Trouble identifying an IC: how to reverse engineer it?

[IntrepidN00b]

Hi

I'm looking at a Hitachi HA13525B on a PCB on an old harddrive.

It is the motor driver controller for the harddrive motor and the harddrive is from around year 2000.

I've found a datasheet that is kind of similar but it is for the HA13501S which seems to have a similar application and it has the right package.

All in all it is a hard chip to find a datasheet for. It may be that it is some kind of bespoke chip.

The package is: MP-26DT

As I'm an intrepid noob  , and I'm doing this to learn things for the fun of learning  .
 
Is there a way to poke at the chip and see if one can figure out what the function of the pins are?
Like I wrote, it is still on the PCB and not all the legs are connected from what I can see.

Thanks for any pointers!

/Chris

[IntrepidN00b]

Here are some photos of the chip in question
EDIT: Sory for the larg file size, I'll try to make it smaller.

[PA0PBZ]

HA13525A, who knows:

[Psi]

What exactly is the drive brand/model?

Maybe one of us has one laying around.

[amyk]

HA13520F HDD, Spindel(1A) & Voice Coil(0.4A) Motor Drv,5V
HA13525FP HDD, Spindel(1.5A) & Voice Coil(1A) Motor Drv,12V

You could try asking Hitachi, but since their website is a mess...

[free_electron]

Fujitsu drive most likely. That chip is a spindle / voice coil driver. you can clearly see the sense resistors for the currents. botom left of the chip is the sense resistor for the spindle current. just above the chip is the sense resistor for the voice coil. the interface is 3 wire digital clk/frame/dio
But without the register map there is no beginning to talk to this thing.

These chips are custom built for their application and datasheets are inexistent or only available under NDA. The schematic above doesnt look right. the pin labeled R2 should be labeled Voicecoil 1 and the pin labeled voicecoil1 schould be labeled Sense
Pin 17 is not phase... it is DMUX . when the drive is up and running this pin will toggle whenever an electrical cycle begins on the rotation. during spinup this pin will 'flutter' seemingly random to the uninitiated but it is actually a communication pin (DMUX = Digital Multiplexer ) that brings out internal control signals for the spindle startup. the processor switches this around depending on what it needs to close the acceleration loop. during startup there is not enough backk-emf to o position sensing so the cpu needs access to a couple of control signals in the state machine that controls the motor.

c4 c5 and r1 are the dampening filter for the vcm control loop

c1 c3 is a boostpump for the top mosfets. they use 4 nmos transistors

pins 12 13 14 and 15 are CLK , FRAME and DIO . DIO is a bidierctional data line. CLK is clokc and FRAME tells you what transaction is going on.
the pin labeled CHARGEPUMP is most likely the system clock . This is the clock that drives the entire mechanics package. this clock can be 'drifted' to sync up the data operations on the drive to the rotation of the platter. it is delivered by the system controller.
the serial clock is totally different and operations there happen out of sync fromt he mechanics clock.

There is much more but i cannot tell you. NDA.

-edit- seems like this is a much older device that has very little intelligence on board.
s1 s2 and s3 operate the bridge . the only thing i am missing is the control psignal for the VCM. there has to be an analog signal somewhere telling the vcm what to do.

is striked through the things that don;t apply to this old clunker. They are valid for modern drives though, like anything made in the last 10 years or so... That hitachi thing is at 10 years old, probably 15 ...

[tesla500]

If you just need to get the drive working, buy another drive and swap the boards. If you want to use the chip for something else, don't. Use something that's available and doesn't require an NDA to obtain he datasheet.

[w2aew]

Hey free electron, sounds like you've worked in the HDD industry.  I did too, in a past life, doing silicon hardware validation of Read Channel IP...