FW for Hitachi hua722020ala330 - 2TB Hdd needed

[woxy]

I assume that you have not attempted to spin up the 2TB HDDs with the IBM PCB, so the 2TB motors would still be suspect.

Well, as for now I'm not 100% sure, but if thing I've tried that too. And if I did, the only result could be "silent" or something like bzzz bzzz.

That said, are you using compatible boards for your testing? One scenario that would fit the symptoms would be if your boards had different pinouts for the motor connector. For example, one could be wired as A/B/C/Common, whereas the other could be Common/B/C/A. The last thing you would want is for your motor to spin backwards.

I had this thought, about different pinouts. But how can it be:

1. 2TB PCB spins up IBM - for few seconds
2. Old Hitachi PCB spins IBM also - for as long as powered

so why IBM PCB could not spin up at least an OLD Hitachi? Why It doesn't work vice versa?
Or maybe the OLD Hitachi PCB spinned up IBM in reverse (and for IBM motor this is OK, but reverse spinning for Hitachi is locked by motor or something) ?

I've also looked with multimeter on motor pinouts, and as i figured it out, there are probably common/a/b/c pinout in all IBM/Hitachi hdd, but needs to be confirmed once again.

[fzabkar]

Could it be that the IBM motor requires less current than the 2TB and old Hitachi motors? If so, then maybe the current limit of the IBM PCB is too low to spin up the old Hitachi motor.

Otherwise, have you tried push-starting the 2TB motor just after applying power to the PCB?

[woxy]

Yes. I've tried to push start 2TB Hitachi on original PCB. It runs for a while, as long as push start is more effective.
Without push, it looks like the PCB is not generating phase change as fast as it is required for motor start.
Because motor goes forward one step, then goes back to previous position. It happens the same, as I was just applying
5V DC directly to another disassembled hdd motor. When 5V applied, the motor stepped forward.

So, either the motor controller chip on 2TB gives less current than it is necessary for 2TB Hitachi and to avoid
overcurrent it stops after few tries (which can be true, because 2TB Hitachi PCB spinned up IBM motor just for a while,
when old Hitachi spinned IBM as long as powered),
or motor driver is faulty and does not generate phase change correctly, and thats why it doesn't start,
or some of motor chip pinout have been faulty/overcurrented and this causes problem.

[fzabkar]

... because 2TB Hitachi PCB spinned up IBM motor just for a while,
when old Hitachi spinned IBM as long as powered ...

I suspect that this may be normal behaviour.

The 2TB MCU may be spinning down the motor after it fails to access the System Area, whereas the firmware in the old Hitachi appears not to care.

[BlackST]

Hi all,
I'm looking for NVRAM for those drives. I've got 3 of them and all suffers from same, trying to start but only makes only bzz bzz for few times. Then motor driver IC is getting hot. I'm assuming corrupted firmware.

1) This is SalvationData support forum
2) Even if you get the NVRAM content from another one, it likely won't work or will work partially (if you have SD tools or any other tool and/or you know how Hitachi / IBM work, you know why)
3) The problem is elsewhere.

I've tryied like with Seagate but got nothing on telnet.

No wonder.

[woxy]

1) This is SalvationData support forum

I'm aware ot that, please move my topic into Hard Disk Drives: data recovery and repair on English forum.

2) Even if you get the NVRAM content from another one, it likely won't work or will work partially (if you have SD tools or any other tool and/or you know how Hitachi / IBM work, you know why)

I know that, this is why I'm looking for best matching NVRAM content.

3) The problem is elsewhere.

And maybe You know where it could be?

[woxy]

I've disassembled whole hdd.
Weird thing is that with original 2TB PCB and also with IBM PCB, hdd motor spinned up
only when all platers of hdd was removed. When platers putted again, none of PCB
spinned up motor, looked like motor didn't have strength to run with platers ballast.

Conclusion, somehow 2TB motor is faulty, even though multimeter didn't show anything wrong
with windings, and motor runs easily by hand (so no stucked bearings), because
none of PCB's spinned up motor with platers attached. And as the motor can't be easily replaced,
last 2 Hitachi's will be sold as donors.

Not a DIY problem unfortunately.

[fzabkar]

At the risk of opening an old wound, here is one of my first posts to this forum:
missing-vcm-current-sense-resistor-servo-calibration-t13400.html

It was greeted with the kind of derision that was customary for that time, but I now feel that you may be able to achieve something by at least experimenting with the motor current sense resistor(s), if you can find it/them. For example, if you add a resistor in parallel with the existing one, then you will effectively be increasing the current drive to the motor.

