< 10 mg/Nm³ @ 6 % O₂ dry
Outlet capability, new media
Technology · Fabric filtration
Bag filter (baghouse)
A pulse-jet baghouse filters flue gas through fabric bags, collecting particulate as a surface cake that itself does most of the filtering. Correctly designed — air-to-cloth 0.8–1.2 m/min on biomass, media matched to temperature and gas chemistry — it holds outlet dust below 10 mg/Nm³ at 6 % O₂ dry on new media, at a pressure-drop cost of 100–200 mmWC across the bag life.
0.8–1.2 m/min net
Air-to-cloth ratio, biomass duty
4–6 bar
Pulse cleaning pressure
100–150 → 150–200 mmWC
Operating ΔP over bag life
01 — Principle
How a pulse-jet bag filter collects dust
The cake does the filtering; the fabric mostly holds the cake.
A baghouse hangs hundreds to thousands of fabric bags, each on a wire cage, inside a compartmented casing. Dust-laden gas flows from the outside of the bag inward; particulate deposits on the bag surface and builds a porous dust cake. Within minutes of start-up it is this cake — not the fabric — doing the fine filtration, which is why a conditioned baghouse collects sub-micron particles that pass through the clean cloth. Cleaned gas leaves through the top of each bag into a plenum; periodically, a solenoid-fired pulse of compressed air at 4–6 bar shoots down the bag, snapping the fabric outward and shedding the cake into the hopper.
Because collection is mechanical, the emission floor is largely indifferent to the dust's electrical properties. A pulse-jet unit on biomass fly ash holds the outlet below 10 mg/Nm³ at 6 % O₂ dry on new media, degrading gently as bags age — a floor an electrostatic precipitator can match only with generous collecting area, and cannot hold at all when ash resistivity moves against it. The price is pressure drop: 100–150 mmWC across new bags, rising to 150–200 mmWC late in bag life, paid continuously in induced-draught fan power.
Within Arrow Energy's gas-cleaning technology range the pulse-jet bag filter is the choice where the permit sits in single-digit mg/Nm³, where ash resistivity is hostile to an ESP, or where fuel flexibility matters more than pressure drop. On biomass it is almost always paired with a pre-collector — the Electrocyclone — for reasons of dust load and embers covered in Section 05.
02 — Sizing
Air-to-cloth ratio and can velocity: the two numbers that size the casing
Get these wrong and no media choice will save the installation.
What air-to-cloth ratio should a biomass baghouse be designed at?
Design at 0.8–1.2 m/min net for biomass fly ash: gas volume flow divided by installed cloth area with one compartment out of service. The gross ratio — all compartments filtering — will read lower, around 0.7–1.0 m/min. Vendors quoting a single undefined "ratio" are usually quoting gross, which flatters the design by 10–20 %.
Air-to-cloth ratio (filtration velocity) is the superficial gas velocity through the fabric. Push it high and the casing gets smaller and cheaper — and pressure drop rises roughly with the square of velocity, dust is driven into the fabric depth instead of resting on the surface, blinding accelerates, and pulse cleaning must fire more often, flexing the bags toward early failure. Biomass ash, fine and char-laden, punishes optimism here more than most dusts; 0.8–1.2 m/min net is the band in which cake filtration stays on the surface where it belongs. The net/gross distinction matters contractually: cleaning and maintenance take compartments offline, and the unit must meet its emission and ΔP figures in that state, not only with every compartment healthy.
Can velocity — the upward gas velocity in the open casing area between bags — is the second constraint. If it exceeds the settling velocity of the pulsed-off agglomerates, shed dust is carried straight back onto neighbouring bags instead of falling to the hopper, cleaning becomes a circulation loop, and ΔP creeps regardless of pulse energy. Keeping can velocity down means limiting bag length or widening bag pitch in the bottom-entry designs common on boiler duty; it is checked explicitly at design, not left to fall out of the bag count. Guaranteed figures for emission and pressure drop are stated per project after the technical assessment, on a stated basis — mg/Nm³, reference O₂, dry or wet, and load range.
03 — Media
Filter media by temperature and gas chemistry
Fibre selection is a chemistry decision first and a temperature decision second.
