Click of Death (HDD discussion)

[stormbind]

Hello

The click of death is a terrible scenario in which a HDD is broken by the read-head being unable to find a starting point on the disk. The click occurs when the head hits a barrier at the the disk's outermost limit, after which the head continues looking from the innermost part of the disk.

Some fixes can be found.

1. One is to freeze the HDD to shrink the rotating parts that may have become stuck, and allow them to turn again. However, this only helps with very old drives and furthermore, it only works for a few minutes before needing another freeze.

2. The more common fix is to turn off the HDD give it a lucky whack. This might free a head that has become abnormally stuck or trapped out of place, allowing it to operate normally once again.

3. Opening the HDD to repair the head at home is not viable because ordinary dust in the air will render the platters useless. However, some specialist laboratories can do this for $$$$.

So my question is: Is it possible to make an amateur safe dust free enclosure, and how would one go about doing that?

I was thinking of making a vacuum space in a glass tank - but then how could I get my hands to the HDD without them exploding?

 

[wabatt]

Maybe you could use drill a hole --> and seal it with a glove -->

 

[PeteAtoms]

Is overheating not an issue?

 

[stormbind]

Maybe you could use drill a hole --> and seal it with a glove -->

Yeah, but there is a problem with plastic and rubber.

A HDD is vacuum sealed - I interpret this as meaning they were sealed in a vacuum environment rather than hosting a vacuum because there have been HDDs with filtered air holes. Either way, there can be no contaminant inside. For example, super-fine airborne and invisible dust touching the platter would destroy the data. Creating a vacuum from already filtered air is a possible solution and plastic and rubber can do that.

However, those thin rubber globes are going to swell like balloons when you suck the air out. A plastic box might also deform. A similar pressure would apply to the hands and skin.

I'm not suggesting that human hands would really blow up because it wouldn't be an absolute vacuum. However, I do think there would be a health and safety concern. It would also be difficult to work on the HDD with obstructing bloated hands

 

[stormbind]

Is overheating not an issue?

No, this is HDD repair rather than daily use

I had backups of backups and the last backup is dead. Western Digital should be shot off this planet. The data is very important research and not replaceable. However, there is no spare cash to pay for a professional recovery.

 

[Genocidicbunny]

Short answer: get professional recovery if the data is that valuable.

Long answer: you wont be able to get anything resembling a clean room in a home environment. Bear in mind, that the bits on the platters are nanoscale, which means even a single stray particle of dust can ruin a good chunk of the platter.

You would not need a vacuum to accomplish any recovery, just a very pure air supply. This simplifies your parts list (enclosure, regular gloves vs pressure gloves, air filter) but at the same time, increases the degree of complexity manyfold. First, you need a perfectly clean enclosure, your gloves must be perfectly clean (not coated with *anything*) and you need a sophisticated air filter. The last one is the main sticking point -- you would need to shell out thousands to get an air filter that can do the kind of job you need it to do.

Here we reach the same conclusion: you just wont be able to do what you need to in a home environment.


If the data is that important, you will need to shell out the money. There's the price tag on the data. Whether or not that price is too high is all up to you.

For future reference, for really important data I suggest you have a RAID 1 array with double backup, one on-site for easy access, and one less regular backup offsite, preferably in a safety deposit box.

 

[stormbind]

Its the data from my expedition on the aboriginal people (photos etc.). I was only a student so there wasn't much money and nobody with any money cares about an entire civilisation facing imminent extinction

I also thought that actually learning to fix a nuked drive (I would try an unimportant drive first) would quickly turn into an asset. So lets say there is a sterile glass tank.

Before allowing air in, I speculate that the air needs to be cleaned in a reservoir-tank and then allowed to enter the operation-tank. How can we theoretically remove nano-sized particles from air? Can we heat or freeze them to make them stick together?

Do you think that making a website and asking for donations would work? I'm not interested in wasting time, oddly

 

[Zelig]

For future reference, for really important data I suggest you have a RAID 1 array with double backup, one on-site for easy access, and one less regular backup offsite, preferably in a safety deposit box.

