Pressure ratio is what the compressor sees; boost is what you read
A boost gauge measures gauge pressure — manifold pressure minus whatever the atmosphere happens to be. A compressor does not work in gauge pressure. It takes air at some absolute pressure and raises it to a higher absolute pressure, and the ratio between those two numbers determines how hard it is working, how hot the air comes out, and where on its map it is running.
The two diverge as soon as anything changes. Fifteen psi of boost at sea level is an absolute manifold pressure of 29.696 psia against 14.696 ambient, a ratio of 2.02. The same 15 psi at 5,000 feet is 27.23 against 12.23, a ratio of 2.23 — 10% more work for the same gauge reading. Nothing about the engine changed; the denominator did.
Restrictions push the number further. A dirty filter or an undersized intake pipe costs pressure before the compressor, so the inlet term shrinks. An intercooler costs pressure after it, so the compressor must produce more than the manifold receives. Both raise the ratio, and both are invisible on a boost gauge because the gauge sits in the manifold, downstream of everything.
Getting this number right is the whole point of a compressor map. Plot the wrong pressure ratio and you pick a turbo that surges, chokes, or runs in a 55% efficiency island where the charge arrives so hot that the density gain you were paying for never materialises.
Each term, and why it sits where it does
Ambient pressure is the starting point and comes from altitude. The standard atmosphere gives P = 14.696 × (1 − 6.87535×10−6h)5.2559 with h in feet, which is the pressure relation that follows from a constant temperature lapse rate in the troposphere. It puts sea level at 14.696 psia, 5,000 ft at 12.23 psia and 10,000 ft at 10.11 psia. Weather moves the real figure by a percent or two either way, so use a barometer reading corrected to station pressure if you have one.
Inlet loss goes in the denominator because it happens upstream of the compressor. The compressor inhales at ambient minus the loss across the filter and piping. A half psi does not sound like much until you notice it raises the ratio by about 3.5% at sea level, and every percent of ratio becomes a temperature rise you then have to intercool away. This is why inlet plumbing on a serious turbo car is so much larger than it looks like it needs to be.
Intercooler loss goes in the numerator because it happens downstream. To see 15 psi in the manifold with a 1.5 psi drop across the core, the compressor must produce 16.5 psi gauge at its outlet. That is real work the compressor does and you never see, and it is the reason an intercooler is a genuine engineering trade rather than a free win: the density you gain by cooling costs you a little in pressure and a little in throttle response from the extra volume.
Boost is referenced to ambient in the numerator because that is how a boost gauge reads. If you have measured absolute manifold pressure directly — from a MAP sensor in kPa, for instance — subtract ambient to get the gauge figure this calculator expects. Note that 100 kPa is close to but not exactly one atmosphere: standard sea-level pressure is 101.325 kPa.
Both loss terms grow roughly with the square of mass flow, since they are dynamic pressure losses. A figure measured at half throttle will badly understate the loss at full flow.
Worked example: 15 psi at sea level, then at 5,000 feet
Target 15 psi in the manifold, with 0.5 psi lost across the filter and inlet piping and 1.5 psi lost across the intercooler.
- Ambient at sea level. 14.696 × (1 − 0)5.2559 = 14.696 psia.
- Compressor inlet. 14.696 − 0.5 = 14.196 psia.
- Absolute manifold pressure. 14.696 + 15 = 29.696 psia.
- Compressor outlet. 29.696 + 1.5 = 31.196 psia.
- Pressure ratio. 31.196 ÷ 14.196 = 2.198.
Without the two losses the ratio would have been 29.696 ÷ 14.696 = 2.021. Two psi of plumbing loss added 0.18 to the ratio, an 8.7% increase in the work the compressor has to do — which is worth roughly 20°F of extra charge temperature at typical efficiencies.
Now run the same car at Denver, 5,000 ft. Ambient becomes 14.696 × (1 − 6.87535×10−6 × 5,000)5.2559 = 14.696 × 0.965625.2559 = 12.228 psia. Inlet is 11.728, outlet is 12.228 + 15 + 1.5 = 28.728, and the ratio is 28.728 ÷ 11.728 = 2.449. The same wastegate setting has moved the operating point up the map by a quarter of a ratio, which on many compressors is the difference between the 74% island and the 68% one.
For unit conversion: 15 psi is 15 ÷ 14.5038 = 1.034 bar and 15 × 6.89476 = 103.4 kPa of boost. Absolute manifold pressure at sea level is then 101.325 + 103.4 = 204.7 kPa, which is what a two-bar MAP sensor would report.
Taking the ratio to a compressor map
A compressor map plots pressure ratio on the vertical axis against mass flow on the horizontal, with closed contours of isentropic efficiency and speed lines running across it. Your operating point is the pair (mass flow, pressure ratio), and where it lands tells you everything.
