Hi , this is how it is supposed to work - the "full" backup is incremental. It makes full versions of files that have changed since last full backup.
As in,
day1 - Full version of the file A
day2 - Block version of the file A
day3 - new Full version of the file A
day4 - no Block version because the file has not changed
day5 - no Full version because the file has not changed compared to the last Full

and so on.

A good write-up here, please check it: https://www.cloudberrylab.com/blog/backup-types-explained-incremental-vs-differential-vs-full-vs-synthetic/

Just to clarify, all file backups are incremental forever; meaning, after the initial Full backup, each subsequent execution only backs up the files that changed. If you use Block-Level, it means CloudBerry will back up only the changes within files (great for larger files that have small changes compared to the overall size). When you schedule Full backups (with Block-Level), you are only telling CloudBerry to perform a full file backup on any files previously backed up with a block-level backup. That way, CloudBerry can properly maintain retention.

If you change targets in the backup plan, you should have CloudBerry run a consistency check on the repository (Register Accounts - Edit - Consistency Check) so it can update the local database with what's in the new backup storage (nothing, presumably). Or you can clone the old plan and edit the target (and then delete the original if it's no longer needed). A repository consistency check can take some time depending on the quantity of files, so it might just be easier to create the new plan as detailed above using the Clone option.

Thanks both for the information and confirmation. @David, I went with your solution and created a new backup plan.

Cheers!