Read HDDs with clicking noise

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View


I am working with PCs for several years and had in the past often broken
disks from my own and out of the stock of acquaintances in my hands.
Most frequently, I had the firm WD HDDs in my hands, their detection by
the BIOS when you start the PC has already proved unsuccessful. Any
general addition to the rhythmic clicking of the write-, read head was
the startup of the PC usually hang at this point.

Since I still have two copies of these disks, I have sought during the
last year for intense sources of error and their correction on the
Internet. Most frequently, I found any information on the lost "track
0", taking the hard drive tries to find the track 0 to read the hard
drive's content. Would it be possible to read a hard drive via the track
10, 20, 50 or 100, because then it would be so easy to make the attempt
to read from a track and create a higher ordinal number generating an
image to another hard disk for the remaining data to save. Similarly,
would probably also the backward reading (ie, from the track with the
highest ordinal number) be possible and it could be so far as to hasting
to the readable part of the data on top of the HDD, and thus produce an

Does anybody know more about this?

Thank you very much


Re: Read HDDs with clicking noise

detlev.e wrote:
Quoted text here. Click to load it
Your best bet is to buy Sprinrite from It will exercse
the drives and if it's possible will restore the bad sectors.
Spinrite costs less that $100 and is well worth it if it
even as few as 2 disks.

Re: Read HDDs with clicking noise

detlev.e wrote:
Quoted text here. Click to load it

The problem is the nature of hard drive controller design and protocol.

There are two potential ways hard drives could have been designed.

1) Controller PCB always responds to requests, no matter how sick
    the mechanical parts of the hard drive are. This is the way
    I would design it. By doing so, it would be possible to
    "talk" to the drive with a diagnostic, download new firmware,
    or similar things.

2) Instead, the way they've chosen to design the drive, is to
    prevent the drive from responding to a probe from the host
    computer, if the drive has any problem initially, reading
    data from the platter. This makes it impossible to debug
    what the problem might be.

The hard drive has data stored at "track -1", a reserved area
just before user data. That track contains identity information,
potentially spare sector information or the like. Lots of
critical data for drive operation. If the controller cannot
read that track, it won't listen to any queries from the PC.
There isn't much you can do to a drive, if it won't listen
to you.

If the controller is able to fetch the critical data from
the reserved area, then, it might be possible to use
a utility like Spinrite. But you have to get some response
from the drive, if you hope to do anything with it.

Sometimes, a drive stops working, because an internal data
structure that the firmware uses, is full. This is a software
problem, but can stop the drive from operating, just
as surely as a mechanical problem would. So not all
drive failures are purely mechanical. Some are caused
by bad firmware design.


Re: Read HDDs with clicking noise

Thank you for your suggestions and your explanations

I wonder there is another track -1 because I thought there is only track
0 which prevents drives from reading.
I was very astonished to notice that several WD drives quit their work
and since that time I only found them producing the clicking noise.
I know the pcb inside the vacuum chamber might have any defect and that
was what I wanted to find out.
I remember there is a chance to connect to the pcb by microcontroller
equipment and I intended to build one for me to be able to check HDDs
perhaps also to read data out of it.
Therefore I asked for a way to read data from the last sector to the
first or counting down reading all the data.

It seems to me you are a very experienced guy in these hardware questions.
Do you know anything else about data structure on the disks and also
whether it would be possible to find a way conneting to a HDD without
using the manufacturer's pcb or is the manufacturer's pcb the only choice?

Paul schrieb:
Quoted text here. Click to load it

Re: Read HDDs with clicking noise

detlev.e wrote:
Quoted text here. Click to load it

There is no vacuum inside the HDA. The HDA is at atmospheric pressure.
The "breather hole" on the HDA, is there to help equalize the pressure.

The airflow pattern is shown here. The airflow helps, by sweeping debris into
the filter pack.

The silver sticker here, hides the servo writer access port, which is
used at the factory to write the tracks. If the HDA was under vacuum,
it would suck that sticker right inside the HDA :-) The latest
high density drives, no longer use an external servo writer, and
the mechanical tolerances are now too fine to allow the usage
of external servo writers. But you'll still see the silver sticker
on your older hard drives.


This article mentions a few things about the hard drive design.
They call the "track -1" the System Area or SA.

I've seen an interface on the controller board, but don't have
any details as to whether it is a defacto industry standard
or not.

There is at least one web forum, for hard drive repair, but
I didn't keep a bookmark for it. The people in the forum
are secretive, and you won't learn a lot from reading their
posts. I suppose that is what happens, when you're making
good money from data recovery.

Some web sites will tease you a little bit. Yes, they're
telling you a few things, but they're not going into
any detail.

There is some gadget with "3000" in the part number, that
I've seen references to. I don't see anything here though,
that looks like it contacts the pads on the PCB. There may
be a few tools like this around, for solving firmware/software
issues or obtaining raw access.

I'm not in the data recovery business. Many years ago, I worked
on a controller for disks, and my first hand experience is too
old to be of much use. The drives back then, had 8" or larger
platters, to give you some idea of how long ago that would be.


Site Timeline