I'm not aware of any published source for this time limit, nor ways to reduce it.

The docs do say, however, "If the volume has been impaired for more than 20 minutes, you can contact the AWS Support Center." [0] which suggests its some expected cleanup/remount interval.

That is, it is something that we regularly encounter when EC2 instances fail, so we were sharing from personal experience.

[0] https://docs.aws.amazon.com/ebs/latest/userguide/work_volume...