Harddrives - nice pictures

[Noopy]

Hi all,

I did some high resolution pictures of two old WD Caviar harddrives (22100 and 22500).



Of course with some die Pictures too. 



If you are interested go to:
http://www.richis-lab.de/HDD_WD_Caviar_22100.htm
http://www.richis-lab.de/HDD_WD_Caviar_22500.htm

Text is german... 

Have fun!

Greetings,

Richard

[simxdx]

 ，Very clear pictures！

[Miyuki]

Nice pictures 

Wonder how much tinier is reading head now in Terabyte sizes and how weak signal it produces compared to back then

[Noopy]

Thanks! 

I have some newer models lying around here.
Coming soon... 

I don´t know if you can compare the signals of an old device to a newer one. The technology changed a lot as far as I´m informed...

[pwlps]

Wonder how much tinier is reading head now in Terabyte sizes and how weak signal it produces compared to back then

The signal is not so much smaller, in fact it was rather the progress in the reading head technology (AMR,GMR,TMR...) giving more signal strength that allowed increasing areal density.  For example the development by IBM of new heads based on the GMR effect in the 90's allowed to break the technological limitations of the previous AMR generation  (still used in WD Caviar models, WD acquired the GMR technology from IBM circa 1998).

[Miyuki]

Wonder how much tinier is reading head now in Terabyte sizes and how weak signal it produces compared to back then

The signal is not so much smaller, in fact it was rather the progress in the reading head technology (AMR,GMR,TMR...) giving more signal strength that allowed increasing areal density.  For example the development by IBM of new heads based on the GMR effect in the 90's allowed to break the technological limitations of the previous AMR generation  (still used in WD Caviar models, WD acquired the GMR technology from IBM circa 1998).

I know most of the capacity progress is trough narrower more dense packed tracks as speed and linear density goes up no more than 10 times from that 90s drives but total density skyrocketed and have exponential growth
But still they must now processing signal in GHz ranges from head as they get about 200MB/s data rate

[dom0]

Interesting.



This kinda reminded me of Hebrew, though it doesn't seem to correspond directly to the Hebrew alphabet. And IBM did do both semiconductor and disk drive research in Haifa.

If you flip it (attached) it looks very much like a stylized kind of Hebrew.

[Noopy]

My first opinion was also hebrew.

Interesting idea to flip the symbols. 

[Noopy]

Hi all!

I have posted some new pictures. This time a hard disk with a GMR-Head.   



Whole story here:
http://www.richis-lab.de/HDD_Samsung_SV4003H.htm

Have fun!


...perhaps you are interested in a floppy drive too:
http://www.richis-lab.de/floppy.htm
Eventually very similar just bigger... 

[Noopy]

News on my homepage: A Western Digital Red 3TB

https://www.richis-lab.de/HDD_WD_RED_3TB.htm





Piezos in the pick up and heaters in the head. What a crazy shit.


A interesting thing:

Till now I had two WD Red 3TB with bad SMART values.

The boards in both hard disks seem to have a corrosion problem:



In my opinion they use galvanic silver. Cheap and good with BGA but problematic regarding corrosion.
Very bad when you want ot use spring contacts for very fast signals:



I would have chosen gold! 


Have fun! 

[Noopy]

And again a new HDD!
See the details of a Seagate ST-177I:

https://www.richis-lab.de/HDD_Seagate_ST-177I.htm


Everything a little bit bigger:




Of course with die Pictures:



 


Have fun! 

[Noopy]

Till now I had two WD Red 3TB with bad SMART values.

The boards in both hard disks seem to have a corrosion problem:



In my opinion they use galvanic silver. Cheap and good with BGA but problematic regarding corrosion.
Very bad when you want ot use spring contacts for very fast signals:



I would have chosen gold! 

Do you remember the WD Red silver tarnish thing?

After a lot of time and a query WD gave me the following answer:

"Thank you for waiting for our reply. Our engineers have studied your reports and inform us thus.
We would like to apologize for any trouble this has caused you. After review the surface of the PCB shows silver tarnish which is not a functional or reliability issue. This is expected as the surface finish is silver. ..."

With a little help of Google you find a lot of information regarding the rising resistance of silver tarnish.
I can´t imaging that shouldn´t be an issue. 

We will see whether I will get a better answer the next days... ...weeks...

[Prehistoricman]

Do you remember the WD Red silver tarnish thing?

After a lot of time and a query WD gave me the following answer:

"Thank you for waiting for our reply. Our engineers have studied your reports and inform us thus.
We would like to apologize for any trouble this has caused you. After review the surface of the PCB shows silver tarnish which is not a functional or reliability issue. This is expected as the surface finish is silver. ..."

