Two numbers, and they fight each other
Dust collection has to satisfy two requirements at once, and they pull in opposite directions.
The first is volume. The machine throws chips and dust into the air at the cutter, and you need enough cubic feet per minute passing the hood to catch them before they escape into the room. A table saw wants somewhere around 400 CFM at the under-table port. A planer throwing shavings at speed wants more, perhaps 500. A wide belt sander producing very fine dust in quantity wants 600 or more. Too little flow and the dust simply walks out around the hood.
The second is velocity. Once the dust is in the pipe, the air has to keep moving fast enough to carry it. Wood chips are heavy compared with air, and below a certain speed they stop being transported and start settling out along the bottom of the duct. The figures commonly quoted for wood dust are 3,500 to 4,000 feet per minute in branch lines, and around 3,000 to 4,000 in the main. These are practice figures from industrial ventilation work, not code requirements, and the number that applies depends on what you are moving — heavy, damp or coarse material needs more, very fine dry dust needs less.
The conflict is that volume wants big pipe and velocity wants small pipe. Air conditioning duct is sized as large as will fit because bigger is quieter and cheaper to move air through. Dust duct is sized as small as will carry the flow, because oversized dust duct fills up. A 6 inch branch needs about 690 to 785 CFM to stay in the transport range, so putting 6 inch pipe on a machine that only needs 400 does not give you margin — it gives you a pipe that slowly silts up.
What each pipe size can and must carry
| Duct | Area (sq ft) | CFM at 3,500 FPM | CFM at 4,000 FPM |
|---|---|---|---|
| 4" | 0.0873 | 305 | 349 |
| 5" | 0.1364 | 477 | 545 |
| 6" | 0.1963 | 687 | 785 |
| 7" | 0.2673 | 935 | 1,069 |
| 8" | 0.3491 | 1,222 | 1,396 |
Read that table as a requirement rather than a capacity. The right-hand columns are not "up to" figures, they are the flow the pipe needs to have in it. This is the single most useful thing on the page, because it explains why the four inch port moulded into the side of a machine is a compromise the manufacturer made for the shop vac market rather than a recommendation. Four inch pipe tops out around 350 CFM at transport velocity, which is below what a table saw actually wants and a long way below a planer.
Why the collector never delivers its number
A dust collector is a fan, and a fan's output depends on what it is pushing against. The CFM on the label is normally measured at zero static pressure, meaning the impeller spinning with nothing attached. Every foot of duct, every elbow, every foot of flex hose, the machine's own hood and internal passages, and the filter all consume pressure, and as the pressure rises the flow falls.
The way to account for that without guessing is equivalent length: express each fitting as the length of straight pipe that would cost the same pressure, add it to the actual straight footage, and apply a friction rate. The multipliers used here are twelve duct diameters for a long-radius 90 degree elbow, six for a 45, and twelve for a wye branch. In 6 inch pipe that makes a 90 worth about six feet. Flex hose gets a straight multiplier — each foot counted as two to three feet of pipe, and worse than that if it is compressed, sagging or bent tight. The corrugations are the reason, and a short length of flex at the machine costs more than people expect.
Add all of it up, multiply by the friction rate for the flow and the pipe size, add an allowance for the machine hood and the filter, and you have an estimate of the static pressure your system puts on the collector. Then the honest step: take that pressure to the fan curve and read off what the collector actually delivers there. Manufacturers who publish curves are telling you something; ones who publish only a headline CFM are telling you something too.
The things that fix a disappointing system
When a collector underperforms, the temptation is to buy a bigger one, and it is usually the least effective move available. In rough order of what actually returns airflow: replace flex hose with rigid pipe everywhere except the last couple of feet at the machine; clean or replace the filter, since a loaded cartridge can consume more pressure than the entire duct run; go up a size on the main so the long haul is cheap and keep the branches at transport velocity; replace tight elbows with long-radius ones and delete the ones the layout does not need; make sure only one blast gate is open at a time so the flow is not divided; and fix the hood at the machine, because a hood that does not enclose where the dust is actually thrown will lose it no matter how much air is moving past. Only after all of that is the impeller the limiting part. A shop that does those six things with a mid-sized collector will outperform a shop that bolts a large one onto the same bad ducting, and it will cost a great deal less.
Questions people ask
How many CFM does a 4 inch dust collection port need?
To stay in the 3,500 to 4,000 FPM range commonly used for wood dust, a 4 inch duct needs roughly 305 to 350 CFM in it. That is a requirement rather than a limit — below it, chips settle in the pipe instead of being carried to the collector. It is also less than several common machines actually want at the hood, which is the fundamental problem with 4 inch ports: a planer or a wide sander needs more air than 4 inch pipe can move at a sensible pressure. This is why upgrading a machine to a 5 or 6 inch port, where the machine will accept one, usually does more than upgrading the collector.
Is a shop vac enough for dust collection?
For a sander, a router or a track saw, often yes — those are small hoods with high resistance and a shop vac is built for exactly that: low flow, high pressure, typically 50 to 150 CFM at a large static pressure. For a table saw, planer, jointer or bandsaw, no. Those need several hundred CFM at low pressure and a shop vac cannot supply it through a 2.5 inch hose. The two machines are not different sizes of the same thing; they sit at opposite ends of the flow-versus-pressure trade, which is why a cyclone separator on a shop vac still will not run a planer. Note also that a plain shop vac without a fine filter or a HEPA cartridge returns the finest dust straight back into the room air.
Why does my dust collector lose suction at the far machine?
Because pressure loss accumulates along the path, and the far machine has the longest one. Every foot of pipe, every elbow, every foot of flex and the machine hood itself takes a share, so the flow at the end of a long run with several fittings can be a fraction of what the same collector delivers three feet from the inlet. Flex hose is usually the biggest single culprit, followed by a clogged filter, then tight elbows, then undersized main duct. Two open blast gates will also split the flow, and a leaking gate somewhere else in the shop leaks flow you never see. Work out the equivalent length of the far run against the near one and the difference is usually obvious.
Should I use bigger pipe to get better airflow?
For the main run from the collector, often yes — a larger main carries the flow with less friction and leaves more pressure available for the branches. For the branch at the machine, no. Branch size has to be matched to the flow so the velocity stays in the transport range, and a branch that is too large for the air passing through it accumulates chips along the bottom until something packs solid. The rule that keeps you out of trouble is to size branches for velocity and mains for the total flow of whatever can be open at once, which is normally one gate in a home shop. Bigger is not universally better here, which is the main thing that separates dust ducting from air conditioning ducting.