Seagate HDD periodically disables/enables

Solution 1:

These symptoms are unfortunately an extremely good indication of USB Seagates that are failing, but not always.

The 500 GB - 1 TB full size models from at least 6 years ago were worse than the recent 2-3 TB drives as well. You will unlikely be able to find the cause, but you should determine if the drive is really dying or there's another reason.

• Make sure that, if there's anything critical on that drive, make a copy on different media, (especially until you know that drive is not failing!)
• Move the drive to another PC, and see if it "cycles connections" there as well.
• On the same PC, move the USB port to a different one.
• Use a different USB cable.
• Use a different power supply.
• Make sure that, if your drive has an external brick power supply, that it is plugged into the same extension cord as your PC. I've seen 5-10v AC differences in grounds, the commons, or the hot 120v from one outlet to the next. You can even see this occasionally with the same wall outlet.

See what special software Seagate has to help diagnose what's really going on.
USB drives purposefully hide errors from the OS, just as most modern non-USB drives do, and could also hide bad sectors from detection. (In the 1980's drives would show sector errors, and the OS had to map these errors out of use. This is why you won't find many image backup technology products from that era, since restoring to a different drive with bad sectors in a different spot would often corrupt the drive, even if you couldn't detect it.)

Esoteric solutions to try, if desperate
If you are trying to retrieve some limited information from the drive, and it won't stay connected long enough, here are some tricks that have worked for me in the past with non-USB drives, but only with certain types of internal drive technologies:

• Try to keep the drive upside down, or in some other strange position. IF the bearings in the drive are failing, I've seen this work with certain types of drives. (NOTE: QUICKLY moving the orientation of ANY drive while spinning, is EXTREMELY stressful to those bearings. [Think--ANY "gyro" resists this.])
• Cool the drive and try it before it warms up. I've pulled them out of the freezer, and that changes the components enough to make a difference. (You'd have to find a way to prevent any condensation however, perhaps by bagging it up with a desiccant before you cooled it, and then during use.)

Background on why you can't necessarily tell where the errors are on any drive, including a USB drive:
in the late 1980's, drive technology changed from ST-506 which showed errors and depended on the OS's remapping errors. As drives were commonly designed to be "logically defect free", like ESDI, SCSI, ATA, and all modern drives today, image backups became fast, easy and reliable. But because the OS could no longer "see" the errors, special software was developed to "talk to the drive's OS" built into the drive itself. Today, that interface is somewhat standardized as S.M.A.R.T.

But, for competitive and marketing reasons, manufacturers like Seagate don't always use, or even accurately report what's really going on. Seagate also modified the actual drive itself inside many USB models, so it would no longer work as the ATA or SATA that the drive really was, directly connected. This was done to prevent the practice of tossing the USB part of the USB drives, enabling the use of these bare drives from the less expensive USB's sold, and were cheaper than the desktop standalone drives themselves!

Anecdotal problems and solutions:
I've had two USB Seagates that failed with precisely these symptoms. I couldn't even keep them on long enough to check them for errors. I haven't cared enough to try some of my esoteric solutions, other than removing the drive from the USB portion and connect it to the MB, (which didn't work at all).

THIS did work recently!
I've had a 3 TB (full size 3.5" with separate power supply), that was disconnecting and reconnecting about every five minutes. I "solved it" accidentally, by moving the drive to a different USB port! I have a powerful laptop with 3 USB ports, and I was using the other USB "power only" for an Alexa Dot, and the other USB for a mouse. Most of the time without powering the Dot, the drive will never disconnect.

Regardless of the lack of bad sectors the drive is reporting to you, it may still have those sector errors, or other errors you can't see.

The BEST BET today for NEVER losing data is to:

• ASSUME the drive WILL DIE, and duplicate the data in MULTIPLE locations. The great news is that today, "buy and replace" beats coaxing old stuff to work again, or figure out why they are, or might be failing.
• BUY another drive often -- every year or two. Just because you paid $200 a few years ago, it isn't a huge expense since that could probably be replaced for $25 today. (Obsolescence - First year: $200 x 50% = $100, second: $100 x 50%, third: $50 x 50% = $25! Plus, it's now likely somewhat worn out.)
• Duplicate and TEST your backups. -- HOWEVER, you STILL have to MOVE and TEST those duplicates!! I remember one huge 15 page review article about a $50,000 software system that worked perfectly AS DESIGNED, but they still lost 99.999% of their data! (The expert reviewers mistakenly backed up ONLY the ROOT DIRECTORY, by missing the ONE CHECKMARK, to "include ALL subfolders"!! LOL!!)
• Plan on which data needs to be duplicated and heavily protected.
• Never assume that your "backup media" will always be able to restore anything. I've got quite a few backups that would require reassembling 1980's PCs, OS's, and that specific drive type to read the data. Plus, after years and normal heat, magnetic backups slowly lose data all by themselves!
• NEVER ASSUME that CLOUD DRIVES are infallible!! They won't likely "die" for the same reasons, but they certainly they will "disappear" if you don't pay attention. You can also irretrievably lose the keys to access the data.