Best practices

From default to validated: the liquid class development workflow

A five-step routine for building a class: understand the liquid, start from a predefined class, inspect, change one parameter at a time, then verify with a correction curve.

Developing a liquid class can feel like turning knobs until the dripping stops. It goes faster, and the result lasts longer, if you follow a routine. The same five steps work across labs and instruments, because the underlying job is always the same: define a liquid, adapt a class to it, and prove it works.

Step 1: understand the liquid

Before you open any software, learn how the liquid behaves. Is it viscous, volatile, cohesive, prone to foaming? Pipette it by hand to get a feel for it, and pull its density and any hazards from the safety data sheet. Everything downstream is easier when you know what you are dealing with.

Step 2: start from a predefined class

Pick the predefined class whose liquid most resembles yours and use it as a baseline. Starting close means your changes are small. Save a copy under a name that says what it is for, so you never modify a default you cannot get back.

Step 3: run a test and look closely

Run a simple transfer that mimics one step of your real method, and watch it. A visual inspection catches most problems before any measurement: droplets on the tip after aspiration or dispense, bubbles, an aspiration or dispense height that is too high or too low, and whether the channels are tracking the liquid level correctly.

Step 4: change one parameter at a time

When something looks wrong, resist the urge to change five settings at once. Adjust one parameter, run it, and see what moved. It is slower for a single iteration and far faster overall, because you actually learn which knob does what for this liquid instead of stumbling onto a combination you cannot reproduce.

Step 5: verify and set the correction curve

Once it looks right, measure it. Weigh or read the delivered volumes across your working range, confirm precision is acceptable, then adjust the correction curve so the mean volume matches the target at every volume you care about. Only now is the class ready to hand to the people who will run it.

  1. Understand the liquid and its properties.
  2. Start from the closest predefined class and save a copy.
  3. Test and visually inspect.
  4. Change one parameter at a time until it looks consistent.
  5. Verify volumes and set the correction curve, then release.
Change one parameter at a time. It is the single habit that separates repeatable liquid class development from lucky guessing.
Piptera

Notes on pipetting calibration, liquid classes, and building an open, vendor-neutral catalog for every liquid handler.

© 2026 Piptera. Built for labs.