Steel Silo Dust Collection Ducting: Design Rules That Actually Work

Dust collection ducting is the network of pipes, bends, and transitions that moves captured dust from silo inlet points to a filter unit. Get it wrong and you're looking at duct blockages every 3 weeks, filter bag replacement costs that double your maintenance budget, and dust emissions that'll cost you €15,000–50,000 in fines depending on your jurisdiction. Get it right and the system runs for years with nothing more than quarterly bolt checks.

Why Ducting Design Gets Silos Wrong

The "Copy-Paste From Another Plant" Problem

I've walked onto too many sites where the ducting layout was clearly lifted from a different project. Same diameters, same routing, same pickup points — just stamped onto a new drawing set. It never works. A cement plant in central Vietnam had 200 mm ducts copied from a grain handling facility in northern China. The grain dust was light and fluffy. The cement dust was heavy and abrasive. They had their first blockage within 48 hours of commissioning.

The material you're handling dictates everything. Bulk density, particle size distribution, moisture content, abrasiveness — each one changes the velocity requirements, the bend radius, even the wall thickness you need. You can't just resize the duct diameter and call it a day.

Velocity: The Number Nobody Agrees On

Here's where the arguments start. Some designers swear by 20 m/s minimum transport velocity. Others push for 25 m/s. I've seen specs that demand 30 m/s for coal dust. The truth is more nuanced than any single number.

For cement and clinker dust, 18–22 m/s keeps particles suspended without grinding the duct walls to nothing. For fly ash, which is lighter and finer, you can drop to 15–18 m/s. Coal dust needs 22–25 m/s because of the higher bulk density — typically 800–1,000 kg/m³ versus 900–1,200 kg/m³ for cement. Go too slow and material settles in horizontal runs. Go too fast and you're replacing elbows every 18 months.

The EN 14491 standard gives guidance on dust explosion prevention in ductwork, but it doesn't prescribe transport velocities. That's left to the engineer. And that's where experience matters more than a textbook.

Pressure Drop: The Hidden Cost

A typical dust collection system for a 5,000-tonne cement silo runs at 1,500–2,500 Pa total pressure drop. The ducting itself accounts for 40–60% of that. Every unnecessary bend, every abrupt transition, every undersized section adds 50–150 Pa that your fan has to overcome.

Over a 20-year silo life, that extra 200 Pa of pressure drop translates to roughly €8,000–12,000 in additional electricity costs at European industrial rates. I ran the numbers for a plant in Poland last year. They had 14 bends in the main duct run where 6 would have done. The fan was working 15% harder than it needed to. Nobody had checked.

Sizing Ducts: Velocity, Pressure, and the Numbers That Matter

Minimum Transport Velocity by Material

This is the table I keep on my desk. Not because I've forgotten it — because contractors keep trying to talk me into lower numbers.

Notice the range. That's deliberate. A 100 µm particle behaves differently from a 10 µm particle, even in the same material. If your process produces a wide particle size distribution, size for the coarsest fraction. The fines will follow.

Duct Diameter Selection

The formula is simple: Q = v × A, where Q is volumetric flow rate, v is velocity, and A is cross-sectional area. The execution is where things go sideways.

A typical cement silo filling operation generates 2,000–4,000 m³/h of dusty air. At 20 m/s transport velocity, that gives you a duct diameter of 185–265 mm. Round up to the next standard size — 200 mm or 250 mm — and you're fine. But if you're handling pneumatic conveying air from a 50-tonne truck unload, the flow rate jumps to 6,000–10,000 m³/h. Now you need 300–400 mm ducting. Size it wrong and the truck unload takes twice as long because the air can't escape fast enough.

I remember a project in Malaysia where the truck unloading duct was sized for 3,000 m³/h. The actual peak flow was 7,500 m³/h during the initial discharge surge. The silo pressure relief valve lifted every single time. Took us two weeks to figure out it wasn't the relief valve that was broken — it was the ducting that was choking the system.

