Is it normal for a laptop to not provide 12V on SATA?

Commander Kinsey wrote:
On Fri, 31 Jan 2020 22:12:59 -0000, nospam wrote:

In article op.0e918e1rwdg98l@glass, Commander Kinsey
wrote:



Why must 3.5" drives be self powered? The caddy could take the 5V
from USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason. No machine made in the last
decade is limited to 500mA per port.

due to the popularity of bus-powered laptop drives and people wanting
to charge smartphones, tablets and other devices, usb ports began to
source more power than they were required to, as much as 2 amps,
eliminating the need for a y-cable (which was never a good idea to
begin with).

It worked perfectly, why do you say it was a bad idea?

even if the port sources 5v!2a (which is not guaranteed on all
computers), it's still not enough to spin up a 3.5" drive

A 12 year old desktop HDD (3.5") that I have lying on my desk (the one I
plugged into the offending laptop) uses 0.5A @ 12V and 0.45A at 5V, this
equates to 8.25W, or less than the 5Vx2A = 10W that you just specified.
And that's an old and presumably inefficient drive.

There is spinup current on the thing too.

That single-point-spec on the label is not the full story.

Your power delivery solution has to cover all
moments of operation after T=0.

Spinup current varies.

Small drives, like 500GB drives with a single platter,
have the highest 12V current. That's because the manufacturer
assumes it's a boot drive, and instantaneous starting
is required.

Larger capacity 3.5" drives have slightly more modest current draw
at spinup. The spinup might take ten seconds. The max allowed
spinup is thirty five seconds. Some Hitachi drives weigh in at
28 seconds until they're "ready" (not all the prep time is spin
related).

I measured a few with my clamp-on DC ammeter, to note
this trend, that the small capacity drives use the
higher current values. If all the specs were available,
I wouldn't have to resort to my ammeter.

Back when you could download detailed specs, there
was even a "current draw waveform" recorded from an
instrument and included in the PDF. The current draw
is a kind of envelope and the motor controller has
probably been programmed with a certain target
value in mind.

There is a separate chip that does motor control. It
controls acceleration, regulates spindle speed (especially
on drives that accept a sync input from other drives). We
found out about motor controllers, when a certain model
of Maxtor drive started failing. (The motor controller
chip was overheating for some reason that wasn't explained
at the time.)

If I was designing a solution, I would shoot for roughly
12V @ 2A and 5V @ 1A as a power footprint, or roughly 30W.
There was a time when the power footprint was 40W officially,
but those days are over. A lesser number is sufficient today,
but I don't know if this is specified by a standards body.

If you look at the adapter that comes with the 3.5" disk enclosure,
it's a 12V @ 2A adapter, and a linear regulator on the enclosure
controller board makes the 5V rail from the 12V input. This suggests
the adapter might have a surge rating which is not printed on the
adapter, in order for the adapter to "provide 30W+ for a short time".
As otherwise, with precise current limiting, the stupid thing
should tip over if used with a 500GB drive.

Paul

Paul wrote:
Commander Kinsey wrote:
On Fri, 31 Jan 2020 22:12:59 -0000, nospam wrote:

In article op.0e918e1rwdg98l@glass, Commander Kinsey
wrote:



Why must 3.5" drives be self powered? The caddy could take the
5V from USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason. No machine made in the
last decade is limited to 500mA per port.

due to the popularity of bus-powered laptop drives and people wanting
to charge smartphones, tablets and other devices, usb ports began to
source more power than they were required to, as much as 2 amps,
eliminating the need for a y-cable (which was never a good idea to
begin with).

It worked perfectly, why do you say it was a bad idea?

even if the port sources 5v!2a (which is not guaranteed on all
computers), it's still not enough to spin up a 3.5" drive

A 12 year old desktop HDD (3.5") that I have lying on my desk (the one
I plugged into the offending laptop) uses 0.5A @ 12V and 0.45A at 5V,
this equates to 8.25W, or less than the 5Vx2A = 10W that you just
specified. And that's an old and presumably inefficient drive.


This is what they were doing in 2018. It's got a battery in it
to aid with spinup.

https://blog.seagate.com/consumers/t...xternal-drive/

Would that be relying on something like this ? Dunno.

Multi-lane SuperSpeed (USB 3.2 Gen x2) 1.5 A 5 V 7.5 W

It can't be this one.

https://www.amazon.ca/Seagate-Deskto.../dp/B07CQJBSQL

It's this one.

https://www.amazon.ca/Seagate-Innov8...ustomerReviews

One reviewer writes:

"This is a beautiful, well engineered, technically advanced and manufactured external
drive at an amazing price! The Innov8 works with Type-C hosts which advertise
Power Delivery of 1.5A or higher. If the host does not advertise 1.5A or greater,
the Innov8 will not spin up and will flash an error LED signal. That unfortunately
is the case with my setup. I have a USB 3.1 Type C port but it is GEN 1 and supports
only 900ma and won't drive this Seagate. I really wanted it to work but sadly I had
to return it but if you have the right USB 3.1 Gen 2 1.5A or better port you are
good to go! I wish Seagate would manufacture a option to have a "brick" provide the
power and use the USB 3.1 for data only. This would open up the market for this drive
and I would have one today!"

