HDD won't spin - Tried replacing PCB

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

Slycer

Reputable
Jan 7, 2015
38
0
4,530
Hi,


there are some serious experts on this board and I've learned a lot by reading fzabkars posts. Hope he sees this thread and chimes in. A month ago three of my drives failed. Five were running. A SSD survived along with a SATA drive, WD 20EFRX. I lost two Samsung HD204UIs and a WD 20EFRX. I don't know what happened. Drives were on mixed power leads so it couldn't have just been a power spike on one cable.

At this point in time I'm really only trying to save the WD 20EFRX. It contains all my study data, 10 years worth of photos, bank certificates and other things I accumulated over the years. Really, it's everything I care about in digital form.

The thing is that the drive will not spin up. I have ordered a replacement PCB from hdd-parts and swapped the BIOS chip. Well I had it swapped. Still NOTHING. If I plug the drive into the PC, it detects it in BIOS. When I started looking into things a month ago before I ordered the PCB I even downloaded a linux distro with hard drive tools and somehow, I have no idea how with the drive not spinning, I was able to BROWSE the drive. Every folder, every file was there. Since replacing the bios I only tested if it's still detected in BIOS and it is. Haven't yet tried to browse the contents again. Of course I wanted to copy the data the moment I saw I was able to browse the contents but I got an error. Probably expected since the drive wasn't spinning.

Does anyone have any idea what I could try doing to get the drive working? I am seriously desperate.


Thanks to anyone in advance.

 
I have tried to wake the disk up with hdat2. When normal IDE is chosen in BIOS I get an error. And at the end a "sorry nothing to do", message.

When I run it with IDE legacy mode enabled I get no errors but still the sorry nothing to do message.

I'm uploading s video of the second process.

BIOS hangs for like a minute on every reboot since it cant detect it.

Do you think that I should try with AHCI mode?
 

Fzabkar, do you think it's possible that this isn't a Windows 10 related faliure? I have no idea what it could be. But I find it incredibly unlikely that a different error would affect two same drives just when a third different one was bricked by Windows 10. These three have to be related, right? And even if it wasn't a Windows 10 related fault, having two drives fail at the same time showing the same symptoms is very strange on it's own.

I want to rule out a hardware fault so I tried checking the diods once again. And I have no clue if I'm doing this the right way. This is how I measure the readings. And still most say OL.


hd204uitvs.jpg
 
The ATA standard states that when PUIS is enabled the drive should still respond to the Identify Device command, although the data may be incomplete. For example, some WD drives report a blank model number or zero capacity. The fact that your screenshot is showing that the drive hangs after receiving the Identify Device command would suggest that the drive has a problem unrelated to PUIS (unless there is a bug in the firmware).

That said, the fact that the drive is sensed at all would suggest that the PCB is at least partly functional, which in turn would suggest that those three components are OK. BTW, your measurement locations are correct. If you select the 200 ohms range, the two diodes should read "OL" (out of range) and the zero-ohm resistor should read 0.0 ohms or close to it.

I'm assuming that nothing changed in the original configuration other than the new OS. Otherwise, if the Samsung drives were shifted to a different controller, then perhaps it would be worth reducing the SATA link rate to 1.5Gbps with a jumper. Some Samsung models have detection issues with some chipsets, eg AMD SB850 and Intel P67. In fact it might be worth trying the jumper anyway.

http://www.techarp.com/review/Samsung/EcoGreen_F4_2TB/jumper.png

If all else fails, then there is a way to retrieve diagnostic information from the drive's serial terminal port. You will require a serial TTL adapter (~US$10).

More information at The HDD Oracle:
http://www.alexsoft.org
 
This is so FUBAR. I'm never testing an OS again.

I've repeated the reading on the diods. I now get proper feedback. Both zero ohm resisotros put out 0 ohm. The 5V dioda return about 1.4 [kilo?]ohm and the 12V diods return 293 [kilo?]ohm. I'm so not sure about the scale since I can't pick 200 ohm on the dial. The multimeter sets the range.

This is a DIODE test, selected on the multimeter:

hd204uitvs-1.jpg


With this configuration I get, 0.465v on the 5V and 0.646 on the 12V. And it's the SAME VALUES for BOTH Samsung drives.When the leads are reversed I also get a reading. Honestly, I don't know if this is normal but it's not easy to get a reading. Unless I hit the sweet spot I get the OL.
 
