Why a Helium Leak Detector Needs Two Pumps
In almost any helium leak detector system you'll find two pumps: a mechanical roughing pump (also called a backing pump) and a turbomolecular pump. The two pumps operate in entirely different flow regimes, and neither can cover the other's range pressure range.
Two Flow Regimes, One Pressure Path
A helium leak detector is, at its core, a mass spectrometer tuned to detect helium-4. For the analyzer cell to work, pressure must drop from atmosphere pressure (760 Torr) to roughly 10⁻⁶ Torr — about a billion-fold reduction. Gas behaves completely differently at the two ends of that range. Near atmospheric pressure, molecules collide with each other constantly and move as a bulk fluid — viscous flow. Below roughly 1×10⁻³ Torr, molecules collide with the chamber walls far more than with each other, and flow becomes molecular. Each pump technology is built for one regime, not both.
The Roughing Pump: Atmosphere to Molecular Flow
The roughing pump is typically either a wet pump such as a rotary vane pump or a dry pump which is likely to be a scroll or lobe pump. (More about Wet Pumps vs Dry Pumps in another article.) Both styles are mechanical, positive-displacement mechanisms that physically shove large volumes of gas out of the system. This type of pump is efficient in viscous flow but loses effectiveness as the gas thins out, so its job stops at roughly 100 mTorr, the pressure where flow transitions from viscous to molecular. Since the turbomolecular pump cannot start at atmospheric pressure: its blades spin at 60,000–90,000 RPM, and at atmospheric density the drag and heat generated would stall or damage the rotor almost instantly, it’s crucial that the backing pump reduces the pressure by 1000 times.
The Turbo Pump: Molecular Flow to High Vacuum
Once the system reaches molecular flow (100mTorr), the turbo pump takes over. Alternating rows of angled rotor and stator blades spin fast enough to strike individual gas molecules and impart directional momentum toward the exhaust — a mechanism that only works when molecules are sparse enough to interact with the blades individually rather than as a bulk gas. From there, the turbo drives pressure down to below 0.001 mTorr (~ 10⁻⁶ Torr), deep enough for the Mass Spectrometer section of the leak detector to function properly and repeatably.
What It Means for Troubleshooting
This two-stage relationship explains a lot of real-world fault behavior. A leak detector that won't reach test-ready vacuum, or that trips a turbo overspeed or overtemperature fault on startup, usually traces back to the backing pump: worn vanes, low oil, or a foreline restriction keeping the system from reaching that 100–500 mTorr handoff. In most modern leak detector designs, if the roughing pump can't establish molecular flow, the turbo just won't spin up at all to protect the turbo and the system will fault.
High Helium Background Challenges
With a clear picture of the two flow regimes inside the leak detector, it's easier to see why systems operating in high-helium environments struggle to hold repeatable, low leak rate measurements. Once gas reaches molecular flow, the turbo's blades strike individual molecules and drive them toward the exhaust, but without viscous flow to carry them along, that motion isn't one-directional. Molecules can just as easily random-walk back upstream. Turbo pumps also compress light gases like helium far less effectively than heavier background gases. The result can be a steady trickle of ambient helium back through the pump to the analyzer cell, which raises the instrument's background signal and makes it harder to resolve a genuine small leak from the surrounding helium noise.