I posted this in a topic covering a similar situation (forcing a drive to 1.5 Gbit/s) albeit with a Western Digital drive. Perhaps this creates some confusion hence I decided it could be better to create a new topic for this drive family only. Apologies for this; learning the customs on this forum! Also wanted to say hello and thank you all for sharing some of the knowledge HDD manufacturers would rather keep in their labs... makes me really thankful that we have the web and some awesome people using it.

I recently purchased one of those Seagate 2.5" Solid State Hybrid Drives, the ST1000LM014. It is able to run at SATA 3 (6 Gbit/s) speeds, but my notebook does not like that at all. According to my research, even though the chipset (nVidia's MCP79) advertises its ability to communicate at 3 Gbit/s, it has troubles actually working at this speed with some drives. Choking them to 1.5 Gbit/s apparently solves all the problems (eg. OS freeze).

So long story short, I'd like to force my ST1000LM014 (firmware SM15) to 1.5 Gbit/s. Some of the previous Seagate models, eg. the Momentus XT family, honoured jumper settings for this (two leftmost pins with HDD facing PCB-side-up) but this drive sadly does not It seems that I need to either invite it to honour the jumper (a "SATA gen I 1.5Gb/s Jumper Supported" software setting mentioned in the other thread on the WD drive?) or somehow force it to only advertise support for 1.5 Gbit/s to the controller.

I'd be fine doing this via the serial interface or any other tool, but wanted to know if it can be done at all without low-level programming? I've read all the Seagate console commands but there is none specifically for this, at least according to the command lists I've acome across...

Hi,
after deploying recently 130 laptops in an edu environment, I can say there are many issues relating to newer hardware.

I don't think you are going to find a solution for that drive. a similar thread, and answer:


sometimes admitting defeat lets you sleep easier at night again a lot earlier

HaQue, thanks for input, good to see someone reply You are right in that the quote you... quoted describes the problem exactly, thou unfortunately in my case a firmware rollback has not been put forth by Apple (an doing this the unauthorized way is apparently not as easy as it sounds, as Apple prohibits you from rolling firmware back) and secondly, according to Apple changelogs, earlier firmwares for my particular model would not disable 3 Gbit/s mode anyway This was only the case with 2009 MacBook Pro models, mine is late 2008.

So all Apple-side LEGOs being taken away from me, all I have left to play with is the drive itself.

Would prefer not to admit defeat without a fight A TTL-USB interface is on its way, will give it a try when it arrives if only "for science" and posterity's sake.

Ok, well once you get the interface it will be interesting to see how you go.

Cheers

A small update. I connected the drive to a USB-TTY converter but are not able to communicate with the drive. My setup is as follows, with 'up' being PCB-side: [EMPTY GNDcable RXcable TXcable] [.......] [..................].

I've tried different COM port speeds and even parity settings but still all I get is random characters in the PuTTY console window when I type in ctrl-z or letters to see if at least it echoes what I write. Any clues on what would the correct way to communicate with it be? Is it possible that this is a newer drive that eg. encrypts serial communication or uses some strange baud rate just to make it harder for anyone to connect?

Seagate uses: 9600 or 38400 OR NONE

None means NO TERMINAL ACCESS = Drive Blocks Terminal Output = Need to Short Certain points in Specific Timing to bypass it (Depends on PCB)

good luck

_________________
Kuwait Data Recovery - UNIX GTC
The only reason for time is so that everything doesn't happen at once. By: Albert Einstein

Hm, how would you interpret this then? Console output of ctrl+z pressed twice, then just the letter "z". Not all key presses had visible output.

At baud 9600:


At baud 38400:

I guess your problem with terminal output is from Putty (Limited options)
have a look here @ the images

_________________
Kuwait Data Recovery - UNIX GTC
The only reason for time is so that everything doesn't happen at once. By: Albert Einstein

einstein9, thank you for your efforts and helping me getting a stab at this drive's speed. Unfortunately, it seems that HyperTerminal gives me the same output (again ctrl-z ctrl-z and then just letter the "z" all the way):

Putty limited options compared to hyperterminal ???

look at the options under the "terminal" menu. If you have played around with options such as local echo, the CR and LF options, or stuff under the "keyboard" sub menu, try getting them back to default.

or as einstein9 suggests, try using HyperTerminal as the less "features" may mess with the terminal less and not give that corruption.. OR your firmware could be dodgy and not working properly I guess

also, you might try this program from:

https://code.google.com/p/baudrate/

ISTM that the USB-serial bridge is OK, but you could confirm this by connecting its Tx and Rx pins and performing a simple loopback test by typing on the keyboard and looking for echos.

