What a Drain Trap Is Doing and Why It Matters
Why this matters
A compressed air drain has exactly two jobs and they pull against each other: get the liquid out, and do not let the air out with it. Every drain type on the market is a different compromise between those two, and every drain in the field is failing at one half or the other. Nobody knows which, because nobody has ever walked the plant and written it down. That walk is the deliverable here, and the artifact it produces is worth more to a customer than any single repair on it.
Isolate before a drain comes off
A drain body, a bowl and a drip leg are all part of a pressurised path. Close the isolation upstream, vent the section through its own bleed, confirm zero on the section gauge with the bleed still open, then lock and tag the isolation under 29 CFR 1910.147, which covers stored mechanical and pressure energy, before a fitting is broken. A drain body unscrewed under pressure comes off as a projectile, and the leg it was fitted to whips.
Condensate from a lubricated machine carries oil. Wear chemical-resistant gloves and sealed eye protection when catching or handling it, and route its disposal to your local sewer authority's pretreatment rules or to a licensed hauler rather than to a floor drain. Do not stand in front of a drain outlet while cycling it: the discharge is liquid at line pressure and it carries scale, and compressed air driven under the skin can embolise. Where a drain has to be test-fired, put a hose on the outlet into a catch container first, then stand to the side of the container rather than over it.
The two halves of one job
State it as a pair, because a drain that is excellent at one half and terrible at the other is a very common object.
Removing liquid. Condensate that is not removed collects in the low point it formed in, then travels down the header as slugs when flow picks up, and arrives at a tool or a coating gun as water. It also drowns whatever filter it reaches, and it rusts the inside of the pipe, which then sends scale downstream forever.
Keeping air. A drain that opens to atmosphere is a hole in the system for as long as it is open. Air leaving through a drain is compressor work thrown away at full specific power, and unlike a leak in a fitting it is usually somebody's deliberate setting.
Everything below sorts drains by which half they fail.
Where the condensate actually forms
The water is not evenly spread through the plant, and drains sized as though it were will be wrong at both ends.
Air leaving a compressor is hot and saturated. The aftercooler drops its temperature sharply, and the large majority of the total water in the system, commonly cited as roughly three quarters, condenses right there and is caught by the separator. The receiver drops more as the air rests and cools further. The dryer takes the remainder down to its rated dew point. Downstream of a working dryer, a properly operating system should be producing almost no new condensate at all, which is why a filter bowl at a point of use filling with water is a finding about the dryer, not about that bowl.
So the aftercooler separator and the receiver handle the big volume, the dryer handles a modest one, and the point-of-use bowls should be handling almost nothing. And the whole picture scales with the weather: intake air that is warmer and more humid puts far more water into the system, so a timer drain tuned in February is under-draining in August, and a plant that has never seen water in winter can flood in a wet July.
Drip legs and takeoffs
Where a branch leaves a header, geometry does half the drain's work for it.
branch rises from the TOP of the header
|
==================+========================
header, sloped so it falls toward the leg
|
drip leg, a
dead-end drop
below the header
|
drain at the
lowest point
A branch taken off the top cannot receive liquid running along the bottom of the header. A branch taken off the bottom or the side receives every slug that passes. The drip leg is a deliberate dead-end at the low point where liquid collects instead of continuing, and the drain sits at the very bottom of it. A drip leg with the drain fitted on its side leaves a permanent sump of standing water underneath, which is where the rust starts.
Which half each drain type fails
| Type | How it decides to open | The half it fails, and how |
|---|---|---|
| Manual valve | A person | Fails at removing liquid, because nobody opens it; or, left cracked, fails at keeping air permanently |
| Timer solenoid | A clock | Fails at both ends of the year: too short an open under-drains in humid weather, too long an open blows air in dry weather |
| Mechanical float | Liquid level lifts a float | Usually fails at keeping air, because debris on the seat stops it closing; sometimes sticks shut on sludge |
| Electronic level sensing | A level sensor opens a valve and closes it before air breaks through | Usually fails at removing liquid, because a fouled sensor or a lost supply stops it firing; it is the type most likely to have a fault indicator |
The organising idea is one line long. A drain that opens on time is guessing at a quantity that varies with ambient humidity and plant load. A drain that opens on level measures it. That is the whole reason level-sensing drains exist, and it is also why they need their sensors kept clean: a measurement device with a fouled sensor is worse than a guess, because it looks authoritative.