With a little help of Google you find a lot of information regarding the rising resistance of silver tarnish.
I can´t imaging that shouldn´t be an issue. 

We will see whether I will get a better answer the next days... ...weeks...

I've seen quite a few HDD PCBs with this look. I think one had issues before I gave the contacts a good scrubbing.

Amazing pics. I love the look of the platter on that old Seagate.

[station240]

I have that silver finish on 4 WD green hard drives, after one completely died, I had to clean all the other drives PCBs.

Failure mode is this:
1. Controller gets bad/weak signals from the heads, assumes it's just a bad sector.
2. Controller moves the data to another part of the drive, and marks the original sector as bad.
3. Error rates get worse, controller moves more files, marks more sectors as bad.
4. After days/weeks of this, whenever the computer is on the drive has activity.
5. Actual disk surface has damage from constant moving of files.
6. Drive has more bad sector than it can mark.
7. Attempt to reformat drive as last resort after saving data.
8. Reports over 10,000 unrecoverable bad sectors, marks them as bad then crashes.
9. Reboot shows drive as totally unavailable, to the BIOS!
10. Totally bricked.

[Noopy]

I've seen quite a few HDD PCBs with this look. I think one had issues before I gave the contacts a good scrubbing.

"Unfortunately" I had to disassemble the HDD before I tried to clean the contacts.  I have four of them left all around five years old. If they ever give bad SMART-values I´ll try some cleaning first.

Amazing pics. I love the look of the platter on that old Seagate.

Thanks! Good feedback is always a reason to do more pictures. 

I have that silver finish on 4 WD green hard drives, after one completely died, I had to clean all the other drives PCBs.

Failure mode is this:
1. Controller gets bad/weak signals from the heads, assumes it's just a bad sector.
2. Controller moves the data to another part of the drive, and marks the original sector as bad.
3. Error rates get worse, controller moves more files, marks more sectors as bad.
4. After days/weeks of this, whenever the computer is on the drive has activity.
5. Actual disk surface has damage from constant moving of files.
6. Drive has more bad sector than it can mark.
7. Attempt to reformat drive as last resort after saving data.
8. Reports over 10,000 unrecoverable bad sectors, marks them as bad then crashes.
9. Reboot shows drive as totally unavailable, to the BIOS!
10. Totally bricked.

Sounds absolutely reasonable.

I´m curious wheter I´ll get a better answer from WD than the last one.
But ok, what should they write? "Hey, we screwed up. Sorry!" 
They design such technical wonderful things and than they mess with the PCB-finish... 

[TheEPROM9]

You are giving me ideas for photos withy my macro lens. Like most of you, I have more HDD's than you can shake a stick at.

[TheEPROM9]

Hi all,

I did some high resolution pictures of two old WD Caviar harddrives (22100 and 22500).



Of course with some die Pictures too. 



What camera kit did you use & also how did you de pot the chips?

If you are interested go to:
http://www.richis-lab.de/HDD_WD_Caviar_22100.htm
http://www.richis-lab.de/HDD_WD_Caviar_22500.htm

Text is german... 

Have fun!

Greetings,

Richard

[TiN]

I have few WD Black 1TB (FAEX) disks that have silver tarnish on PCB pads as well. Used to take controller PCB out, clean pads with erases, put it back together and until next few month it worked fine. After doing this for few times, ditched the drives and bought Toshiba HDDs and Intel SSDs. 

[Wimberleytech]

Very nice work!!  Have you reversed any old Quantum drives?

What are you using to decap the plastic packages?  HF?

[Noopy]

Thanks!

A quantum drive would be interesting too! Till now I had none but Ebay will help... 


The decapping was done with a torch. With a little practise that gives 70-80% good dies depending on the type of chip.
The pics were taken with a "normal" dslr with a reversed 10mm objective and "balgen"-rings.

I have posted the how-to:

https://www.eevblog.com/forum/projects/decapping-and-chip-documentation-howto/msg2666622/
 

[magic]

Not even sure if HF would work at all, hot nitric or sulfuric is the usual choice.

Even if it works, it would certainly eat the glass passivation layer afterwards and possibly the aluminium underneath(?)

[Noopy]

The pics were taken with a "normal" dslr with a reversed 10mm objective and "balgen"-rings.

The magnefication and the quality of the pictures is not as good as possible with a >10k€-Zeiss-Monster but I can easily tilt the objects.
The focus depth is very small but often I can take very nice pictures this way:


https://www.richis-lab.de/IC_02.htm

[magic]

Nice dust spots in the upper left

[legacy]

what do use for taking pics like these? 

[Yansi]

The pics were taken with a "normal" dslr with a reversed 10mm objective and "balgen"-rings.