Warming a Steam Line Back Up Without Breaking It

Purpose

To bring a steam line from cold to operating pressure at a rate the drainage can keep ahead of. The controlling variable is not the position of a valve, it is the rate of pressure rise, and every step below exists to slow that rate and to prove, before each increase, that the line ahead is clear. A warm-up done correctly is a sequence of holds with short openings between them.

The sibling article on why a cold start is the most dangerous moment sets out why the load and the drainage capacity are at opposite extremes during this window; that arithmetic is not repeated here.

Scope

Applies to bringing an isolated steam main, branch or header from ambient temperature up to operating pressure after a shutdown of any length, on systems at any pressure that a shop would encounter in building service or light process work.

Does not apply to a line whose joints have been broken. Re-pressurising piping after a repair is a first pressurisation and is governed by the test requirements of ASME B31.1 Power Piping in the edition your jurisdiction or contract has adopted, which binds the installation, and that testing precedes anything in this procedure. Does not apply to boiler start-up itself, which is the boiler manufacturer's procedure and, where your state's boiler law adopts it, the National Board Inspection Code, in the edition adopted by your state's boiler law, which binds the owner and reaches the service contractor through the jurisdictional inspector.

Roles and responsibilities

On a short run one person may hold the walker and lead roles, but not during the first admission of steam. During that step the lead is at the valve and someone else is on the line. Where that walker stands is part of the step: out of the plane of every flange and threaded joint and off the outboard side of every elbow, for the whole admission, taking far-end readings from a position picked during the cold walk. A joint that has been cold for two weeks is most likely to part on the first admission, so readings that need somebody standing at a fitting wait for a hold rather than an admission.

Procedure

1. Establish why the line was down. A line isolated for a shutdown and a line isolated for a repair get different treatment, and only the first is covered here. If any joint, gasket, valve or section was disturbed, stop and confirm with the authorising supervisor that the code testing above has been completed and accepted. Skipping this step means a warm-up doubles as an unwitnessed pressure test on a joint nobody proved, at head height, in an occupied building.

2. Remove isolation under the site's energy control procedure. Locks applied under 29 CFR 1910.147 come off only by the person who applied them or under the site's documented transfer procedure. Where the line's isolation includes electrical elements such as a motor-operated valve or a tracing circuit, that half is governed by 29 CFR 1910.333(b)(2) rather than 1910.147, and any proving of dead conductors follows the live-dead-live sequence in NFPA 70E-2021, 120.5 as adopted by your employer's electrical safety program.

3. Walk the entire run before admitting anything. The walker confirms, in this order: no personnel working on or under the line; every walkway clear of every point that will discharge; every manual drain and blowdown pointed somewhere nobody stands, re-routed if not; hearing protection available where a discharge is expected above the action level in 29 CFR 1910.95; nothing stored under the line that a discharge would reach.

This walk is where the re-routing gets done. Once steam is admitted, standing in front of an outlet to redirect it is not an option.

4. Open every manual drain on the run to a safe discharge and leave them open. Open each one slowly, standing to the side of the outlet rather than in front of it. Expect water. A drain producing nothing at this stage is either blocked or on a section that has already drained by gravity, so log it and continue.

The drains carry the warm-up load while the trap differential is still near zero. Traps alone at this stage pass a small fraction of their rated capacity.

5. Confirm the vents are open and the trap isolation valves are open. A closed trap isolation valve left over from the last job produces a section that never drains, and the venting duty is at its maximum in the first minutes rather than later.

6. Admit steam through the warming valve or bypass, never through the main isolation valve. Operate it from the side, out of the line of the bonnet. Where no bypass exists, open the main isolation a small fraction of a turn and leave the handwheel there.

The bypass limits the rate mechanically rather than by judgment. A large isolation valve cracked by hand admits far more steam than the operator intends, and the operator does not find out for several minutes.

7. Hold at minimum pressure for at least 20 minutes on a line that has been cold overnight, and longer after a shutdown of several days. Twenty minutes is a common starting point rather than a rule; tune it to your line and lengthen it whenever the far end is still climbing at the end of the hold. During this hold, heavy water discharge from the drains is correct and expected. This is the whole warm-up load leaving.

8. Stop on any hammer, at any point, without exception. A single report during warm-up means condensate has collected somewhere in a quantity the drainage did not take. Close the warming valve, keep the drains open, get everyone out of the plane of the flanges and the outboard side of the elbows, and let the line drain before any further admission. Pushing through hammer to get to pressure is the specific decision that parts joints, and it is usually made because someone is behind schedule.

