A Vacuum Pump Can Be Working Correctly While the System Still Will Not Reach Vacuum
A vacuum system that refuses to reach its expected pressure does not automatically have a bad vacuum pump.
The pump may be operating normally while outside air enters through a seal, fitting, valve, hose, feedthrough, flange, shaft seal, weld, or damaged sealing surface.
At lower pressures, there is another complication: gas can also be released from surfaces and materials inside the chamber. This is called outgassing.
That means poor vacuum performance can come from at least three different places:
The pump itself.
A real leak into the system.
Gas being released from inside the system.
Correctly identifying which one you have can save hours of unnecessary pump repair.
Leybold identifies leaks and gas liberation from vessel walls and seals as two of the primary reasons a vacuum system may fail to reach its desired ultimate pressure. (Leybold)
What Is a Vacuum Leak?
In everyday language, a vacuum leak is simply a path that allows gas to enter an evacuated system.
Vacuum engineering usually expresses the size of that leak as a leak rate, often in:
mbar·L/s
The basic idea is gas throughput: how much gas enters the system over time.
Leybold defines a leak rate of 1 mbar·L/s as the amount of gas flow that would cause the pressure in an isolated 1-liter vessel to rise by 1 mbar in one second. (Leybold)
That definition becomes extremely useful because it means you can estimate the leak rate of a chamber without immediately using a helium leak detector.
You need the chamber volume, a vacuum gauge, and a stopwatch.
First Determine Whether the Pump or the System Is the Problem
Before chasing every O-ring on the machine, separate the pump from the rest of the vacuum system.
If practical for the equipment, isolate or blank off the pump inlet and determine whether the pump can approach the pressure expected for that pump under appropriate test conditions.
If the pump performs normally when isolated but performs poorly when connected to the chamber, the problem is likely somewhere in the system, process, or connecting plumbing rather than inside the pump.
If the pump still cannot reach an acceptable pressure when properly isolated, investigate the pump itself: oil condition, gas-ballast position, filters, internal seals, vanes, valves, or other maintenance issues.
This is an important distinction because replacing a repair kit will not fix a leaking chamber flange.
Likewise, replacing every chamber O-ring will not fix a worn vacuum pump.
Our earlier article on Vacuum Pump CFM vs. Ultimate Vacuum explains why the pressure measured in a complete system can differ significantly from the pump’s published ultimate-pressure specification. (Vacuum Pump Supply)
Vacuum Pump CFM vs. Ultimate Vacuum
The Pressure-Rise Test: One of the Most Useful Vacuum Troubleshooting Tools
A pressure-rise test is one of the simplest ways to determine how well an evacuated chamber holds vacuum.
Pump the chamber down as far as reasonably possible.
Then isolate the chamber from the vacuum pump.
Record the pressure immediately after isolation.
Continue recording pressure versus time.
If gas is entering or being released inside the chamber, the pressure will begin to rise.
The rate and shape of that pressure rise provide useful information about what is happening.
Leybold describes this method as closing the valve between the evacuated vessel and pump, measuring the pressure increase over a known period, repumping the system, and repeating the test. (Leybold)
How to Calculate Vacuum Leak Rate From a Pressure-Rise Test
Once chamber volume is known, the basic leak-rate equation is:
qL = V × Δp / Δt
where:
qL = leak rate
V = chamber volume
Δp = change in pressure
Δt = elapsed time
If volume is expressed in liters, pressure in mbar, and time in seconds, the resulting leak rate is expressed in mbar·L/s. (Leybold)
For example, suppose you have a 50-liter vacuum chamber.
After isolating it from the pump, pressure rises from:
0.010 mbar
to:
0.025 mbar
in:
60 seconds
The pressure change is:
0.015 mbar
So:
qL = 50 × 0.015 / 60
The calculated gas load is approximately:
0.0125 mbar·L/s
That number is far more useful than simply saying, “The vacuum comes up pretty fast.”
