Comparing Coalescers
When you compare coalescers side by side, a pattern emerges: the designs that look great in drawings rarely survive in dirty water.
The difference isn’t marketing, it’s physics, geometry, and maintenance reality.
At Mercer, we’ve tested, serviced, and replaced every type on the market. We’ve seen what lasts, what fails, and why.
This page breaks it all down so you can ask better questions and make better choices.
| Design Type | In Theory | In Reality | Cleaning Ability | Solids Control | Lifecycle Cost |
| Parallel Corrugated (Sinusoidal) | Efficient laminar flow and quick drainage | Troughs trap solids; fouls rapidly | Low | Weak | High |
| Plastic Honeycomb Blocks | Compact, high surface area | Traps fines; impossible to clean | None | Poor | Very High |
| Secondary Mesh Packs (Screens) | Final “polish” for fine droplets | Plugs first; drives bypass flow | Low | None | Very High |
| Vertical Tubes | Solids drop out, oils rise cleanly | Sludge bridges; high maintenance | Low | Weak | High |
| M-Pack | M-shaped plates form peaks and valleys to direct oil upward and solids downward | Small holes restrict flow; tight stacks foul quickly and limit separation efficiency | Low | Weak | High |
| Pizza Box | Flat, slotted panels offer high surface area in a compact, low-cost module for light-duty coalescence. | Performs acceptably in clean, low-solids streams; narrow channels still collect films and require periodic rinsing. | Moderate | Fair | Moderate |
| Coalescing Balls | Free-floating spheres encourage droplet coalescence and self-cleaning motion | Reality: they clump, abrade, and retain oil; difficult to clean or contain consistently | Fair | So-So | Moderate |
| Mercer Multi-Pack™ | Controlled flow, cleanable, consistent | Maintains efficiency long-term | High | Strong | Low |
Takeaway: Surface area doesn’t equal performance. Serviceability does.
In Theory:
Angled, wavy plates form alternating crests and troughs, creating narrow channels for oil droplets to coalesce and rise. These are commonly marketed as CPI separators, an acronym that morphed into meaning Corrugated Plate Interceptor, but was originally meant to imply a Coalescing Plate Interceptor. Over time, the industry quietly dropped the coalescing function from reality while keeping the label.
On paper, the design still promises laminar flow and efficient droplet capture along the corrugated surface, oil up, solids down, and minimal maintenance.
In Reality:
That “C” might as well stand for clogged. The troughs quickly turn into sediment traps. Solids settle, bridge, and blind half the available flow area. Velocity doubles through the remaining open paths, shearing droplets back into suspension and destroying separation efficiency. Within months, operators are battling fouling, not filtering oil.
Result: Performance collapse long before the first maintenance cycle. What started as a “coalescing” system devolves into a sludge-packed maze that’s nearly impossible to service without full teardown.
Pull Quote: “If you have to pull plates every month, it’s not a separator, it’s a subscription.”
Why It Fails:
The original CPI was intended for clean refinery wastewater with minimal solids — not today’s complex industrial effluents. As solids and variable flow enter the equation, the geometry backfires.
Ignoring solids management and the velocity limits set by API 421, this design assumes laminar flow in turbulent conditions.
The truth: the “C” in CPI hasn’t meant coalescing in decades — it now just means corrugated, and that single word swap tells you everything about why it fails in the field.
In Theory:
Lightweight plastic blocks with massive surface area promise better coalescence in a smaller footprint. Manufacturers market them as low-maintenance and inexpensive to replace.
In Reality:
Each tiny cell becomes a dirt magnet. Fines, waxes, and emulsified oil bind to every corner, reducing effective area within days.
Cleaning beyond the surface is impossible, pressure washing only polishes the first inch.
Result:
Chronic fouling and “planned obsolescence.” Replacements become the business model for the supplier, at the detriment of the customer’s OPEX.
If your separator has a parts catalog thicker than its spec sheet, it’s not saving you money.
Why It Fails:
Coalescence requires contact and release, honeycombs offer the first, never the second.
Once blinded, they behave like filters, not separators.
In Theory:
A fine mesh “polishing” layer behind the main pack captures remaining droplets to improve effluent clarity.
In Reality:
If the primary pack underperforms, this “secondary” mesh becomes the first to foul.
Backpressure increases, forcing flow to bypass or lift media from its supports.
Operators end up cleaning or replacing it more often than filters, at far higher cost.
Result:
Clogged mesh, erratic flow, and no measurable gain in oil removal.
Polishing” doesn’t help if your first stage never separated in the first place.
Why It Fails:
It misplaces the work. True separation happens in the first stage, where solids control and proper velocities are critical.
In Theory:
Vertical polypropylene tubes create upward flow channels where oils rise and solids fall out naturally.
