Is it me or are there many unreliable USB flash drives?

In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Ant
wrote:


Fry right away or after a while?

usually quickly, but it depends how much over 5v it is. chances are
that unless you bought a noname charger, the usb host is within spec.

Well, the USB ports are inside the computers. ;P

Do they get hot in one computer but not another?

--
__ __
#_ |\| | _#

In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev wrote:
In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Ant
wrote:


Fry right away or after a while?

usually quickly, but it depends how much over 5v it is. chances are
that unless you bought a noname charger, the usb host is within spec.

Well, the USB ports are inside the computers. ;P

Do they get hot in one computer but not another?

The memory sticks, that is.

--
__ __
#_ |\| | _#

In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev wrote:
In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev wrote:
In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Ant
wrote:


Fry right away or after a while?

usually quickly, but it depends how much over 5v it is. chances are
that unless you bought a noname charger, the usb host is within spec.

Well, the USB ports are inside the computers. ;P

Do they get hot in one computer but not another?

The memory sticks, that is.

Correct in memory sticks in all computers.
--
Quote of the Week: "I really believe I've been a good person. Not
perfect - forget about perfect - but just learning by what I was taught
and living by my own values. I might have stepped on a few ants - and a
few other things as well - but I've never hurt anybody." --Kiri Te
Kanawa
Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
/\___/\ Ant(Dude) @ http://antfarm.home.dhs.org (Personal Web Site)
/ /\ /\ \ Ant's Quality Foraged Links: http://aqfl.net
| |o o| |
\ _ / Please nuke ANT if replying by e-mail privately. If credit-
( ) ing, then please kindly use Ant nickname and AQFL URL/link.

In article , Ant
wrote:

In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev
wrote:
In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev
wrote:
In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Ant
wrote:

Fry right away or after a while?

usually quickly, but it depends how much over 5v it is. chances are
that unless you bought a noname charger, the usb host is within spec.

Well, the USB ports are inside the computers. ;P

Do they get hot in one computer but not another?

The memory sticks, that is.

Correct in memory sticks in all computers.

There's your problem ... "Memory Sticks" should only be used in Sony
devices. ;-)

Computer Nerd Kev wrote
nospam wrote
Computer Nerd Kev wrote

There are USB voltage meters sold cheaply on Ebay to
show if the USB voltage from a computer is too high/low.
If the USB voltage was far too high, the extra energy
would cause excessive heating of the drive.

if the voltage was too high, not only would it not
be usb compliant, but it would likely fry anything
plugged into it, which is expecting a regulated 5v.

I don't mean by design (although anything's possible
with Chinese knock-off stuff, if that's involved), I mean
due to a failure of the computer hardware.

Now the 5V at the USB ports being high would often only make sense
if they were supplied by a separate line from the power supply.

They are in desktops. That's the whole point of the +5V standby line.

If the internal computer components and the USB ports were
both subjected to a failure of the 5V voltage regulation, it
would be very likely to destroy the computer without the
user having any time to worry about their USB accessories.

Depends on how much its out by. Certainly if its out by enough
to see what normally stays at room temp get very hot when
doing nothing with just the USB stick plugged in, doing nothing.

Most desktop PCs have standard power supplies that
provide the USB 5V from the same lines as everything else,

That is just plain wrong. Its usually supplied by the +5V standby
line, so you can wake the system by moving the mouse etc.

so it is unlikely that an over-voltage situation
would only affect the USB ports.

You have that completely backwards.

A laptop, however, may have a separate power supply circuit
for the USB 5V. If this failed, and supplied too high a voltage, it
may not sevierely affect the normal operation of the computer.

In fact laptops are more likely to have a
separate +5V line for the same reason.

A power-boosted USB hub with a faulty or incorrectly
specified mains adapter powering it could also be a culprit.

the amount of current sourced is limited by
what the device negotiates from the host.

Current limiting only protects against an over-voltage situation
if the device requires more than the minimum current limit (500mA)
to be damaged. If an IC in the memory stick normally draws 20mA
at 5V, it may draw 28mA at 7V (this based on rough calculations:
5V / 0.02A = 250R 7V / 250R = 0.028A, but it proves the point),
however the IC may only be rated to sustain 5.5V, so it may be
damaged while only consuming 28mA. The rest of the components
in the memory stick will also be drawing additional power, but
you see that the total power draw is not increased enough to
hit the 500mA maximum if the stick normally draws, say, 100mA?

Another possibility, if you only ever use one USB port (or a group
of similarly affected ones) for testing these memory sticks, is
that a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector. Here, not only
the heat, but also the coresponding reduced power to the memory
stick, may induce irrecoverable write errors or other failures.

unlikely.

