The CFM Number on the Data Sheet Is Only One Point in the Story
A vacuum pump may be advertised as:
8.4 CFM.
20 CFM.
100 m³/h.
Those numbers are useful, but they do not tell you how the pump performs across its entire operating range.
A vacuum pump does not necessarily move gas at its headline pumping speed from atmospheric pressure all the way down to ultimate vacuum. As pressure changes, the pump’s effective gas-moving performance can change as well.
That is why pump manufacturers publish performance curves.
A performance curve answers a much more useful question than:
“How many CFM is this pump?”
It answers:
“How much pumping speed does this pump provide at the pressure where my process actually operates?”
That distinction becomes extremely important when sizing a pump, comparing two models, troubleshooting slow pump-down, or deciding whether a larger pump will actually improve a system.
Our earlier article on CFM versus ultimate vacuum explains why pumping speed and ultimate pressure are different specifications. A performance curve takes the next step and shows how those two ideas interact across the operating range. (Vacuum Pump Supply)
Vacuum Pump CFM vs. Ultimate Vacuum
What Is a Vacuum Pump Performance Curve?
A typical vacuum pump performance curve plots:
Inlet pressure on the horizontal axis
against
Pumping speed on the vertical axis
Pressure is usually shown in units such as:
mbar, Torr, Pa, or micron.
Pumping speed may be shown as:
m³/h, L/s, or CFM.
Vacuum pressure covers an enormous numerical range, so the pressure axis is usually logarithmic rather than linear.
That means the spacing between:
10 mbar and 1 mbar
is visually similar to the spacing between:
1 mbar and 0.1 mbar.
Each step represents a factor of ten.
Understanding that logarithmic scale is one of the first keys to reading the graph correctly.
If the manufacturer uses unfamiliar pressure units, the VPS conversion chart can help translate Torr, micron, mbar, Pa, kPa, and inHg. (Vacuum Pump Supply)
Vacuum Pressure Conversion Chart
Start With the Horizontal Axis: Inlet Pressure
The horizontal axis represents the pressure at the pump inlet.
On many vacuum-pump curves, atmospheric or relatively high pressure appears toward the right side of the graph and deeper vacuum toward the left.
For example:
100 mbar is higher pressure than 10 mbar.
10 mbar is higher than 1 mbar.
1 mbar is higher than 0.1 mbar.
And 0.001 mbar represents a much deeper vacuum than all of them.
This sounds basic, but it prevents a common mistake.
With vacuum, the smaller absolute pressure number represents the deeper vacuum.
When comparing two curves, always confirm that both use the same pressure units before drawing conclusions.
Then Read the Vertical Axis: Pumping Speed
The vertical axis tells you the pump’s volumetric pumping speed at a particular inlet pressure.
For the Edwards RV12, for example, Edwards publishes a peak pumping speed of 12 m³/h at 50 Hz, while the same RV family is also presented with performance curves showing how pumping speed changes as inlet pressure falls. Edwards publishes separate curves for RV3, RV5, RV8 and RV12 models and distinguishes high-vacuum, high-throughput and gas-ballast operating modes. (Edwards)
The important word is peak.
A pump rated at 12 m³/h does not necessarily provide exactly 12 m³/h at every point on the pressure axis.
The curve tells you what happens away from the headline specification.
The Flat Part of the Curve Is the Pump’s Strongest Operating Region
Many mechanical vacuum-pump curves contain a broad area where pumping speed remains relatively stable.
This is the operating region where the pump is delivering something close to its rated pumping speed.
If your process operates inside that region, the headline CFM or m³/h specification may be a reasonably useful approximation.
As the pump approaches its ultimate-pressure region, however, the curve typically begins to fall.
That decline is important.
The pump is approaching the limit of what its design, internal sealing, lubricant, gas load, and compression ratio can achieve.
Eventually pumping speed approaches zero at the pump’s ultimate pressure because the pump can no longer produce a meaningful net reduction in pressure.
That is why ultimate pressure should never be confused with a normal working pressure.
A pump may technically reach a particular ultimate vacuum while providing relatively little useful pumping capacity very close to that limit.
The “Knee” of the Curve Can Matter More Than the Ultimate Vacuum
The area where the performance curve begins to turn downward is often particularly useful when evaluating a process.
Suppose Pump A and Pump B both advertise similar ultimate pressure.
Pump A maintains strong pumping speed much deeper into the pressure range.
Pump B begins losing speed considerably earlier.
If your process operates in that region, Pump A may outperform Pump B substantially even though their catalog headline numbers appear similar.
This is why experienced vacuum-system designers care about the shape of the performance curve, not only the endpoints.
