Replacing a Disk that is "About to Fail"

I've replaced disks several times, but mostly by sort of staggering through the procedure. I usually power off my machine, remove the failed disk, replace it with a new one, power on, and then wander through SST screens until I remember how to re-configure the new drive.

I'd like to do this correctly for once. I have a model 270 with all the disks in the main enclosure and I do *not* have hot swap. I have six disks in a RAID5 configuration, and the system is complaining that one of them (DD003 in position D02) has an "Impending DASD failure". Is there a manual which specifically tells me the steps to replace this disk drive?

What I'd LIKE to do is tell the machine that a drive is bad, then power down, then replace the drive, then power up, then rebuild. My issues with the process usually center on the fact that after I've installed the new drive, the system doesn't know about it until I somehow prod it a little bit. I can't "Include unit in device parity protection" because the system still thinks the old disk is in there, and so I get the message "no diskunits are eligible". Typically I've found I had to actually do something to the disk such as "Analyze Disk Unit Surface" to get it to recognize the new disk.

Anyway, any help would be appreciated, even if it's just RTFM.

Joe