How to test a steam trap: temperature, sound and sight

You test a steam trap three ways: temperature across it, sound inside it, and what comes out downstream. No single one of those is conclusive on its own — a hot trap can be dead, a noisy trap can be healthy. Two methods agreeing is what gives you an answer, and all three can be done with the line running.

Here is the sequence we use on survey work, and what each reading actually tells you.

Why a failed trap is worth finding

A trap that fails open does not stop the process. That is exactly why it goes unnoticed for years. It quietly passes live steam into the condensate return, and the plant pays for it three times over: in the fuel to make the steam, in the makeup water and treatment chemicals lost through the flash tank vent, and in the return line itself, which starts hammering and eating its own fittings.

A trap that fails closed is louder about it. The heat exchanger or coil behind it floods with condensate and stops holding setpoint, and someone opens the bypass to keep production moving — which is worse than the failed trap, because a bypass is a trap failed wide open with nobody logging it.

In plants that have gone three to five years without a survey, finding 15 to 30 percent of the trap population failed is normal, not unusual.

Method 1: temperature across the trap

Take a surface reading about six inches upstream of the trap and another six inches downstream, on bare pipe, with an infrared gun or a contact probe. What you compare it against is the saturation temperature for the line pressure — not some fixed number.

Read it like this:

• Inlet at saturation, outlet noticeably cooler. Normal. The size of the drop depends on the trap type and the return system, but you want to see a difference.
• Inlet and outlet both at saturation, no drop. Suspect blow-through. The trap is passing live steam.
• Inlet well below saturation. The trap is not the problem — you have a plugged strainer, a closed isolation valve, or a line that lost its steam supply upstream.
• Both sides near ambient. Failed closed, plugged, or the equipment behind it is simply not running.

The trap that fools people is a thermostatic trap. It is supposed to hold condensate back until it subcools, so a cool inlet on a balanced-pressure or bimetallic trap can be perfectly correct behaviour. Know what type you are standing in front of before you condemn it.

Method 2: listen to it

An ultrasonic gun with a contact probe is the right tool. A long screwdriver with the handle against the bone behind your ear will get you most of the way there, and every steamfitter over fifty knows this.

What you are listening for is the discharge pattern, not the volume:

• Inverted bucket: distinct, intermittent bursts. Continuous rushing means the bucket lost its prime or the seat is cut.
• Thermodynamic disc: should snap open and shut somewhere in the range of four to ten cycles a minute. Faster than about ten and the disc or seat is worn — that is rapid cycling, and it wastes steam on every cycle.
• Float and thermostatic: a modulating, more or less continuous flow. Silence on a loaded F&T is a bad sign, not a good one.

A steady, high-pitched rush that never stops, on any trap type, is live steam going through.

Method 3: look at the discharge

If the trap discharges to atmosphere, or the return has a sight glass, this is the fastest check there is. The skill is telling flash steam from live steam.

Flash steam is lazy. It forms an inch or two off the outlet, billows, and dissipates within a couple of feet. It is condensate re-evaporating as it drops to return pressure, and it is completely normal — often 10 to 15 percent of the condensate by mass.

Live steam is a sharp, high-velocity jet, bluish-white, and it holds its shape well past where flash steam would have disappeared. It is continuous rather than surging. Once you have seen the two side by side you will never mistake them again.

What a failed steam trap looks like from the plant floor

You do not always need a gun to know something is wrong. These are the symptoms that should put a survey on the schedule:

• Water hammer in the condensate return, especially on startup
• A heat exchanger, jacket or coil that can no longer hold its setpoint at full steam demand
• A return line that is too hot to touch along its entire run instead of just near the traps
• The flash tank vent blowing a continuous plume rather than a lazy intermittent one
• Makeup water consumption climbing with no change in production
• A bypass valve that somebody cracked open and nobody closed

How often to survey

Once a year is the floor for a low-pressure system. Quarterly is reasonable above 100 psig, where a single failed trap is expensive enough to pay for the whole survey. What matters more than the interval is that the results get written down — tag number, trap type, size, pressure, verdict, date. The second survey is worth far more than the first, because a trap that failed twice in two years is telling you something about sizing, water hammer or dirt that a replacement will not fix.

If you want to practise reading the symptoms before you are standing in front of a live header, our interactive steam trap survey trainer walks through the same decisions with simulated readings. The downtime cost calculator is useful for putting a number on what the failed ones are costing.

When to call someone

Call if you have never surveyed the system and do not know how many traps you own, if the return line is hammering hard enough to move pipe, or if the same trap has failed twice. Keystone runs steam trap surveys on-site across Southern California — Long Beach, the South Bay, Orange County and the Inland Empire — as part of a preventative maintenance program or as a standalone job, and we hand back a tagged inventory rather than a verbal count.