A pump should ideally provide useful pumping speed at the pressure where the process actually runs.
Do Not Size a Pump to Operate at Its Absolute Ultimate Pressure
Suppose a pump is rated to an ultimate pressure of:
2 × 10⁻³ mbar.
And your process must operate continuously at:
2 × 10⁻³ mbar.
On paper, it looks like a match.
In practice, you are asking the pump to operate at the very edge of its published capability.
There is almost no margin for:
chamber outgassing,
tiny leaks,
process gas,
oil contamination,
conductance losses,
temperature changes,
or normal pump wear.
A better selection normally provides adequate pumping speed below the process pressure, rather than merely possessing an ultimate-pressure specification equal to the target.
The Edwards RV family demonstrates why these are separate numbers. Edwards publishes peak pumping speeds from 3.3 to 12 m³/h at 50 Hz across RV3 through RV12, while the hydrocarbon-oil ultimate-pressure specification is approximately 2 × 10⁻³ mbar. Those specifications describe two very different aspects of performance. (Edwards)
Why Are There Sometimes Several Curves for the Same Pump?
A performance graph may contain multiple lines for one pump model.
Do not assume they are duplicate measurements.
The manufacturer may be showing different:
gas-ballast positions,
operating modes,
motor frequencies,
pump speeds,
gas species,
or configurations.
The Edwards RV curves are a good example.
Edwards identifies three operating conditions on its current published performance charts:
-
High-vacuum mode with gas ballast closed.
-
High-throughput mode with gas ballast closed, or high-vacuum mode with the first gas-ballast setting.
-
High-throughput or high-vacuum operation with the second gas-ballast setting.
Those curves are deliberately different because the pump is being operated differently.
Reading the legend is therefore just as important as reading the axes.
Why Gas Ballast Moves the Curve
Opening gas ballast intentionally allows additional gas into the compression process.
That helps prevent condensable vapor—especially water vapor—from condensing into the pump oil.
The tradeoff is poorer ultimate pressure.
Edwards publishes the RV series at approximately 2 × 10⁻³ mbar ultimate pressure in its normal high-vacuum configuration, but considerably higher ultimate-pressure values when gas ballast is operating. (Digital Brochure)
Nothing is wrong with the pump.
The operating condition changed.
This is why someone comparing pump curves must make sure the ballast setting is the same.
Comparing one manufacturer’s closed-ballast curve with another manufacturer’s open-ballast curve would produce a misleading result.
For a deeper explanation of the operating principle, VPS already has a dedicated gas-ballast guide.
What Does a Gas Ballast Valve Do?
Why 50 Hz and 60 Hz Curves Are Different
AC motor-driven vacuum pumps may turn at different speeds depending on supply frequency.
That can change displacement and pumping speed.
Edwards, for example, publishes different 50 Hz and 60 Hz data for many rotary vane pumps. The E2M40 is rated at 37 m³/h pumping speed at 50 Hz and 44 m³/h at 60 Hz, while its published ultimate pressure remains approximately 3 × 10⁻³ mbar without gas ballast. (Edwards)
Likewise, Edwards’ RV performance charts contain separate 50 Hz and 60 Hz curves.
So if you are comparing a European 50 Hz installation with a North American 60 Hz replacement, do not assume the pumping speed is identical.
Check the correct curve for the actual motor and frequency.
Displacement and Pumping Speed Are Not Always the Same Number
Some manufacturer data sheets show both:
maximum displacement
and
pumping speed.
These numbers can differ.
For example, Edwards publishes the E2M40 with maximum displacement of 42.5 m³/h at 50 Hz, while measured pumping speed is 37 m³/h. At 60 Hz, maximum displacement is 50.5 m³/h while pumping speed is 44 m³/h. (Edwards)
Displacement is closely related to the pump’s geometric ability to move volume.
Pumping speed represents actual measured performance under the manufacturer’s specified test method.
When comparing pumps for an application, measured pumping speed is generally the more useful number.
Do not automatically compare one manufacturer’s displacement figure with another manufacturer’s pumping-speed figure.
You may be comparing different specifications.
Nominal Pumping Speed Is Not Necessarily the Speed at Your Chamber
This may be the most important concept in the entire article.
The pump manufacturer measures pumping speed at or near the pump inlet.
Your process chamber is usually connected through:
vacuum hose,
pipe,
elbows,
valves,
inlet filters,
traps,
reducers,
and fittings.
Every one of those components creates resistance to gas flow.
In vacuum engineering that resistance is described using conductance.
