Dust Extraction CFM by Machine: Duct Diameter, Static Pressure Loss and What Your Collector Actually Delivers
A table saw needs about 350 CFM at its port for chips and 550 for fine dust; a 15in planer needs 600 to 800. A 4in duct physically cannot carry more than about 350 CFM at the 4,000 FPM velocity wood dust requires, which is why upgrading the collector without upgrading the pipe changes almost nothing.
Every CFM figure on this page is extraction volume — air pulled out of a machine's dust port by a collector. It has nothing to do with compressed air supplied to a tool, which is a completely different measurement with different units of pressure and different sizing rules. If you arrived here looking for air supply, this is the wrong table.
The short answer: a table saw wants about 350 CFM at the port for chip collection and 550 for fine dust capture. A 12in benchtop planer wants 400 to 550. A 15in stationary planer wants 600 to 800. A drum sander wants 500 to 700.
The number that actually determines whether you hit those figures is duct diameter, not collector horsepower. Wood dust needs roughly 4,000 feet per minute of air velocity to stay suspended in a vertical branch, and a 4in pipe at 4,000 FPM carries 349 CFM. That is a physical ceiling. Bolt a 5 HP cyclone onto 4in flex hose and you get about 450 CFM at the machine, which is barely more than a 1.5 HP single-stage delivers through the same hose. The pipe, not the impeller, is the bottleneck in nearly every home shop.
How much each machine needs at its own port
Two columns, because they are two different jobs. Chip collection stops the shop filling with shavings. Fine dust capture catches the sub-10-micron fraction that stays airborne for hours and is the part that matters for your lungs. The second number is always higher, and most stock ports were designed around the first.
Machine Stock port CFM for chip collection CFM for fine dust capture Duct size that supports it Table saw, cabinet port only 4 in 350 550 5 in or 6 in Table saw with an overarm guard 4 in plus 2.5 in 450 650 combined 6 in main, 2.5 in to the guard 12-13 in benchtop planer 4 in 400 550 5 in minimum 15 in stationary planer 5-6 in 600 800 6 in 6-8 in jointer 4 in 350 450 5 in 14 in bandsaw, lower port 4 in 350 400 4-5 in, and add an upper guide port Router table, fence plus below 2.5 in plus 4 in 300 400 4 in below, 2.5 in at the fence Drum sander, 16-18 in Two 4 in 500 700 6 in split to two 4 in branches Lathe with a floor hood 4 in 350 500 5 in, with the hood within 6 in of the tool rest Mitre saw in a hood 4 in 450 600 plus 6 in. The worst capture geometry in any shop Drill press 2.5-4 in 200 300 Chips only; the fine fraction is negligible except in MDF Take the fine-dust column as your design figure for any machine you stand next to for more than a few minutes. Use the chip column only for machines you visit briefly. The lathe entry deserves a note, because it is the machine most often left out of a collection plan. A lathe throws shavings in an arc away from the tool, so ducting to a fixed port catches almost nothing. It is the hood position that does the work — within about 6in of the cut, angled to intercept the throw — and no amount of CFM compensates for a hood in the wrong place. Sanding at the lathe is the reverse: fine dust goes everywhere and a well-placed hood earns its keep.
Duct diameter sets a ceiling nothing else can lift
Airflow equals velocity times cross-sectional area. Wood dust needs about 4,000 FPM in vertical runs and branches to stay entrained, and about 3,500 FPM in horizontal mains. Fix the velocity and the pipe size decides the CFM. There is no negotiating with it.
Duct diameter Cross-section, sq ft CFM at 3,500 FPM (horizontal main) CFM at 4,000 FPM (branch or vertical) Friction loss per 10 ft of smooth pipe 4 in 0.087 305 349 About 1.0 in w.g. 5 in 0.136 477 545 About 0.7 in w.g. 6 in 0.196 687 785 About 0.5 in w.g. 7 in 0.267 936 1,069 About 0.4 in w.g. 8 in 0.349 1,222 1,396 About 0.35 in w.g. Compare the 4,000 FPM column against your machine's fine-dust requirement. If the pipe cannot carry the number, no collector will deliver it, and the fix is pipe rather than horsepower. Look at the 4in row against the machine table and the problem is immediate. A 4in duct tops out around 349 CFM. A table saw needs 550 for fine dust. A 15in planer needs 800. The stock 4in ports on home-shop machines were sized for chip clearance in an era when nobody measured the fine fraction, and enlarging them at the machine is often the highest-value change in the whole system.
The other half of that trade: going too large drops velocity below the entrainment threshold and dust settles in the pipe. An 8in main serving one 4in machine is not better, it is worse. Size the main for the largest single machine plus a little, not for the sum of everything.
Every fitting costs you feet
Pressure loss through fittings is conventionally expressed as equivalent feet of straight pipe, because it makes the arithmetic additive. Add up every real foot and every equivalent foot, multiply by the loss per foot, and you have the static pressure the collector has to overcome.
