That depends on how long you find the disk usable and what the cause of the failure is.
For example, if you have some broken sectors, chances are that your drive is still quietly going on for a year or longer.But although there is a chance that it is going well, chances are that you are going to lose more and more data unnoticeably.And that applies to most types of failure that do not directly render the disk unusable.
That varies quite a bit.
On a small number of (SSD based external hard drives start to make an entrance fee but are still used at this moment) contain almost all external hard drives so-called “traditional hard drives” that means Concretely a steel box with one to three (2.5 inch disks) stacked on top of each other magnetically recordable metal discs, with read heads on it and, in several slices, also between them.
This picture shows it well: a hard drive is one of the last really mechanical parts that we encounter in PCs in addition to fans.
If you realise that the enclosure is in the picture (a notebook model):
- Only 2.5 inches, or 6.35 inches, is wide
- Contains up to 10 billion writable sectors (4TB, Sector size 512 bytes)
- The discs (platters) in question generally run with 5400 revolutions per minute, either about 65 km/h at the outer edge of the disc.
- The distance between the read head and the disc surface is Anno 2019 only a few nanometers.
To put that into perspective: that is lower than the height of the skin fat of a fingerprint, which means that the read head can therefore literally collide with a fingerprint on a platter.
The above makes it clear that hard drives are actually surprisingly little broken for such a delicate part with such high performance and tolerances.
Or, to turn it around, it’s not surprising that hard drives go smoothly. Unlike traditional hard drives, modern Ssds do not contain moving parts and are therefore much less common for that reason.
You mention “failing” in your question.It is important to know that a hard drive is actually “failing” in the first days of its life. A hard drive contains billions of writable sectors as mentioned above, but almost 10 percent of the usable disk surface is reserved for so-called “ECC data” and is therefore not free for us to write down data. ECC stands for “Error correction codes” and is a clever way to deal with broken sectors, whether they are already present if the disc is new or originated during use. To explain how it works is a bit complicated, but let’s see the ECC sectors as a “reserve of soldiers” that replenish losses with the “main power” when soldiers are killed or injured. Sometimes soldiers of the main power are also replaced preventively if they suspect they are going to fall out in the near future.
Anyway, a new hard drive is already broken and often contains millions of broken sectors and new sectors are broken every day, which will be replaced by “fresh” by ECC.This is normal and by ECC you will not notice it as an end user. One should not confuse this term with the concept of “bad sectors”: that is the term used to describe sectors on a hard disk which for some reason cannot be repaired with ECC. To stay with the soldier’s example: if there are too many soldiers at the same time in a small area, the reserves cannot be brought to the front quickly enough to avoid the total defeat.
“Bad sectors” are always a reason for replacing the hard drive: the operating system can no longer be on the disk: every time it tries to write data from or to the broken sectors, it delivers delays and malfunctions. So-called “Checkdisk” (ScanDisk), the operating system can detect these “bad sectors” and “repair” them.Only the sectors are not really repaired, but are simply considered as permanently unreadable, so the operating system will no longer attempt to approach these physically broken sectors in the future. If “bad sectors” are present on a hard drive, it should be seen as a sign that the hard disk is likely to give the whole mind in the short term, but that too is not a law of Medes and Persians: I have seen discs with bad sectors that s have been used intensively and years later have still not given the spirit definitively. More on that Later.
Back to real hard drives and “real failure”: except “bad sectors” There are still some other reason for the “real failure” of a hard drive: all the parts in the picture above can be broken.
The (slow) failure of the engine of the read heads, the (slow) failure of the reading heads themselves, the coming into contact with the heads of the surface of the platters, the (slow) failure of the controller, the (slow) failure of the engine that The Platters revolve around: it can all happen.
Symptoms are a ticking sound (motor read heads) lack of sound (Motorplatters does not do it anymore) “saking sound” (contact between read heads and disk surface, this is reasonably terminally), it suddenly does not “spin up the hard disk” when Booting the PC (Controller or engine read heads) or no longer recognizing the drive in the operating system (software error controller, hardware failure controller, various other reasons).
In all cases always applies: Directly turn off the PC and send the disk to a reliable so-called “data-recovery” Company as the data on the disk:
(A) is critical (value > circa 鈧?000,-)
B) The data is not backed up.
In other cases, depending on the symptoms and the value of the data, of course, some more risk can be taken, although contact between disk surface and read heads is always terminally.
And for all the above symptoms applies: the hard drive as a whole can give the mind or have already given it, but except for the terminal “contact between disk surface and read heads” scenario also applies here again: it can last for months and/ Or the disc can come back to life spontaneously.
To give some examples for external hard drives and also (finally) answer your question:
Ticking disc: The disc is no longer readable.