Leybold states that the effective pumping speed available at a vessel is generally lower than the pump’s inlet pumping speed because intermediate piping, valves, traps, separators and other components restrict flow. (Leybold Content)
The relationship can be written:
1 / Sₑff = 1 / S + 1 / C
where:
Sₑff = effective pumping speed at the chamber
S = pumping speed of the pump
C = conductance of the connecting vacuum line
This equation explains something that surprises many users:
Installing a pump with twice the CFM may not provide twice the pumping speed at the chamber.
If the plumbing is the restriction, the larger pump cannot overcome all of that lost conductance.
A Simple Conductance Example
Consider a vacuum pump with a pumping speed of:
20 L/s
connected to a system through plumbing with a conductance of:
10 L/s.
Using:
1 / Sₑff = 1 / S + 1 / C
we get:
1 / Sₑff = 1 / 20 + 1 / 10
So the effective speed at the chamber is only about:
6.7 L/s.
Now replace the pump with a 40 L/s model while leaving the same restrictive plumbing.
The effective speed becomes only about:
8 L/s.
The pump capacity doubled.
The effective chamber pumping speed increased only about 20%.
That is why hose diameter and vacuum-line design can sometimes matter more than buying a larger pump.
The catalog curve tells you what the pump can do.
Conductance tells you how much of that performance actually reaches the process.
What Is Vacuum Pump Throughput?
Pumping speed is also different from gas throughput.
Leybold defines vacuum throughput with the relationship:
Q = p × S
where:
Q = gas throughput
p = inlet pressure
S = pumping speed at that pressure.
Throughput is commonly expressed in units such as mbar·L/s. (Leybold Content)
This matters because the same volumetric pumping speed represents dramatically different quantities of gas at different pressures.
A pump moving 10 L/s at 100 mbar is handling much more gas mass than when it is moving 10 L/s at 0.001 mbar.
That is why process gas load must be considered when selecting a pump for continuous operation.
Performance Curves Are Especially Important for Continuous Processes
Pump-down applications and continuous processes ask different questions.
For a pump-down application, you may primarily care about:
How quickly can this chamber go from atmospheric pressure to the target pressure?
For a continuous process, the more important question may be:
Can the pump remove the process gas continuously while maintaining the required pressure?
That is where the performance curve and throughput relationship become particularly important.
If your process continuously generates gas at a pressure where the pump’s speed has already fallen sharply, the system may stabilize at a higher pressure than expected.
The pump has not necessarily failed.
The operating point may simply be where:
gas entering the system = gas the pump can remove.
How Leaks Shift the Operating Point
A vacuum leak adds gas load.
So does outgassing.
So does vapor released by the process.
The system pressure stabilizes when the pump removes gas at the same rate it enters or is generated.
This means a pump can have an excellent ultimate-pressure specification but never reach that pressure when attached to a real system with significant gas load.
The performance curve helps explain why.
At the process pressure, the pump has a certain pumping speed.
That speed multiplied by pressure determines its throughput capability.
If the gas load is too high, the pressure rises until the pump can keep up.
Our vacuum leak-detection article explains how pressure-rise testing can help distinguish true leaks from outgassing when a system does not reach the expected vacuum.
Why Pump Curves Are Useful for Troubleshooting
A performance curve is not only a buying tool.
It can also help determine whether a pump is behaving normally.
Suppose a system historically reached 0.01 mbar easily.
Now it stalls at 0.5 mbar.
The manufacturer’s curve indicates that the pump should still have substantial pumping speed at 0.5 mbar.
That suggests something in the system has changed.
Possible causes include:
oil contamination,
gas ballast left open,
inlet leakage,
process contamination,
worn internal components,
a clogged exhaust system,
or an increased gas load.
The curve gives you a performance expectation against which the real system can be compared.
Without it, troubleshooting becomes guesswork.
Why an Exhaust Problem Can Affect the Inlet Performance Curve
The inlet is not the only side of the pump that matters.
A vacuum pump must also discharge the gas it compresses.
Excessive exhaust backpressure can raise operating temperature, increase motor load, affect oil behavior, and interfere with proper pump operation.
This is why exhaust mist filters and separators are maintenance items rather than permanent installations that can be ignored indefinitely.
Some modern industrial pumps even monitor exhaust-filter condition through backpressure sensing; Leybold’s SOGEVAC NEO documentation, for example, describes pump monitoring that includes exhaust-filter condition. (Vacuum Pump Supply)
If a pump that historically followed its expected performance suddenly does not, check both sides of the system.
A Real Edwards RV12 Example
Consider the Edwards RV12.
Edwards publishes:
12 m³/h peak pumping speed at 50 Hz
and approximately:
2 × 10⁻³ mbar ultimate pressure with hydrocarbon oil in the standard high-vacuum condition. (Edwards)
The performance graph then shows the actual pumping-speed curve between those extremes.
