I've signed off on roughly 380 compressed air dryer units since 2021 — refrigerated, heatless desiccant, heat-of-compression, and a couple of hybrids I'd rather forget. About 12% of first deliveries get rejected at my desk. The reasons are rarely dramatic. Wrong dew point class. Missing test documentation. A timer drain where the spec called for zero-loss. Boring things. Boring things are exactly what blow up a commissioning schedule.

These are the questions buyers actually send me, answered the way I'd answer them in an email at 6pm.

  • What's the first thing you check on a datasheet?
  • Refrigerated or desiccant — how do I pick?
  • Why do two dryers with the same rated flow perform so differently?
  • What's the most common spec mistake you see?
  • How much does the dryer actually cost to run?
  • What do you ask a sourcing partner before you trust their quote?
  • What do buyers consistently overlook?

1. What's the first thing you check on a dryer datasheet?

Not price. Not the rated flow. I look for the ISO 8573-1 class string — the three numbers that describe particle, water, and oil content in the air the dryer is supposed to deliver.

It reads like 2:4:1. Particles, water, oil. The water class is where most of the argument happens:

  • Class 1 — pressure dew point of −70°C
  • Class 2 — −40°C
  • Class 3 — −20°C
  • Class 4 — +3°C
  • Class 5 — +7°C
  • Class 6 — +10°C

A refrigerated dryer usually lands at Class 4 to 6 depending on design and operating conditions. A desiccant dryer gets you to Class 2, or Class 1 with a bit of extra purge.

Now the catch. That class string is meaningless without the conditions attached to it. −40°C at what inlet temperature? What ambient? What flow? If a datasheet says Class 2 and nothing else, I send it back. ISO 7183:2007 exists precisely so manufacturers publish capacity at defined reference conditions instead of whatever number reads best in a brochure.

2. Refrigerated or desiccant — how do I pick?

Pick the dew point first. The technology is downstream of that decision.

The test I use: find the coldest surface your compressed air will ever touch. If nothing in the line drops below roughly +10°C (+50°F), a refrigerated dryer is usually enough, and it'll be cheaper to buy and far cheaper to run. If any part of the piping runs outdoors in winter, through an unheated mezzanine, or into a machine that chills its own internals, you need −40°C or lower. That's desiccant territory.

Instrument air, paint lines, powder conveying, food packaging, anything feeding a process that can't tolerate condensation — most of those land on desiccant, or on a refrigerated dryer plus a point-of-use desiccant unit.

But desiccant isn't free. A heatless (pressure-swing) dryer typically consumes 15–20% of its rated flow as purge air, and that air is already-paid-for compressed air. The desiccant bed also degrades fast if oil carryover reaches it, which means the pre-filter matters as much as the dryer. More on that below.

3. Why do two dryers with the same rated flow perform so differently?

Because the rating conditions aren't the same, and most buyers never read them.

The common reference is CAGI ADF 100: 100 psig inlet, 100°F inlet temperature, 100°F ambient. Plenty of manufacturers publish at those conditions because they're favorable. Now look at your actual plant in August. Your compressor might be discharging at 105–110°F after the aftercooler, and the dryer room might be sitting at 110°F with no ventilation.

Here's the part that catches people: a refrigerated dryer rated at 1,000 scfm at 100°F inlet may only handle 600–650 scfm if the air arrives at 120°F. The correction curve is steep. I'm not going to hand you exact factors — get those from the manufacturer's datasheet — but the shape of that curve is consistent across the industry and it's brutal.

The question everyone asks is what's the cfm rating? The question they should ask is what's the cfm rating at my worst week, not my average week?

4. What's the most common spec mistake you see?

Confusing pressure dew point with atmospheric dew point. I've watched this derail a sign-off twice.

Pressure dew point (PDP) is measured at line pressure. Atmospheric dew point is measured after expanding the air to ambient. They are not the same number and the gap is large. A dryer delivering −40°C PDP at 7 bar is roughly equivalent to −58°C to −60°C at atmospheric pressure. Flip it around: if someone tells you a dryer hits −40°C and they measured it with an atmospheric dew point meter, the actual pressure dew point is somewhere around −23°C. That is a completely different product.

So when you specify, write it as pressure dew point at X bar, and write the ISO 8573-1 class next to it. Otherwise you will spend a Tuesday afternoon on a conference call arguing about which instrument was calibrated to which reference.

Second most common: sizing for average demand instead of peak. A dryer that runs at 100% capacity all summer has no margin for a second shift.

5. How much does the dryer actually cost to run?

Almost nobody puts this in the purchase comparison. They should.

For a refrigerated dryer, the recurring cost is mostly pressure drop — typically 2–3 psi across the unit, which your compressor pays for every hour of every day. Call it 1–1.5% of compressor energy.

For desiccant, it's the purge. Take a 500 scfm heatless dryer with 15% purge: that's 75 scfm of compressed air vented continuously to regenerate the bed. At roughly 20 kW per 100 cfm of compressor output, you're burning about 15 kW around the clock. At $0.10/kWh and 8,760 hours, that's somewhere near $13,000 a year — before you've bought anything.

Don't hold me to that number. Your power rate, duty cycle, and whether the dryer runs weekends will move it a lot. But the order of magnitude is the point: the purge air can cost more over five years than the dryer did.

This is why I push back when someone asks for desiccant 'just to be safe.' If your dew point requirement is Class 4, you're paying a five-figure annual premium for margin you don't need. To be fair, there are processes where that margin is cheap insurance. You have to know which one you're in.

6. What do you ask a sourcing partner before you trust their quote?

I ask what they don't do.

Anyone can forward a spec sheet and a price. The suppliers worth keeping are the ones who tell you where their coverage ends. In 2023 we shortlisted three partners for a 600 scfm desiccant skid. The one that won wasn't the cheapest and it wasn't the fastest. It sent back a four-page letter listing which parts of our spec were underspecified — including an oil carryover limit on the pre-filter that we'd left blank — and it flagged one item where it flatly said another vendor was the better call.

That last part is the tell. A partner who says this isn't our strength, here's who does it better has just told you their other answers are probably honest too. A partner who says they can supply everything from the compressor to the drain valve to the leak-detection survey is telling you something else, and I've learned to hear it.

Concretely, I want four things before I approve a sourcing partner: a datasheet with reference conditions stated, the factory acceptance test documentation, a written warranty boundary (what's covered, what voids it), and a named contact for spare parts three years out. If any of those is vague, the quote is a brochure.

7. What do buyers consistently overlook?

The drain. Every time.

Condensate drains are the least glamorous component on the skid and the one that causes the most callbacks. A timer drain opens on a schedule regardless of load — in a light-load week it's dumping compressed air, and in a heavy-load week it's not dumping enough. A zero-loss or level-actuated drain costs more up front and pays for itself in months. I've rejected deliveries over this. It feels petty until you see the utility bill.

Three more things that don't show up in most comparisons:

  • Ambient conditions in the dryer room. A refrigerated dryer in an unventilated corner will trip on high refrigerant pressure every July. Nobody budgets for the ventilation.
  • Filtration upstream. Especially ahead of desiccant. One slug of oil carryover can poison a bed, and the bed is the expensive part.
  • Verification after install. Put a calibrated portable dew point meter on the outlet at pressure and log it across a full load cycle. Not a spot check on day one. A full cycle.

I can only speak to what I've seen across industrial B2B supply — mid-size plants, predictable production schedules, mostly domestic sourcing. If you're running seasonal demand with six-week swings or buying through international logistics, the sizing math and the risk profile are different, and my rules of thumb will bend.

But the drain thing holds everywhere. Check the drain.