Fitting or run Equivalent feet of straight 4 in Equivalent feet of straight 6 in Note 10 ft of smooth metal duct 10 10 The baseline everything else is measured against 10 ft of flexible hose 25-30 25-30 Count flex at roughly three times smooth pipe. This is the single biggest avoidable loss in most shops 90 degree long-radius elbow 6 9 Centreline radius of 1.5 times diameter or better 90 degree short-radius or segmented elbow 9 14 Standard on cheap fittings; never put one at the collector inlet 45 degree elbow 3 4.5 Two 45s beat one 90 every time you can fit them 45 degree wye branch 3-5 5-7 The correct way to join a branch to a main 90 degree tee branch 20 plus 30 plus Do not fit one. Ever Blast gate 2-3 2-3 Self-cleaning gates hold their figure; sliding plate gates get worse as dust packs the slot Machine hood or cabinet entry Adds 1-2 in w.g. outright Same Not a length. A fixed pressure penalty on top of everything else Total your real feet plus equivalent feet, multiply by the per-foot loss from the diameter table, then add the hood penalty. That total is what your collector's fan curve has to beat. Work an example. A table saw 18 ft from the collector, joined by 12 ft of 6in smooth pipe, one long-radius 90, and 6 ft of 6in flex at the machine. Real and equivalent feet: 12 plus 9 for the elbow plus 18 for the flex, so 39 equivalent feet of 6in. At roughly 0.05 in w.g. per foot that is about 2 in of duct loss, plus 1.5 in for the saw cabinet entry, plus filter resistance of 2 to 4 in on a loaded cartridge. Call it 6 in w.g. total.
That number is the one to look up on the collector's fan curve — not the free-air figure on the box. Most single-stage collectors have collapsed well below their rated flow by 6 in of static pressure.
Rated against delivered
Badge CFM is measured at a bare inlet with no ducting, frequently no filter, and always no machine. It is not a lie, but it describes a configuration nobody has ever used.
Collector Badge figure Measured at a bare inlet Through 20 ft of 4 in flex and two 90s Through 25 ft of 6 in smooth and two long-radius 90s 1 HP single-stage, 30-micron bag 650-800 CFM 500-600 300-350 Bag resistance chokes it before the duct does 1.5 HP single-stage, 1-micron pleated filter 1,100-1,250 CFM 700-800 350-450 550-650 2 HP single-stage with a lid separator 1,200 CFM 650-750 300-400 500-600 3 HP cyclone, 0.5-micron filter 1,500-1,800 CFM 1,100-1,300 400-450 800-1,000 5 HP cyclone 1,800-2,200 CFM 1,400-1,600 450-500 1,000-1,300 Shop vacuum, 2.5 in hose 150-190 CFM 90-120 Not applicable 55-80 CFM through 12 ft of 2.5 in hose Read the fourth column down the page. The whole point of this table is that it barely moves — five times the horsepower buys you about 40 percent more air through a 4in hose. That fourth column is the most expensive lesson in home-shop dust collection. A shop that upgrades from a 1.5 HP single-stage to a 3 HP cyclone and keeps the 4in flex has spent a lot of money to gain perhaps 50 CFM at the machine. The same shop, keeping the 1.5 HP unit and running 6in smooth pipe instead, gains around 200. Pipe first, then horsepower.
The shop vacuum row is there because it answers a common question. A vac is a high-pressure, low-volume machine — brilliant at a sander pad or a router base where the port is small and the pressure drop is enormous, useless at a table saw cabinet where you need volume. They are not smaller versions of the same tool.
The 4in port is the problem, not the collector
Machine manufacturers fit 4in ports because 4in is what fits in the casting and what the market expects. It is not a considered airflow decision.
Enlarging a port is usually straightforward on a table saw cabinet or a bandsaw, less so on a planer with an integral chip hood. A 5in or 6in port on a table saw cabinet, combined with a matching duct, moves the machine from about 350 CFM to something near 550 without touching the collector at all.
Second port, second gain. Below-blade collection catches maybe half of what a table saw produces; the rest comes off the top of the blade and goes straight past you at head height. An overarm guard with a 2.5in port, teed into the same run, is the difference between a saw that looks clean and a saw that is clean.
Where you genuinely cannot improve the port — a benchtop planer, a portable machine — accept it and put the effort into a short, straight, smooth-walled run instead. Three feet of flex at the machine is a reasonable compromise. Fifteen feet of it is throwing away more than half your air.
Filters and the pressure you cannot see
Filter resistance is invisible, continuous and gets worse every hour the machine runs.
A new 1-micron pleated cartridge adds around 1 to 2 in w.g. A loaded one, caked on the inside, adds 4 to 6. Since that sits on top of every other loss in the system, a neglected filter can halve delivered airflow with no change to anything you can see or hear. The impeller note stays the same; the shop just gets dustier.
Bag ratings are worth understanding too. A standard 30-micron felt bag passes essentially all of the fine dust straight back into the room, which means a collector with one is a chip mover, not a dust collector. It moves shavings from the machine into a bag and returns the dangerous fraction to the air at head height. Upgrading to a 1-micron or better cartridge is the cheapest genuine health improvement available in a home shop.
Clean the filter on a schedule rather than when it looks bad. Pleated cartridges with an internal paddle want a spin every few hours of runtime; more often in MDF or sanding work, which loads a filter faster than anything else in the shop.
Trueness Index: measuring your own system
The Trueness Index question for extraction is simple — how much of the published figure survives to the machine port? Answering it takes an anemometer and about twenty minutes.
Measure at the port with the gate open and the machine off. A vane anemometer held in the duct gives velocity in FPM; multiply by the duct's cross-sectional area in square feet to get CFM. A 6in duct reading 3,600 FPM is delivering 0.196 times 3,600, or about 706 CFM. Compare that against the fine-dust column in the first table.
Do it for each machine, one gate at a time, and write the numbers on the gate in marker. Nearly every shop finds one branch far worse than the rest, and it is almost always the one with the most flex hose or the tee somebody fitted because they had one spare.
Repeat it after cleaning the filter. The gap between those two readings is the honest cost of a dirty cartridge in your specific system, and it is usually larger than anyone expects.