| Fibre | Continuous limit | Strengths | Caveats on boiler flue gas |
|---|---|---|---|
| Polyester (PES) | 135 °C | Lowest cost, good abrasion resistance | Hydrolyses in hot moist gas — marginal on 50 %-moisture biomass fuels; keep well below limit |
| Aramid (m-aramid) | 190 °C | Good temperature margin, strong fabric | Hydrolysis-sensitive above ~160 °C in moist gas; poor acid resistance below dewpoint |
| PPS (polyphenylene sulphide) | 190 °C | Excellent acid and hydrolysis resistance | Attacked by oxygen-rich gas above ~160 °C at high O₂; watch excess-air excursions |
| P84 (polyimide) | 240 °C | Fine-fibre surface, high capture efficiency | Hydrolysis-sensitive; often used as blend or surface layer rather than alone |
| PTFE | 250 °C | Near-universal chemical resistance | Highest cost; often applied as membrane lamination on other fibres |
| Fiberglass | 260 °C | Highest temperature, dimensionally stable | Poor flex/abrasion tolerance — pulse energy and can velocity must be conservative |
Two caveats govern every row. First, the acid dewpoint: biomass and process flue gases carry SO₃ and water vapour, and below the acid dewpoint condensing acid attacks aramid and polyester aggressively while PPS and PTFE shrug it off. Cold-end excursions — start-up, shutdown, low load — put every bag below dewpoint temporarily, so the media must tolerate what the operating discipline cannot prevent. Second, hydrolysis: with bagasse at roughly 50 % moisture as fired, flue-gas water content is high, and the moisture-plus-temperature combination degrades polyester and aramid well below their dry-gas ratings. The continuous limits in the table are ceilings, not operating points; a 20–30 °C margin between operating temperature and fibre limit is normal design practice.
04 — Operation
Cleaning, pressure drop and the fan bill
A baghouse is bought once and paid for continuously at the ID fan.
How does baghouse pressure drop change over bag life?
New, conditioned bags run at 100–150 mmWC total unit pressure drop. As residual dust embeds permanently in the fabric depth over years of service, the baseline drifts to 150–200 mmWC at end of bag life. The induced-draught fan must be sized for the end-of-life figure; sizing for the commissioning figure trades boiler capacity for capital saved.
Pulse-jet cleaning fires compressed air at 4–6 bar through a blowpipe above each bag row, typically on a ΔP-triggered cycle rather than a fixed timer: cleaning on demand minimises both compressed-air consumption and bag flexing. Over-cleaning is as damaging as under-cleaning — every pulse flexes the fabric against the cage and strips cake that was doing useful filtration, briefly opening the emission floor. A well-tuned unit lets ΔP oscillate in a narrow band around its setpoint and fires only the rows that need it.
The pressure-drop numbers translate directly to money. Each 10 mmWC of ΔP at a given gas flow is fan power in proportion — on a mid-size biomass boiler, of the order of several kW continuous per 10 mmWC — so the difference between a baghouse at 180 mmWC and an ESP train at a fraction of that is a permanent operating-cost line, not a rounding error. This is the structural trade against the electrostatic precipitator: the ESP buys its low ΔP with collecting area and high-voltage power; the baghouse buys its resistivity-proof emission floor with fan power and periodic media replacement. Section 07 tabulates the full comparison.
05 — Biomass risk
Sparks, embers and the case for a pre-collector
Fabric and glowing char do not share a casing peacefully.
Can a baghouse survive on a spark-prone biomass boiler?
Yes, with layered protection — never bare. Bagasse and wood boilers carry glowing char that burns holes in fabric within seconds. A pre-collector such as the Electrocyclone removes embers along with roughly 90 % of the dust load, spark arrestors and drop-out boxes intercept stragglers, and inlet ΔT monitoring isolates the unit on temperature-rise events.
High char carryover is intrinsic to bagasse and rice-husk firing, and a single ember landing on a polyester or aramid bag is a hole; a shower of them during a load swing can be a compartment. The defence is layered because no single measure is complete. First, remove the source: an Electrocyclone or multicyclone first stage captures and quenches embers in the pre-collector hopper while removing the bulk of the dust — on the two-stage design basis, cutting the load from 6,000 to 720 mg/Nm³ before the fabric sees it, which reduces cake mass per cycle by ~88 % as a side benefit. Second, intercept: spark-arrestor screens and low-velocity drop-out sections at the baghouse inlet. Third, detect and act: thermocouples across the inlet with rate-of-rise alarms, bypass or isolation dampers, and — per project assessment — hopper fire detection, since a smouldering hopper is the failure mode that takes out whole units.
DESIGN BASIS — ILLUSTRATIVE CALCULATION, NOT A GUARANTEE · all concentrations @ 6 % O₂ dry · 60 t/h bagasse-fired boiler · raw dust 6,000 mg/Nm³ · Electrocyclone first stage at 88 % → 720 mg/Nm³ at the final-collector inlet
06 — Failure modes
Why bags fail, and the design answer to each mode
Bag replacement is the dominant opex line; failure analysis is where it is won.
Abrasion at the inlet. The first bag rows facing the inlet duct are sandblasted by the incoming stream — on biomass, by silica-rich ash that is genuinely abrasive (rice-husk ash runs 85–90 % amorphous silica). Failed bags cluster near the inlet with wear on the impingement side. Remedies: inlet baffles and diffusers that kill jet velocity before it reaches fabric, generous inlet plenum design, and pre-collection of the coarse abrasive fraction upstream.
Chemical attack. Hydrolysis thins polyester and aramid wherever hot moist gas exceeds the fibre's real moist-gas limit, and acid condensation below the dewpoint attacks the same fibres during every cold start. Failures appear as general fabric weakening — bags that tear at the cage on normal pulsing. The remedy is honest media selection against actual gas chemistry and temperature excursions (Section 03), not against the nameplate operating point.
