Hi. Nice script.

1. For a 24h calculation cycle the server update time alternates between 23:00 and 00:00. In the latter case that could potentially cause the task scheduled at 00:00 to be missed as the trigger event had already passed. By adding isToday check to isTsAfterChkT should prevent that.
2. All the clients are sending update requests to server during the first minute of an hour. Adding some random minutes to isChkT would reduce the server load in case the script turns out to be very popular :)
3. The next heating calculation time printed in the log was not correct (the current hour was reported instead). The proposed patch contains a modified nextChkHr() function that corrects the issue.