For that one, not only do you need the right connector, you need the right connector.

I believe Seagate bought LaCie some time ago (boutique peripheral design house),
so if anyone has the staff to use the new PD spec (whatever version it is
this week), they do.

Paul

On 01/02/2020 00:46, Commander Kinsey wrote:
On Fri, 31 Jan 2020 23:40:31 -0000, David wrote:

On 31/01/2020 22:03, Commander Kinsey wrote:
On Fri, 31 Jan 2020 21:35:04 -0000, Carlos E.R.
wrote:

On 31/01/2020 21.55, Commander Kinsey wrote:
On Fri, 31 Jan 2020 20:27:04 -0000, nospam
wrote:

In article op.0e9w9oyowdg98l@glass, Commander Kinsey
wrote:

Is it normal for a laptop to not provide 12V on SATA?

yes, since laptop drives don't use 12v.

I take it this is true of all 2.5" drives, eg. SSDs?

yes.

Oh well, the laptop in question now has an external supply to power
the
desktop hard disk I needed to use it with.* There's maybe 12V
somewhere
inside the laptop, but I couldn't be bothered finding out where or
making a
hole in the case to access it.* I thought there was a good chance it
might
not have any 12V, it might be the 19V from the power brick gets
converted to
5V, 3.3V, 1V etc for the components.

external 3.5" drives must be self-powered.
external 2.5" drives can be either bus-powered or self-powered.

Why must 3.5" drives be self powered?* The caddy could take the 5V
from
USB and convert it up to 12V.

Not enough amperage available.

Depends on the drive and the version of USB.* Some USB can give 20 volts
at 5 amps.

You are 100% correct, no matter WHAT 'nospam' claims!

https://www.wikiwand.com/en/USB#/overview

Good reference material there!

HTH

Nospam is Rod Speed, an Aussie.* They're a bit backwards over there.

THAT explains a lot! ;-)

So does this ......

https://www.reddit.com/r/australia/c..._to_rod_speed/

On Sat, 01 Feb 2020 10:46:14 -0000, David wrote:

On 01/02/2020 00:46, Commander Kinsey wrote:
On Fri, 31 Jan 2020 23:40:31 -0000, David wrote:

On 31/01/2020 22:03, Commander Kinsey wrote:
On Fri, 31 Jan 2020 21:35:04 -0000, Carlos E.R.
wrote:

On 31/01/2020 21.55, Commander Kinsey wrote:
On Fri, 31 Jan 2020 20:27:04 -0000, nospam
wrote:

In article op.0e9w9oyowdg98l@glass, Commander Kinsey
wrote:

Is it normal for a laptop to not provide 12V on SATA?

yes, since laptop drives don't use 12v.

I take it this is true of all 2.5" drives, eg. SSDs?

yes.

Oh well, the laptop in question now has an external supply to power
the
desktop hard disk I needed to use it with. There's maybe 12V
somewhere
inside the laptop, but I couldn't be bothered finding out where or
making a
hole in the case to access it. I thought there was a good chance it
might
not have any 12V, it might be the 19V from the power brick gets
converted to
5V, 3.3V, 1V etc for the components.

external 3.5" drives must be self-powered.
external 2.5" drives can be either bus-powered or self-powered.

Why must 3.5" drives be self powered? The caddy could take the 5V
from
USB and convert it up to 12V.

Not enough amperage available.

Depends on the drive and the version of USB. Some USB can give 20 volts
at 5 amps.

You are 100% correct, no matter WHAT 'nospam' claims!

https://www.wikiwand.com/en/USB#/overview

Good reference material there!

HTH

Nospam is Rod Speed, an Aussie. They're a bit backwards over there.

THAT explains a lot! ;-)

So does this ......

https://www.reddit.com/r/australia/c..._to_rod_speed/

ROFL! And I then got distracted reading about some Aussie cyclist moaning because a car dared to overtake him. FFS the car left him about a metre gap. You won't get that if I'm driving.

On Sat, 01 Feb 2020 01:04:55 -0000, Paul wrote:

Commander Kinsey wrote:
On Fri, 31 Jan 2020 22:12:59 -0000, nospam wrote:

In article op.0e918e1rwdg98l@glass, Commander Kinsey
wrote:



Why must 3.5" drives be self powered? The caddy could take the 5V
from USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason. No machine made in the last
decade is limited to 500mA per port.

due to the popularity of bus-powered laptop drives and people wanting
to charge smartphones, tablets and other devices, usb ports began to
source more power than they were required to, as much as 2 amps,
eliminating the need for a y-cable (which was never a good idea to
begin with).