Hopefully my tutorial will show you what to look for:
http://www.users.on.net/~fzabkar/HDD/Tu ... 0411N.html

[BlackST]

I've disassembled whole hdd.
Weird thing is that with original 2TB PCB and also with IBM PCB, hdd motor spinned up
only when all platers of hdd was removed. When platers putted again, none of PCB
spinned up motor, looked like motor didn't have strength to run with platers ballast.

Conclusion, somehow 2TB motor is faulty, even though multimeter didn't show anything wrong
with windings, and motor runs easily by hand (so no stucked bearings), because
none of PCB's spinned up motor with platers attached. And as the motor can't be easily replaced,
last 2 Hitachi's will be sold as donors.

Not a DIY problem unfortunately.

Hint : Why replace the motor ? ?

[woxy]

At the risk of opening an old wound, here is one of my first posts to this forum:
missing-vcm-current-sense-resistor-servo-calibration-t13400.html

It was greeted with the kind of derision that was customary for that time, but I now feel that you may be able to achieve something by at least experimenting with the motor current sense resistor(s), if you can find it/them. For example, if you add a resistor in parallel with the existing one, then you will effectively be increasing the current drive to the motor.

Hopefully my tutorial will show you what to look for:
http://www.users.on.net/~fzabkar/HDD/Tu ... 0411N.html

Will try to search an re medium for 2 last Hitachi's. Not sure if I'll come with any success.
Great thanks, for Yours support, will write if/or if not any luck

Hint : Why replace the motor ? ?

I've sit, thought about it and here are conclusions:

1. If 2TB Hitachi PCB's are all able to run IBM motor, and MCU is cool, so PCB's are fine (at least as for IBM motor).

2. If after disassembled whole 2TB Hitachi, IBM PCB can spin up Hitachi's motor in same direction,
then motor windings of Hitachi's looks to be fine, but Hitachi's motors needs more power to run.

3. 2TB Hitachi's original PCB, can run original Hitachi's motor only without any platers mounted on motor,
and this spinup looks like this: bz bz bz spiiiiiiiiiiinn, then in one sec the spin noise is getting higher, but motor stops spinning during the noise pitch up, and after that high noise, the motor stops.
So in my opinion, this can be caused by:
- Hitachi drive failure some way, and PCB senses that, so it stops
- PCB sees the the overcurrent and stops
- on this short spin try, the voltage on all motor pins was about 6.12V

4. IBM original PCB, can spin Hitachi's motor too, but also only without any platers. When spinning voltages on all Hitachi's motor pins were 6.12V. After 15sec of spinning on IBM PCB, the MCU is getting very hot.
So from this i can conclude:
- Hitachi's motor is failing somehow, and that's not allowing to run with platers
- Hitachi's motor needs more current to spin with platers, that's why IBM PCB MCU is getting hot.

5. So if from IBM PCB and Hitachi PCB, conclusion is that Hitachi's motor needs more power, and on original HDD (as from stock) it runs OK, then obviously there is something with Hitachi's motor. Or if not, so at least tehere is something wrong with power distribution to Hitachi's drive.

So if i would Have any working 2TB Hitachi drive, I could check any of my conclusions.

[fzabkar]

FWIW, here are photo clips showing the location of the motor current sense resistor for a Hitachi HDS722580VLAT20 drive:

http://www.users.on.net/~fzabkar/HDD/HD ... ense_R.JPG
http://www.users.on.net/~fzabkar/HDD/HD ... r_cont.JPG

The sense resistor is on the underside of the PCB.

Attachments

HDS722580VLAT20_bot_sense_R.JPG (34.13 KiB) Viewed 19603 times

HDS722580VLAT20_comp_motor_cont.JPG (26.69 KiB) Viewed 19603 times

[fzabkar]

BTW, I should add that a great tool for testing windings for shorted turns is Bob Parker's ESR meter (http://members.ozemail.com.au/~bobpar/esrmeter.htm). It is available in kit form. Although it was designed for measuring the ESR of electrolytic capacitors in-circuit (in which case it has little application in HDD work), it is also an ideal tool for the stated purpose. In fact I find it more convenient than an inductance meter.

For example, the resistances of the windings on a motor in a WD1600BB drive measure approximately 0.8 ohms on a multimeter, whereas my ESR meter reads 33 ohms (phase-to-phase). This is because it injects an 8us pulse into the component being tested, so inductances look more like reactances than resistances.

If 2TB Hitachi PCB's are all able to run IBM motor, and MCU is cool, so PCB's are fine (at least as for IBM motor).