Too far left and you hit the surge line. Surge is flow reversal through the wheel, and it sounds like a rapid fluttering or a chuffing noise on throttle lift. It is hard on thrust bearings. A compressor that is too large for the engine surges at low rpm and part throttle, which is why simply fitting the biggest turbo you can afford is a mistake.
Too far right and you hit choke, where flow stalls at sonic velocity through the inducer and efficiency collapses. A compressor that is too small chokes at high rpm, and the symptom is boost that will not hold at the top end however hard the turbo spins.
Aim for the highest efficiency island you can reach at the point where you spend the most time at full load. For a road car, that is usually mid-range rather than redline. For a drag car it is the top end. The difference between running at 76% and 62% efficiency at the same pressure ratio is worth 40 to 60°F of charge temperature, which the boost horsepower calculator will turn into a hard number in horsepower.
To get the mass flow coordinate, work out the engine's air consumption at your target rpm and volumetric efficiency — the same arithmetic as induction sizing, converted from volume to mass at the density in the manifold. Then check that the fuel system covers the resulting power with the injector calculator.
Ambient pressure and the ratio needed for 15 psi by altitude
| Altitude (ft) | Ambient (psia) | Manifold (psia) | Pressure ratio at 15 psi |
|---|---|---|---|
| 0 | 14.696 | 29.696 | 2.021 |
| 2,000 | 13.665 | 28.665 | 2.098 |
| 4,000 | 12.692 | 27.692 | 2.182 |
| 6,000 | 11.777 | 26.777 | 2.274 |
| 8,000 | 10.916 | 25.916 | 2.374 |
| 10,000 | 10.107 | 25.107 | 2.484 |
Absolute manifold pressure falls with altitude even at constant gauge boost, so the engine makes less power despite the compressor working harder.
Mistakes and assumptions
- Treating boost and pressure ratio as the same thing. Boost is a difference; pressure ratio is a quotient. They only track each other at fixed ambient pressure and zero losses.
- Ignoring the inlet. Loss before the compressor divides the denominator and raises the ratio disproportionately. A restrictive filter is the cheapest power you will ever find.
- Measuring losses at part throttle. Pressure drops scale roughly with the square of flow, so a figure taken at half flow understates the full-flow loss by about four times.
- Using standard-atmosphere pressure on an unusual day. Weather moves station pressure by a couple of percent. A barometer beats a formula when you have one.
- Confusing 1 bar with 1 atmosphere. One bar is 100 kPa; standard atmosphere is 101.325 kPa. The difference matters when comparing published boost figures.
- Forgetting that the ratio changes with the weather and the day. A wastegate set by gauge pressure gives a different pressure ratio in winter than in summer at the same track.
Key terms
- Pressure ratio
- Absolute compressor outlet pressure divided by absolute compressor inlet pressure. Dimensionless, and always greater than 1 when the compressor is doing work.
- Surge
- Unstable flow reversal through the compressor when it is asked for high pressure at low flow. Audible as fluttering, and damaging to bearings.
- Choke
- The flow limit where velocity through the inducer reaches sonic conditions and efficiency collapses. It is the right-hand boundary of a compressor map.
- Station pressure
- Actual absolute pressure at a location, as opposed to the sea-level-corrected pressure reported in weather forecasts.
How pressure ratio connects to everything else in a boosted build
Pressure ratio is the hinge between the hardware and the physics. On the hardware side it selects the compressor. On the physics side it sets the temperature rise: an isentropic compression raises absolute temperature by PR0.2857, divided by the compressor's efficiency, which is exactly the calculation behind the boost horsepower estimate. A pressure ratio of 2.2 through a 70% efficient wheel raises absolute inlet temperature by about 28%, so 80°F air leaves the compressor near 230°F.
That temperature is why the two loss terms deserve real attention rather than a shrug. Every 0.1 of pressure ratio you waste on plumbing is heat you then spend intercooler capacity removing, and heat you cannot remove is timing you cannot run. The order of attack for a car that is not making the power its boost suggests is almost always: fix the inlet, fix the intercooler, then look at the turbo.
Supercharged engines use the identical arithmetic. A centrifugal blower has a compressor map exactly like a turbo's; a Roots or screw blower does not have one in the same form, but the pressure ratio still governs its discharge temperature and its parasitic drive power. The one difference is that a belt-driven blower's ratio is tied to engine speed rather than to exhaust energy, so it varies predictably with rpm rather than with load.
Finally, keep the compression ratio honest against the pressure ratio. Peak cylinder pressure is roughly proportional to the product of the two, so a high static compression ratio and a high pressure ratio together are what lift head gaskets. When you have settled on hardware, verify the finished car's output against its trap speed — it is the one measurement in this whole chain that cannot be argued with.