A good experiment is designed to survive an unexciting result. Define the main outcome before collecting data: for example, energy delivered to a specified resistor over a complete daily light cycle, using a fixed illuminated footprint.
Build comparisons into the experiment
Compare the inoculated cell with an otherwise matched cell containing medium and electrodes but no living culture. Compare illuminated and dark operation while recording temperature. Inspect electrode changes and record water, nutrients and maintenance. These are proposed controls; their practical implementation must suit the organism and apparatus.
Keep the raw record
Log timestamps, terminal voltage, current, load, illumination, temperature and interventions. Identify the electrode batch, cell geometry and exposed area. Preserve failed runs. Independent cells estimate device-to-device variation; repeated readings from one cell do not provide the same evidence.
Make uncertainty useful
State instrument range, calibration basis, resolution and uncertainty at the actual signal level. Check offsets and leakage with a known test circuit before interpreting a biological signal. State the numerical rule for subtracting a control and report the raw results as well.
NIST's policy connects traceability to a documented calibration chain with uncertainty contributions, and says traceability alone does not guarantee suitability for a measurement. That distinction is particularly relevant when an apparent improvement is similar in size to measurement uncertainty.
Access and claim limit: NIST policy text was inspected. The protocol here is an original proposed MaxwellX workflow, not a published standard, completed experiment, independent replication or certification.