The bubble point test is the fastest integrity check on a sterilizing-grade filter — and the one that fails most quietly when the membrane isn't what you think it is. Here's how it works, how to set limits, and where it sits in the method set.
What the number actually tells you
Wet the membrane, apply rising pressure, and watch. The pressure at which the first continuous stream of air flows through is the bubble point. It tells you the size of the largest flow-through pore in the wetted membrane — not the pore size printed on the datasheet, which is the as-supplied condition, but the condition the membrane is in right now, after sterilisation, after handling, after whatever the day threw at it.
That distinction is the whole value of the test. The datasheet number is true once. The bubble point is true every time you run it.
How the test runs, start to finish
The sequence is short, and each step carries a failure mode worth knowing.
Wet the membrane with a degassed wetting fluid — water for most aqueous filtration, an alcohol when the fluid system is alcohol-compatible. Incomplete wetting is the classic false result: air trapped in a pore releases early, and you read a bubble point that's lower than the membrane actually supports.
Apply pressure and increase it at a controlled rate. The first continuous stream of bubbles is the point. It's not one bubble — one bubble is noise. It's a sustained flow, which means air is opening the largest wetted pore and pushing through it.
Record the pressure. Compare it to the acceptance limit. Pass, fail, or investigate. The whole cycle takes seconds to a minute, which is exactly why it's the method you can run on every fill without the line noticing.
Setting limits you can defend
The limit doesn't come from the instrument. It comes from the membrane's qualification data — the as-supplied pore-size distribution, the dry-bubble or diffusive-flow values from the time the membrane was qualified — and from your own procedure's risk assessment. The instrument measures. The procedure decides.
If your limit is a copy of a sister product's limit, you're borrowing someone else's risk assessment. It usually works. Then the membrane lot, the fluid, or the sterilisation cycle changes — and the limit you inherited is the number that fails, and the deviation report says you set it.
Work backward from the membrane. Qualify the as-supplied condition, apply the acceptance criteria your procedure defines, and validate against it. One membrane, one set of limits, one documented chain.
Where bubble point sits in the method set
Bubble point is one of five methods in the USP <43> / EP 11.0 set, and it's the one the guidance reaches for most. Under the 2008 EU GMP Annex 1 text, the filter was to be verified before use and confirmed immediately after use by an appropriate method "such as a bubble point, diffusive flow or pressure hold test." The 2022 revision keeps pre-use verification and adds a non-destructive post-use requirement — and bubble point is non-destructive, which is exactly why it stays in the procedure.
Its job in the set: the quick in-line check on each fill. The slower methods carry the deeper work — dry bubble at qualification, diffusive flow for post-use sensitivity, pressure hold for leak confirmation. Bubble point is the one that runs every time and keeps the record moving.
Why it's the in-line workhorse
Three properties do the work. It's fast — seconds to a minute per test, so the line doesn't wait. It's non-destructive — the filter that passes keeps filtering, which matters when the same membrane runs a full batch. And it's repeatable — same membrane, same wetting, same conditions, same number.
A result that's the same twice is a result you can trend. And a trend is what shows the membrane changing before it fails. A single pass tells you nothing about the membrane's future. A run of passes with a slowly drifting bubble point tells you the membrane is wetting differently, or fouling, or ageing — and that's a decision you make before the deviation, not after.
When a result is wrong
The test fails quietly, and the usual suspects are unglamorous.
The membrane dried out between wetting and testing, and the reading comes in low. The wetting fluid wasn't degassed, and bubbles appear early. The wrong wetting fluid for the system. A pressure ramp too fast, so the "first continuous stream" was actually a burst. A temperature move that the reading followed.
Each of these is a procedure problem, not an instrument problem — which is why the method lives in the SOP, not in the machine. The instrument gives you the number. The procedure makes the number true. When the two disagree, the SOP wins, and the record should say so.
What a pass doesn't tell you
One more thing, because it's the question the auditor asks and the procedure rarely answers. A pass on bubble point tells you the largest wetted pore is still within your limit. It does not tell you the membrane is sterile — it never did that. It does not tell you the filter didn't lose a small patch of support layer, as long as that patch doesn't change the largest pore. It does not tell you the seal at the housing is intact, which is a separate check entirely.
That's why bubble point runs next to the other methods, not instead of them. It's the fastest, cheapest signal you have on the membrane's physical condition, and the one you can run on every fill. But the signal is one of several, and the program is the set. A procedure that runs only bubble point and calls the filter verified has answered one question out of five — and the four it didn't ask are the four a deviation finds out later.
The instrument side
On the hardware, the Integtest V6.5 and V8.0 are the working instruments for the set. The V6.5 runs on a 7-inch color touch screen, stores more than 15,000 sets of historical records, weighs 8 kg and runs at ambient +5 to +40 °C with 10–80% relative humidity. It's specified around a very sensitive measurement system, with conditions developed in accordance with GAMP guidelines, full 21 CFR Part 11 compliance, and the IQ/OQ/PQ documentation package included.
That last item is the one that keeps the number auditable: the instrument that arrives is the instrument the validation file describes. The bubble point is only as good as the procedure around it — and the procedure is only as good as the documentation behind it.
Standards and sources
This guide draws on the primary documents below. Read them with your procedure open — the gap between the two is where the program lives.
USP <43> — Filter Units · EU GMP Annex 1 — Sterile Medicinal Products (EMA) · FDA Guidance for Industry — Sterile Drug Products Produced by Aseptic Processing (2004)
Running bubble point in-process?
Tell us your membrane and your procedure — we'll come back with configuration, limits guidance and documentation.
Talk to Our Team