It worked perfectly, why do you say it was a bad idea?

even if the port sources 5v!2a (which is not guaranteed on all
computers), it's still not enough to spin up a 3.5" drive

A 12 year old desktop HDD (3.5") that I have lying on my desk (the one I
plugged into the offending laptop) uses 0.5A @ 12V and 0.45A at 5V, this
equates to 8.25W, or less than the 5Vx2A = 10W that you just specified.
And that's an old and presumably inefficient drive.

There is spinup current on the thing too.

That single-point-spec on the label is not the full story.

Your power delivery solution has to cover all
moments of operation after T=0.

Spinup current varies.

Small drives, like 500GB drives with a single platter,
have the highest 12V current. That's because the manufacturer
assumes it's a boot drive, and instantaneous starting
is required.

Larger capacity 3.5" drives have slightly more modest current draw
at spinup. The spinup might take ten seconds. The max allowed
spinup is thirty five seconds. Some Hitachi drives weigh in at
28 seconds until they're "ready" (not all the prep time is spin
related).

I measured a few with my clamp-on DC ammeter, to note
this trend, that the small capacity drives use the
higher current values. If all the specs were available,
I wouldn't have to resort to my ammeter.

Back when you could download detailed specs, there
was even a "current draw waveform" recorded from an
instrument and included in the PDF. The current draw
is a kind of envelope and the motor controller has
probably been programmed with a certain target
value in mind.

There is a separate chip that does motor control. It
controls acceleration, regulates spindle speed (especially
on drives that accept a sync input from other drives). We
found out about motor controllers, when a certain model
of Maxtor drive started failing. (The motor controller
chip was overheating for some reason that wasn't explained
at the time.)

If I was designing a solution, I would shoot for roughly
12V @ 2A and 5V @ 1A as a power footprint, or roughly 30W.
There was a time when the power footprint was 40W officially,
but those days are over. A lesser number is sufficient today,
but I don't know if this is specified by a standards body.

If you look at the adapter that comes with the 3.5" disk enclosure,
it's a 12V @ 2A adapter, and a linear regulator on the enclosure
controller board makes the 5V rail from the 12V input. This suggests
the adapter might have a surge rating which is not printed on the
adapter, in order for the adapter to "provide 30W+ for a short time".
As otherwise, with precise current limiting, the stupid thing
should tip over if used with a 500GB drive.

A capacitor could aid the spinup. And can't you draw a bit too much current for a while from USB anyway? Just like if I stick a 3A fuse in a plug that runs a 3A motor, the fuse doesn't blow with the 10A surge to start it.

On Sat, 01 Feb 2020 00:46:06 -0000, nospam wrote:

In article op.0e94shobwdg98l@glass, Commander Kinsey
wrote:

Why must 3.5" drives be self powered? The caddy could take the 5V from
USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason.

nope. usb 2, not that it matters since both have the same power specs.

Wow, they didn't bother increasing the power? Ridiculous. But most manufacturers made them much higher current anyway, because by the time we had USB 2, people wanted to charge phones faster than 500mA.

No machine made in the last
decade is limited to 500mA per port.

some still do, although not many.

It would have to be really cheap ****.

due to the popularity of bus-powered laptop drives and people wanting
to charge smartphones, tablets and other devices, usb ports began to
source more power than they were required to, as much as 2 amps,
eliminating the need for a y-cable (which was never a good idea to
begin with).

It worked perfectly, why do you say it was a bad idea?

it doesn't work perfectly in every case, it's non-compliant with the
usb spec and it's basically a hack that risks potential damage to the
host and peripheral.

Bull****. It just means it can take twice th current. It would be the same as wiring a stove to two 13A plugs. You now have 26A.

even if the port sources 5v!2a (which is not guaranteed on all
computers), it's still not enough to spin up a 3.5" drive

A 12 year old desktop HDD (3.5") that I have lying on my desk (the one I
plugged into the offending laptop) uses 0.5A @ 12V and 0.45A at 5V, this
equates to 8.25W, or less than the 5Vx2A = 10W that you just specified. And
that's an old and presumably inefficient drive.

10w is not guaranteed on all hosts. most can source 1a (5w), however,
that too is also not guaranteed.

If it's in the USB spec, the socket must be able to provide it.

your specs sound like idle power,

It's just what it says on the label.

not maximum power draw, in particular
spinning up the drive, which may need 20w or more, and if there is not
enough power to spin up the drive, it won't be particularly useful.

A temporary surge won't matter, just as a motor starting doesn't blow a plug fuse in the mains.

drive makers do not want to deal with compatibility issues, especially
when it comes to potential data loss.

I'd design my caddy to have an optional PSU connector. You try it without, if it doesn't work, plug in the mains. Lots are actually like this.

you also have only 5v from a usb port, and boosting it to 12v would
incur losses, so even in an ideal situation, you still won't have 10w.

