condensate state, trap state, air-side or water-side temperature rise. Name the instrument because a contact probe and an infrared thermometer are not interchangeable here: bare shiny steel reads low on infrared, sometimes by tens of degrees, which fakes exactly the fault being looked for.

A pass or a fail is not a record

This is where most trap sheets fail, and it fails quietly because the sheet looks complete.

"Trap OK" throws away everything the next visit needs. It cannot be compared, it cannot be trended, and it cannot be re-interpreted when the surrounding conditions turn out to have moved. Record the state as one of failed open, failed closed, good, or out of service with no load, and record the readings that produced that call next to it, including the two pressures. Year-over-year comparison of a trap against its own history is far stronger evidence than any single reading, because it removes model, size, pressure and installation differences at once, and that comparison is only available if the numbers survived.

The same applies to a temperature written without its pressure. A coil return header at 243 F is excellent at 15 psig and a serious finding at 60 psig, and there is no way to tell which from the number.

The filled-in record, and what the third visit does with it

A steam unit heater in a warehouse bay, tag UH-7, fed at 10 psig.

Recorded once, on the first visit:

• Trap T-118, float and thermostatic, orifice designation as read from the nameplate, serving UH-7.
• Compound gauge G-22 fitted at the unit heater inlet, reading vacuum and pressure.
• Lift, trap outlet to return main, measured with a tape: 12 feet.
• Return main pressure read at gauge G-3, at the receiver in the basement. Return is nominally vented.
• Vacuum breaker fitted at the top of the unit heater header. No separate air vent; the thermostatic element in T-118 is the venting path.

Visit one, January, first call, outdoor 18 F. G-22: 10 psig running. G-3: 1 psig. Return header metal, contact probe on an already-bare fitting: 234 F. Trap T-118: discharging, good. Air temperature rise across UH-7: 42 F.

Visit two, the following January, same hour and a comparable outdoor temperature. G-22: 10 psig. G-3: 2 psig. Return header metal: 234 F. Trap T-118: discharging, good. Rise across UH-7: 40 F.

Visit three, the third January, same condition. G-22: 10 psig. G-3: 3 psig. Return header metal: 233 F. Trap T-118: discharging, good. Rise across UH-7: 33 F.

What the record says that no single visit could. Saturation at 10 psig is about 240 F, so the return header at 233 to 234 F sits 6 to 7 F under saturation across all three years, inside the ten degree working limit that says a space is genuinely full of steam. The supply pressure never moved. The trap tested good every time and, since it is being compared against its own history rather than against a table, that call is trustworthy rather than reassuring.

What moved is the return main and the output. G-3 went 1, 2, 3 psig. The rise across the unit heater went 42, 40, 33 F.

Convert the return main figures into what the trap actually faces. The outlet side is the return main pressure, plus the 12 foot lift at 2.31 feet of water per psi which is 5.19 psi, plus about 1 psi for friction. Year one: 1 plus 5.19 plus 1 is 7.19 psig required, against 10 psig at the unit, leaving 2.81 psi of margin. Year two: 8.19 required, 1.81 psi of margin. Year three: 9.19 required, 0.81 psi of margin.

For a trap passing subcooled condensate through a fixed opening, flow scales roughly with the square root of the differential, so the square root of 0.81 over 2.81 is 0.54: the trap has about 54 percent of the capacity it had in year one, through no change in the trap. Where condensate arrives at saturation and flashes across the seat, real capacity falls below that, so 54 percent is the optimistic end.

And the record predicts the date. At a return main of about 3.8 psig the required figure reaches 9.99 psig and the margin is gone, at which point UH-7 stops draining at any load. One more season of the same drift gets there.

The fault is not on this unit heater. A nominally vented return main climbing 1 psig a year is live steam entering that return, which on a common return is somebody else's failed-open trap on a higher-pressure branch, and it is eating the margin of every device discharging into it. The record turns a vague "this heater is not as good as it used to be" into a specific instruction to go find a trap somewhere else in the building.

What a normal record would have produced instead. Three sheets each reading "trap tested good, unit heating," and a fourth visit that replaces T-118, finds no improvement, and quotes a new unit heater. Nothing on those three sheets is false. All three are simply unreadable next to each other, which is the only place the finding exists.

References

• ASME Steam Tables, or an equivalent saturated-steam table, for the saturation temperature at each recorded gauge pressure
• Trap manufacturer capacity data at the actual differential, which owns the real capacity where the square-root approximation is only a bound
• 29 CFR 1910.147 for control of hazardous energy in general industry, the construction energy control requirements of 29 CFR Part 1926 on a construction site, 29 CFR 1910.132 for PPE hazard assessment, and 29 CFR 1926.1101 with 29 CFR 1910.1001 for presumed asbestos-containing thermal system insulation
• See related: How to Test a Steam Trap Without Taking It Apart; How to Survey a Trap Population and Rank What to Fix; What a Vacuum Breaker Is For