How to Read a Head-Flow CFM Number (and What It Actually Means for Your Build)

Every cylinder head ad leads with one number: peak CFM. "Flows 320 CFM!" "400+ on the intake!" It's printed in bold, it's bigger than the price, and it's the first thing most buyers compare. It's also the single most misread spec in the entire engine-parts catalog.

A flow number isn't a horsepower rating, it isn't measured the way most people assume, and the peak figure is usually the least useful part of the whole flow sheet. This is a walk through what a CFM number actually is, why two heads with the same peak can behave nothing alike, and how to turn a flow sheet into a real decision for your specific build.

What "CFM" is actually measuring

CFM stands for cubic feet per minute — the volume of air a port will pass in one minute under a controlled pressure difference. A flow bench pulls (or pushes) air through the port with the valve held open a fixed amount, and a meter reads the volume moving through.

Three things make that single number mean something:

• The valve is held at a specific lift — say .500" open — not "wide open." Flow is measured in steps across the whole lift range.
• A fixed pressure drop is applied across the port. This is the "depression" the bench pulls, and it's the part that trips everyone up (next section).
• It's airflow, not power. A port that flows more air can make more power, but only if the rest of the engine — cam, displacement, RPM, intake, exhaust — can use it.

Think of CFM as the size of the straw, not how hard the engine drinks. A bigger straw only helps if you're actually trying to drink that fast.

Test pressure is everything: 28" vs 25" vs 10" of water

This is the number-one reason flow comparisons go wrong. CFM is only meaningful alongside the pressure it was measured at, and that pressure is quoted in inches of water (the depression the bench pulls).

The common standards:

• 28" of water — the modern industry standard for performance cylinder heads. Almost every aftermarket head sheet you'll see today is at 28".
• 25" of water — an older standard, still used by some manufacturers and in a lot of legacy data.
• 10" of water — common in some OEM and older domestic testing.

Here's the trap: the same port flows a bigger number at a higher test pressure. You cannot compare a head rated at 28" to one rated at 10" by reading the bold number — the 28" head will look far better even if the ports are identical.

Flow scales with the square root of the pressure ratio. To convert a flow figure from one test pressure to another:

CFM_new = CFM_known × √(Pressure_new / Pressure_known)

Worked example — converting 10" data up to 28":

A head measured at 235 CFM at 10" of water:

CFM_28 = 235 × √(28 / 10)
       = 235 × √2.8
       = 235 × 1.673
       = ~393 CFM at 28"

That's why an old OEM-style number can look small next to a modern aftermarket sheet — it's quoted at a lower depression. Always convert both heads to the same test pressure before you compare them. If a seller won't tell you the test pressure, the number is unusable.

The flow curve matters more than the peak

A flow sheet isn't one number — it's a table of CFM at rising valve lift: .100", .200", .300", .400", .500", .600", and so on. The shape of that curve tells you far more than the peak.

Two heads can both peak at 320 CFM and behave completely differently:

Illustrative figures for comparison — not from a specific casting.

Head A makes its flow early and holds it. Head B only catches up at the very top. If your cam only lifts the valve .550", Head A is the better head for your engine even though they tie at peak — because your engine spends almost all its time in the low- and mid-lift region, and only crosses peak lift for a fraction of a degree at the top of the cam's travel.

The rule: match the flow curve to where your cam actually operates. A street build with a .520"-lift hydraulic roller lives in the .200"–.450" band. Big peak-lift numbers from a race head are flow you literally never reach. Read the CFM at your cam's lift, not the headline.

Intake vs exhaust: the I/E ratio

A flow sheet has two columns — intake and exhaust. The relationship between them, the exhaust-to-intake ratio (E/I), tells you whether the head is balanced.

• A healthy street head usually has an exhaust that flows roughly 70–80% of the intake.
• Below ~65% and the exhaust side may bottleneck at high RPM, which can call for more exhaust duration in the cam to compensate.
• A very high ratio isn't automatically better either — it can mean the intake is under-developed.

