USB Stabilizer Features

Windows automatically writes to mounted storage devices to update/repair filesystem metadata, which can cause permanent data loss, especially when working with unstable drives. All USB Stabilizer editions ensure that this does not happen by blocking all writes in all circumstances.

The 10Gb device has 3 independent channels, so it can work on 3 cases at the same time.
Channel 1: USB-C or M.2 M-key NVMe.
Channel 2: USB-A.
Channel 3: SATA or USB-A.

This is the limit of the connection speed between the device and the Host PC. Tech and Pro hardware connects to the PC over USB3.0 5Gb, while the 10Gb device connects over USB3.2 10Gb. The total speed of the 10Gb device is divided among its three channels, so for example it could read a SATA SSD at 500 MB/s, a USB flash drive at 300 MB/s, and a USB HDD at 200 MB/s at the same time.

The Read Timeout is how long the tool will wait for the storage device to respond to any read command before giving up and aborting it by issuing a reset or repower. Giving the storage device more time to respond increases the probability of a successful read at the cost of additional drive stress and wasted time. There are diminishing returns and working too slowly may cause the storage device to fail completely before the necessary data is recovered. For typical HDDs it makes sense to wait for 300-500ms and for typical SSDs for 100-200ms. The read timeout being fixed on 2000ms is a very large limitation of the Tech device which directly affects the types of cases it can handle. It is much better than having no hardware and therefore no read timeout at all, but 2000ms is quite slow, which only allows the Tech unit to provide a modest increase to bad sector processing speed, and a moderate improvement to the rate of further drive degradation.

The error handling speed is how long the tool takes to work through failed reads (bad sectors). It is the combination of the Read Timeout threshold being used and the reset processing time. Tech and Pro are the same hardware device, so their reset processing time is the same, and the Pro unit is faster only due to its ability to use a lower Read Timeout. The 10Gb device is fundamentally different as it uses novel types of USB resets which work much faster.

Error handling speed is the primary performance metric of any read instability handling tool as it directly affects the amount of data that will be recoverable before the unstable drive suffers a final crash. The real-world impact of this speed difference is even more significant than that. Slower error handling gives the drive more time to spend on internal processes within its firmware, which is highly detrimental. It is quite common to see drives with non-critical firmware issues where almost nothing is recoverable with slower tools because between each block read the drive has enough time to start trying to update various corrupt logs within its firmware, and hang as a result, getting stuck in a vicious cycle where most blocks fail to read. Faster error handling means that the tool will be more likely to send the next read command before the drive gets stuck on internal processes, causing the outcome to be completely different.

SATA resets are inherently much faster than USB resets. Due to having to work through a SATA to USB adapter, Tech and Pro editions can only use USB resets even when working with SATA drives. The 10Gb device has native SATA support, so it can use native SATA resets when working with SATA HDDs, which work multiple times faster.

Flash storage devices usually do not respond well to resets and have to be repowered instead, so their error handling speed is the combination of the Read Timeout threshold plus repower processing time. The Pro unit is once again faster than the Tech unit entirely due to its ability to use a lower Read Timeout. The 10Gb unit is faster than the Pro due to its hardware being quicker to establish a new connection to the storage device after a repower.

There are many different ways to reset a hanging HDD to force it to abort a failed read. Tech and Pro editions use only PHY reset to accomplish this, meaning that their hardware physically reconnects the data bus of the drive upon reaching the Read Timeout threshold. As soon as the drive loses the connection to host, the read request becomes irrelevant, so the drive stops trying to process it and allows us to move on. In addition to PHY reset, the 10Gb device offers 3 other types of native SATA and native USB resets which work much faster. It is impossible for any single type of reset to work well for all cases, so having the choice of different reset methods is critical for some drives.

If the connected source drive has a short circuit, all editions of USB Stabilizer will quickly cut off the power, greatly reducing the probability of causing further damage.

USB protocol requires the PC and storage device to be in constant synchronization with one another. If the storage device fails to respond quickly enough, or responds irregularly due to bad sectors or other read instability issues then it may lose its current state and become desynchronized with the PC, causing Windows to freeze, crash, and/or drop the USB storage device. USB Stabilizer hardware stands between the PC and storage device, making sure that the connection to the PC is always well-maintained, regardless of the storage device status. This way even if the storage device becomes desynchronized, only its connection to USB Stabilizer hardware will be reinitialized, which will not have any impact on the PC, allowing data recovery to proceed. USB Stabilizer hardware maintains a connection even if the storage device is physically disconnected during recovery. All three editions are the same in this respect.

Prevents Windows from mounting the filesystem (assigning drive letters) of the connected drive, saving time and eliminating the possibility of drive errors causing Windows to hang during mounting. Windows Partition Manager is entirely blocked, making the drive accessible even if critical file system elements, such as the MBR, have turned into bad sectors.

If the source drive stops responding to all commands and all resets, as a last resort USB Stabilizer will automatically repower it to seamlessly continue the data recovery process. With the Tech edition this will happen only after 60 seconds of non-response while with Pro and 10Gb editions the timeframe can be configured to any number of seconds.

