From: David W. Hodgins on
On Fri, 09 Apr 2010 23:30:43 -0400, Nicholas Dreyer <gurfler(a)q.com> wrote:

> I was leaning that way, but my only question regarding that approach
> would be: Is there any risk to the motherboard from regularly turning
> back on the low-level 5v "soft power"?

You will get different answers depending on peoples prior experiences.

My preference is to leave all electronics running 24/7, with
something like boinc running, to keep the cpu at 100% usage, all
of the time.

That way, everything stays at the same temperature.

My cpu (with a 14" fan blowing into the open case) runs at a steady
49C when clean. When it hits 55C, I clean the cpu heatsink, without
having to shut the system down.

My two hard drives run at 36C and 34C. They were running in the
high forties, before I started using the big fan, with the case
cover removed.

While the bearings on my hard drives may fail sooner, the
circuit boards do not undergo any thermal expansion/contraction,
so the rest of the computer should last longer.

As always, make sure all critical data is backed up offsite.

Regards, Dave Hodgins

--
Change nomail.afraid.org to ody.ca to reply by email.
(nomail.afraid.org has been set up specifically for
use in usenet. Feel free to use it yourself.)
From: Massimo on
On Sat, 10 Apr 2010 02:39:00 -0400, "David W. Hodgins"
<dwhodgins(a)nomail.afraid.org> wrote:

>On Fri, 09 Apr 2010 23:30:43 -0400, Nicholas Dreyer <gurfler(a)q.com> wrote:
>
>> I was leaning that way, but my only question regarding that approach
>> would be: Is there any risk to the motherboard from regularly turning
>> back on the low-level 5v "soft power"?
>
>You will get different answers depending on peoples prior experiences.
>
>My preference is to leave all electronics running 24/7, with
>something like boinc running, to keep the cpu at 100% usage, all
>of the time.
>
>That way, everything stays at the same temperature.
>
>My cpu (with a 14" fan blowing into the open case) runs at a steady
>49C when clean. When it hits 55C, I clean the cpu heatsink, without
>having to shut the system down.
>
>My two hard drives run at 36C and 34C. They were running in the
>high forties, before I started using the big fan, with the case
>cover removed.
>
>While the bearings on my hard drives may fail sooner, the
>circuit boards do not undergo any thermal expansion/contraction,
>so the rest of the computer should last longer.
>
>As always, make sure all critical data is backed up offsite.
>
>Regards, Dave Hodgins

I cannot believe my eyes reading this posting.
Are you serious?
This will cost you much extra money, only 'to protect the
motherboard', you significantly shorten the lives of your devices and
from the vuepoint of the environment...

Massimo

From: ~misfit~ on
Somewhere on teh intarwebs Nicholas Dreyer wrote:
> On Fri, 09 Apr 2010 19:08:51 -0400, Strobe wrote:
>
>> On 09 Apr 2010 03:26:08 GMT, Nicholas Dreyer <gurfler(a)q.com> wrote:
>>
>>> I keep my PC on for at most a few hours a day. After it has powered
>>> down, is it safer to shut off all power to the motherboard using the
>>> toggle switch on the power supply, or is it better to leave it on,
>>> or does it not matter at all?
>>>
>>> Thanks for any advice, or pointers to some useful studies on the
>>> matter.
>>
>> For ultimate power economy, use the switch - modern PSUs continue to
>> provide 5v when the PC is 'off'..
>>
>> To *really* economise, switch off with a cheap external power strip
>> - it's cheaper to replace that than a PSU when the switch eventually
>> wears out!
>>
>> If you live in an area prone to lightning storms, it's safer to
>> actually unplug the PC when not in use - a nearby lightning strike
>> on the power line can easily jump over a power switch and *might*
>> fry your PC.
>>
>> Of course, at most you'll save a few dollars a year... And many other
>> devices (TVs, microwaves, cable boxes) also suck power when they're
>> nominally 'off'.
>
> Thanks all for the feedback that points to the advantages of complete
> power off.
>
> I was leaning that way, but my only question regarding that approach
> would be: Is there any risk to the motherboard from regularly turning
> back on the low-level 5v "soft power"?

Not so much to the motherboard but there's more stress put on the power
supply as the big capacitors 'fill', there's a big inrush current. In fact
the only times (twice in 15 years) I've had catastrophic failure (BANG!) of
(one of) the large capacitor/s in the PSU was when applying mains power to
it.

