Cactus hdd
 

Cactus in AU speak means broken.

This one, a Maxtor Basics External Desktop 1TB USB drive took a dive, only 18" off a low tabletop to a timber carpetted floor but it was enough to see it dead and gone.

Electronically it seems to be fine. The motor driver beeps cyclically trying to sppol the motor up.

I guess my luck ran out because I have previously had a main drive fall 6ft onto pavement with the computer it was in and fly out when it hit the ground.

On an initial visit to see if the swingarm had jumped out I found no such lucky break.

The rotor has locked up. Likely the hub is an interference fit on the motor shaft or the internal hub of the motor is an interference fit on the shaft and something shifted during the hard landing.

If anyone has any recommendations, these will be appreciated. The archive is not mission critical but there are a few things I would like to retrieve off the drive if I can.

"Cactus"? Australians have interesting slangs. "Sacked" means "fired". "Buggared" means "messed up." "Holiday" means "vacation." I worked with an Australian for several years and always found the slight difference in language interesting.

The seek head of a hard drive sits very close to, but never touches, the platter. The gap between the platter and the head is so fine that it was compared to having a 747 flying a few inches above the surface of the water.

A cushion of air separates the head from the platter. At high enough altitudes, the thin air could make the drive unreliable due to head crashes.

So if the platter's hung up, that's it as far as an economic repair is concerned. And if you've opened up the case, be aware that even tiny bits of dust can wreck the workings and make it unusable. Therefore, I believe it's totally buggared? :-)

Dean.


Thanks for your response.

I managed to dismantle the guts out of the drive and hopefully the disk stack has remained half aligned at least as I immobilised them as best I could. It is all beautiful precise engineering in there, "up-edded" by a "PLAIN" bearing, not the ball set I expected to find.

The word "ed" is foulsafe workshop strine for fuck-"ed". This one may have become a bit dated by now. It was around in the seventies. Foulsafeing is the practice of gentrifying or disarming colourful language.

The bearing appears it is not an oilite bearing, the traditional sintered bronze type but looks like aluminium with possibly a chrome shaft going through it.

My guess is the thing is minimally lubed to avoid oil creep and contamination of the platters and maybe relies on a super polished finish on the shaft to keep friction down. However as we all know, aluminium "picks up" when things get a bit warm and loaded, comes off the parent metal, sticks to the passing surface then tears back into itself on the next turn.

It was a real problem with sleeve valve radial engines and was resolved by tinplating the wearing surface. GM Australia also did it with the Holden motor pistons. Maybe this has been done in the HDD bearing, maybe not. Whatever, I won't be finding out. The thing has seized solid, a perfect welded joint.

The platters, although heavy should not load the bearing laterally, only as an endthrust load when they are oriented horizontal. If the drive is vertically oriented it would be another matter but remain a constant load. I would expect the bearing to wear sooner of heat up more in this position.

Things however change if the drive is moved and gyroscopic precession comes into play. At 7,200rpm, the loading and surface friction on the shaft would increase dramatically and so would the surface temp from added friction. If it is already hot and close to "pick up" occurring, maybe enough to push it high enough to melt spots in the bearing surface and weld to the shaft.

My theory is that the bearing fails under a massive precession load as the drive is turned by its fall to ground and not by actual impact.

I cannot pull the shaft out of the bearing so I have no way of observing what retains the rotor on the shaft. It may be that a retaining ring or very thin shoulder surface has forced into a bearing surface thus jamming it. I would expect though that this could be released by pressing or driving the shaft back through the bearing. This one does not move at all.

Images on the web show typical "pick up" failure.

My pet theory is that these drives will be far less tolerant than older drives of any movement. This thing did not drop far and it landed on underfelted carpet. I have had drives take a much harder one for the team without failing. This one did not fall hard enough to cause a head crash. The platter surfaces are perfect, now contaminated with dust of course.

I am a bit vexed. There is no point to creating a portable 1TB drive for archiving if it is not going to last the distance. By the nature of its tasking it is going to take a thump or two.

I don't have a lot of confidence in this design of plain bearing. It is certainly not going to endure in a continuous duty environment.

There are comments on one forum where on testing, the operator discovered lesser performance when the drive was oriented vertically. This might be attributed to loading of the bearing laterally with the full weight of the platters and a reduction in rpm therefore readback speed.

I guess these things are "consumer" devices and that you get what you pay for.

The drive is a Maxtor Basics External Desktop Hard Drive. The case is not the MyBook style with dead flat surfaces but stylised with curved surfaces, just begging to spin around and work its way off a shelf or tabletop when cords hang tight and heavy.

The actual HDD drive insde the enclosure is a Seagate Barracuda 7200.11. ST31000333AS.

There is some adverse comment about this series of drives on various forums and it seems that Seagate has a problem on its hands.

I have most of my stuff safetied on three drives, one the worker within the computer which gets cleared, the two archive drives, both 1TB Maxtor Basics. I am now becoming a bit nervous.

Whilst the platters and heads are in apparent perfect condition, I don't think I am going to be able to re-build this thing.

So, to quote Australian workshop slang, the drive looks like it is "cactus dictus", which more or less means irretrievably ruined.

Bob...

I've been using IBM/Hitachi SATA drives in a Firmtek housing for the past several years and haven't had a single failure yet. Of course none of the drives have been dropped, either.

For portable storage I'm using GTech Firewire drives. The mechanisms are the smaller 3.5" drives. Less mass. Less to torque if dropped.