Blinding. Sticky, hygroscopic or condensing-salt ash lodges irreversibly in the fabric depth; ΔP climbs and no pulse pressure recovers it. Common on fuels with high alkali content or when operating near dewpoint. Remedies: surface-treated or membrane media that keep the cake on the surface, temperature-window control, and conservative air-to-cloth ratio so the cake — not the depth of the fabric — carries the filtration duty.
A disciplined bag-failure log — position in the casing, fibre condition, wear pattern — turns replacement from a recurring cost into a diagnostic. Clustered inlet failures point to gas distribution; uniform weakening points to chemistry; one compartment failing early points to its pulse system or poppet valve.
07 — Selection
Baghouse or ESP: the twelve-parameter decision
Neither is generically better; the fuel, the permit and the fan decide.
| Parameter | Pulse-jet baghouse | Dry ESP |
|---|---|---|
| Capital cost | Lower at small–mid gas flows | Lower at large flows; scales well with size |
| Pressure drop / fan power | 100–200 mmWC — the dominant opex | 10–30 mmWC across the casing |
| Sensitivity to ash resistivity | None — mechanical collection | High — resistivity sets migration velocity and back-corona risk |
| Temperature ceiling | 135–260 °C, set by media fibre | ≤ 200 °C dry-type; design point 150–180 °C on bagasse |
| Moisture / dewpoint tolerance | Poor — condensation blinds media and feeds acid attack | Moderate — corrosion risk, but no media to blind |
| Load turndown | Good — cake filtration holds at low flow | Good — efficiency typically improves at reduced flow (higher SCA) |
| Maintenance profile | Bag and cage replacement campaigns; valves, compressed-air system | Rappers, insulators, TR sets; no consumable media |
| Footprint | Compact for the emission achieved | Larger; collecting area scales with ln²(1/P) of the duty |
| Emission floor | < 10 mg/Nm³ on new media, resistivity-independent | 30–50 mg/Nm³ typical retrofit basis; ≤ 10 mg/Nm³ with adequate SCA |
| Media / internals replacement opex | Full bag set on a multi-year cycle | No scheduled media; electrode and plate life is long |
| Ember tolerance | Low — needs pre-collection, arrestors, ΔT monitoring | High — embers meet steel plates, not fabric |
| Wet or near-saturated gas | Unsuitable | Dry type unsuitable; wet ESP variant exists for saturated gas |
On a bagasse boiler chasing a limit in the tens of mg/Nm³, high moisture, embers and abrasive ash push toward the ESP; a single-digit permit, hostile resistivity or a compact plot push toward the baghouse. In either case the two-stage architecture holds: an Electrocyclone first stage shrinks the final collector, whichever technology it is, and removes the ember risk that most threatens the fabric option.
FAQ
Engineering questions, answered
What emission level can a baghouse achieve?
A correctly sized pulse-jet baghouse holds outlet particulate below 10 mg/Nm³ at 6 % O₂ dry on new media, and low tens of mg/Nm³ as media age. Unlike an ESP, the emission floor is nearly independent of dust electrical resistivity, because collection is mechanical filtration through the dust cake.
What air-to-cloth ratio should a biomass boiler baghouse use?
Design at 0.8–1.2 m/min net air-to-cloth ratio for biomass fly ash — net meaning with one compartment isolated for cleaning or maintenance. Ratios pushed above this band raise pressure drop, drive dust deeper into the fabric, accelerate blinding, and shorten bag life faster than the saved casing cost justifies.
Which filter bag material is best for biomass flue gas?
Match fibre to temperature and chemistry: polyester to 135 °C, aramid or PPS to 190 °C, P84 to 240 °C, PTFE to 250 °C, fiberglass to 260 °C. For biomass with moisture around 50 % in the fuel and acid dewpoint exposure at start-up, PPS or PTFE resist hydrolysis and acid attack best.
Can a baghouse handle sparks and embers from a biomass boiler?
Not unprotected — glowing char burns holes in most media. Mitigation is layered: a mechanical or Electrocyclone pre-collector removes embers with roughly 90 % of the dust load, spark arrestors intercept the remainder, and inlet-temperature-rise monitoring triggers isolation. With these measures baghouses run reliably on bagasse and wood-fired boilers.
What pressure drop should I expect across a pulse-jet bag filter?
Plan for 100–150 mmWC across new, conditioned bags, drifting to 150–200 mmWC toward end of bag life as residual dust embeds in the fabric. The induced-draught fan must be specified for the end-of-life figure, not the commissioning figure, or the boiler loses capacity as the bags age.
Why do filter bags fail early?
Three dominant modes: abrasion where inlet gas sandblasts the first rows with silica-rich ash, chemical attack — hydrolysis of polyester and aramid above their moisture-temperature limits and acid attack below the dewpoint — and blinding on sticky or hygroscopic ash. Each has a design remedy: inlet baffling, correct fibre selection, and temperature-window control.
Related engineering pages
Send us your plant data
Fuel, boiler capacity, gas flow, current emission and the limit you must meet. An Arrow engineer replies with a technical assessment basis — not a brochure.