Seagate Data Recovery

Tierra’s ongoing research and development program revealed the range of hard drive recovery problems encountered during Seagate data recovery. Hard drive recovery is needed when a hard disk fails and the data files and folders become either partially or totally inaccessible. Seagate Technology is one of the world’s largest hard drive manufacturers, with 52,000 staff and annual revenue in excess of US$11 Billion. In 2006 they bought out Maxtor, which at the time was itself the world‘s third largest maker of hard disks.

The data for this article is based on the case notes from the data recovery work carried out on Seagate Barracuda and laptop hard drives received in our labs over the last 6 months.

First some information showing which models we have seen in the labs. Approximately 80% of Seagate drives were among the 24 hard drive models listed in the table below:

The remaining 20% were composed of the following models:

So, what are the Problems:

Bad Sectors

By far the most frequently encountered problem in Seagate data recovery, and this is common to all manufacturers of hard drives is what is referred to as “bad sectors”.

Over time and with heavy usage the ability of the platters (the spinning disks within the hard drive that hold the data) to maintain the electromagnetic signal that is the user data wanes (just as a permanent magnet over time will slowly lose its magnetism).  It is however also true to say that we routinely see drives of all types which have been used lightly, are still under warranty and yet are suffering from bad sectors.

This sort of platter damage is the cause of unreadable (or “bad”) sectors (every hard drive is divided up into a certain number of sectors, for example a 500GB hard drive will typically have 976,773,163 sectors available to store your data). A sector is essentially the smallest readable area on the data platter, it is usually 512 bytes in size. All hard drives have some bad sectors and will develop more over their lifetime, they have a built-in store of spare sectors which they will allocate as bad sectors develop. However this pool of spare sectors is not infinite and once exhausted the total number of usable sectors on the drive will start to decline. This can be more of a problem than may initially appear, a single unreadable sector if located in a suitable part of the drive can prevent your operating system from starting at all. More typically users will initially notice that the access time to the hard drive is slowing down, often accompanied by an increase in operating noise from the hard drive.

If a drive with bad sectors is received early enough (in other words before the bad sectors become widely spread across the drive) then it is almost always possible to recover data. The amount of data recoverable is entirely determined by how many bad sectors have developed.  The greatest enemy of a drive in this condition is the Windows CHKDSK program. This will often run automatically on boot-up unless prevented and will scatter and over-write the user data in an effort to get back to a condition where the Operating System is working again, but it will happily overwrite user data in the process.

The other great enemy to Seagate data recovery of a drive in this condition is recovery software. This will thrash the Seagate hard disk in an effort to locate the data, all the time adding more and more bad sectors.

The only sensible approach to a drive with bad sectors is to clone all of the accessible parts of the drive in an environment that will not change the contents of the drive. Once the clone has been obtained then recovery software can be run without making the situation any worse. There are various methods and equipment used to clone drives with bad sectors. It is also often the case that a drive with problems other than this will also turn out to have bad sectors when those other initial problems have been solved.

Ratio of Drives With Bad Sectors Found In Seagate Data Recovery

From the Seagate drives that have passed through our labs in the period covered by this article approximately 35% of those received were suffering from bad sectors.

Firmware Issues:

The next most common failing for our Seagate data recovery sample was firmware problems.

Firmware is data stored internally within the hard drive in order to get itself started. It cannot be accessed without specialised hardware. Any professional data recovery service will have access to and proficiency in the use of such equipment. It will include (among many other pieces of information):

• Drive Model
• Serial Number
• The location of the storage areas on the data platters that will be utilised to store user data
• Over the drive’s lifetime as individual sectors become unreadable the drive will reallocate the space lost with new space from a limited bank of spare sectors, the firmware keeps a map of these re-allocations.

To complicate matters this firmware is stored in part on the hard drive’s printed circuit board (usually either on a discrete EPROM chip or else embedded within the PCB processor chip) and in part on the data platters. At initial power-up of the hard drive the contents of the PCB firmware store are read, this portion of the firmware will then direct how the drive starts up, this includes where it will look on the data platters for the remaining part of the firmware.

The symptoms exhibited by a drive with damaged firmware will vary widely depending upon the exact nature of the firmware damage. The drive may spin-up, click repeatedly and not appear in the BIOS or may seem to be detected by the PC or Mac but not allow access to user data. One of the more common symptoms of damaged firmware (especially for Seagate drives) is the drive spinning-up and sounding normal (that is to say no repetitive clicking from the heads) but simply not appearing in BIOS or appearing but shown as having zero capacity.

Generally speaking, if firmware becomes corrupted the drive will fail to initialise. Seagate drives, just like those of other manufacturers suffer both from such corruption and sometimes from known firmware “bugs” which just like bugs in other software programs crop-up occasionally.

Most of the time, with the appropriate knowledge, experience and technology it is possible to re-build or work around these faults.

Of the Seagate drives received in the last 6 months, in the region of 15% were affected by one firmware problem or another.