Maybe this is a common issue? http://forums.storagereview.com/index.php/topic/29715-samsung-hd204ui-doesnt-wake-from-sleep/

I'll leave in connected externally via usb during the night. Hope something moves.
 
The diodes and resistor are OK (I was more concerned with the resistor).

What we really need to see now is the drive's terminal log.

Here is an example:

http://en.elektroda.pl/rtvforum/topic2091947.html

AIUI, the "U" indicates that the drive is spinning up.

"SATA PLL cal done" suggests that the drive has autonegotiated a mutally agreeable SATA link speed with the host's SATA controller.

The "PA" lines are confirming that the preamp on the headstack is present and compatible.

I'm hoping that we will see something related to PUIS.

ActiveFW : 00
FWVer : 0001
SATA PLL cal done

DDR size detected = 32MB
*PA VID=0000 PN=0009 Rev=0001- 785x Found
*PA VID=0000 PN=0009 Rev=0001- 785x Found
U
 
And for this I need a serial TTL adapter? I just don't have any idea what exactly that is. I've looked at that oracle board but can't find the exact model of what I need to order.

Googling for samsung terminal cable doesn't return anything. Looked at ebay too. I see the small cable here but I don't understand the connecting schemes. http://malthus.mooo.com/viewtopic.php?f=116&t=189

My HD204UI only has two rows of pins on the back. 4 pins.
 
There are three pins, Transmit (Tx), Receive (Rx), and Ground.

Either of those eBay serial adapters should work, but sometimes you may need a slight modification because the HDD's I/O level is 2.0V whereas the adapter's is 3.3V.

http://img15.hostingpics.net/pics/217313dku518.jpg
http://img11.hostingpics.net/pics/228167mod18V.jpg

You will also need to install the COM port driver (software) that comes with the adapter.
 

This is what I'll order today. Is it OK? http://www.ebay.de/itm/CP2102-USB-2-0-an-RS232-TTL-COM-6-Pin-Module-Konverter-Converter-Adapter-Seriell-/271712367348?pt=DE_Computing_USB_Kabel_Hubs_Adapter&hash=item3f4351b2f4

And what you're saying is that I'll also need a small modification on the power adapter to switch it from 3.3 to 2V? There is a link to the drivers on the eBay page.

I'm sorry for being daft, I'm really swimming in deep water with this. Is the first picture an end product the way it should look like? I see a resistor soldered onto the board. Or is it just something to show me how to reduce the voltage. Looking at the picture I dont know what the black/white/green adapter does. I thought I only needed the RX and TX cable from the adapter. I did not know that I needed to bring power through the serial port. Is the hard drive connected to PSU power when doing these readings? Or is the 2V from this adapter enough to get the data? I kinda basically get the idea what to get the readings I need only one cable is connected to the drive. The one I need to make. No 12V power, no sata cable.

This is the back of my drive with the pins. Do you know where which cable goes when I get everything?

samsung-f4eg-front.jpg


And thanks again for taking all this time to help me fzabkar. I REALLY appreciate it.
 

Hear hear. I had to deal with a HDD with a blown diode a couple years back (ultimately unsuccessfully). I spent half a day searching the Internet trying to find more info on what was going on. The info fzabkar is providing is just pure gold.
 
The HDD's diagnostic interface format is RS232 with LVTTL signalling, whereas the USB port uses a different protocol and different signalling. The adapter PCB incorporates a USB-serial bridge IC which acts as an interpreter for the two dissimilar devices. To put it simply, the HDD and USB port speak different languages and the adapter handles the translation.

The HDD's diagnostic port has 3 pins (transmit Tx, receive Rx, ground) which connect to the adapter as follows:

HDD Tx -------> adapter Rx
HDD Rx <------- adapter Tx
HDD Ground <--> adapter Ground

I have been communicating with a colleague who will shortly upload a tutorial that should help you. It might take a few days, though.

Don't bother with the resistor and diode just yet. Even when there are problems, in most cases you will still be able to read the diagnostic output, but there may be one or two spelling errors.
 

No problem, since I won't be getting the USB TTL adapter until next week probably. Ordered yesterday two, just in case I do something bad with the first. They're so cheap at 2 euros each. Can't wait to see what the drive outputs. I really hope I'll be able to revive it. Both of them.

But really, I can't belive that the WD was so much easier to rescue. Could be it because it's a newer drive? 2014 instad of 2011? Or does Samsung do something whacky with their firmware. I'm still sticking with the idea what all this is a software error. Can't have two drives fail at the same time with a hardware error. Statistics gotta back me up on that.