Such failures are known to occour due to strain on solder joints
causing them to break, or joints that weren't correctly soldered
in the first place. It is a less common failure, but it's also
uncommon to consume large quantities of memory sticks.

But doesn’t explain why all the sticks get hot.

In article , Computer Nerd Kev
wrote:

There are USB voltage meters sold cheaply on Ebay to show if the
USB voltage from a computer is too high/low. If the USB voltage
was far too high, the extra energy would cause excessive heating
of the drive.

if the voltage was too high, not only would it not be usb compliant,
but it would likely fry anything plugged into it, which is expecting a
regulated 5v.

I don't mean by design (although anything's possible with Chinese
knock-off stuff, if that's involved), I mean due to a failure of
the computer hardware.

Now the 5V at the USB ports being high would often only make sense
if they were supplied by a separate line from the power supply. If
the internal computer components and the USB ports were both
subjected to a failure of the 5V voltage regulation, it would
be very likely to destroy the computer without the user having
any time to worry about their USB accessories.

if the internal power supply failed then computer wouldn't work, not
just the ports.

Most desktop PCs have standard power supplies that provide the
USB 5V from the same lines as everything else, so it is unlikely
that an over-voltage situation would only affect the USB ports.

exactly the point.

A laptop, however, may have a separate power supply circuit for
the USB 5V. If this failed, and supplied too high a voltage, it
may not sevierely affect the normal operation of the computer.

no.

A power-boosted USB hub with a faulty or incorrectly specified
mains adapter powering it could also be a culprit.

no hub was used.

the amount of current sourced is limited by what the device
negotiates from the host.

Current limiting only protects against an over-voltage situation
if the device requires more than the minimum current limit (500mA)
to be damaged. If an IC in the memory stick normally draws 20mA
at 5V, it may draw 28mA at 7V (this based on rough calculations:
5V / 0.02A = 250R 7V / 250R = 0.028A, but it proves the point),
however the IC may only be rated to sustain 5.5V, so it may be
damaged while only consuming 28mA. The rest of the components
in the memory stick will also be drawing additional power, but
you see that the total power draw is not increased enough to
hit the 500mA maximum if the stick normally draws, say, 100mA?

usb devices don't work that way.

Another possibility, if you only ever use one USB port (or a group
of similarly affected ones) for testing these memory sticks, is
that a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector. Here, not only
the heat, but also the coresponding reduced power to the memory
stick, may induce irrecoverable write errors or other failures.

unlikely.

Such failures are known to occour due to strain on solder joints
causing them to break, or joints that weren't correctly soldered
in the first place.

broken joints don't get hot. they simply don't work.

It is a less common failure, but it's also
uncommon to consume large quantities of memory sticks.

it's not uncommon at all. memory sticks are cheap enough to be
considered disposable. sometimes they're even given away for free.

In comp.sys.ibm.pc.hardware.storage Rod Speed wrote:
Computer Nerd Kev wrote
nospam wrote
Computer Nerd Kev wrote

There are USB voltage meters sold cheaply on Ebay to
show if the USB voltage from a computer is too high/low.
If the USB voltage was far too high, the extra energy
would cause excessive heating of the drive.

if the voltage was too high, not only would it not
be usb compliant, but it would likely fry anything
plugged into it, which is expecting a regulated 5v.

I don't mean by design (although anything's possible
with Chinese knock-off stuff, if that's involved), I mean
due to a failure of the computer hardware.

Now the 5V at the USB ports being high would often only make sense
if they were supplied by a separate line from the power supply.

They are in desktops. That's the whole point of the +5V standby line.

No it isn't. The 5V Standby line is to enable the motherboard
hardware to detect when the power switch is pressed, and on
many motherboards it is used to power additional functions
while the computer is off as well.

The Standby line was only specified to 10mA before the ATX 2.2
spec. and 2A afterwards. That means that an ATX 2.2 PSU could
only supply power for four USB 2 ports at 500mA maximum output,
or two USB 3 ports as 900mA maximum output. Before ATX 2.2 no
USB port could have been provided with remotely enough power
from a PSU meeting the 10mA 5V standby minimum requirement.

http://pinouts.ru/Power/atx_v2_pinout.shtml

I know some PCs supply power to USB devices when off (I find
this a pain), but it's not the case in many designs, and the
required use of the 5V standby line by the motherboard's
circuitry means that damaging voltage on this line will affect
the computer at the same time as the USB devices, leaving the
user with more to worry about than just their memory sticks
dying.

If the internal computer components and the USB ports were
both subjected to a failure of the 5V voltage regulation, it
would be very likely to destroy the computer without the
user having any time to worry about their USB accessories.