9. Raise pressure in steps of no more than about a quarter of the final gauge pressure. After each step, hold until the far-end pipe surface temperature stops rising. The acceptance is the temperature levelling off, not the temperature matching a table: a surface reading taken at a bare fitting sits a few degrees below the internal saturation temperature because of the film and the metal gradient, so a plateau a few degrees under saturation for that step is normal. A reading that plateaus well below saturation, more than about ten degrees, means air or water is still occupying the far end and the step is not finished.

Read that temperature with a contact probe or on a patch of high-emissivity tape. Bare shiny steel reads low on an infrared thermometer, sometimes by tens of degrees, which would make a cold section look acceptable.

10. Close the manual drains one at a time, from the source end toward the far end, as each changes from producing water to producing a clear flash plume. Observe that change from outside the discharge path and never by hand. Closing them all at once, or closing from the far end backward, leaves a section with no escape route while water is still arriving at it.

11. Open the main isolation valve fully only once the line is at pressure and the bypass has equalised the two sides. Opening a large valve against a differential produces its own transient and damages the seat.

12. Walk the line hot. Confirm every drip trap is discharging, confirm guides and anchors have moved and nothing is bottomed out, and look for leaks against a dark background from a distance. Do not search for a steam leak by hand or with a bare arm; steam at usable pressure is invisible for the first stretch out of a leak and it cuts before it feels hot.

13. Do not tighten a bolted joint that is hot. A weeping flange at temperature gets isolated, depressurised to zero confirmed on a gauge known to be good, drained, and cooled below 120 F verified with a non-contact reading from arm's length before a wrench touches it. Re-torquing hot puts a person's hands and face beside a joint that is being disturbed while it still contains energy, and the gasket relaxation that made it weep is not corrected by adding load to hot bolts.

14. Record the log. Times, pressures at each step, far-end temperatures, which drains produced what, and anything found on the hot walk. The log is what tells the next person whether twenty minutes was enough.

Worked example: 400 ft of 4 in at 15 psig after a two-week shutdown

Final pressure 15 psig, so steps of about a quarter of that: 4, 8, 11, then 15. Saturation temperatures read from a steam table for each step: about 224 F at 4 psig, 235 F at 8 psig, 242 F at 11 psig, 250 F at 15 psig.

Reading the log. The first hold ran 35 minutes rather than 20, and the extension was the right call: at the 20-minute mark the far end was at 118 F and climbing, which is the condition step 9 says means the step is not finished. Every subsequent gap between surface reading and saturation was 2 to 3 F, comfortably inside the ten-degree limit, which is what a line with no air pocket and no flooded section looks like.

Total elapsed from first admission to full pressure was 1 hour and 25 minutes, from the start of the line walk to full pressure 1 hour and 40 minutes, and to the end of the hot walk 1 hour and 55 minutes. Compare that against what a timeclock start does: full boiler pressure onto a cold 400 ft main in the time it takes the control valve to stroke. The load is identical in both cases, because it is set by the mass of the pipe. The rate is not.

The one entry that was a finding rather than a step. The bottomed-out hanger was found on the hot walk and not corrected there. A hanger out of travel at operating temperature is carrying the line's expansion as a restraint, and correcting it on a hot pressurised line means releasing a support on pipe that will move when it is released. It goes on the list for the next cold shutdown, isolated and locked out under 29 CFR 1910.147, with the line drained and below 120 F before anything is loosened.

What would have changed the procedure. A longer shutdown, colder ambient or a larger pipe size all lengthen the holds rather than changing the steps. A line with no manual drains at its drip points changes it fundamentally: the traps alone then have to carry the warm-up through the low-differential period, so the answer is a far slower ramp and, before the next season, drains added at the drip legs while the line is down.

References

• ASME B31.1 Power Piping, in the edition your jurisdiction or contract has adopted, which binds the installation, for post-repair pressure testing before any warm-up
• National Board Inspection Code, as adopted by your state's boiler law, and the boiler manufacturer's own start-up procedure, for the boiler side
• 29 CFR 1910.147 for isolation and lock removal, 29 CFR 1910.333(b)(2) with NFPA 70E-2021, 120.5 as adopted by your employer's electrical safety program for any electrical isolation, and 29 CFR 1910.95 for hearing protection at open discharges
• ASME Steam Tables, or an equivalent saturated-steam table, for the saturation temperature at each pressure step
• See related: articles in this library on why a cold start is the most dangerous moment, and on why a drip leg exists and what happens without one