Nowadays those losses are minimal.

tl;dr 3.5" drives are not bus-powered.

What is "tl;dr"?

some specs-

western digital green, a very popular drive, ranges from needing 1.2a @
12v for 500 gb up to 1.75a for 4tb, and that's just for the mechanism,
not the bridge.
http://products.wdc.com/library/SpecSheet/ENG/2879-771438.pdf

That's odd, since Green is the low power version, used for Sky TV boxes etc. They run very cool. I don't believe they use anything like that much. It's way more than other drives.

seagate skyhawk needs 1.8-2a @12v to spin-up the drive, depending on
capacity, or about 22-24 watts.
https://www.seagate.com/www-content/...wk-3-5-hddDS19
02-7-1711US-en_US.pdf

usb-c can source more power, but that's not widespread, yet.

I have one on my two year old desktop. It's a red socket.

red is a sleep/charge usb-a port, which remains powered if the computer
is in sleep/standby and may also source more current than the other usb
ports since it's intended for charging.

usb-c is an entirely different connector and does not need a colour.

usb-c on the left, usb-a on the right:

Didn't realise they were two different things. Apparently my board has one of each:
https://www.gigabyte.com/Motherboard...P-rev-10/sp#sp

Anyway, you said "usb-c can source more power, but that's not widespread, yet."
Clearly it is widespread if I have it on a 2 year old board.
And now you're saying it's sleep/charge with more power, not USB-C.
My god it's amazing anything works at all with 10 different sub-standards.
In fact my phone charges faster on it's own mains charger than on a USB port which has more than enough available. Looks like even the phone doesn't understand the specs. (I tested the current draw with a meter: charger supplies 750mA of it's 1A rating, USB 3 port supplies 500mA of it's 900mA rating, WTF?)

https://s.yimg.com/aah/pchcables/3-f...en-1-type-c-ma
le-to-type-a-male-cable-5gbps-2a-4.png

Why didn't you do this? Put the link on one line so I didn't have to paste it together? Wrapping lines to 80 chars is irritating, but wrapping a single word (or URL) is absurd! No word or link should ever split. Who cares if it goes off the edge, at least I can click it.
https://s.yimg.com/aah/pchcables/3-f...5gbps-2a-4.png

On 01/02/2020 00.58, Paul wrote:
Commander Kinsey wrote:
On Fri, 31 Jan 2020 21:35:04 -0000, Carlos E.R.
wrote:

Not enough amperage available.

Depends on the drive and the version of USB.* Some USB can give 20
volts at 5 amps.

The table here looks to be split into two pieces.

https://en.wikipedia.org/wiki/USB_hardware#Power

Low-power device*********************** 100 mA** 5 V** 0.50 W
Low-power SuperSpeed (USB 3.0) device** 150 mA** 5 V** 0.75 W
High-power device********************** 500 mA** 5 V** 2.5 W
High-power SuperSpeed (USB 3.0) device* 900 mA** 5 V** 4.5 W
Multi-lane SuperSpeed (USB 3.2 Gen x2)* 1.5 A*** 5 V** 7.5 W

...* Battery charging or laptop powering solutions +
************************************************** |
************************************************** v

Power Delivery 1.0 Micro-USB************ 3 A*** 20 V*** 60 W
Power Delivery 1.0 Type-A/B************* 5 A*** 20 V*** 100 W
Power Delivery 2.0/3.0 Type-C*********** 5 A*** 20 V*** 100 W

There is one description of a "powering and data" solution
in the text below the table, but it involves multiple
connectors. I guess that's a win of some sort.

I doubt a laptop will have a _data_ USB connector providing high power
to whatever connects there. Laptops are intended to use peripherals that
use as low power as possible (because laptops are designed to run on
batteries), and if the peripherals needs more, they have to come with
their own power supplies.


The way I looked at this was, if the PD spec was going to
be thick and incomprehensible, without a "rationale" section
copied by the media for our consumption, I'd just wait until
some fool built one for sale as proof the capability was there.

If you think that's bad, look at ESATAp, where there's no spec.
There are two "ear" pins and some flavor of connector has USB
pins on it as well. There might be four different connectors
involved (if someone was counting). It's because of this mess,
that my one good computer store carries *zero* ESATA cables,
and has a whole wall full of SATA cables.

https://en.wikipedia.org/wiki/Serial_ATA#eSATAp

** Paul


--
Cheers, Carlos.

On Sat, 01 Feb 2020 12:33:42 -0000, Carlos E.R. wrote:

On 01/02/2020 00.58, Paul wrote:
Commander Kinsey wrote:
On Fri, 31 Jan 2020 21:35:04 -0000, Carlos E.R.
wrote:

Not enough amperage available.

Depends on the drive and the version of USB. Some USB can give 20
volts at 5 amps.

