Well, that was silly

Just to confirm, does the drive have a SATA connector on it, with a PCB that plugs in to give it a USB port? Or does it just have a USB port integrated into the drive PCB?

When connecting directly, are you going via SATA or USB?

If the drive has a SATA connector I would try that, since it'll eliminate the USB-SATA converter board.

If the drive has a SATA power connector I'd plug that into the PC's PSU, which will eliminate the external power supply and connector.

Theo

It was always my understanding is that was dust that would cause the problems. The heads are only a few microns above the platter and so any dust down to this size can get stuck between the head and platter and damage the platter as it spins past the head at 5400 to 7200 rpm.

A quick google produces

You can pay data recovery compnay to suck data off it for you .... not cheap. I had a drive fail (like yours plain dead) and as it was under warranty WD did a full recovery onto a new HD for me for free.

A long time ago, a professional recovery company used to say that for things that were not worth the cost of professional recovery, spraying a mist of water in a room, to drop dust out of the air, would often give enough time to open a hard disk casing, free up a stuck head or platter and recover at least some of the data.

Naaah. Youtube has many videos of HDDs opened, heads unstuck and moved by hand, and then the data read off the HDD -- this in the 500 GB range.

So they are not *that* sensitive -- not that I would trust a once-opened drive!

Thomas Prufer

I once took the lid off an 80MB HDD, and it kept working for about two weeks afterwards. Back then they weren't so fussy.

Theo

If the HD had been powered down the heads should be parked. I would also expect there to be a last gasp style of activity where power is removed to park the heads.

I would be tempted to see if this is a actual HDD problem or something broken in the external enclosure. Say by mounting the HDD mechanism itself in your computer or on a USB SATA adaptor.

(is it a 2.5 or 3.5" external drive)?

I think you will be *very* lucky if the drive still works. A lot of drives use glass platters coated in oxide, and a really big mechanical shock will shatter the platters.

I did this with the 4 TB USB drive that I use for backing up all our TV recordings, camcorder films etc. That drive gets a lot of wear and tear being moved every day or so between the PC where it is being used for the backup and the garage where it is kept separate from the computer. One time the drive slipped out of my hand and then the USB cord which I still had hold of unplugged itself from the drive. The drive only fell about a foot onto a carpeted wooden floor, but I heard the unmistakable tinkle of glass fragments when I picked it up.

I didn't lose anything - except £80 for a new drive just the same, and the (unattended) time for the backup program to make a new copy of about

I remember when I started working in 1986 the HDDs in the server had big warnings printed on which said something like "this drive will be irreparably damaged if it is dropped from more than 2 centimetres onto a table or other hard surface".

Things have improved since then, but for some level of shock, spindles will come off bearings or (glass) platters will shatter - or heads will strike the spinning platters and scrape the oxide off in a ring.

Funny story: I heard of a company which destroyed HDDs with sensitive information on, equivalent to shredding paper documents. Most companies trusted the data-shredding company, but the MOD (apparently) insisted on being handed back a bag with the oxide on it, which conjures up a mental image of someone nefarious trying to glue back the oxide dust particles into the correct locations on the disk.

Early HDDs had open platters. When they were in use they were kept in a "spin drier" unit which was sealed and pumped with clean, filtered air. But the platter-and-spindle unit could be removed and replaced with another one, to change between discs. There was a plastic cover which was screwed into the "spin driver", locked onto the spindle which was then unlocked from the motor, and a base was quickly fitted.

There were single platters (2.5MB) enclosed in plastic shells, which slid into a rack-mount drive whereupon the shell opened and the platter slid between the heads.

I took apart a glass platter drive recently (I wanted to dispose of it without the effort of finding a spare interface too hook it up to for wiping). It took a number of blows to shatter the platters. They were surprisingly tough. Damage is much more likely to be to the PCB, the heads or to the magnetic surface on the glass.

Yes, we had some of those, too. There were four washing-machine-sized drive units, each of which could hold one of the disks you describe, for a total of

Even once we had it, along with online editing and job submission (this was the early-mid 70s), there were still hold-outs. One chap had his entire program on punched cards - he'd wander around pushing what looked like a large tea-trolley, only it had 12 trays of cards on it (for a total of 24,000 cards) rather than a teapot. Luckily we had a reasonably fast card-reader, so it only took 20 minutes to read his job into he computer.

Every lab with a DEC PDP 11/34 in the 1970's had one of these. The DEC drives had one fixed 2.5 Mb drive and one removable platter.

Thanks for all the new answers - too many to reply to individually, I'm afraid. Here's an update:

I'm delghted to report that the drive is working again. Actually, despite what I thought and reported, the drive was probably working all the time: the connection inside the caddy seems to have been the cause of the problem. While I'm investigating this I've temporarily mounted the bare drive in an old desktop machine using the SATA connections and it's behaving perfectly. I'm backing up the crucial files as I'm typing this.

The caddy clearly isn't designed ever to come apart after its initial assembly so it might be a case of simply replacing it, but I'd prefer to fix it if I can

I'm grateful to everyone for the sympathy, help, and encouragement.

Thanks for that. I just posted an update saying that this is exactly what I tried and that happily the drive is (and probably always has been) working perfectly: the problem must be with the caddy.

Likely. You have nothing to lose opening it to see it a connector has come off or something is obviously physically broken and can be repaired, but since the led comes on, the chances aren't great

Yes, you have nothing to lose but some time.

What are you like at soldering ? It can be fixable if you can.