Why a Refrigerated Dryer Fails Quietly
Why this matters
A refrigerated air dryer almost never fails the way a compressor fails. It does not trip, it does not alarm, it does not stop. It slides, and while it slides it keeps running, keeps its light green, and keeps handing the plant air that is wetter every week. By the time anyone connects the water at a tool to the machine in the compressor room, the shop has usually replaced consumables twice, blamed a hose, and rebuilt something that was never wrong.
The reason is structural, not a design defect: nothing on a standard refrigerated dryer measures the thing the dryer is for.
Isolate before you open a pressurised path
Everything below involves reading pressures and temperatures on live equipment. Nothing here requires opening a joint. Where a diagnosis does require opening one, close the upstream isolation, vent the section, confirm zero on the section gauge with the vent still open, and lock and tag the isolation under 29 CFR 1910.147, which governs stored mechanical and pressure energy. Where a cover has to come off a control panel, that is electrical work: de-energize at the disconnect, lock and tag under 29 CFR 1910.333(b)(2), and prove dead live-dead-live per NFPA 70E-2021, 120.5, in the edition your employer's electrical safety program has adopted. Do not aim a vent or a bleed at skin; compressed air driven under the skin can embolise.
Nothing on the machine watches the outlet
A refrigerated dryer works by chilling the air until water condenses out, separating that water, then reheating the air against the incoming stream. Its outlet dew point is set by how cold the evaporator holds the air, and that is set by how much heat the machine can reject, how much load is coming in, and how hot the incoming air is.
Here is what the machine actually tells you, and what each indicator does not cover.
| Indicator | What it reports | What it cannot see |
|---|---|---|
| Green run light | The controller has power and no tripped fault | Anything about outlet moisture |
| Refrigerant gauge or dew point dial | Evaporator condition, on a scale often marked as a coloured band | Actual outlet dew point at line pressure, under load |
| Separator sight glass | Whether liquid is present at that instant | Whether the water is being carried on past it as droplets |
| Drain discharge | That the drain fires | Whether it fires often enough for the current condensate rate |
| Amp draw at the compressor | The refrigeration circuit is running | Whether it is running against a fouled condenser |
Every row on that table is a health indicator for the box. The gate that actually matters, measured pressure dew point at line pressure under real load compared against the coldest pressurised metal in the plant, is covered in a sibling article and is not re-derived here.
The three routes a quiet failure takes
Quiet failures cluster into three routes, and they need different fixes.
Heat rejection. A matted condenser, a fan that has lost blades or reversed, a filter mat nobody knew existed, or a compressor room that has become an oven. All of them raise condensing pressure, which raises evaporator temperature, which raises outlet dew point. The mechanism and the symptom point the same way: worse heat rejection means warmer, wetter air out. At the other end of the same range, a clean condenser in a cool room gives an outlet dew point at or below the nameplate, and the limit there is the freeze protection, because the evaporator cannot be allowed below the freezing point of the condensate it is making.
Load growth. Nobody re-sizes a dryer. Plants add a line, add a shift, add a second compressor, and the dryer that was sized correctly seven years ago is now sitting below the flow it sees at peak. Capacity ratings are published at reference conditions and de-rated by the manufacturer's correction factors outside them, so a dryer can be simultaneously within nameplate and under-sized for the day it is having.
Air-side bypass. A three-valve bypass left cracked open blends untreated air into the outlet, and the dryer will be perfect while the plant gets wet air. An internally leaking air-to-air heat exchanger does the same thing invisibly. This route is the one where the machine is genuinely faultless and the pipework is lying.
The call: consumables at the far end of the shop
A fabrication shop reported doubled consumable consumption at a plasma cutting station and intermittent water in the filter bowl feeding it. The station is the longest branch in the building. The dryer had never been serviced, had never alarmed, and had a green light.
Eliminated: the point-of-use drain. First thing checked, because it is the cheapest and it is the usual suspect. It was discharging on cycle, with a visible slug of condensate each time. A drain that is firing does not explain water arriving downstream of it, and drains are covered in their own article rather than here.
Eliminated: low refrigerant charge. The temptation was to call it low charge because the outlet was warm. But the direction is wrong. Losing charge lowers evaporator pressure and temperature, and on a refrigerated dryer that shows up as icing and a rising air-side pressure drop, not as a warm outlet. This machine had no ice, no unusual air-side drop, and a suction reading that sat high rather than low. Every one of those points away from charge and toward heat rejection.