The table here looks to be split into two pieces.

https://en.wikipedia.org/wiki/USB_hardware#Power

Low-power device 100 mA 5 V 0.50 W
Low-power SuperSpeed (USB 3.0) device 150 mA 5 V 0.75 W
High-power device 500 mA 5 V 2.5 W
High-power SuperSpeed (USB 3.0) device 900 mA 5 V 4.5 W
Multi-lane SuperSpeed (USB 3.2 Gen x2) 1.5 A 5 V 7.5 W

... Battery charging or laptop powering solutions +
|
v

Power Delivery 1.0 Micro-USB 3 A 20 V 60 W
Power Delivery 1.0 Type-A/B 5 A 20 V 100 W
Power Delivery 2.0/3.0 Type-C 5 A 20 V 100 W

There is one description of a "powering and data" solution
in the text below the table, but it involves multiple
connectors. I guess that's a win of some sort.

I doubt a laptop will have a _data_ USB connector providing high power
to whatever connects there. Laptops are intended to use peripherals that
use as low power as possible (because laptops are designed to run on
batteries), and if the peripherals needs more, they have to come with
their own power supplies.

A laptop is often used plugged into a power supply. I would prefer if I used a high power device (especially for a short period or while the laptop was on my desk and plugged in), that it provided the proper power output. Also, is it not common to charge your phone off a laptop while out and about? A laptop has a bigger battery than a phone.

Commander Kinsey wrote:


A capacitor could aid the spinup. And can't you draw a bit too much
current for a while from USB anyway? Just like if I stick a 3A fuse in
a plug that runs a 3A motor, the fuse doesn't blow with the 10A surge to
start it.

The ATX supply, if you look at the label on the side of the
supply, only 5VSB @ 2.5A or 5VSB @ 3A is available.

Modern supplies were not designed for the "futuristic"
application of iPad charging.

All the USB ports on a lot of computers, they're connected
to +5VSB, and thus, the power footprint is limited.

+5VSB is preferred, because it supports "keyboard/mouse wakeup".
It might allow a Wifi adapter with "keep-alive" in the MAC,
to continue working when the PC OS is shut down.

You could have some USB ports designated as "high power ports",
but if they're wired to +5V or +12V or whatever, those rails
only run while the computer is running. When the fans go off,
those rails go off.

The Seagate Innov8 drive uses batteries for +12V. A boost
pump converts +5V @ 1.5A to 7.5W of long term power.
The drive cannot draw (long term), more than 7.5W, or
the drive will spin down when the battery is drained.

If you replace the batteries in the Innov8 with a number
of 100F Supercaps in series, yes, you might manage to make
the drive spin up once. All that the usage of Supercaps does,
is "give you a battery with a shorter amp-hour rating".

If you replace the Supercaps with Ultracaps, those have
excellent short-term current flow capability. You can
draw 3000 amps for 30 milliseconds from an Ultracap.
Perhaps causing the shaft of your screwdriver to
disappear into a plasma of molten chromium steel.

Example of an ultracap application:

https://www.windpowerengineering.com...mw-for-60-sec/

You can see the size of the straps in their bank are
pretty significant in size.

The ultracap would be "too much of a good thing" for
usage in a consumer good. It would give you better
"quick off the line" behavior on an electric car perhaps.

But as long as the long-term input power is limited
to 7.5W, these fantasies about energy storage are
for naught. If you made the lithium batteries large
enough, you might run the drive for the entire session
off battery. But now the unit is getting larger than
most people would want. Volumetrically, the lithium battery
has an advantage over a supercap, in this sort of
application. The Lithium battery may not have a very
high "peak current draw" rating, but it's got the
Ah or Wh advantage.

I'll stick with a wall wart, which isn't likely to
wear out or malfunction. (None of mine have so far.
They get plugged into the UPS when I use them.)

Paul

On Sat, 01 Feb 2020 18:31:23 -0000, Paul wrote:

Commander Kinsey wrote:


A capacitor could aid the spinup. And can't you draw a bit too much
current for a while from USB anyway? Just like if I stick a 3A fuse in
a plug that runs a 3A motor, the fuse doesn't blow with the 10A surge to
start it.

The ATX supply, if you look at the label on the side of the
supply, only 5VSB @ 2.5A or 5VSB @ 3A is available.

Modern supplies were not designed for the "futuristic"
application of iPad charging.

All the USB ports on a lot of computers, they're connected
to +5VSB, and thus, the power footprint is limited.

+5VSB is preferred, because it supports "keyboard/mouse wakeup".
It might allow a Wifi adapter with "keep-alive" in the MAC,
to continue working when the PC OS is shut down.

Is it not possible to connect them to either? I.e. when the computer is powered up, the normal 5V rail is connected also?

It's rather absurd a modern PSU only has a few amps available on standby.

I've never encountered it anyway, as a desktop is normally on when you're fast charging something off it. If it's overnight when it's off, then you don't need high current for a slow charge.