At higher pressures, the pump operates over a relatively broad speed range.
As pressure approaches the pump’s ultimate capability, pumping speed falls.
Gas ballast and high-throughput mode shift the curves.
And 60 Hz operation produces a different curve than 50 Hz operation.
That one chart tells you more about the pump’s real behavior than the words:
“12 m³/h, 2 × 10⁻³ mbar.”
Those two numbers are the endpoints.
The curve tells you what happens between them.
How to Compare Two Vacuum Pump Curves Correctly
Before comparing two pumps, make sure the curves use equivalent test conditions.
Check the pressure units, pumping-speed units, gas being pumped, electrical frequency, gas-ballast setting, operating mode, oil type, and whether the specification shows measured pumping speed or geometric displacement.
Then find the pressure your process actually requires.
Move vertically from that pressure until you intersect each pump’s curve.
Then read horizontally to determine the approximate pumping speed available at that pressure.
That is a much more meaningful comparison than simply asking which pump has the larger maximum CFM figure.
Quick Guide: What Each Specification Tells You
| Specification | What It Actually Tells You |
|---|---|
| Peak pumping speed | Maximum or near-maximum volumetric gas-moving capability |
| Pumping-speed curve | Pumping speed available across the pressure range |
| Ultimate pressure | Lowest pressure the pump can approach under specified test conditions |
| Displacement | Theoretical/geometric gas-moving capacity |
| Effective pumping speed | Pumping speed actually available at the chamber after line losses |
| Throughput | Gas quantity being handled at a specific pressure |
| Maximum continuous inlet pressure | Pressure above which the manufacturer may restrict continuous operation |
| Gas-ballast curve | Performance when ballast is intentionally introducing additional gas |
| 50/60 Hz curve | Performance at the applicable motor frequency |
Frequently Asked Questions
Why does vacuum pump pumping speed fall near ultimate pressure?
As a mechanical pump approaches its pressure limit, internal backflow, sealing behavior, lubricant vapor, compression limits, and other gas loads become increasingly significant relative to the small amount of gas remaining at the inlet. Net pumping speed therefore decreases as the pump approaches its ultimate pressure.
Is the highest point on the curve the pump’s CFM rating?
Often it is close to the manufacturer’s published peak pumping speed, but always check how the manufacturer defines the specification. Some literature publishes displacement while other literature publishes standardized measured pumping speed.
Can I use the performance curve to calculate pump-down time?
Yes, but an accurate calculation should account for the fact that pumping speed can change with pressure. A simple constant-speed pump-down equation is useful for rough estimates, while more accurate modeling integrates the actual pumping-speed curve and includes system conductance, leaks, outgassing, and process gas.
Why is the pumping-speed curve different with gas ballast open?
Gas ballast intentionally introduces additional gas into the compression stage to improve condensable-vapor handling. This raises attainable ultimate pressure and changes pump performance in the lower-pressure region. Edwards publishes separate RV curves for these operating modes.
Why does my chamber not see the pumping speed shown on the curve?
The manufacturer’s curve generally describes performance at the pump inlet. Hose, pipe, valves, traps, filters, and fittings reduce conductance, so effective pumping speed at the chamber will usually be lower. (Leybold Content)
Does doubling CFM cut pump-down time in half?
Not necessarily.
If the pump is the dominant limitation and operates in a suitable pressure range, additional pumping speed can significantly reduce pump-down time.
If the vacuum line is conductance-limited, the process releases significant gas, or the larger pump’s speed falls in the relevant pressure range, the improvement may be much smaller.
The Bottom Line
A vacuum pump performance curve tells you something a single CFM specification cannot:
how the pump actually performs at your pressure.
When reading a curve, look at:
the pressure axis,
the pumping-speed axis,
the shape of the curve,
the pressure where pumping speed starts to fall,
the operating mode,
the gas-ballast setting,
the electrical frequency,
and the target pressure of your process.
Then remember that the curve is still describing the pump.
Your chamber sees effective pumping speed, which can be reduced substantially by hoses, valves, traps, filters, and other restrictions between the pump and the process. (Leybold Content)
That is why the best vacuum-pump question is rarely:
“How many CFM is it?”
A better question is:
“How much pumping speed will I actually have at the pressure where my system needs to operate?”
Vacuum Pump Supply maintains pump manuals, schematics, technical specifications, cross references, and Resource Centers for many Edwards, Leybold, Busch, Welch, Alcatel/Adixen, and other commonly serviced vacuum pumps. The current VPS catalog also includes performance-curve information on pump pages such as the Edwards E2M18 and E2M40, making it easier to compare real operating performance rather than headline specifications alone. (Vacuum Pump Supply)