Depends on how much its out by. Certainly if its out by enough
to see what normally stays at room temp get very hot when
doing nothing with just the USB stick plugged in, doing nothing.

The maximum voltage ratings would probably only be known by the
memory stick IC manufacturers and their clients, however 5.5V
is a common design maximum. 5.25V is the maximum PSU output
under the ATX specification.

Most desktop PCs have standard power supplies that
provide the USB 5V from the same lines as everything else,

That is just plain wrong. Its usually supplied by the +5V standby
line, so you can wake the system by moving the mouse etc.

so it is unlikely that an over-voltage situation
would only affect the USB ports.

You have that completely backwards.

A laptop, however, may have a separate power supply circuit
for the USB 5V. If this failed, and supplied too high a voltage, it
may not sevierely affect the normal operation of the computer.

In fact laptops are more likely to have a
separate +5V line for the same reason.

A power-boosted USB hub with a faulty or incorrectly
specified mains adapter powering it could also be a culprit.

the amount of current sourced is limited by
what the device negotiates from the host.

Current limiting only protects against an over-voltage situation
if the device requires more than the minimum current limit (500mA)
to be damaged. If an IC in the memory stick normally draws 20mA
at 5V, it may draw 28mA at 7V (this based on rough calculations:
5V / 0.02A = 250R 7V / 250R = 0.028A, but it proves the point),
however the IC may only be rated to sustain 5.5V, so it may be
damaged while only consuming 28mA. The rest of the components
in the memory stick will also be drawing additional power, but
you see that the total power draw is not increased enough to
hit the 500mA maximum if the stick normally draws, say, 100mA?

Another possibility, if you only ever use one USB port (or a group
of similarly affected ones) for testing these memory sticks, is
that a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector. Here, not only
the heat, but also the coresponding reduced power to the memory
stick, may induce irrecoverable write errors or other failures.

unlikely.

Such failures are known to occour due to strain on solder joints
causing them to break, or joints that weren't correctly soldered
in the first place. It is a less common failure, but it's also
uncommon to consume large quantities of memory sticks.

But doesn?t explain why all the sticks get hot.

"a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector."

The socket is the bit in the computer.

If the same socket/s are used, all the USB memory sticks drawing
the same, or similar, currents will get hotter than normal. They
will also receive less voltage than otherwise due to the voltage
drop over the resistance of the failed solder joint - which is
waht causes the heat.

--
__ __
#_ |\| | _#

In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev wrote:
In comp.sys.ibm.pc.hardware.storage Computer Nerd Kev wrote:
In comp.sys.ibm.pc.hardware.storage Ant wrote:
In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Ant
wrote:


Fry right away or after a while?

usually quickly, but it depends how much over 5v it is. chances are
that unless you bought a noname charger, the usb host is within spec.

Well, the USB ports are inside the computers. ;P

Do they get hot in one computer but not another?

The memory sticks, that is.

Correct in memory sticks in all computers.

If the USB insert preferred term sticks get hot in different
computers then the failure isn't in the computer after all. unless
the stress under the 5V high voltage has causes parts of them
to short and now they're heating up at normal voltage, but
it's unlikely that different models would all fail in that
way.

--
__ __
#_ |\| | _#

In comp.sys.ibm.pc.hardware.storage nospam wrote:
In article , Computer Nerd Kev
wrote:

There are USB voltage meters sold cheaply on Ebay to show if the
USB voltage from a computer is too high/low. If the USB voltage
was far too high, the extra energy would cause excessive heating
of the drive.

if the voltage was too high, not only would it not be usb compliant,
but it would likely fry anything plugged into it, which is expecting a
regulated 5v.

I don't mean by design (although anything's possible with Chinese
knock-off stuff, if that's involved), I mean due to a failure of
the computer hardware.

Now the 5V at the USB ports being high would often only make sense
if they were supplied by a separate line from the power supply. If
the internal computer components and the USB ports were both
subjected to a failure of the 5V voltage regulation, it would
be very likely to destroy the computer without the user having
any time to worry about their USB accessories.

if the internal power supply failed then computer wouldn't work, not
just the ports.

Most desktop PCs have standard power supplies that provide the
USB 5V from the same lines as everything else, so it is unlikely
that an over-voltage situation would only affect the USB ports.

exactly the point.

A laptop, however, may have a separate power supply circuit for
the USB 5V. If this failed, and supplied too high a voltage, it
may not sevierely affect the normal operation of the computer.

no.

Yes. If the 5V supply circuit was built separately for use by
the relatively high current USB devices, it's failure (in this
way) would not stop the computer operating. In some cases it
might (I won't spend the time spelling it out), but if no USB
devices are connected, it won't.

A power-boosted USB hub with a faulty or incorrectly specified
mains adapter powering it could also be a culprit.

no hub was used.