However, sometimes the symptom can be completely gone after restarting or turning the PC off and sometimes staying away for a long time. However, In most cases this is not the case and keeps tapping. Cooling or proper heating can help. Keeping upside down can help. A relatively soft touch with a ring or screwdriver can help. Only if the data on the disk is less important and/or there is a backup of the data and you just want to try “quick and dirty” to make the contents of a disc safe. Your actions can cause extra damage.
Bad sectors: Can take months to years before the disc finally breaks down, if that happens.However, sometimes the “bad sectors” are a result of another, more serious, underlying problem, such as a production defect in the magnetic coating of a platter, bad read heads or a short-lived contact between disk surface and Heads. In these cases, the gradient goes much faster.
No sound: (Engine platters not running) The disc is no longer readable.However, sometimes the symptom can be completely gone after restarting or turning the PC off and sometimes staying away for a long time. However, In most cases this is not the case and the engine does not start. Cooling or proper heating can help. Keeping upside down can help. A relatively soft touch with a ring or screwdriver can help. (The electric motor of the Platters then stands “between two coils”) Again: only if the data on the disc is less important and/or a backup copy of the data and you just want to try “quick and dirty” to make the contents of a disc safe. Your actions can cause extra damage.
Saating sound: Almost always a lost cause: even data-recovery companies can often not do much more with one or more platters with a thick circular scratch on the disk surface.The only hope is that the data is on a circular part of the platter (s) that is not compromised, but by building a filesystem on a hard drive that chance is small. Moreover, this is labor intensive work for a data recovery company, which often makes recovery very costly, if it is already feasible.
Faulty controller: Can be good for months if the controller sometimes does and sometimes doesn’t work.
(Temperature can affect. Often: the colder the better.) Often, however, the symptom is simply a “dead” hard drive that no longer makes sound and no longer vibrates and is no longer detected by the operating system. As well, this is often quite easy to solve, even for the “do-it-yourself”: Since the controller is almost invariably on the outside of the metal casing of the hard drive itself, it is often relatively easy to replace. (On the inside of a hard drive you don’t want to come, that can only be in a so-called “clean room” since even tiny dust particles in the metal casing of a hard drive are enough to allow the operation to stop/cause true damage after Re-assembling and running again.) Case of course is a controller of exactly the same type and even the same overhaul to mount.
Finally, we still have to deal with the “external hard drive enclosure” that is around the hard drive itself.
The controller chip that takes care of the USB to SATA (Sata is as a rule the interface of the hard drive itself) can go broke. Sometimes the chip gives the mind in one go, sometimes it takes months to years from the first symptoms to finally break completely. Also, this is relatively easy to solve by extracting the hard drive from the enclosure and using, for example, a hard disk docking station to read out.
Conclusion: The question “if an external hard drive fails, how long can I generally continue to use the drive before it completely breaks down?” has no simple answer.
It can be minutes, it can be years. The motto is almost always: replace, if it is not too late. Do not hesitate to put professionals at work, preferably at an early stage.
And beyond: Make backups.External hard drives (already no backup media, more suitable for data transport than for backup purposes) therefore do not use to MOVE files to, but only to COPY data to. Because of its nature, an external hard drive is more vulnerable than a hard drive built into a PC.
OFF-topic/Bonus: “Backup” Wisely meets the 3 -2 -1 rule:
- Have 3 copies of a file.
- 2 copies of the file stored locally but on separate media.
- 1 copy of the file is offsite saved.
(In a remote location)
The simplest way to accomplish this for SMB/privately is an Online backup that also writes to a NAS or external hard drive.
This can already be for a few euros per month, but unfortunately hardly anyone does it. If one is already back-upt one does that manually to a NAS, external hard drive or USB stick. Often one comes to the discovery that a backup for example is already months old in a crash, files (mail is a known one) have been forgotten or, worse, for example Ransomware and/or a voltage spike (lightning strikes and the like) not only the hard Disk IN the computer, but also the computer has wrecked it. After the backup unplugging helps, but as a rule it shows a backup to a local medium as an external hard drive or NAS very sensitive and in practice some labor intensive solution that also requires a lot of discipline, which is probably the cause that It still goes wrong so often. Dutch (we have traditionally always had some trouble with ICT related costs such as software licences and subscriptions to services such as virus scanners and online back-up products) seem to be a bit more sensitive here than people in our neighbouring countries .
If he starts to fail, it is important to stop using the drive but to buy a new one and transfer the data while still working.
Those old ones will save you another time unused in case the new problems get.
Hard drives support SMART.That helps you to find out early that the condition of the disc is deteriorating and replacing them before it severely fails. With an SSD you probably don’t get a warning beforehand.