Question Hard Drive not recognized through ANY SATA port but recognized via USB after NVME SSD installation

[Oyavo]

@Oyavo, you've done everything except assessing the health of the drive. That's what SMART is for, and that's where you should have started.

Unless I'm mistaken, I was under the impression that the following information was part of the drive health:

C:\Windows\system32>wmic diskdrive get status
Status
OK
OK

Where the first line was my NVME and the second was the 2TB hard drive, with nothing else plugged in at the moment.
Otherwise, here is a full export from CrystalDiskInfo:

 (02) ST2000LM003 HN-M201RAD
----------------------------------------------------------------------------
           Model : ST2000LM003 HN-M201RAD
        Firmware : 2BC10001
   Serial Number : S321J9FG100479
       Disk Size : 2000.3 GB (8.4/137.4/2000.3/2000.3)
     Buffer Size : 32767 KB
     Queue Depth : 32
    # of Sectors : 3907029168
   Rotation Rate : 5400 RPM
       Interface : UASP (Serial ATA)
   Major Version : ATA8-ACS
   Minor Version : ATA8-ACS version 6
   Transfer Mode : SATA/600 | SATA/600
  Power On Hours : 16823 hours
  Power On Count : 3165 count
     Temperature : 28 C (82 F)
   Health Status : Good
        Features : S.M.A.R.T., APM, AAM, NCQ, GPL
       APM Level : 0080h [OFF]
       AAM Level : FE00h [OFF]
    Drive Letter : F:

-- S.M.A.R.T. --------------------------------------------------------------
ID Cur Wor Thr RawValues(6) Attribute Name
01 100 100 _51 000000000007 Read Error Rate
02 252 252 __0 000000000000 Throughput Performance
03 _91 _90 _25 000000000AF5 Spin-Up Time
04 _97 _97 __0 000000000E27 Start/Stop Count
05 252 252 _10 000000000000 Reallocated Sectors Count
07 252 252 _51 000000000000 Seek Error Rate
08 252 252 _15 000000000000 Seek Time Performance
09 100 100 __0 0000000041B7 Power-On Hours
0A 252 252 _51 000000000000 Spin Retry Count
0C _97 _97 __0 000000000C5D Power Cycle Count
BF 100 100 __0 000000000020 G-Sense Error Rate
C0 252 252 __0 000000000000 Power-off Retract Count
C2 _64 _46 __0 0036000E001C Temperature
C3 100 100 __0 000000000000 Hardware ECC recovered
C4 252 252 __0 000000000000 Reallocation Event Count
C5 252 252 __0 000000000000 Current Pending Sector Count
C6 252 252 __0 000000000000 Uncorrectable Sector Count
C7 200 200 __0 000000000000 UltraDMA CRC Error Count
C8 100 100 __0 0000000000EC Write Error Rate
DF 100 100 __0 00000000008E Load/Unload Retry Count
E1 _81 _81 __0 000000031298 Load/Unload Cycle Count

 

[fzabkar]

The non-zero read and write error rates suggest that the drive has started to degrade. The UltraDMA CRC Error Count is zero, so that would discount interface errors.

Your drive is actually a Samsung model with a Seagate rebrand. It has a serial diagnostic port that would tell us a lot more, but that would be something to pursue in the event of data recovery.

 

[Oyavo]

I would connect the drives in question to other known working systems simply see if the drive works. Most likely with an external USB enclosure. One drive at a time to focus on that drive.

"Works" meaning constantly readable/writable and no errors. If the errors follow the drive then the drive is at fault.

At the moment, while I unfortunately do not have another system to connect to via sata port, it works via USB adapter on MacOS read/write no problem whatsoever.

As for the damage: could have been small or minor begin with and simply has become worse to to heat related expansion/contractions. Or just being handled no matter what level of care is taken. Maybe vibrations are another factor.

Intermittent problems are difficult to troubleshoot. Take your time, be methodical.