In Reality:
Real wastewater contains variable solids loads and fluctuating flow rates. The tight vertical spacing quickly builds a sludge column at the bottom, while sticky hydrocarbons coat the interior.
Cleaning requires full removal, often after total failure.
Result:
Bridging, channeling, and high labor costs.
Why It Fails:
Limited cross-sectional area and zero solids evacuation.
Once the first few tubes blind, the rest follow.
In Theory:
The M-shaped pack design features peaks and valleys with small holes at each point. Upper holes let oil rise; lower holes let solids settle. Modules are stackable for flexible installation.
In Reality:
The small holes don’t provide enough space for effective rise or settling. Flow paths clog and fouling spreads rapidly. Performance drops as maintenance requirements rise.
Result: Compromised separation and frequent cleanouts that negate its “low-profile” advantage.
If your ‘innovation’ adds more holes than results, it’s not progress — it’s punishment.
In Theory:
“Pizza Box”-style media—flat, slotted polypropylene or PVC panels stacked in a grid—were adapted from air-scrubber and condensate mist-elimination designs, not wastewater separators. System modules were engineered for gas-to-liquid contact, where there are no solids, no sludge, and no emulsified oils. In those applications, the geometry works: thin-film spreading, high surface area, and uniform droplet capture under clean conditions.
In Reality:
Drop those same internals into an oil-water separator handling drilling mud, tank bottoms, or refinery washdown, and the results are disastrous. The narrow channels clog instantly with suspended solids. Instead of promoting droplet coalescence, they behave like filters—trapping fines, blinding off flow paths, and forcing bypass through every available gap. What was supposed to enhance separation now creates turbulence, shears droplets, and drives carry-over downstream.
Result:
Frequent shutdowns, premature replacement, and a steady side-business for anyone selling spare media kits. The system becomes a maintenance treadmill—fighting physics that were never meant for dirty water.
If your separator internals were born in an air scrubber, they’ll drown in dirty water.
Why It Fails:
In Theory:
Free-floating coalescing balls are designed to increase oil-droplet contact by gently rolling and shifting with the flow. In light-duty applications where solids are minimal, the spherical geometry promotes droplet collisions, helps break surface tension, and encourages coalescence without requiring complex structures or tight channel spacing.
Manufacturers position them as a low-cost, low-profile option for improving removal efficiency in compact separators.
In Reality:
When used in the right wastewater profile—light oils, low solids, and relatively clean influent—coalescing balls can provide a modest improvement over an empty chamber. They tolerate uneven flow better than honeycomb blocks or mesh packs, and they don’t create the deep, inaccessible fouling layers that cripple densely packed media.
However, they are not a magic solution. In higher-solids service, the balls can accumulate sludge films or tangle together in low-velocity zones. Periodic removal and rinsing is still required, though the process is generally simpler and more forgiving than cleaning stacked plates or blinding honeycomb cells.
Result:
Predictable performance within their intended range, better resilience than most low-cost media options, and easier maintenance for operators—especially in budget-sensitive or light-duty applications. In the economy segment, they represent one of the few configurations that provide value without locking facilities into constant replacement cycles.
Not perfect—but far better than anything else at the low end of the market.
Why They Work (and Where They Don’t):
Used in the right place, coalescing balls deliver a practical, economical lift in separation performance—making them the most reliable of the budget-friendly options.
In Theory:
Controlled-flow geometry designed from first principles of API 421, Stokes’ Law, and real-world solids behavior. Self-sustaining, cleanable packs that maintain an effective area throughout their service life.
In Reality:
Mercer Multi-Pack™ systems maintain predictable performance year after year. Solids are managed, not ignored and internal surfaces are accessible for full inspection and cleaning.
No disposable plastic blocks. No more clean what you can and hope for the best.
Result:
Stable discharge quality, fewer shutdowns, and verifiable compliance.
If it can’t be cleaned, it can’t be trusted.
Proof in the Field:
Clients routinely report decades of operation without replacement media. Routine inspections confirm consistent removal efficiency under varied flow conditions.
Coalescence efficiency isn’t a guess. It’s governed by velocity, viscosity, and droplet rise rate.
Mercer applies real design equations, not copy-and-paste assumptions.
We size and tune every unit around these variables because that’s what keeps separators working in the real world.
If your separator internals were born in an air scrubber, they’ll drown in dirty water.
Why It Fails:
If you only remember one section, make it this one.
Before signing off on any separator design, ask:
If those questions make a vendor uncomfortable, you’ve already uncovered the most important data point of all.
Good engineering stands up to scrutiny. Marketing collapses under it.
You don’t have to settle for systems designed to fail. Mercer builds separators that follow physics, respect operators, and outlast expectations.
Talk to an Engineer