Next I would measure the voltages at the jumper pins on the drive to determine whether the interface is 3.3V or 2.5V.

Then do the same with your bridge's Tx and Rx pins, with it disconnected from the drive.

That said, it does look like a baud rate or framing issue.

_________________
A backup a day keeps DR away.

Did this before connecting my drive to make sure the bridge all of my soldering is fine; passed without any problems.

Seems you may have nailed it here! I measured 5.0V and 4.8V on the bridge pins (Tx and Rx, respectively), but voltages were as low as 1.76V on the HDD...

From what you're saying I need to lower my voltage to around 2.0V on both the Tx and Rx? (Assuming that I have not yet burned the drive's serial interface and just lowering voltages will make it happy enough to talk to me.) From your collective experience, what sort of a current the interface may draw? Need this to estimate what resistor i should plug in between my interface and the drive.

Try also the 57600 (same config) - This is for Samsung, which i guess your drive is

good luck again

_________________
Kuwait Data Recovery - UNIX GTC
The only reason for time is so that everything doesn't happen at once. By: Albert Einstein

Does the drive carry an APPLE logo ?

"ST1000LM014 (firmware SM15)" is a Seagate drive, AFAIK. The part number (1EJ164) also looks like Seagate.

_________________
A backup a day keeps DR away.

The four jumper pins would be wired to the MCU's I/O logic. This would be powered from the drive's Vio supply. It appears that Vio may be +1.8V.

I don't know the specs for your MCU, but many of these devices can tolerate 5V signal levels. Nevertheless, I would play it safe and use a 1.8V or 2.0V zener diode to clamp the input to the drive.

The USB-serial bridge IC can usually be programmed for 5V or 3.3V I/O, so I would try to jumper it for 3.3V, or 2.5V if available. A 5V TTL interface will usually have an indeterminate range between 0.8V and 2.0V, so it could be that a 1.8V input is not being reliably detected.

_________________
A backup a day keeps DR away.

You may be able to determine the input range of your USB-serial bridge by means of a potential divider in the loopback path.




R1 = 3K3, R2 = 1K2 --> Rx = 1.33V
R1 = 3K3, R2 = 1K5 --> Rx = 1.56V
R1 = 3K3, R2 = 1K8 --> Rx = 1.76V
R1 = 3K3, R2 = 2K2 --> Rx = 2.00V
R1 = 3K3, R2 = 3K3 --> Rx = 2.50V
R1 = 2K2, R2 = 3K3 --> Rx = 3.00V

Keep reducing the voltage until your keystrokes are no longer echoed reliably.

_________________
A backup a day keeps DR away.

after 1 month and counting only to determine if the terminal works or not, maybe it's better to know if the terminal will be of any aid in this case (simpler solution : look at a working system to see which p/n and release work).
Apple are picky without APPARENT reason.

Apologies for not replying earlier, for some reason I stopped getting thread reply notifications Came back just to make sure and what a nice surprise to see so many replies!

Same thing as before - non-letter characters with the letter "z" echoed every now and then.

Worse, instead of letter "z" I get random letters interleaved with non-letter characters.

With hardware flow control off I get the same result as above.

Nope, I purchased this drive from a computer shop, Apple was not involved... until I tried to put the drive in an Apple product that is

Thank you for this suggestion, will pop in to an electronics store tomorrow to get the parts and try to at least discover the input range. The bridge I'm using is based on PL2303, whose datasheet does mention its ability to work with 5V, 3.3V or 3V signals. Will inspect the board to see what can be done (there's no configuration jumpers unfortunately).

I'd love to know that but as you can see at the beginning of the thread no-one was able to offer a definite conclusion ie. "yes it is possible" or "no not possible on this model". That's why I embarked on this little quest to check this myself... and so far am enjoying the ride as much as a possible outcome