I'm somewhat skeptical of the utility of RAID 1. It can have some utility if you know what you're doing, but cheap (ie, almost anything onboard a motherboard) RAID controllers are junk, and RAID 1 is not a form of backup, it's simply a way of managing (a) risk.

 

[Cutlass]

Yeah, my motherboard allegedly supports RAID. Except that after running for a month both HDDs scrambled

 

[Genocidicbunny]

I'm somewhat skeptical of the utility of RAID 1. It can have some utility if you know what you're doing, but cheap (ie, almost anything onboard a motherboard) RAID controllers are junk, and RAID 1 is not a form of backup, it's simply a way of managing (a) risk.

I never said it was. But if you have the option, its a decent first line of defense. I did say that there should also be a double backup. The RAID 1 is mainly for convenience, in case one of the drives goes, so you dont have to immediately retreive the backup (and thus possibly compromise it)

Its the data from my expedition on the aboriginal people (photos etc.). I was only a student so there wasn't much money and nobody with any money cares about an entire civilisation facing imminent extinction

I also thought that actually learning to fix a nuked drive (I would try an unimportant drive first) would quickly turn into an asset. So lets say there is a sterile glass tank.

Before allowing air in, I speculate that the air needs to be cleaned in a reservoir-tank and then allowed to enter the operation-tank. How can we theoretically remove nano-sized particles from air? Can we heat or freeze them to make them stick together?

Do you think that making a website and asking for donations would work? I'm not interested in wasting time, oddly

You need a filter. Heating or freezing the air supply will do little to clean it. You need to actually pass it through a filtration system that only allows atmospheric gases and nothing else through.

Yeah, my motherboard allegedly supports RAID. Except that after running for a month both HDDs scrambled

Probably a crappy JMicron controller. If you wanna do raid, you would want to look into something besides JMicron.

 

[stormbind]

I hypothesise that water could be an effective 'membrane' to remove most particles, using one body of water with two surfaces (a partition splits the surface). Gasses are absorbed into the water on one side and released into an enclosed low-pressure environment on the other.

However, water vapour would need to be controlled without introducing dust. That might not be too hard because the molecules are quite big. The worst part is that it might take a very long time for sufficient gasses to be absorbed and then released!!

Yeah, that was a dumb idea. How do air scrubbers work?

 

[PeteAtoms]

What about a damp cloth?
What about charcoal filters?
What about a gas mask?
An ionic breeze or air purifier setup?

Actually, would ionized air coming from an ionic breeze have a negative, positive or neutral affect on electronics?

 

[stormbind]

I thought about charcoal but its dusty.

I'll investigate the gas masks, thanks - but to give an idea of how tiny the particles are - if you can smell anything then the air isn't clean enough.

You made a very good point. Maybe that's why the HDDs are vacuum sealed to start with?

 

[Genocidicbunny]

I hypothesise that water could be an effective 'membrane' to remove most particles, using one body of water with two surfaces (a partition splits the surface). Gasses are absorbed into the water on one side and released into an enclosed low-pressure environment on the other.

However, water vapour would need to be controlled without introducing dust. That might not be too hard because the molecules are quite big. The worst part is that it might take a very long time for sufficient gasses to be absorbed and then released!!

Yeah, that was a dumb idea. How do air scrubbers work?

Passing the air supply through a detergent (such as water) is one way of scrubbing the air, but it does not guarantee a clean air supply as there will still be evaporation. Most use detergents targeted for specific particulate.

What about a damp cloth?
What about charcoal filters?
What about a gas mask?
An ionic breeze or air purifier setup?

Actually, would ionized air coming from an ionic breeze have a negative, positive or neutral affect on electronics?

Damp cloth is too porous.
Charcoal filters are good for liquids, but not as good for gases. They will certainly trap some of the dust but not nearly enough.
Gas mask is a possible one. You need a really good gas mask though.
Your regular retail air purifier cannot do nearly a good enough of a job.