No kidding. Before I posted this thread I looked around various sites and on this forum, fzabkar is a hdd guru whose knowledge is only surpassed by his willingness to help people like me.
 
Here is the power up sequence for a HM321HI drive when PUIS is disabled:


=========================================================================
*PA VID=0007 PN=0000 Rev=0004
*PA VID=0007 PN=0000 Rev=0004
ChipRev = 63B0
U
S_0Shock Sensor Circuit Enabled
SO_1
Shock Sensor Circuit Enabled
IS=6
CLA #36/56
Handoff RPM=667
SPOK
mS1 00000003
A1O=6 A1G=421
O=509 S=0
SK C: 5005 H:0
Loaded FIT ( 0: 0: 1)
CalibTable Loaded. Rev:0x17
ResoTable Loaded. Rev:0x01
Bump Limit Loaded. Rev:0x00
RRO1xTable Loaded. Rev:0x01
Ovly loaded to 0x0001CD00
Ovly loaded to 0x1005B800
FdtTable Loaded. Rev:0x02
[GL][90PT:0FF0,0204]
[GL][FULL:0FF0,0204]
Reading Serial Num Pass
Up MC


PwrOn RRO1x @ H0
Table) cos = -1605632, sin = -1520128
Coeff) cos = 167211, sin = 168953


DiskSlip : 202 [SrvTrk]
DiskSlip : 181 [SrvTrk]
TgtCyl: 842
Hd: 0 Zn: 0 Avg.: 183
TgtCyl: 176473
Hd: 0 Zn: 1 Avg.: 1106


SVCAL(0080,0000)-->PASS
RecordValid Ok : 0107E47D 0107E41D
P11FFS Enabled
ReadyTime = 2029160 us
ENG>mS1 00000003
SK C:134751 H:0
SoUL!
PK_4 C: 8496 H:0
=========================================================================


Here is the same drive when PUIS is enabled:


=========================================================================
*PA VID=0007 PN=0000 Rev=0004
*PA VID=0007 PN=0000 Rev=0004
ChipRev = 63B0
U
S_0ENG>
=========================================================================

BTW, HDAT2 is able to disable PUIS for this model (after launching it with the /w switch).
 
So I received the terminal and the cables now. How do I connect them properly? Which pin on the hard drive is TX and which is rx? Do I need to connect the power to the hard drive? I still need to solder the power lead on the terminal, the 3.3v. I solder the 3.3v lead to the RX lead on the terminal?
 
I've seen a trick with covering the three pin motor contacts. I may be completely off mark here. Is this by any chance when I use the 12v lead connected to the hard drive itself to power the terminal?

I just have this theory that I need to connect the 3.3v when I don't want to use the 12v lead. But if I do want to use the 12v lead I need to cover the three pin motor contacts.

Am I completely off mark? Or do I need the 3.3v from the terminal either way? My soldering gun broke.. So I'm looking for alternatives or I'll have to find someone to solder this on for me.
 
Sorry, I didn't see your replies.

I would start by connecting the adapter to your PC on its own. Install the drivers and access the COM port using a tool such as HyperTerminal or PuTTY. Select the appropriate baud rate, connect the Tx and Rx wires together, then start typing. You should see your typed characters being echoed on your screen. This means that the adapter is working.

Now connect the Tx and Rx pins between the adapter and the HDD according to the diagrams at the HDD Oracle. It doesn't matter if you connect the Tx and Rx pins incorrectly -- you won't break anything, it just won't work.

The adapter is powered from the USB port. There is no need to fiddle with the HDD, or anything else.
 
How do I know which chip I have? Samsung/Marvel. A round one has M on it, a rectangular one has Samsung written on it.

Anyway, I'm trying to test the TTL.. I've connected it. Had to fiddle around to get the thing working in Windows (drivers issue). Connected to it via Putty. And it doesn't send anything back. I write hello then press enter and nothing happens. Or atleast it looks like nothing happens. Shouldn't two "hellos" appear?

I will need to bridge the TX and 3.3V line on the TTL before I connect it to the hard drive, yes?

 
Don't touch the 3.3V pin. It is not required, and you'll seriously damage your drive or adapter or both if you connect a power pin to a Tx/Rx signal pin.

To test the adapter in Windows, you need to connect its Tx pin to its Rx pin. This means that anything you type will be reflected back through the port. This is called a "loopback test".
 
Status
Not open for further replies.