It gives you a measurable baseline.
You can repair a suspected leak, repeat the test, and see whether the system actually improved.
A Pressure Rise Does Not Automatically Mean You Have a Leak
This is one of the most important points in vacuum troubleshooting.
An isolated chamber can rise in pressure even when there is no conventional hole allowing room air into the vessel.
Materials inside the system may release trapped or adsorbed gas.
Elastomers can release gas.
Moisture can desorb from chamber surfaces.
Porous materials can contain trapped gas.
Process residue can continue evaporating.
This is outgassing.
Leybold notes that a true leak tends to produce a more constant gas load, while gas liberation from surfaces normally decreases with time. That difference can often be seen in the shape of the pressure-rise curve. (Leybold)
Leak or Outgassing? Look at the Shape of the Curve
Imagine plotting pressure vertically and time horizontally.
A real leak with relatively constant gas flow tends to produce a pressure-rise curve that remains more nearly linear.
Outgassing usually behaves differently.
As surfaces release gas and gradually become depleted, the rate of pressure increase slows. The curve begins to flatten.
Real systems often contain both effects simultaneously, so the distinction is not always perfect. But repeating the pressure-rise test after additional pump-down time can help.
Leybold recommends repeating the measurement after repumping the system. If the time required for the same pressure increase remains roughly constant, that supports the presence of a real leak. If the pressure rise becomes progressively slower, declining outgassing is likely contributing significantly. (Leybold)
This distinction matters because the solutions are completely different.
A damaged O-ring needs repair.
A wet chamber may simply need more pump-down time, warming, cleaning, or process correction.
What Is a “Virtual Leak”?
Technicians sometimes describe trapped gas or slowly released gas as a virtual leak.
There may be no actual passage from atmosphere into the chamber.
Instead, gas is trapped somewhere inside the vacuum system and slowly escapes into the evacuated volume.
Examples can include blind threaded holes, trapped volumes behind seals, porous materials, contaminated surfaces, or spaces inside poorly designed assemblies.
To the vacuum gauge, this gas load can look remarkably similar to a real leak.
That is another reason simply watching the vacuum gauge during pump-down does not always tell you what is wrong.
Where Do Vacuum Leaks Usually Occur?
Start with anything that was recently disturbed.
Common locations include flange O-rings, KF centering rings, clamps, valve seals, threaded fittings, hose connections, flexible vacuum tubing, feedthroughs, gauge connections, viewports, chamber doors, shaft seals, welds, and scratched sealing surfaces.
An O-ring does not need to be visibly destroyed to leak.
A hair, small particle, twist, nick, or scratch across the sealing surface can create a significant gas path at vacuum.
Before using more sophisticated equipment, visually inspect accessible sealing surfaces and confirm that components are assembled correctly.
Do not automatically tighten every clamp harder.
Vacuum seals work through correct geometry and sealing-surface contact, not unlimited clamping force.
Why Gauge Location Matters
The pressure displayed by a gauge is the pressure where the gauge is installed.
It is not necessarily identical to the pressure at the pump inlet or elsewhere in the chamber.
Restrictions between the gauge, chamber, and pump can create pressure differences while gas is flowing.
This is the same conductance effect discussed in last week’s pump-sizing article.
During a pressure-rise test, however, the system is isolated from the pump and there is no continuous pumped flow, making the gauge especially useful for tracking the chamber’s gas load over time.
Make sure the gauge itself and its connection are not creating the leak you are trying to find.
Also remember that different gauge technologies cover different pressure ranges and may respond differently to various gases.
VPS maintains a conversion resource for comparing Torr, micron, mbar, Pa, kPa, inHg, and mmHg when equipment uses different pressure units. (Vacuum Pump Supply)
Vacuum Pressure Conversion Chart
When a Pressure-Rise Test Is Not Enough
A pressure-rise test tells you that gas is entering or being released inside the system.
It does not necessarily tell you where.
For that, a tracer-gas leak detector is often the better tool.