You could have some USB ports designated as "high power ports",
but if they're wired to +5V or +12V or whatever, those rails
only run while the computer is running. When the fans go off,
those rails go off.

The Seagate Innov8 drive uses batteries for +12V. A boost
pump converts +5V @ 1.5A to 7.5W of long term power.
The drive cannot draw (long term), more than 7.5W, or
the drive will spin down when the battery is drained.

If you replace the batteries in the Innov8 with a number
of 100F Supercaps in series, yes, you might manage to make
the drive spin up once. All that the usage of Supercaps does,
is "give you a battery with a shorter amp-hour rating".

If you replace the Supercaps with Ultracaps, those have
excellent short-term current flow capability. You can
draw 3000 amps for 30 milliseconds from an Ultracap.
Perhaps causing the shaft of your screwdriver to
disappear into a plasma of molten chromium steel.

Example of an ultracap application:

https://www.windpowerengineering.com...mw-for-60-sec/

You can see the size of the straps in their bank are
pretty significant in size.

The ultracap would be "too much of a good thing" for
usage in a consumer good. It would give you better
"quick off the line" behavior on an electric car perhaps.

I think the huge bank of Li-Ion batteries in cars is quite enough. You're limited by the motor.

But as long as the long-term input power is limited
to 7.5W, these fantasies about energy storage are
for naught. If you made the lithium batteries large
enough, you might run the drive for the entire session
off battery. But now the unit is getting larger than
most people would want. Volumetrically, the lithium battery
has an advantage over a supercap, in this sort of
application. The Lithium battery may not have a very
high "peak current draw" rating, but it's got the
Ah or Wh advantage.

I'll stick with a wall wart, which isn't likely to
wear out or malfunction. (None of mine have so far.
They get plugged into the UPS when I use them.)

Hmph, I've worn out loads. Some are pretty **** quality, those are the ones that get rather hot when in use. I've got one powering an old monitor here - it must be cold cathode lit because the monitor gets damn hot at the back. Its own power supply was also damn hot, and gave up. I bought a generic supply with twice the current that it needed, and that's damn hot. They never think to put vents on these things....

On 01/02/2020 11:40, Commander Kinsey wrote:
On Sat, 01 Feb 2020 10:46:14 -0000, David wrote:

On 01/02/2020 00:46, Commander Kinsey wrote:
On Fri, 31 Jan 2020 23:40:31 -0000, David wrote:

On 31/01/2020 22:03, Commander Kinsey wrote:
On Fri, 31 Jan 2020 21:35:04 -0000, Carlos E.R.
wrote:

On 31/01/2020 21.55, Commander Kinsey wrote:
On Fri, 31 Jan 2020 20:27:04 -0000, nospam
wrote:

In article op.0e9w9oyowdg98l@glass, Commander Kinsey
wrote:

Is it normal for a laptop to not provide 12V on SATA?

yes, since laptop drives don't use 12v.

I take it this is true of all 2.5" drives, eg. SSDs?

yes.

Oh well, the laptop in question now has an external supply to
power
the
desktop hard disk I needed to use it with.* There's maybe 12V
somewhere
inside the laptop, but I couldn't be bothered finding out where or
making a
hole in the case to access it.* I thought there was a good
chance it
might
not have any 12V, it might be the 19V from the power brick gets
converted to
5V, 3.3V, 1V etc for the components.

external 3.5" drives must be self-powered.
external 2.5" drives can be either bus-powered or self-powered.

Why must 3.5" drives be self powered?* The caddy could take the 5V
from
USB and convert it up to 12V.

Not enough amperage available.

Depends on the drive and the version of USB.* Some USB can give 20
volts
at 5 amps.

You are 100% correct, no matter WHAT 'nospam' claims!

https://www.wikiwand.com/en/USB#/overview

Good reference material there!

HTH

Nospam is Rod Speed, an Aussie.* They're a bit backwards over there.

THAT explains a lot! ;-)

So does this ......

https://www.reddit.com/r/australia/c..._to_rod_speed/


ROFL!* And I then got distracted reading about some Aussie cyclist
moaning because a car dared to overtake him.* FFS the car left him about
a metre gap.* You won't get that if I'm driving.

I get distracted too - that's half the fun of playing on Usenet! ;-)

Have you ever been involved in a road traffic accident?

--
David

In article op.0fa4wuxhwdg98l@glass, Commander Kinsey
wrote:

Why must 3.5" drives be self powered? The caddy could take the 5V
from
USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason.

nope. usb 2, not that it matters since both have the same power specs.

Wow, they didn't bother increasing the power? Ridiculous. But most
manufacturers made them much higher current anyway, because by the time we
had USB 2, people wanted to charge phones faster than 500mA.

the usb 2 spec was approved in 2000 and appeared on hardware shortly
thereafter. by 2001 or 2002, pretty much everything was usb 2.

charging phones and other devices became common many years later,
starting in the late 2000s, and much more so in the early 2010s when
smartphones and tablets became ubiquitous, the latter of which *really*
needs more than 500ma.