Failed or Weak Read/Write Heads:

Just as the magnetic platter loses its ability to store an electromagnetic signal over time and with heavy use it is also the case that the ability of the heads to read that data can similarly diminish. In Seagate data recovery, such problems can arise through time as the drive ages or can be the result of a physical event such as the drive suffering an impact.

The symptoms are typically a heavy clicking heard and felt after the drive platters have spun-up. In this condition the drive will usually not be identified in BIOS and is very unlikely to get as far as loading an Operating System or file listing. It is important to bear in mind however that many other problems other than failed heads can also cause a drive to click at start-up.

After eliminating other possible causes of these sounds the usual procedure is to carry out an internal inspection of the drive to establish whether there has been any damage to the data platters themselves, where this is not the case then a closely matching donor drive would be acquired from which to transplant the read/write heads.  This work must of course be carried out in a clean-area environment.

In instances where the heads have damaged the data platter it is usually the case that the damaged area of the platter will wreck the donor heads, it is therefore seldom worthwhile to attempt a head-swap with damaged platters. This is because the heads float on an air cushion the tiniest fraction of a millimetre above the data platters which are spinning at typically 5400 or 7200 RPM, as you can imagine, the smallest disruption to the platter surface will impact and severely damage the head.

Our work in Seagate data recovery showed a little over 12% of the cases were found to be due to failed or weak read/write heads.

Seized Platter Motor:

The data platters inside the hard drive are spun by the platter motor. If the platter motor becomes seized then the platters will not spin-up. This is usually the result of the drive having being subjected to an impact of some description. It can be surprising how little force in such an impact is needed to seize the platter motor. In Seagate data recovery, we have seen more than one instance of an external hard drive merely falling onto its side being sufficient.

The symptoms (as you might imagine) are that on application of power the drive does not spin-up. There is usually some indication that the drive is attempting to spin-up, usually a slight vibration every second or so.

There are other causes of these symptoms, for example where the read/write heads have become  stuck to the data area of the platters, the friction alone can be sufficient to prevent the platters from spinning-up.  Accurate diagnosis, as well as access to a clean area environment are essential in both cases.

Where the cause has been found to be a seized platter motor it is necessary to carry out a platter swap. This is a procedure whereby the data platters, printed circuit board and the read/write head assembly are moved to a closely matching donor hard drive chassis in order to utilise the donor drive’s platter motor (which is built into the chassis) in an attempt to access the user data.

Approximately 10% of our Seagate drives were suffering from seized platter motors.

Printed Circuit Board (PCB) Problems:

The 2 most common problems associated with Seagate data recovery PCB failures are:

1. Over-voltage supply to the drive, either through a bad connection or a spike from the PC /Mac power supply.
2. The failure of the chip on the PCB that controls the platter motor.

In the case of over-voltage supply the path to recovery depends of course upon how much damage has been done. In some instances it is possible temporarily to repair the PCB at least for long enough to recover the user data. In other cases it may be necessary to replace the PCB (which is almost never a straight swap, it almost always involves transferring firmware data unique to the individual drive from the old PCB to the new). In many instances the damage may have gone even further, it is common for the so-called pre-amplifier to suffer damage in the event of an over-voltage supply, this is a part of the read/write head assembly inside the drive and will also require replacing if it has been damaged.

In the case of platter motor controller chip failure, for Seagate data recovery, it is usual procedure to swap the PCB with that from closely matching hard, but as stated above the unique firmware from the defective PCB will almost certainly need to be transferred in order to gain access to the drive data.

Around 10 % of the Seagate drives in the sample were suffering from PCB faults.

The remaining 18% of the Seagate sample of drives were essentially intact but had suffered so-called logical damage. That is to say that the hard drive itself was fully operational but the data had been lost due to accidental re-format, deletion or file system corruption.

In this situation the drive will usually boot into the operating system as usual but, of course, the required data cannot be accessed.

Seagate Momentus 5400 Hard Drives With Firmware 7.01 or 3.CAE

These are a rather unfortunate special case. The drives with these specific firmware versions (shown on the drive label as  FW:7.01 or FW:3.CAE) are particularly prone to problems with the magnetic substrate that makes up the storage platters. The surface of the platters appears to degrade much more than other models. For this reason Seagate data recovery rates for these drives are very poor compared to other drives.

A word on Seagate data recovery rates:

You will note that we do not give Seagate data recovery rates in this article for the sample of Seagate drives under analysis. The reason for this is that recovery rates are essentially meaningless. They are rather like survival rate league tables for hospitals, at first sight they seem to offer a precise and valuable indication of expected outcomes but with closer examination and a little thought are at best misleading.  With any competent, professional data recovery company the outcome is essentially determined by the condition of the patient on admission.

As a simple example – if a hard drive is suffering from bad sectors and a customer requires a particular critical document, then the outcome will be 100% successful if the bad sectors do not occur in the region of the platters where that crucial document is stored, but may well be 0% if the bad sectors have affected that particular area.

To give a recovery rate across all drives received simply averages away any meaningful information.

For more information on Seagate data recovery, fill in our contact form or call us on 0845 094 0027.

No related posts.