pjp wrote:
In article ,
lid says...
On Thu, 17 Jan 2019 08:30:32 +1100, Peter Jason wrote:

I am processing movies and the CPU is indicating
100%.

Where can I see the temperature of the CPU?
Speccy tells me that my Intel Core i5 4690 is sitting on 37° C

Seeing as we are talking about temps.

I use an external usb cable allows me to connect either IDE or Sata
drive to it and it shows up as a removable disk same as any other
external. I have more than a few hard disks sit on a shelve I use for
various purposes mostly arching data.

One 1Tb drive gets very warm, I'd actually say hot in fact. I've seen it
over 60 degrees C without doing anything but plugged in. Drive operates
normally without any funny noises or anything like that. SMART data
indicates nothing amiss with the drive. I can leave it plugged in
overnight and nothing seems changed in it's behaviour, e.g. it works as
expected.

Any thoughts?

With your clamp-on DC ammeter, you'd check which rail
has abnormal power consumption. A high +5V consumption
would mean a problem with the controller board. A high +12V
could be a motor problem.

With an infrared camera, you'd check the thermal profile and
see if you can determine where the heat originates.

A dry motor bearing could cause increased friction. Normally,
there would be sound effects to go with it. Maybe a "squeak"
at startup or shutdown. And the high temperature of the drive,
will work to evaporate fluid from the motor bearing, so the
high drive temperature only accelerates the failure mechanism.

The controller board has a three-phase motor controller chip.
It converts DC at 12V, into sine-like waves to drive the
three phases on the motor. This is intended to reduce "torque
ripple" in the rotation of the platter. If the motor has high
friction, eventually the motor IC will hit its current
limit, and some stunted waveform will be fed to the coils.
This could cause more "hum" than normal. If the IC gets
hot enough while doing this, it should shut off.

The controller board has two surge arrestors. One across +5V
and one across +12V. A failure of one of those could draw
excess DC current. Those can be found on occasion, to be
burned to a crisp (due to abuse they've suffered from
a bad power supply). Those should be nearer the power
connector on the controller board. Surge arrestors are
needed to support "hot swap" while current is still
flowing.

You could have a short circuit in the 15 pin SATA power, but
it's unlikely the thermal output would travel all the way
to the drive casing. The connector could burn up. There are
some power supplies with flaky "compressed fit" power connectors
on the cabling. I haven't seen a root cause (a picture of
an undamaged one), to understand how that is happening.
If it was a tin whisker, the PSU could likely blow it out
and remove it.

Anything that's "driven" by a silicon chip, is likely
to have some sort of power limiter for device safety.
And if the power is that high, some part of the drive
could shut off, as a hint how high it's getting.

One of the reasons there would be an incentive for a motor
controller design to check, is some Maxtor drives in the past,
the motor controller was poorly made and it ran hot all the
time. And those chips could burn out from the stress,
which suggested at the time that they had no protection
on their outputs at all. If they used a "Class D" approach,
there's no particular reason the chip has to be boiling
hot all the time.

There have been a few cases, where components on computer
motherboards got hot enough to melt the solder, and then
if the machine receives a "bump", the component "floating"
in its solder can get bumped off the pad it sits on. I'm not
aware of any hard drives having circuits similar to those :-)
The root cause there, was regulators designed as linear
regulators, using an opamp and MOSFET pass transistor, and
such circuits have little in the way of protection features.
They've stopped doing that. It was a "P4 era" thing. And
there was at least one Asus motherboard, where a certain
tiny regulator on the board, ran at "100C" all the time
as its normal operating condition. Someone spotted that
with a thermal camera, on some Athlon motherboard.

Paul