Setting a Pipette Calibration Interval You Can Defend

Almost every laboratory we visit calibrates pipettes every six or twelve months, and almost none of them can explain why. The interval was inherited, copied from a template SOP, or set by whichever vendor quoted first. That is a soft spot in an otherwise solid quality system, and auditors have learned to press on it.

The standard asks you to decide

ISO 8655 sets maximum permissible errors. It does not set your calibration interval. Neither does ISO/IEC 17025. Both expect the user to establish an interval appropriate to the instrument, the application, and the observed performance — and to be able to show the reasoning.

That is good news. It means an interval justified by your own data is stronger than one copied from someone else's SOP, even if the number ends up identical.

Four inputs that should move the number

• Use intensity. A pipette cycled two thousand times a week is not the same asset as one used twice a month. Seal wear is a function of cycles, not calendar days.
• What is being pipetted. Organic solvents, concentrated acids and bases, and viscous or high-vapor-pressure reagents attack seals and O-rings far faster than aqueous buffer.
• Consequence of error. If a volumetric deviation reaches a patient result, a released batch, or a regulatory submission, the interval should be shorter than the one protecting an exploratory experiment.
• Number of users. Shared instruments accumulate handling damage and technique variation. A pipette used by twelve people in a teaching lab needs more attention than a personally assigned one.

Let as-found data set the interval

Here is the part most programs skip. Every calibration produces an as-found result — what the instrument was actually delivering before anyone touched it. That is a direct measurement of whether your interval is right.

If instruments consistently arrive in tolerance with margin to spare, your interval is probably conservative. If a meaningful fraction arrive out of tolerance, your interval is too long — and you have unquantified risk sitting in completed work.

Track the as-found pass rate by instrument class, by lab, by application. After two or three cycles you have real evidence. A program where 98% of instruments pass as-found at twelve months has earned that interval. A program where 15% fail at six months has a finding waiting to happen.

Shorten selectively, not globally

The instinct after a failure is to shorten the interval for everything. Resist it. It multiplies cost across your entire pool to address risk that is usually concentrated in a specific subset — a particular model, a particular bench, a particular reagent, or a particular technique problem.

Segment instead. Put the high-consequence, high-use, harsh-chemistry instruments on a shorter interval and leave the rest where the data says they belong. That is a defensible, proportionate program, and it usually costs less than a blanket change.

What to write down

Your SOP should state the interval, the rationale, the inputs you considered, and the mechanism by which as-found data can change it. One page. When an auditor asks why six months, you hand them the page and the trend data behind it, and the conversation is over.

If you want help building that trend from your existing certificates, send them to us. Reconciling an inherited asset register and turning it into an interval you can defend is a normal first step for our customers.