Helium is widely used because it is inert, readily detectable, and can pass through very small leak paths. Mass-spectrometer helium leak detectors can locate leaks far smaller than those that could be identified using simple mechanical methods. (Leybold)
How Helium Vacuum Leak Detection Works
In vacuum-mode helium leak testing, the test object is evacuated.
A helium-sensitive leak detector is connected to the vacuum system.
A small amount of helium is then applied externally around suspected leak locations.
If helium enters through a leak, the detector senses the helium and displays a response.
This allows a technician to move methodically around flanges, welds, seals, feedthroughs, valves, and other possible leak points until the location is identified.
Leybold recommends applying helium slowly and methodically rather than flooding the entire test piece. Too much helium can make it difficult to determine where the gas entered the system. (Leybold)
Vacuum Mode vs. Sniffer Mode
Helium leak detectors can often operate in two fundamentally different configurations.
Vacuum mode places the component under vacuum and looks for tracer gas entering from outside.
Sniffer mode usually involves a component containing tracer gas at higher pressure while a handheld probe searches externally for escaping gas.
Edwards describes its ELD500 as capable of vacuum mode for precise leak-rate measurement and sniffer mode for locating leaks. (Edwards)
The correct method depends on the component and how it operates in service.
Vacuum equipment should generally be tested in a configuration representative of actual vacuum operation whenever possible. Leybold similarly distinguishes the vacuum method—gas moving from outside into an evacuated component—from positive-pressure methods where gas escapes outward. (Leybold)
How Sensitive Is a Helium Leak Detector?
Very sensitive.
The current Edwards ELD500 specification lists minimum detectable helium leak rates down to 5 × 10⁻¹² mbar·L/s for certain versions in vacuum operation, while sniffer operation is specified down to 7 × 10⁻⁹ mbar·L/s. (Edwards)
Vacuum Pump Supply currently carries Edwards ELD500 and Leybold PHOENIX helium leak-detection equipment for applications requiring quantitative leak testing and precise localization. (Vacuum Pump Supply)
Edwards ELD500 Wet Helium Leak Detector
Leybold PHOENIX Quadro Leak Detector
Do You Always Need a Helium Leak Detector?
No.
Many routine vacuum problems can be diagnosed by systematically isolating sections of the system, inspecting seals, verifying the pump independently, and performing pressure-rise testing.
Helium becomes especially valuable when the required leak tightness is much more stringent, the leak cannot be found visually, the chamber contains many possible leak points, quantitative leak-rate documentation is required, or downtime makes faster localization economically important.
A $10 O-ring can stop an expensive machine.
Finding which $10 O-ring is leaking is often the difficult part.
Why Installing a Bigger Pump Can Hide a Leak
There is an interesting connection between leak rate and pumping speed.
At steady state, additional pumping speed can sometimes lower the observed operating pressure even though the leak itself has not changed.
Leybold notes that, in principle, increasing effective pumping speed can compensate for a fixed leak rate enough to reach a specified operating pressure. In practice, that can become impractical or expensive. (Leybold)
This is important because installing a larger pump can sometimes make a leaking system appear to perform better.
But the leak is still there.
If the process requires genuine leak integrity—not merely a lower gauge reading—more pump is not the same thing as repairing the system.
What If the Pump Itself Is Leaking?
Vacuum pumps contain their own sealing components.
Shaft seals can wear.
O-rings and gaskets can harden.
Valve components can become damaged.
Oil leaks can sometimes accompany air leakage depending on the pump design and location.
If isolation testing points toward the pump, use the manufacturer’s service documentation and exploded parts diagrams to identify the correct seals, gasket sets, repair kits, filters, vanes, and other components.