All editions of USB Stabilizer allow connecting the source drive to an unpowered USB port, and then turning the power on within the software. This helps ensure that the drive does not spend time being needlessly powered on and reduces the chance of damage from short circuits and electrostatic shock during connection.

USB protocol includes an error reporting system which allows the storage device to communicate general issues to the host, which USB Stabilizer then reports in its log.

The interaction of the precise timings of various state changes of USB protocol with the internal timings of each flash storage device being read can have a significant impact on speed. USB Stabilizer hardware & driver work together to employ a complex algorithm that automatically synchronizes with each flash storage device to reduce the down time between block reads, speeding up reading of healthy areas of flash storage devices.

Degraded flash storage devices, like SATA/NVMe SSDs, require a different read instability handling algorithm than degraded HDDs, relying more heavily on repowering instead of resetting, which all USB Stabilizer editions can be configured to do.

There are absolutely no recurring costs of any kind for any edition of USB Stabilizer, everything works offline, and we always support all of our customers, regardless of the age of their tool. (Please note that R-Studio does have a recurring cost to access new updates.)

A project folder is generated for every drive which contains the sector addresses of every successful read, failed read, write attempt, USB-SCSI errors, as well as the drive’s raw ATA DiskID, SCSI Inquiry Data, SCSI VPD Page Serial Number, and USB Descriptor Serial Number. The drive’s raw SMART logs can also be saved at any time, for example at the beginning and end of the imaging session to record whether the number of reallocations has changed.

Many applications and Windows itself automatically perform read retries after failed read attempts. USB Stabilizer can be instructed to immediately respond with errors when bad sectors or bad blocks are retried. For example, if an application/Windows tries to read a block of sectors, the drive responds with an error, and then that same block is retried one sector at a time then USB Stabilizer can immediately respond to each of those single sector retries with errors without reading the source drive, which greatly speeds up the recovery process at the cost of losing some good sectors within bad blocks.

Shows graphically the status of each read operation on the sector level to communicate whether read commands are successful, timing out, or being responded to with errors. Also shows the graph of the speed, making it easy to identify unexpected speed drops which point to physical degradation of the storage device.

Complex read instability cases require a customized approach to achieve results by setting algorithms in the form "if X happens then do Y.” There is a nearly limitless number of use cases for this. For example, if the drive often fails 10 reads in a row then setting an algorithm “if get 2 consecutive failed reads then skip 8 reads” will increase speed and drive longevity by 5 times. As another example, some storage devices, particularly SSDs, can enter the so-called “offline state” where they immediately respond to all reads with errors without actually trying to read anything and repowering is the only way to force them to exit this state. In this situation the read commands are not timing out, so Tech edition will not be repowering the drive and therefore not handling the case. This problem can be easily solved by setting an algorithm “if get 1 error then repower drive.”

Practically all modern drives start self-scan procedures after as little as 20 seconds of idling. Self-scan is the drive reading its own surface to find bad/weak sectors to reallocate. This is good for maintenance of healthy drives, but once the drive has bad sectors in the user data area, this process causes far too much physical degradation and will likely create new bad sectors faster than it reallocates old ones. More importantly, data will often be lost because most modern drives will still reallocate sectors even if they can’t first retrieve their contents; in other words, weak sectors will be losing their contents without any user input. This option allows turning off source drive power after a customizable period of inactivity, preventing self-scans from happening.

Data can be hidden within Host Protected Area (HPA), Device Configuration Overlay (DCO), and/or Accessible Max Address Configuration (AMAC), which Pro and 10Gb editions can detect and remove to regain access.

Drives which are locked with an ATA Password or a WD Smartware Password can be unlocked with a known password. If the password is unknown then these drives cannot be unlocked.

The 10Gb edition reports the real time current consumption of all connected drives. This is primarily useful for diagnostic purposes. For example, a lack of current consumption indicates electronic failure which must be resolved by swapping the PCB or doing manual electronic repairs. In contrast, unusually high current consumption may indicate that the drive’s TVS diode has failed and must be removed.

While all editions of USB Stabilizer will cut power to the drive in case of a short circuit, only the 10Gb edition will report what happened, making diagnostics much easier. The specific power line is identified, for example the log would say “short circuit on SATA port’s 12V line.”

The 10Gb device offers additional options for configuring its behavior in case the source drive fails to respond after a reset or repower. The options allow sending more resets/repowers after a customizable time frame, and/or aborting the recovery process and turning off the drive. This increases safety for long-term projects that must be left unattended for extended periods of time.

The 10Gb device is able to configure the firmware of supported SATA drives to use an internal read timeout, so that the drive self-aborts all reads which take longer without the necessity to send a reset or repower. This works multiple times faster and can be extremely significant, however this functionality is only supported for older drives as modern drives do not have this firmware feature. The 10Gb device also allows disabling the SMART subsystem on connected SATA drives, which helps lighten the load on the drive as updating SMART logs requires the drive to write to its service area (located on the platters).