Contact

SECOND in Collabora Office Calc: Extract Rounded Seconds with Clear Precision Limits

Date & Time Intermediate Collabora Office Calc

Use Collabora SECOND with its documented nearest-second rule, whole-second controls and explicit holds for ties and minute-boundary carry.

Verification status: See the evidence and test scope described in this guide. How formula examples are checked

Quick answer: SECOND(number) returns a second component, documented as an integer from 0 through 59. Collabora 26.04 explicitly says it rounds to the nearest second rather than simply taking the integer part. Do not assume a truncation rule from another spreadsheet.

Availability and version history

  • Introduced: original Collabora release and launch date are not established by the sources reviewed.
  • Availability: documented in Collabora Office 26.04 help, scoped to that desktop-help release.
  • Platform scope: separately deployed Online and mobile versions are not inferred from desktop documentation.
  • Checked: October 14, 2026. Direct documentation and independent calendar/time controls; no native Collabora test was performed.

Syntax and rounding distinction

=SECOND(number)

The input is a numeric internal time value. Ordinary whole-second controls are straightforward; fractional seconds require attention to the product-specific rounding statement.

The source does not resolve exact half-second ties or the carry behavior when rounding a value near the end of a minute. This guide avoids certifying those native boundary outputs and uses fractional controls safely away from a half-second tie.

Example 1: reproduce the documented clock component

=SECOND(TIME(12;20;17))

Expected result: 17. This corresponds to the source’s 12:20:17 example. Display the output as a plain number, not as a full clock time.

The adjacent components are HOUR=12 and MINUTE=20. SECOND alone does not preserve those values or identify which date the original time belonged to.

Example 2: test the stated nearest-second rule

=SECOND(12.4/86400)
=SECOND(12.6/86400)

Under the documented nearest-second rule, independent expected results are 12 and 13. Division by 86400 converts a count of seconds into a spreadsheet day fraction.

The two controls deliberately avoid 12.5. They distinguish ordinary nearest rounding from truncation without assuming how the implementation breaks an exact tie. These are documented-model expectations, not newly captured native results.

Advanced example: add a fractional second to a clock value

=SECOND(TIME(12;20;17)+0.4/86400)
=SECOND(TIME(12;20;17)+0.6/86400)

Expected second components under the stated rule are 17 and 18. The hour and minute stay well away from a carry boundary in these examples.

Formatting the original timestamp with no fractional-second digits does not necessarily change its stored value. The function’s documented rounding is a calculation rule; a number format is a display rule.

Advanced example: distinguish component seconds from total seconds

For A2=TIME(0;2;17):

=SECOND(A2)
=A2*86400

Expected results are 17 and 137. The first returns a within-minute component; the second converts the entire clock fraction to seconds since midnight.

For a numeric duration spanning hours or days, multiplying the duration by 86400 retains those larger units in the total. SECOND deliberately cannot report an arbitrary total because its output is a bounded component.

Advanced check: reconstruct a whole-second clock value

For A2=TIME(12;20;17):

=TIME(HOUR(A2);MINUTE(A2);SECOND(A2))

Expected reconstructed time: 12:20:17. Its numeric day fraction is approximately 0.514085648. The total number of seconds since midnight is 44417.

If the original value includes 17.4 seconds, this component reconstruction loses the 0.4-second fraction. If it includes 17.6 seconds, the documented model reconstructs 18 seconds instead. Do not describe extraction and reconstruction as a universally lossless round trip.

Advanced example: keep subsecond differences in the original values

Consider two timestamp values that differ by 0.2 second while remaining in the same clock minute. Their rounded SECOND results can be identical, so subtracting those components can give zero despite a nonzero interval.

For the specific 12.1- and 12.3-second controls, both nearest-second components are 12, while the independent full-value difference is 0.2 second. Subtract original numeric timestamps and multiply by 86400 when that elapsed precision is required.

Unresolved boundaries and troubleshooting

Near 59.5 seconds, the nearest-rounding statement and the documented 0–59 component range require a carry convention that the short page does not explain. Exact half-second ties, rollover into a new minute and extreme floating-point behavior remain unverified. No guessed output of 60, 0 or an error is assigned here.

Check numeric input units first: a bare 12.6 is a count of days, not 12.6 seconds. Use TIME for whole clock components or divide an explicit seconds quantity by 86400. The official date/time model does not encode timezones, and no native cross-deployment equivalence is claimed.

References and verification

The dedicated product help was read directly. Calendar and numeric outputs are independently calculated expectations, with stated parsing and boundary limits. No native runtime or universal cross-deployment certification is claimed.