Assign values to 24-bit fields in GCC without running afoul of -Werror=conversion [closed]

When compiling code that assigns a value to a 24-bit field, GCC fails with an error if using -Werror=conversion. Other then turning off the warning explicitly, is there a way to cast the assignment to avoid the warning being generated?

using DayOffsetLocal = uint32_t; class TimerPeriod { private: DayOffsetLocal startTime: 24; }; startTime = (temp16*60U) & 0xFFFFFFU;

I tried using

startTime = static_cast<decltype(startTime)>((temp16*60U) & 0xFFFFFFU);

But that still generates

error: conversion from 'Timers::DayOffsetLocal' {aka 'long unsigned int'} to 'unsigned int:24' may change value [-Werror=conversion]