I suspect that you are referring to the motor controller, not the MCU (the main microcontroller).

[fzabkar]

I have identified several points on your PCB:
http://www.users.on.net/~fzabkar/HDD/HU ... _sense.jpg

I believe the R470 (0.47 ohm) resistor is the VCM current sense resistor.

I believe that three of the four pins that I have identified as "spindle current sense" would be connected together somewhere on the circuit board. That "somewhere" should be one end of the current sense resistor, assuming it exists. The other end would be connected to ground.

The component that I have identified as "???" looks like a capacitor, but I'm not certain.

Attachments

HUA722020ALA330_current_sense.jpg (55.95 KiB) Viewed 19597 times

[Amarbir[CDR-Labs]]

BTW, I should add that a great tool for testing windings for shorted turns is Bob Parker's ESR meter (http://members.ozemail.com.au/~bobpar/esrmeter.htm). It is available in kit form. Although it was designed for measuring the ESR of electrolytic capacitors in-circuit (in which case it has little application in HDD work), it is also an ideal tool for the stated purpose. In fact I find it more convenient than an inductance meter.

For example, the resistances of the windings on a motor in a WD1600BB drive measure approximately 0.8 ohms on a multimeter, whereas my ESR meter reads 33 ohms (phase-to-phase). This is because it injects an 8us pulse into the component being tested, so inductances look more like reactances than resistances.

If 2TB Hitachi PCB's are all able to run IBM motor, and MCU is cool, so PCB's are fine (at least as for IBM motor).

I suspect that you are referring to the motor controller, not the MCU (the main microcontroller).

Franc ,
I Am Thier Indian Distributor .If Any Person In HDD guru Wants It I Will Send It To any Country With Free Shipping and Will Personally Take Care Of The Entire Process .

The Link -> http://www.lynx-india.com/index.php?pro ... ohms-meter

PS : i Never Sell Anything In HDDguru ,But As you got the topic out i though as i am a a crazy collector of gadgets ,might be someone can put it to some use here too if he really can understand this

[woxy]

The component that I have identified as "???" looks like a capacitor, but I'm not certain.

This element is a capacitor. I've made some measurements of voltages on almost whole 2TB PCB.
All measurements where made with multimeter GND connector connected to GND of PCB.
On this resistor as You mentioned "I believe the R470 (0.47 ohm) resistor is the VCM current sense resistor."
there is 6.46V when powered on, and when bzzz comes, it drops to almost 0V.
Also I've noticed that on NVRAM chip during bzzz time, there are several 2.5V on pins, and some 0V pins. But after bzzz
stops 0V pins becomes 2.5V pins. And the another thing, during voltage measurements on NVRAM chip, I've discovered
that PCB starts to bzzz again, if measured on pin 7.

Since the fragments of PCB are over 1MB files and over 1024pixels in width, here are links to them:
http://woxy_ivanhoee.republika.pl/2tb_pcb1_1fr.jpg
http://woxy_ivanhoee.republika.pl/2tb_pcb1_2fr.jpg
http://woxy_ivanhoee.republika.pl/2tb_pcb1_3fr.jpg
http://woxy_ivanhoee.republika.pl/2tb_pcb1_4fr.jpg
http://woxy_ivanhoee.republika.pl/2tb_pcb1_5fr.jpg

[fzabkar]

I've looked at your results several times, but I still can't see how your board senses motor current. I would have liked to see the voltage on each of the spindle sense pins, but probing these points with a multimeter has a very high risk of catastrophic damage. If your multimeter's probe tip is too large, you could try affixing a sewing needle to the end. At the very least, it would be useful to know the resistances between each pin and ground.

The R470 resistor does appear to sense the VCM current. When the voice coil is stationery, the voltage at R470 should sit at about the midpoint of the supply rail, ie 6V. The following diagram depicts the H-bridge configuration of the four MOSFET drivers (Q1-4) inside the motor controller. When Q1 and Q4 are switched on, current flows from left to right through the VCM. Conversely, when Q3 and Q2 are switched on, current flows from right to left. The voltage, V, across R470 reflects the VCM current as follows:

Ivcm = V / 0.47 ~ 2 x V

Code:

                               +12V
                                 o
                                 |
                                 |
                       +---------+----------+
                       |                    |
                       |                    |
                    ||-+                    +-||
                    ||<- Q1              Q3 ->||
                 ---||-+                    +-||---
                       |                    |                                 
                       |                    |
                   o----)-------+           |
                       |   ___  |           |
                   o---+--|___|-+--o VCM o--+---o
                       |  current           |
                       |   sense            |
                    ||-+                    +-||
                    ||<- Q2              Q4 ->||
                 ---||-+                    +-||---
                       |                    |
                       |                    |
                       +---------+----------+
                                 |
                                 |
                                ===
                                GND

Your observation in regard to the NVRAM pins suggests to me that the motor controller may be responding to an overload condition by sending a Power-ON Reset (POR) signal to the MCU.