No machine made in the last
decade is limited to 500mA per port.

some still do, although not many.

It would have to be really cheap ****.

yep, and there are such products.

some people even seek them out.

due to the popularity of bus-powered laptop drives and people wanting
to charge smartphones, tablets and other devices, usb ports began to
source more power than they were required to, as much as 2 amps,
eliminating the need for a y-cable (which was never a good idea to
begin with).

It worked perfectly, why do you say it was a bad idea?

it doesn't work perfectly in every case, it's non-compliant with the
usb spec and it's basically a hack that risks potential damage to the
host and peripheral.

Bull****. It just means it can take twice th current. It would be the same
as wiring a stove to two 13A plugs. You now have 26A.

host ports are not designed to be connected together and it's also
non-compliant with the spec.

https://compliance.usb.org/index.asp?UpdateFile=Policies#72
Use of a 'Y' cable (a cable with two A-plugs) is prohibited on any
USB peripheral. If a USB peripheral requires more power than allowed
by the USB specification to which it is designed, then it must be
self-powered.

even if the port sources 5v!2a (which is not guaranteed on all
computers), it's still not enough to spin up a 3.5" drive

A 12 year old desktop HDD (3.5") that I have lying on my desk (the one I
plugged into the offending laptop) uses 0.5A @ 12V and 0.45A at 5V, this
equates to 8.25W, or less than the 5Vx2A = 10W that you just specified.
And
that's an old and presumably inefficient drive.

10w is not guaranteed on all hosts. most can source 1a (5w), however,
that too is also not guaranteed.

If it's in the USB spec, the socket must be able to provide it.

the usb spec guarantees only 100ma.

higher currents are on request, which can be denied if the host cannot
provide it.

typically, providing 500ma isn't an issue, but that's not always the
case.

your specs sound like idle power,

It's just what it says on the label.

idle power doesn't matter if there's not enough power to spin up the
drive.

not maximum power draw, in particular
spinning up the drive, which may need 20w or more, and if there is not
enough power to spin up the drive, it won't be particularly useful.

A temporary surge won't matter, just as a motor starting doesn't blow a plug
fuse in the mains.

it does matter, and a motor start can easily blow a fuse due to the
surge.

drive makers do not want to deal with compatibility issues, especially
when it comes to potential data loss.

I'd design my caddy to have an optional PSU connector. You try it without,
if it doesn't work, plug in the mains. Lots are actually like this.

connecting it to mains power means it's no longer bus-powered.

note that there are 3.5" bus-powered enclosures that have an auxiliary
battery for spin-up, which allows it to work on any usb port.

https://blog.seagate.com/consumers/t...wered-desktop-
external-drive/
On top of the power from a USB-C computer, an internal battery gives
the drive a boost during its spin-up phase.
....
³One of the biggest drawbacks of these high-capacity drives is that
they take so much power. You need to plug them into the wall.
....
Early on, when the drive is still cool, the battery supplies
supplemental power to the hard drive, adds Phil Yin, a
principal hardware technologist.

³But once the drive has warmed up, its power requirement is
reduced and the host system is able to recharge the battery,²
explains Yin.

you also have only 5v from a usb port, and boosting it to 12v would
incur losses, so even in an ideal situation, you still won't have 10w.

Nowadays those losses are minimal.

the laws of physics haven't changed.

tl;dr 3.5" drives are not bus-powered.

What is "tl;dr"?

https://www.urbandictionary.com/define.php?term=tl%3Bdr
A) "Too long; didn't read.", meaning a post, article, or anything
with words was too long, and whoever used the phrase didn't read
it for that reason.

B) Also used by someone who wrote a large posts/article/whatever to
show a brief summary of their post as it might be too long.

some specs-

western digital green, a very popular drive, ranges from needing 1.2a @
12v for 500 gb up to 1.75a for 4tb, and that's just for the mechanism,
not the bridge.
http://products.wdc.com/library/SpecSheet/ENG/2879-771438.pdf

That's odd, since Green is the low power version, used for Sky TV boxes etc.
They run very cool. I don't believe they use anything like that much. It's
way more than other drives.

idle isn't that much power, however, spin-up definitely is.

seagate skyhawk needs 1.8-2a @12v to spin-up the drive, depending on
capacity, or about 22-24 watts.
https://www.seagate.com/www-content/...wk-3-5-hddDS19
02-7-1711US-en_US.pdf

usb-c can source more power, but that's not widespread, yet.