Late note: In your SATA photos, one of the pins appears "split". Leftmost connector, right end.

I can't quite see any damage on the pins from what I can make out though I could be mistaken. Unfortunately the pins are so small that I'm having trouble getting any camera to focus well enough to make it out. I can attempt to clean the pins, but otherwise I'll have to keep trying alternatives as you mentioned.


Furthermore, I also tried to re-attach the drive back into my pc rather than connected on my desk in regards to vibrations, however this did not have any influence from what I saw.

 

[Oyavo]

The non-zero read and write error rates suggest that the drive has started to degrade. The UltraDMA CRC Error Count is zero, so that would discount interface errors.

Your drive is actually a Samsung model with a Seagate rebrand. It has a serial diagnostic port that would tell us a lot more, but that would be something to pursue in the event of data recovery.

Currently, the drive is completely empty and all that was on it has since been transferred. Would accessing this port still be worthwhile? Otherwise, I'm not quite sure what to make of the read//write error information as it seems strange for such an error to show up only once an SSD was mounted to the m.2 slot.

 

[fzabkar]

There would be no point accessing the diagnostic port, unless you are curious and are willing to purchase a USB-Serial_TTL adapter.

When a drive powers up, it loads its firmware ("discware") from a reserved area of the platters. Samsung calls this area the Maintenance Cylinders (MC). If the heads have started to degrade, then the drive may struggle to read the MC, in which case the BIOS times out while waiting for the drive to become ready. The bridge controller in a USB enclosure may allow more time for the drive to identify itself, so that could explain the difference in behaviour. Moreover, even if BIOS doesn't see the drive, the drive could come ready by the time the OS searches for it.

 

[Oyavo]

There would be no point accessing the diagnostic port, unless you are curious and are willing to purchase a USB-Serial_TTL adapter.

I'm not opposed to the idea, they aren't too expensive so I might get one anyways. Thanks for the heads up

If the heads have started to degrade, then the drive may struggle to read the MC, in which case the BIOS times out while waiting for the drive to become ready. The bridge controller in a USB enclosure may allow more time for the drive to identify itself, so that could explain the difference in behaviour. Moreover, even if BIOS doesn't see the drive, the drive could come ready by the time the OS searches for it.

Is there any way to prevent BIOS from timing out the drive without the adapter regardless? The parts that are confusing me are specifically that this all occurred precisely when I installed the m.2 SSD, and how seamlessly the drive is picked up via USB adapter.

Would the drive's slower ability to read really impact only sata connection by upwards of 10-20 minutes, yet not even affect USB connection by a second? I'm not doubting what you've said, I just find it fairly strange and disappointing if that's the case.

 

[fzabkar]

I'm not sure that the problem is related to the MC and head degradation. I'm just advancing one possible explanation that is consistent with what I've seen in the professional data recovery forums.

How to connect a terminal adaptor on a Samsung drive ?

https://www.hddoracle.com/viewtopic.php?p=432#p432

This post has a sample of the output from the diagnostic port:

https://forum.hddguru.com/viewtopic.php?p=293671#p293671

This post will hopefully explain what goes on during the drive's power-up sequence:

The hard drive -- a computer-within-a-computer:
https://www.hddoracle.com/viewtopic.php?p=19087#p19087

 

[evermorex76]

I would connect the drives in question to other known working systems simply see if the drive works. Most likely with an external USB enclosure. One drive at a time to focus on that drive.

The drive has been shown to work perfectly fine on the same system using a USB enclosure, as has been mentioned more than once. Which would make one think it's the host. But other drives work fine on that host, which would make one think it's the drive. But the drive works fine in the enclosure, rinse and repeat. Testing on another system with a SATA internal connection might actually be the best thing.

Late note: In your SATA photos, one of the pins appears "split". Leftmost connector, right end.