Bend Radius and Fitting Losses

A 90° elbow with a 1.5D radius (where D is the duct diameter) adds roughly 25–35 Pa of pressure drop at 20 m/s. A 2.5D elbow? 12–18 Pa. The difference seems small until you multiply it by 8 elbows across the system. That's 100–140 Pa saved just by specifying a gentler bend.

The rule I follow: minimum 2D radius for abrasive materials, 1.5D for non-abrasive. For cement and clinker, I push for 2.5D or even 3D where space allows. The extra steel costs maybe €200–400 per elbow. The energy savings over 10 years pay for it twice over.

Conical reducers instead of abrupt transitions. Y-branches instead of tee junctions. These aren't exotic — they're just details that most fabricators skip because they're not explicitly called out on the drawing. Specify them. In writing. With dimensions.

Material Selection and Wear Considerations

Wall Thickness: Where Thinner Costs More

Standard ducting for dust collection is typically 2–3 mm mild steel. For cement and fly ash, that's adequate. For clinker, slag, and coal, I specify 3–4 mm minimum, with 6 mm wear plates on the outer radius of bends.

Here's the thing about abrasive wear: it's not uniform. The outer wall of a bend takes the hit because particles don't turn as fast as the air. Centrifugal force throws them to the outside. I've seen 3 mm ducting wear through to nothing at a single elbow in 14 months. The rest of the duct was fine. Just that one bend.

Wear-resistant options: ceramic-lined elbows (€800–1,500 each for 300 mm diameter), rubber-lined sections, or simply thicker steel with a sacrificial liner. For most applications, a 6 mm wear plate welded to the outer bend radius is the sweet spot. It adds €150–300 to the fitting cost and extends service life from 14 months to 5–7 years.

Corrosion vs. Abrasion: Pick Your Enemy

In coastal plants — Vietnam, Indonesia, coastal China — you're fighting both abrasion and corrosion. Salt air accelerates external corrosion, especially on uninsulated ducting where condensation forms overnight. I've seen 4 mm ducting perforated by external corrosion in 3 years at a plant 500 m from the South China Sea.

Solution: hot-dip galvanized ducting for external surfaces, with epoxy coating (250–300 µm DFT) on internal surfaces where condensation is expected. The galvanizing adds 15–20% to the material cost. The alternative is a full duct replacement every 5–7 years. The math isn't close.

For gypsum and some chemical powders, you might need 316L stainless steel ducting. That's 4–5× the cost of mild steel. But when the material is hygroscopic and slightly corrosive, mild steel ducting turns into a rust-coloured sculpture in 2 years. I've been there. It's not pretty.

Flange and Connection Details

Flanged connections every 3–4 metres of straight duct. Always. Threaded connections leak. Welded connections make maintenance a nightmare. Slip-fit connections with clamps are fine for temporary installations but they'll leak within 6 months on a permanent system.

I specify PN6 flanges with EPDM gaskets for most dust collection applications. The gaskets compress to 2–3 mm and seal against the low positive pressures involved — typically 500–2,000 Pa. For explosion isolation per EN 14491, you need the flanges to be gas-tight enough to prevent flame propagation. A 1 mm gap is enough to defeat an explosion isolation valve. Torque the bolts. Check them quarterly.

Layout, Routing, and the Stuff Drawings Miss

Horizontal Runs: The Settlement Problem

Horizontal ducting is a necessary evil. You can't always run ducts vertically from a silo top to a filter unit. But every horizontal run is a potential blockage point. The rule: horizontal runs should slope at least 3° toward the silo or toward a dust return point. That's roughly 50 mm per metre of run.

I've seen horizontal ducts installed dead level because the installer didn't read the slope callout on the drawing. After 6 months, 40 mm of cement dust had built up on the bottom of a 250 mm duct. The effective cross-section dropped by 50%. Transport velocity in the remaining space doubled. Wear accelerated. The elbow at the end of that run failed in 11 months instead of the expected 5 years.