I have one on my two year old desktop. It's a red socket.

red is a sleep/charge usb-a port, which remains powered if the computer
is in sleep/standby and may also source more current than the other usb
ports since it's intended for charging.

usb-c is an entirely different connector and does not need a colour.

usb-c on the left, usb-a on the right:

Didn't realise they were two different things. Apparently my board has one
of each:
https://www.gigabyte.com/Motherboard...P-rev-10/sp#sp

Anyway, you said "usb-c can source more power, but that's not widespread,
yet."
Clearly it is widespread if I have it on a 2 year old board.

usb-a first appeared about 25 years ago and is just about everywhere
now, not just computers, but also in cars, mains outlets, furniture and
much more.

usb-c first appeared about 4-5 years ago and has a *long* way to go to
catch up.

https://ae01.alicdn.com/kf/HTB1lgkGX...2-1A-USB-Wall-
Socket-for-US-UK-AU-Universal-Plugs-Dual-USB-Port-Charger-High.jpg

https://m.media-amazon.com/images/I/610iC9sVh9L._SR500,500_.jpg

https://www.homecrux.com/pons-easy-t...es-with-usb-po
rts-for-charging/64070/
https://www.myfurnitureplace.com/pro...-sofa-and-love
seat-with-usb-ports

https://blogmedia.dealerfire.com/wp-.../188/2016/10/2
017-Nissan-Versa-B8_o.jpg
https://cf-images.us-east-1.prod.bol...40004208001/16
1269bd-60ac-4f46-9898-18364895a69d/a78c6fef-6475-4c19-8344-9caf655d9fd3/
1280x720/match/image.jpg

And now you're saying it's sleep/charge with more power, not USB-C.
My god it's amazing anything works at all with 10 different sub-standards.
In fact my phone charges faster on it's own mains charger than on a USB port
which has more than enough available. Looks like even the phone doesn't
understand the specs. (I tested the current draw with a meter: charger
supplies 750mA of it's 1A rating, USB 3 port supplies 500mA of it's 900mA
rating, WTF?)

https://s.yimg.com/aah/pchcables/3-f...en-1-type-c-ma
le-to-type-a-male-cable-5gbps-2a-4.png

Why didn't you do this? Put the link on one line so I didn't have to paste
it together? Wrapping lines to 80 chars is irritating, but wrapping a single word (or URL) is absurd! No word or link should ever split. Who cares if it goes off the edge, at least I can click it.
https://s.yimg.com/aah/pchcables/3-f...5gbps-2a-4.png

using delimiters eliminates any problems with embedded whitespace,
including line breaks and spaces.

the url should be directly clickable in the newsreader. if not, find a
better newsreader.

your example url does not have delimiters and can break.

this is an extreme example, which is still clickable and works fine
even with the extra added whitespace:
https://s.yimg.com/aah/pchc
ables/3-foot-usb-3-0-usb-3-1-gen-
1-type-c-ma
le-to-type-a-male-cable-5gbps-2a-4.png

On 2/1/20 5:04 PM, nospam wrote:
In article op.0fa4wuxhwdg98l@glass, Commander Kinsey
wrote:

Why must 3.5" drives be self powered? The caddy could take the 5V
from
USB
and convert it up to 12V.

3.5" drives need more current than a usb port can provide.

I've seen a drive with two USB plugs for that very reason.

that's not the reason.

the y-cables are for bus-powered laptop drives and standard usb ports,
which can source up to 500ma. by using a y-cable to connect two usb
ports, the drive can theoretically get 1000ma, or 1a, enough to spin it
up (idle doesn't need as much).

You're talking about USB 1 for some reason.

nope. usb 2, not that it matters since both have the same power specs.

Wow, they didn't bother increasing the power? Ridiculous. But most
manufacturers made them much higher current anyway, because by the time we
had USB 2, people wanted to charge phones faster than 500mA.

the usb 2 spec was approved in 2000 and appeared on hardware shortly
thereafter. by 2001 or 2002, pretty much everything was usb 2.

No. There were many things other than USB 2. For example, if you ordered
a hard chicken taco it could come with a shell, lettuce, tomato,
chicken, a sauce and maybe more. But never USB 2.






Just being silly, of course.


--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.

On 02/02/2020 00:14, Snit wrote:
Just being silly, of course.

Haha! :-) I just LOVE the way you run rings around 'nospam'.

If he's REALLY an Aussie, he'll get cross with you. I used to
think my mates over there were well-balanced - until I realised
that they had a chip on EACH shoulder! ;-)

On 2/1/20 5:18 PM, David wrote:
On 02/02/2020 00:14, Snit wrote:
Just being silly, of course.

Haha! :-)Â* I just LOVE the way you run rings around 'nospam'.

If he's REALLY an Aussie, he'll get cross with you. I used to
think my mates over there were well-balanced - until I realised
that they had a chip on EACH shoulder! ;-)



He is an interesting fellow.

--
Personal attacks from those who troll show their own insecurity. They
cannot use reason to show the message to be wrong so they try to feel
somehow superior by attacking the messenger.

They cling to their attacks and ignore the message time and time again.