I thought about charcoal but its dusty.

I'll investigate the gas masks, thanks - but to give an idea of how tiny the particles are - if you can smell anything then the air isn't clean enough.

You made a very good point. Maybe that's why the HDDs are vacuum sealed to start with?

What do you mean by vacuum sealed? HDD's have air filters and they allow atmosphere inside. As for during manufacture, it is very likely done in a clean room, not in a vacuum. Imagine the logistics of having a highly technical assembly line function in a vacuum. You would need to overhaul your hydraulics, your robots would need to be functional in a vacuum. Then you'd have your whole chamber which would need to withstand atmospheric pressure. In other words, its not really useful.

 

[uppi]

Imagine the logistics of having a highly technical assembly line function in a vacuum. You would need to overhaul your hydraulics, your robots would need to be functional in a vacuum. Then you'd have your whole chamber which would need to withstand atmospheric pressure. In other words, its not really useful.

All of that isn't a real problem. The pain is the maintenance. Every time you have to fix something inside you need to pressurize the chamber and then you need several weeks of pumping until you have a good vacuum again.


Trying to fix the HDD by yourself is probably a bad idea. As long as the head is only stuck, the data is still recoverable. But if you do something wrong during the repair and you get a head crash, then there will be unrecoverable data loss.

If you really want to do it, the best way to get clean air would probably be to buy purified nitrogen. This should be both cheaper and better than trying to purify air yourself. Then you should operate whatever your substitute for a clean room is at overpressure. With overpressure, the air is always flowing out and has no chance to bring dirt inside. A vacuum is a bad idea, as any leak will result in air with all its dirt streaming in.

The clean room substitute and all tools you need to use will have to be thoroughly cleaned beforehand. When we're working on a vacuum (near the XHV range) we bake out all our tools by putting then in an 400°C oven for days. If you just want a clean room, you're probably not that sensitive to outgassing, so bakeout might be overkill. All your tools that are needed in proximity to the actual storage material must not be magnetic in any way (don't use iron!), because otherwise you could destroy the magnetically stored information.

For yourself you would need lab-grade surgical masks, head cover and gloves. A full clean room suit would be even better.

Disclaimer: I have no experience in repairing hard drives and have no idea how clean everything needs to be. The above may not be enough.

 

[Genocidicbunny]

All of that isn't a real problem. The pain is the maintenance. Every time you have to fix something inside you need to pressurize the chamber and then you need several weeks of pumping until you have a good vacuum again.

Thats what logistics is.

Disclaimer: I have no experience in repairing hard drives and have no idea how clean everything needs to be. The above may not be enough.

You need ISO 5 as a bare minimum, ISO 4 is better, ISO 3 even more so.

 

[stormbind]

oh, I like the idea of buying canned gasses. Why is nitrogen preferred to others?

I don't know about the discussion on vacuums. I just read that they were 'vacuum sealed' and then went about interpreting that in the most achievable sense.

 

[uppi]

oh, I like the idea of buying canned gasses. Why is nitrogen preferred to others?

- cheapest gas to purify
- non-toxic
- no disposal required, you can just let it escape into the air (don't do this in a closed room, though. If you release a lot of nitrogen, there could be a lack of oxygen which you won't notice until you pass out)
- pretty much chemically inert at room temperature, it neither combusts like hydrogen nor feeds a fire like oxygen.

A purified air-like mixture of nitrogen and oxygen would also work, but I don't know whether anybody sells that.

 

[GoodGame]

I'd go with an aluminum case with its own fans built in, so unless you work near a machine shop that doesn't mind you swiping aluminum, I'd just buy one.

 

[Samus_au]

Clicking can also be the controller board gone bad on the drive. If you can find another drive manufactured around the same time as the bad one you can sometimes switch these out and have things return to normal.

As for a the vacuum stuff, I've opened hard drives up before (outside a safe room) put them back together and had them work long enough to get data back off. Not something I'd advise doing to a drive that has important data on it, but it works.

They are also very cool to watch in motion.