Aeration poke holes every 2–3 metres on horizontal runs help. A 15 mm air line with a check valve, blowing into the duct at 2–3 bar, keeps material fluidized. It's a €50 fitting that prevents a €5,000 shutdown. Not optional in my book.

Routing Around Structural Obstacles

On a real plant, ducting doesn't run in straight lines. It weaves around columns, under cable trays, over pipe racks. Every obstacle adds bends, transitions, and length. The temptation is to just add more fan power to overcome the extra resistance. Resist that temptation.

Instead: reroute. Yes, it means more ducting. But 20 metres of straight duct with two bends beats 12 metres with eight bends every time. I've had arguments with piping contractors who wanted to "just add another elbow" to avoid rerouting around a column. One extra bend costs €300. The extra pressure drop costs €50/year in electricity. Over 20 years, that's €1,000 for a €300 fitting. It doesn't add up.

Where you can't avoid obstacles, use long-radius bends (3D minimum) and conical transitions. A 45° transition instead of a 90° elbow saves roughly 40% of the pressure drop. It's not free — the fitting is longer and needs more support — but it's cheaper than oversizing the fan.

Support and Thermal Expansion

Steel ducting expands roughly 1.2 mm per metre per 100°C temperature change. For dust collection ducts operating at 60–80°C, that's less of an issue. But if you're handling clinker dust at 150–200°C, a 20-metre run expands by 25–35 mm. Without expansion joints, that stress goes into the supports, the flanges, and the welds.

I specify expansion joints every 15–20 metres on runs exceeding 100°C. Bellows-type, stainless steel, rated for the system pressure. They cost €400–800 each. A failed flange from thermal stress costs €2,000–5,000 in downtime and repair. Again, the math isn't close.

Support spacing: 3–4 metres for ducting up to 300 mm diameter, 4–6 metres for 300–600 mm. Use saddle supports, not point loads. A point load on thin-wall ducting creates a dent that becomes a turbulence point, which becomes a wear point, which becomes a hole.

Explosion Protection and Regulatory Compliance

EN 14491 and Ducting Requirements

Dust explosions in silos are rare but catastrophic. When they happen, the ducting is often the propagation path — flame and pressure traveling from the silo through the ductwork to the filter unit, or vice versa. EN 14491 specifies requirements for explosion isolation in ductwork, and it's not optional if you're handling combustible dust in the EU.

The standard requires explosion isolation devices (chemical or mechanical) on any duct connecting a protected vessel to other equipment. For a cement silo, that means isolation between the silo and the filter unit, and between the silo and any venting device. The isolation device must close within milliseconds of detecting a pressure rise — typically triggered by a pressure sensor set at 0.1–0.2 bar overpressure.

Cost: a mechanical isolation valve for a 300 mm duct runs €2,500–4,000. A chemical suppression system is €1,500–3,000 per installation. Against the cost of a silo explosion — €500,000 to €5,000,000 in damage, plus potential fatalities — it's not a difficult decision. Yet I've seen plants skip it because "cement dust isn't explosive." Tell that to the operators at the plant in Brazil where a cement silo exploded in 2019. The investigation found the ignition source was a hot bearing on a conveyor. The propagation path was the ducting.

Venting and Suppression in Ductwork

Explosion venting panels on ducting are sized per EN 14491 calculations based on duct volume, dust Kst value, and predicted reduced pressure. For cement dust (Kst typically 120–180 bar·m/s, St1 class), a vent area of 0.05–0.1 m² per cubic metre of duct volume is typical.

The vent panel must discharge to a safe area. Not into a walkway. Not under a cable tray. Not where a operator walks twice a shift. I've had to relocate vent discharges three times on a single project because the initial location was "convenient" for the ducting layout but lethal for anyone standing nearby when it fires.