Known now, yes.

the amount of current sourced is limited by what the device
negotiates from the host.

Current limiting only protects against an over-voltage situation
if the device requires more than the minimum current limit (500mA)
to be damaged. If an IC in the memory stick normally draws 20mA
at 5V, it may draw 28mA at 7V (this based on rough calculations:
5V / 0.02A = 250R 7V / 250R = 0.028A, but it proves the point),
however the IC may only be rated to sustain 5.5V, so it may be
damaged while only consuming 28mA. The rest of the components
in the memory stick will also be drawing additional power, but
you see that the total power draw is not increased enough to
hit the 500mA maximum if the stick normally draws, say, 100mA?

usb devices don't work that way.

OK, so the device can specify roughly the current it needs via
the USB protocol, which if you're lucky will be respected by the
computer hardware. The current specification isn't accurate
enough to prevent damage though, for the reason shown above.
The current increase as voltage is increased isn't great enough
to trip the limiting.

Another possibility, if you only ever use one USB port (or a group
of similarly affected ones) for testing these memory sticks, is
that a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector. Here, not only
the heat, but also the coresponding reduced power to the memory
stick, may induce irrecoverable write errors or other failures.

unlikely.

Such failures are known to occour due to strain on solder joints
causing them to break, or joints that weren't correctly soldered
in the first place.

broken joints don't get hot. they simply don't work.

Sometimes they simply form a high resistance. Look it up - dry
joints.

It is a less common failure, but it's also
uncommon to consume large quantities of memory sticks.

it's not uncommon at all. memory sticks are cheap enough to be
considered disposable. sometimes they're even given away for free.

As I remember it, OP figures that he's bought good ones - hence
his confusion about the cause of the failure rate.

--
__ __
#_ |\| | _#

In article , Computer Nerd Kev
wrote:


A laptop, however, may have a separate power supply circuit for
the USB 5V. If this failed, and supplied too high a voltage, it
may not sevierely affect the normal operation of the computer.

no.

Yes. If the 5V supply circuit was built separately for use by
the relatively high current USB devices, it's failure (in this
way) would not stop the computer operating. In some cases it
might (I won't spend the time spelling it out), but if no USB
devices are connected, it won't.

usb devices are not 'relatively high current'.

usb originally specified a maximum of 500ma. the usb charging spec can
go to about 2a, and at 5v, is just 10w.

A power-boosted USB hub with a faulty or incorrectly specified
mains adapter powering it could also be a culprit.

no hub was used.

Known now, yes.

it was known before.

the amount of current sourced is limited by what the device
negotiates from the host.

Current limiting only protects against an over-voltage situation
if the device requires more than the minimum current limit (500mA)
to be damaged. If an IC in the memory stick normally draws 20mA
at 5V, it may draw 28mA at 7V (this based on rough calculations:
5V / 0.02A = 250R 7V / 250R = 0.028A, but it proves the point),
however the IC may only be rated to sustain 5.5V, so it may be
damaged while only consuming 28mA. The rest of the components
in the memory stick will also be drawing additional power, but
you see that the total power draw is not increased enough to
hit the 500mA maximum if the stick normally draws, say, 100mA?

usb devices don't work that way.

OK, so the device can specify roughly the current it needs via
the USB protocol, which if you're lucky will be respected by the
computer hardware.

luck has absolutely *nothing* to do with it.

the usb spec *requires* that a usb peripheral device use *up* *to*
100ma and then negotiate higher current requirements, which the usb
host can deny.

anything else is not compliant with the usb spec.

The current specification isn't accurate
enough to prevent damage though, for the reason shown above.
The current increase as voltage is increased isn't great enough
to trip the limiting.

no.

Another possibility, if you only ever use one USB port (or a group
of similarly affected ones) for testing these memory sticks, is
that a high resistance in the computer's USB socket is causing
the _socket_ to heat up, and the heat is being transferred
to the drives themselves via the USB connector. Here, not only
the heat, but also the coresponding reduced power to the memory
stick, may induce irrecoverable write errors or other failures.

unlikely.

Such failures are known to occour due to strain on solder joints
causing them to break, or joints that weren't correctly soldered
in the first place.

broken joints don't get hot. they simply don't work.

Sometimes they simply form a high resistance. Look it up - dry
joints.

they're commonly called cold solder joints, and will cause a device to
fail not get hot.

It is a less common failure, but it's also
uncommon to consume large quantities of memory sticks.

it's not uncommon at all. memory sticks are cheap enough to be
considered disposable. sometimes they're even given away for free.

As I remember it, OP figures that he's bought good ones - hence
his confusion about the cause of the failure rate.

it wasn't due to heat.