There is another consideration. If the machine is turned off at the back, or
unplugged, the small battery on the motherboard that retains BIOS settings
will die more quickly. However they're only a couple bucks (depending on the
brand and where you live).

What could be more of a PITA is if you have custom BIOS settings such as an
'on the edge' overclock or 'just so' RAM timings, or even if you're running
your HDD/s as AHCI (and the default is 'IDE emulation'). Then, when the
battery does go you have to know these settings so that you can re-enter
them when you replace the battery.
--
Shaun.

"When we dream.... that's just our brains defragmenting" G Jackson


From: Nicholas Dreyer on
On Mon, 12 Apr 2010 02:28:18 +0200, Massimo wrote:

> On Sat, 10 Apr 2010 02:39:00 -0400, "David W. Hodgins"
> <dwhodgins(a)nomail.afraid.org> wrote:
>
>>On Fri, 09 Apr 2010 23:30:43 -0400, Nicholas Dreyer <gurfler(a)q.com>
>>wrote:
>>
>>> I was leaning that way, but my only question regarding that approach
>>> would be: Is there any risk to the motherboard from regularly turning
>>> back on the low-level 5v "soft power"?
>>
>>You will get different answers depending on peoples prior experiences.
>>
>>My preference is to leave all electronics running 24/7, with something
>>like boinc running, to keep the cpu at 100% usage, all of the time.
>>
>>That way, everything stays at the same temperature.
>>
>>My cpu (with a 14" fan blowing into the open case) runs at a steady 49C
>>when clean. When it hits 55C, I clean the cpu heatsink, without having
>>to shut the system down.
>>
>>My two hard drives run at 36C and 34C. They were running in the high
>>forties, before I started using the big fan, with the case cover
>>removed.
>>
>>While the bearings on my hard drives may fail sooner, the circuit boards
>>do not undergo any thermal expansion/contraction, so the rest of the
>>computer should last longer.
>>
>>As always, make sure all critical data is backed up offsite.
>>
>>Regards, Dave Hodgins
>
> I cannot believe my eyes reading this posting. Are you serious?
> This will cost you much extra money, only 'to protect the motherboard',
> you significantly shorten the lives of your devices and from the
> vuepoint of the environment...
>
> Massimo

Yes, I would never want to go through with all that Dave suggests - and
doubt anyone else would, unless they have some extraordinarily valuable
motherboard and hard drive setup - and didn't care about the racket,
expense and environmental consequences of constantly running fans, but to
get back to my unanswered question:

Assuming CPU shutdown is to happen after more or less daily use of a few
hours at most, does daily 5v "soft power" re-energizing of motherboard
via power supply rocker switch prior to reboot add significant, or even
measurable wear and tear on its circuitry?
From: Nicholas Dreyer on
On Mon, 12 Apr 2010 13:32:23 +1200, ~misfit~ wrote:
>
> Not so much to the motherboard but there's more stress put on the power
> supply as the big capacitors 'fill', there's a big inrush current. In
> fact the only times (twice in 15 years) I've had catastrophic failure
> (BANG!) of (one of) the large capacitor/s in the PSU was when applying
> mains power to it.
>
> There is another consideration. If the machine is turned off at the
> back, or unplugged, the small battery on the motherboard that retains
> BIOS settings will die more quickly. However they're only a couple bucks
> (depending on the brand and where you live).
>
> What could be more of a PITA is if you have custom BIOS settings such as
> an 'on the edge' overclock or 'just so' RAM timings, or even if you're
> running your HDD/s as AHCI (and the default is 'IDE emulation'). Then,
> when the battery does go you have to know these settings so that you can
> re-enter them when you replace the battery.

Ah, what you say here is very interesting to know. It's sounding like
the slight energy consumption of leaving the PSU rocker switch on
permanently is not as bad as the possibility of a blown out PSU from
repeated powering-up, so I will continue to leave it on permanently.

The reason I was focusing on the rocker switch question was because my
last PSU blew after a mere four years with the rocker switch left on all
the time except for occasional cleaning, maintenance and moving: One
day, after CPU shutdown the previous night, nothing would start until
installation of a replacement PSU.

Thanks for sharing all your experiences,

Nick