The Vacuum Pump Supply technical library currently includes manuals, exploded diagrams, schematics, and parts references for Edwards RV and E2M, Busch R5, Leybold TRIVAC and SOGEVAC, Welch DuoSeal and CRVPro, Alcatel/Adixen, and other common pump families. (Vacuum Pump Supply)
Vacuum Pump Schematics, Manuals & Parts Diagrams
Quick Diagnostic Guide
| What You Observe | What It May Suggest | Next Step |
|---|---|---|
| Pump reaches normal vacuum when isolated | Problem likely in system or process | Leak-test chamber and plumbing |
| Pump cannot reach acceptable pressure when isolated | Pump maintenance issue possible | Check oil, gas ballast, filters, seals and service condition |
| Pressure rises nearly linearly after isolation | Constant leak load more likely | Locate leak mechanically or with helium |
| Pressure rise slows with time | Outgassing may be significant | Continue pump-down, clean/dry system, repeat test |
| Vacuum improves after replacing an O-ring or reseating flange | Local sealing problem confirmed | Repeat pressure-rise test and document improvement |
| Larger pump improves pressure but chamber still will not hold vacuum | Pump may be masking gas load | Perform isolated pressure-rise test |
| Leak location remains impossible to identify | Very small or inaccessible leak possible | Use helium leak detection |
Frequently Asked Questions About Vacuum Leak Detection
How do I know if my vacuum chamber has a leak?
Pump the chamber down, isolate it from the pump, and observe how the pressure changes over time. A repeatable pressure rise can indicate gas entering or being released inside the chamber. Repeating the test after additional pump-down time helps distinguish a constant leak from declining outgassing. (Leybold)
What is a good vacuum leak rate?
There is no universal “good” leak rate.
The acceptable leak rate depends on the operating pressure, chamber volume, process gas load, required stability, and application. A system operating at modest rough vacuum can tolerate a leak that would be completely unacceptable in a high- or ultra-high-vacuum instrument. Leybold specifically notes that leak-tightness requirements become more stringent as the required pressure becomes lower. (Leybold)
Can a vacuum pump overcome a leak?
A pump can remove gas entering through a leak, and a sufficiently high pumping speed may maintain an acceptable operating pressure despite some leakage. That does not make the system leak-tight, and increasing pump size is not a substitute for correcting a leak when system integrity matters. (Leybold)
Why does my chamber lose vacuum after I shut off the pump?
Possible causes include a real leak, outgassing, vapor evaporation, permeation through materials, or a leaking isolation valve. A pressure-rise test combined with repeated measurements can help distinguish the dominant cause.
Why does the pressure-rise test improve each time I repeat it?
If the pressure rises more slowly after additional evacuation, declining outgassing may be responsible for part of the original gas load. A true fixed leak tends to remain more consistent from test to test. (Leybold)
Why is helium used for vacuum leak testing?
Helium is an effective tracer gas because it is inert, readily detectable with mass-spectrometer equipment, and can pass through extremely small leak paths. This allows helium systems to locate very small leaks rapidly and quantitatively. (Leybold)
The Bottom Line
When a vacuum system cannot reach or hold the expected pressure, do not immediately blame the pump.
First determine whether the pump can perform normally when separated from the system.
Then determine whether the chamber has a measurable gas load.
Use a pressure-rise test to quantify that load.
Watch how the pressure-rise curve changes over time to help distinguish a constant leak from outgassing.
Inspect recently disturbed seals and connections.
And when the leak is too small or too difficult to locate mechanically, helium leak detection provides a far more sensitive method of finding the actual problem.
The goal is not simply to get the gauge reading lower.
It is to understand where the gas is coming from.
That is the difference between repairing the vacuum system and merely treating the symptom.
If troubleshooting points back to the pump, Vacuum Pump Supply’s technical library can help identify the correct manual, schematic, seal, filter, vane set, or repair kit for the exact pump model. If the system requires quantitative leak testing, VPS also carries Edwards and Leybold helium leak-detection equipment.
For assistance, send the pump manufacturer, complete model number, chamber type, target vacuum, current measured pressure, and what happens during an isolated pressure-rise test. That information can narrow the problem considerably before parts are replaced.