There is one legal fixed point in this. 29 CFR 1910.169(b) requires a drain at the lowest point of an air receiver and requires that the receiver be drained, so a receiver whose only drain is a manual valve that nobody opens is not merely poor practice.
The artifact: a drain inventory, filled in
One row per drain, walked in order from the compressor outward, with the evidence that produced the verdict. An ultrasonic leak detector is what hears a drain that is not seating; the ear alone will not, over a running plant.
| Location | Type | Test | Verdict |
|---|---|---|---|
| Aftercooler separator | Timer solenoid | Watched a full cycle at peak humidity: 5 second open every 2 minutes, liquid for the first 2 seconds, audible air for the last 3 | Fails at keeping air |
| Receiver | Manual valve | Found closed, no record of last opening, slug of rusty water on cycling | Fails at removing liquid |
| Dryer separator | Mechanical float | Cycled on the test lever, discharged, seated clean, silent on ultrasonic | Pass |
| Coalescer bowl | Mechanical float | Continuous flow audible on ultrasonic, does not seat | Fails at keeping air |
| Afterfilter bowl | Electronic level sensing | Fault indicator lit, valve holding closed, sensor coated | Fails at removing liquid |
| East branch drip leg | Manual valve | Found cracked open and zip-tied in position | Fails at keeping air |
| Paint bench point of use | Manual valve | Cycled, nothing discharged, port packed with sludge | Fails at removing liquid |
Seven drains. One pass. Six findings, split three and three: three failing at keeping air, three failing at removing liquid.
Reading the filled-in card
The aftercooler timer is the biggest single item on the sheet and it is a tuning problem, not a hardware failure. Three of the five open seconds are moving air rather than liquid, which is 60 percent of the open time wasted, and that path is wide open for 3 seconds in every 120, so about 2.5 percent of the plant's running hours have a full-bore hole in the system. The immediate action is to shorten the open time until air no longer follows the liquid, then re-check it in the wettest month of the year, because the setting that is right in February will under-drain in August. The permanent action is a level-sensing drain, which stops the seasonal re-tuning from being a task at all.
The receiver is the compliance item. It carries the largest condensate volume in the plant behind the aftercooler, its only drain is manual, and 29 CFR 1910.169(b) requires a drain at the lowest point of the receiver and requires it be drained. A level-sensing drain goes on, and the existing manual valve stays as the isolation and the backup rather than being removed.
The two float drains are the same part in opposite states, which is the useful pairing on this sheet. The dryer's seats and is silent. The coalescer's does not seat and runs continuously. Same mechanism, same age, and only the ultrasonic test separates them, because a float drain that is not sealing sounds like nothing at all from three feet away over a compressor.
The electronic drain's fault light was doing its job and nobody was reading it. It is the only drain on the sheet that told anyone it had failed. Clean the sensor, restore it, and put the fault indicator on the walk list, because the failure mode of this type is silence at the bowl and a lit LED nobody looks at.
The east branch drip leg is the one that gets handed off. A cracked-open manual drain is not a drain problem, it is a permanent, sized-by-nobody hole in the system, and it is covered in its own article rather than solved on this sheet.
The paint bench port is the one to be careful about. Nothing came out, which reads like a pass to somebody in a hurry. A port packed with sludge is a drain that has been failing for a long time, and it sits at the point of use that is least tolerant of contamination in the whole building. Clear the port with the section isolated, vented and verified at zero, and then ask why a bowl downstream of a working dryer was accumulating anything at all, because that answer is upstream.
Two things the card deliberately does not record. It does not record how long each drain has been in that state, because nobody can know that and a guess in a column becomes a fact by the time it reaches the customer. And it does not price anything, because the sheet's job is to establish which half of the job each drain is failing, and the repair decisions follow from that rather than from the sheet.
References
- 29 CFR 1910.169(b), which requires a drain at the lowest point of an air receiver and requires the receiver to be drained, and 29 CFR 1910.147 for isolating and verifying depressurisation before a drain or a bowl is opened
- Local sewer authority pretreatment rules for oil-bearing compressed air condensate disposal
- Drain and dryer manufacturer documentation for cycle settings, sensor cleaning intervals and fault indication
- See related: Why a Manual Drain Left Cracked Open Is a Permanent Leak; How to Tell Whether a Dryer Is Doing Its Job