You don't always need to chase a perfect ratio, but it tells you something the peak number hides: a head with monster intake flow and a weak exhaust will want cam timing that favors the exhaust, and it may not make the power the intake number implies.

Flow without velocity is a trap

Bigger ports flow more air — at peak. But port size and port velocity trade against each other, and velocity is what fills the cylinder at the RPM you actually drive.

• A large runner flows huge peak numbers but slows the air down. On a small-displacement, lower-RPM engine, that slow-moving air doesn't pack the cylinder well — you lose low-end torque and throttle response.
• A smaller runner flows a lower peak but keeps air moving fast, which improves cylinder filling and torque everywhere you actually drive a street engine.

This is why a 600-hp race head can make less power than a "smaller" head on a mild 350 street motor — the big runner kills velocity at street RPM. Match runner volume to displacement and RPM, not to the biggest peak CFM you can buy. Runner cross-sectional area and the engine's airspeed target matter more than the headline flow.

Turning CFM into a horsepower estimate

You can get a rough power ceiling from intake flow. The common rule of thumb:

Estimated peak HP ≈ Intake CFM (at 28") × ~1.9 to 2.0 × number of cylinders ÷ ... 

In practice the simplest working version most builders use is:

Naturally aspirated peak HP ≈ peak intake CFM (at 28") × 2.0, per cylinder's worth of flow scaled to the engine — i.e., a set of heads flowing ~250 CFM each can support roughly 500 hp on a typical V8, all else being right.

Important caveats on that estimate:

• It assumes the cam, intake, exhaust, and compression are all matched to use the flow. A great head choked by a stock cam won't hit it.
• It's a ceiling, not a promise — it's what the heads could support, not what your combination will make.
• It's for naturally aspirated engines. Boost changes the math entirely; under boost, airflow demand per CFM of head flow goes up, and head flow becomes less of the limit.
• Different references use slightly different multipliers (1.9–2.0+). Treat it as a sanity check, not a dyno sheet.

The honest use of this formula is elimination: if a head only flows enough to support 450 hp and your target is 600, that head is out — no cam will save it. It tells you what's impossible far better than it predicts what's achievable.

Bore size limits the flow you can actually use

A flow bench tests the head on a fixture with a bore adapter — and bore size changes the result. A valve that's close to the cylinder wall is shrouded: the wall blocks airflow on that side of the valve.

• Big valves on a small bore shroud badly. You're paying for flow the cylinder wall won't let you have.
• The same head on a larger bore unshrouds the valve and the real-world flow comes up.

This is why dropping a big-valve performance head onto a small-bore truck block disappoints — the valves shroud and you never see the numbers on the sheet. Confirm the head's valve size suits your bore before you read its flow chart as gospel. Flow figures are usually published on a generous test bore.

A quick checklist for reading any flow sheet

1. Find the test pressure first (28", 25", 10"). No pressure = unusable number. Convert everything to a common pressure before comparing.
2. Read the CFM at your cam's actual lift, not the peak. That's the number your engine will see.
3. Check the curve shape — does it build flow early (street) or only up top (race)?
4. Look at the E/I ratio — is the exhaust ~70–80% of intake, or is one side a bottleneck?
5. Match runner volume to displacement and RPM — velocity over peak for street engines.
6. Sanity-check power with the CFM × ~2.0 ceiling — does this head even reach your target?
7. Confirm valve-to-bore fit so you're not buying shrouded flow.

What to do next

• Pull the full flow sheet, not the headline — if a seller only quotes peak CFM, ask for the lift-by-lift table and the test pressure
• Read flow at your cam's lift, then sanity-check against the CFM × ~2.0 power ceiling
• Match runner volume to your displacement and RPM target before chasing peak numbers
• Confirm the head's valve size fits your bore so the flow on paper is flow you can actually use
• Pick the cam and intake to match the head's curve — a head only makes the power the rest of the combination lets it

Related reading

• Cylinder Heads
• Camshafts
• Valve Springs
• Intake Manifolds
• Symptoms & Fixes — Tech Library