As for your other voltage measurements, the 1R1 coil and XY/XZ MOSFETs constitute the +1.26 Vcore regulator for the MCU.

The 100uH coil to the left is the +2.6V Vio supply for the MCU, NVRAM, and SDRAM.

The two 100uH coils at the bottom constitute a Cuk converter that supplies -5V for the preamp.

The two white rectangular components are piezo shock sensors. The adjacent ICs appear to be op-amps that amplify their signals.

The 18V capacitor is associated with a charge pump. It appears to be working correctly.

I believe that the IC near the motor terminals may be a quad op-amp. I suspect that it may provide second stage amplification for the shock sensors, plus two additional functions (?).

LTC6079, Linear Technology, Micropower Precision, Quad CMOS Rail-to-Rail Input/Output Amplifiers, 16-pin SSOP, 2.7V to 5.5V:
http://cds.linear.com/docs/Datasheet/60789fa.pdf

[woxy]

I've looked at your results several times, but I still can't see how your board senses motor current. I would have liked to see the voltage on each of the spindle sense pins, but probing these points with a multimeter has a very high risk of catastrophic damage. If your multimeter's probe tip is too large, you could try affixing a sewing needle to the end. At the very least, it would be useful to know the resistances between each pin and ground.

Hi, I've measured sense pins according to GND, ad all seems to be direclty connected to GND.
As for voltages on those pins they look like this:

Looking from top:
1. 0V
2. 12.47V
3. 0V
4. 12.47V



The R470 resistor is 470 Ohm or 0.47 Ohm cos I'm confused because, when I tried to measure R470 resistance between R470 and GND the resistance was different on different setting of scale in Ohms on multimeter:
1. in scale of 20k - it showed 6.7kOhm
2. in scale of 200k - it showed 15.2kOhm
3. same R470 measured on both end shows 0.47 Ohm

Weird thing.

[fzabkar]

The "R" in "R470" is a decimal point.

R470 = 0.47 ohms
470R = 470 ohms
47K = 47,000 ohms
4K7 = 4.7K
4R7 = 4.7 ohms

Your resistance measurements confirm that R470 is "floating", ie it is not connected to ground. My ASCII circuit diagram illustrates this. Perhaps ASCII is not the best format for such diagrams ...

Anyway, the R470 resistor is now just a distraction. We have already eliminated it as a suspect.

That said, I confess that I'm out of ideas. I don't understand how the motor controller senses the spindle motor current, except perhaps by some internal means. BTW, the voltage across a motor sense resistor would be zero when the motor is not being driven, so your static voltage measurements would be inconclusive.

All I can suggest now is for you to measure the voltage across the motor sense resistor in your DeskStar drive, both during spinup and in run mode. That should give you some insight into how this part of the circuit works.

[woxy]

Hello again,
today I found my previous mistake.
All spindle sense pin's are not connected to GND.

1 and 3 are to GND - point 1
2 and 4 are to + 12V - point 2

and between them there are 2 capacitors (I think capacitors) maybe the are current sensing thing?



between points 1-2 there is 12V
between points 2-3 there is about 6V
between points 1-3 there is about 4.7V

[fzabkar]

Sorry, it appears that those pins are in fact supply pins, not sense pins as I suspected.

Test point #3 does not stand out as a sense pin, either. I would think that if your board did have a spindle current sense resistor, then it would be of the order of 0.5 ohm or less, in which case the voltage during spinup would be about 1V (assuming a startup current of 2A), and about 0.5V during normal running (assuming a run current of 1A).

I really don't know where else to look.

You could use the following datasheet as a guide to the function of a typical motor controller:

L7250, SMOOTH, spindle motor + VCM controller, ST Microelectronics:
http://wandrew.regruppa.ru/PCInfo/TechDoc/L7250(Smooth).pdf
http://www.datasheetcatalog.org/datashe ... Xyuswx.pdf

See page 43 for an application circuit, and page 21 for a block diagram.

Here is my database of motor controller datasheets:
http://www.users.on.net/~fzabkar/HDD/HD ... ntrol.html