I don't see it, but I also don't think I've seen a worse resolution photo in the last 10 years. It's more clear in the other photo. If it was actually split or something I'd expect it to be getting crushed and warped severely after being connected repeatedly, and it would be very obviously damaged if it was doing something like shorting to a data pin. That's also a ground pin so it should be somewhat less important than others, unless they are not connected internally on the drive side and the controller actually detects continuity on all 3 of them separately before allowing data flow.

Don't try to get so close in with the camera that it can't properly focus and don't use digital zoom. Just take the highest resolution picture possible from the closest point that it can focus, then viewing it at full size should be more clear.

 

[evermorex76]

That "G-Sense Error Rate" of 20 seems concerning. I don't think you'd get that from the simple vibrations of sitting on top of a PC case with the cables hanging out, or normal operation inside the case. That would require a significant impact while the system was running, but not severe enough to cause an actual had crash.

Whether those read and write errors are significant kind of depends on when they happened, which you can't tell. The drive has been powered on for nearly two years and they could have happened over a year ago, or just in January. They could have occurred only during whatever "shock" caused the G-Sense errors. You don't know if the number has risen over time or not. And it's not entirely unexpected for a mechanical drive to have a few errors like that over its lifetime which don't necessarily indicate it's failing, but since you ARE having problems with it, you may as well just assume that's the case. A 2TB mechanical drive isn't terribly expensive to replace, and you don't really even seem to need it as you've gone without it for several months. You're certainly not going to be able to FIX anything that's damaged even if you buy that serial adapter.

Oooh, I just noticed the Load/Unload Retry Count. That could be another indicator that it's having trouble moving the heads onto the platters. I wonder if the drive HAS actually been exhibiting the "click of death" but it's simply soft enough that you haven't been able to hear it. It's a low performance drive, low RPM, maybe the actuator isn't designed to move as quickly as higher performance drives and just has a softer return to base.

 

[fzabkar]

That "G-Sense Error Rate" of 20 seems concerning. I don't think you'd get that from the simple vibrations of sitting on top of a PC case with the cables hanging out, or normal operation inside the case. That would require a significant impact while the system was running, but not severe enough to cause an actual had crash.

Whether those read and write errors are significant kind of depends on when they happened, which you can't tell. The drive has been powered on for nearly two years and they could have happened over a year ago, or just in January. They could have occurred only during whatever "shock" caused the G-Sense errors. You don't know if the number has risen over time or not.

It's difficult to know how Samsung records their raw values. I know that Seagate's error rate attributes are mostly rolling averages rather than lifetime counts (Seagate's seek error rate is one exception). If Samsung uses rolling averages, then their error rates would be a reflection of recent events. As for G-sense errors, I have seen drives accumulate these errors even when users swear that their computers were not subject to shock events.

 

[Oyavo]

Don't try to get so close in with the camera that it can't properly focus and don't use digital zoom. Just take the highest resolution picture possible from the closest point that it can focus, then viewing it at full size should be more clear.

Sorry about that, hopefully this should be a little clearer.

A 2TB mechanical drive isn't terribly expensive to replace, and you don't really even seem to need it as you've gone without it for several months. You're certainly not going to be able to FIX anything that's damaged even if you buy that serial adapter.

To be honest it's probably more out of stubbornness and confusion in that a drive which I nearly ~$200 CAD on after tax for extra storage has somehow resulted in effectively the same storage, especially when the drive appears to be completely fine on the surface.

At the moment I've decided to purchase a male-to-female sata adapter/extension that's similar to the USB adapter I've been using as one last attempt to salvage the drive. The only thing I can think of is unless somehow the slight amount of damage to the plastic surrounding the pins or one of the pins themselves is impeding power-flow (hence why it's slow to show up, still spins, yet also inconsistent on when it shows up) that somehow is bypassed using the adapter, however if I'm being honest this is all just a guess.

 

[fzabkar]

The damage is at the 3.3V end. That isn't used.

https://i.ebayimg.com/images/g/TisAAOSwxHlhhJvf/s-l1600.webp

https://pinoutguide.com/Power/sata-power_pinout.shtml