Separate device calibration from a daily chemical dose
Calibrating an aquarium dosing pump checks or configures how the device relates its operation to delivered liquid. It does not determine how much supplement the aquarium needs. Begin with the current manual for the exact model, because controls, test sequences and accepted calibration inputs differ. NumUtility is a calculation worksheet, not a pump controller, and cannot perform the device's calibration procedure for you.
Keep three quantities separate: actual liquid collected during a test, the volume commanded for that same test if there was one, and an independently chosen amount used for a later runtime illustration. The third quantity is not a substitute for the test command. Mixing these fields can produce a large apparent error even when the pump was never instructed to deliver that amount during the observation.
Follow the actual setup and collection procedure
For its documented device, Reef Factory describes priming the tubing without air, collecting outside the sump and measuring a sixty-second calibration run. It also describes calibrating individual heads separately. That is an example of manufacturer procedure, not a universal command sequence for all pumps. Use your own model's manual rather than copying another brand's timing or interface controls.
Prepare a suitable collection vessel and record the liquid, tubing arrangement, head and measurement method. Keep test liquid out of the aquarium as directed by the actual procedure. If the instructions call for entering the collected amount into device software, do so in that device's own units. A website result labeled mL/minute is not automatically compatible with a field expecting seconds, steps or another calibration parameter.
Calculate flow from a timed collection
Measured flow in mL/second is collected volume divided by elapsed seconds. Multiply by 60 for mL/minute. If a test collects 50 mL in 60 seconds, flow is 50/60 = approximately 0.833333 mL/second, or 50 mL/minute. These are equivalent descriptions of one observed collection, not two independent calibration tests.
Write both the numerator and denominator with units. Entering a duration of one into a seconds field for a one-minute test would overstate flow sixtyfold. Similarly, grams from a scale cannot be entered as mL without a relevant liquid density. A supplement is not guaranteed to have a density of 1 g/mL. Follow the manufacturer's supported measurement method and account for density where mass is used.
Estimate runtime for a separately chosen volume
For chosen volume V and measured rate q in mL/second, projected runtime is V/q. At the example rate of 50 mL/minute, an independently selected 10 mL delivery takes 10/(50/60) = 12 seconds. A 5 mL illustration gives six seconds. These calculations do not recommend either chemical amount or assert that the device supports arbitrarily short runs with equal accuracy.
Starting and stopping effects, supported settings and the actual setup can make a short delivery differ from a long collection's average rate. Confirm delivered volume under the real device procedure rather than treating the projected time as a validated command. NumUtility does not schedule the pump, know the supplement concentration or enforce an aquarium's safe chemical-change limits.
Compare commanded and collected volumes from the same test
If the original test actually commanded 40 mL and collected 50 mL, signed difference is 50 − 40 = +10 mL. Signed percentage error is (collected − commanded)/commanded × 100, giving +25%. A different test that commanded 40 mL and collected 36 mL gives −4 mL and −10%. Positive means observed over-delivery relative to that command; negative means under-delivery.
A timed-only calibration run may have no commanded volume. In that case, leave the comparison field blank rather than enter the desired daily amount or assume a zero-percent error. The timed rate can still be calculated. Also do not use an expected volume from another head, an earlier configuration or a different elapsed interval as the denominator for the current observation.
Do not turn signed error into a universal correction setting
A +25% observed error does not mean every manufacturer's calibration field should be decreased by 25%. Some controls may represent delivered volume, others rate or a different internal parameter. The relationship and adjustment direction belong to the device documentation. NumUtility intentionally does not produce a firmware correction factor or claim to know the setting that should be entered.
After an adjustment performed according to the actual manual, collect and measure a verification delivery. Repeating the same original calculation with unchanged inputs does not test whether the device changed. Keep the before-and-after records with their settings and collection method. A single matching result is useful evidence about that test, but not certification of long-term repeatability or fail-safe operation.
Respect measurement resolution and operating changes
Suppose a 50 mL observation could plausibly have been read as 48 or 52 mL because of measurement limitations. Those alternatives imply different measured rates even when the timer is exact. This is an illustration of resolution sensitivity, not a formal confidence interval. Displaying six decimal places for the resulting rate cannot recover information absent from the collection measurement.
Changes in liquid, tubing, trapped air or pump condition can make a prior record less representative. Use the manufacturer's checks and recalibration guidance rather than assuming one universal interval. The current calculator does not save history, average trials or compare against a model-specific tolerance. Record each head separately and independently verify unexpected results rather than merging unlike observations into one misleading average.
Use the worksheet as an arithmetic cross-check
Enter actual collected mL, elapsed seconds and the independently chosen volume for a runtime scenario. Enter commanded test mL only when that command belongs to the same collection. The outputs show measured rate, projected runtime and optional signed comparison. Zero or invalid required quantities are rejected; missing command data remain unavailable rather than masquerading as a perfectly calibrated test.
This is educational arithmetic, not animal-care approval, supplement advice or device programming. It does not choose a daily chemical amount, verify compatibility, connect to equipment or provide alarms. Use the exact manual for configuration and an appropriate aquarium-care process for chemical decisions. The website's contribution is a transparent calculation that can be checked against the actual collection, not a replacement for either responsibility.