Chemical suppression is an alternative where venting to a safe area isn't practical. A 250 mm duct with a suppressant bottle costs €2,000–3,500 installed. It discharges within 50 milliseconds of detection, quenching the flame before it propagates. The bottle needs replacement after each activation — roughly €300–500 per shot. But it beats replacing a silo.

Grounding and Static Protection

All metallic ducting must be bonded and grounded. Static electricity from pneumatic conveying can reach 10–50 mJ — enough to ignite most dust clouds. A grounding strap every 6–8 metres of ducting, with a resistance to earth of less than 10 ohms, is the minimum.

Non-metallic ducting (rubber-lined sections, flexible connections) needs conductive bonding. I've seen rubber flexible connections with no bonding at all. The static builds up on the rubber surface, discharges to the nearest metal flange, and — if the dust concentration is right — ignites. It's a small detail. It matters.

Case Study: 8,000-Tonne Clinker Silo — Ducting Redesign

A cement plant in central China had an 8,000-tonne clinker silo with chronic dust collection problems. The original ducting was 350 mm diameter, 2 mm wall thickness, with 1.5D elbows throughout. Blockages every 2–3 weeks. Filter bags failing every 4 months instead of the expected 18–24 months. Dust emissions visible from 500 metres on humid days.

Our assessment found three root causes. First, the transport velocity was 16 m/s — too low for clinker dust with a bulk density of 1,400 kg/m³. Material was settling in three horizontal runs totalling 28 metres. Second, the 1.5D elbows were wearing through every 10–12 months. Third, there was no explosion isolation between the silo and the filter unit.

Redesign: upsized to 400 mm diameter with 3.5 mm wall thickness and 2.5D elbows. Added 6 mm wear plates to all elbow outer radii. Installed aeration poke holes on all horizontal runs at 2-metre spacing. Added a mechanical explosion isolation valve per EN 14491. Total ducting cost increase: €18,000 over the original design. Annual maintenance savings: €35,000 (fewer bag replacements, no blockage cleanouts). Payback period: 6 months.

The plant manager told me they'd been living with the problems for 3 years before calling us. Three years of unnecessary cost and visible emissions. That's the thing about ducting — when it works, nobody thinks about it. When it doesn't, it's the first thing everyone complains about.

Frequently Asked Questions

What duct velocity should I use for a mixed-material silo?

Size for the heaviest, coarsest material you'll handle. If your silo stores both cement (1,200 kg/m³) and fly ash (800 kg/m³), design for cement at 20 m/s. The fly ash will travel fine at that velocity — it's lighter and won't settle. The reverse isn't true: if you size for fly ash at 16 m/s, the cement will settle in horizontal runs. I've seen plants try to split the difference at 18 m/s. It works for cement but accelerates wear on the ducting because the velocity is higher than necessary for the fly ash that fills the silo 70% of the time. Pick one material and commit.

How often should ducting be inspected?

Quarterly visual inspections of all accessible sections. Annual thickness measurements on elbows and transitions using ultrasonic testing. For abrasive materials like clinker and slag, check the outer radius of every elbow annually. I've found 6 mm wear plates down to 2 mm in 18 months at a coal handling plant. The duct was still structurally sound, but another 6 months and it would have been a hole. Budget €500–1,500 per year for inspection on a typical system. It's cheap insurance against a €20,000 emergency repair.

Can I use PVC or fiberglass ducting for dust collection?

For non-combustible, non-abrasive materials at low temperatures — maybe. For anything else, no. PVC melts at 80°C and burns, producing toxic fumes. Fiberglass can be used for corrosive applications but has zero abrasion resistance and is a static electricity nightmare. I've seen FRP ducting used for gypsum handling. It lasted 3 years before the internal surface was roughened enough to cause material hang-up. Stick with steel for industrial bulk solids. It's predictable, repairable, and you can weld a patch on it at 2am when something goes wrong. Try welding PVC at 2am.