Eliminated: the bypass. Both bypass valves were confirmed shut and the centre valve confirmed open, physically, by hand, not by looking at the handle position from across the room. A bypass valve that has been cracked for a year has usually been tagged by somebody who meant to come back.
The number that split the day
Measured pressure dew point at the dryer outlet, at line pressure, under load, following the method in the sibling article, including its 18 F margin below the coldest pressurised metal:
- 9 am, plant demand about 60 percent of the dryer's nameplate flow, compressor room at 78 F: outlet 36 F.
- 3 pm, plant demand about 88 percent of nameplate flow, compressor room at 108 F: outlet 61 F.
The coldest pressurised metal anywhere in the building measured 55 F, all runs indoors, so the requirement is 55 minus 18, which is 37 F.
The morning reading passes by 1 F. The afternoon reading fails by 24 F. Same machine, same day, no fault code between the two.
That is the whole shape of a quiet failure in two lines. A tech who arrives at 9 am, measures correctly, and reports a pass has done honest work and produced a wrong answer, because the plant does not run at 9 am conditions.
Why the room was the fault
The correction table in the manual, read at the room temperature recorded in the afternoon, put the unit at roughly four fifths of its nameplate capacity. Afternoon demand was about 88 percent of nameplate. Demand above capacity means the evaporator cannot pull the air down to its design temperature, and the outlet dew point rises until the machine reaches a new balance. That is not a fault. That is the machine reporting the room.
The room got that hot because the compressor's own heat had nowhere to go. A rotary screw rejects the large majority of the electrical energy it draws as heat, and this room's only ventilation was a louvre that had been blocked by pallet storage. The dryer's condenser was drawing the compressor's discharge heat straight back in. Nothing in that chain is a dryer defect.
There was a second, smaller contribution: the condenser coil face was matted, which the shop had never inspected because the dryer had never asked for anything.
What was done and what would have changed it
Three actions, in the order they were taken, because the order was itself the finding. Clear the louvre and restore the room's ventilation path. Clean the condenser face with the unit de-energized, locked and tagged under 29 CFR 1910.333(b)(2), wearing eye protection and a respirator selected under a written program meeting 29 CFR 1910.134, because a matted coil in a fabrication shop holds grinding dust and the route of that hazard is inhalation, which a glove does not address. Then re-measure at the worst hour.
What would have changed the recommendation: if the afternoon reading had come back at or below the requirement after the room was fixed, the story ends there and the dryer stays. If it had come back still above the requirement with the room cool and the coil clean, the finding becomes load growth rather than heat rejection, and the conversation moves to a larger dryer or a second unit staged with the second compressor. Those are different quotes, and the only thing that separates them is a second measurement after the cheap fix.
How they confirmed it
Re-measured at 3 pm on a comparable production day, at line pressure under load, with the room logged: room at 84 F, outlet dew point 39 F. That is 2 F above the 37 F requirement, which is a fail, not a pass, and the shop was told so in those words rather than being handed an improvement percentage.
The remaining 2 F came out by fitting the plasma station with its own point-of-use dryer, which is the standard answer when one branch needs air that is drier than the plant needs. Consumable consumption at that station returned to its previous rate over the following month, which is corroboration rather than proof: it is consistent with the finding and it is not, by itself, a dew point measurement.
Report the honest version. The plant now meets its requirement at the worst hour with a small margin, and it will drift again the next time somebody stacks pallets in front of a louvre.
References
- 29 CFR 1910.147 for isolating stored pressure energy before opening an air path, and 29 CFR 1910.333(b)(2) with NFPA 70E-2021, 120.5, in the edition your employer's electrical safety program has adopted, for de-energizing and proving dead before a control panel or a condenser fan is worked on
- 29 CFR 1910.134 for the written respiratory protection program covering respirator selection when cleaning a coil loaded with shop dust
- Dryer manufacturer documentation for rating conditions, ambient and inlet-temperature correction factors, and freeze protection limits
- See related: How to Tell Whether a Dryer Is Doing Its Job; What a Drain Trap Is Doing and Why It Matters