You are not logged in. Login Now
 0-20          
 
Author Message
kaplan
To leave it on or to turn it off? Mark Unseen   Apr 8 04:01 UTC 1994

I don't remember where or how long ago I got this advice, but I've been
assuming that there is significant wear on a hard disk every time it is
powered up.  If I want my computer to last, it's better to let it idle
for an hour or two.  Don't keep turning it on and off.  Electronic gizmos
of all kinds are harmed every time they cool down and heat back up, so
leave the monitor on for a couple hours too.  The electricity is cheaper
than repair or replacement.

The new baby in our house (a portable computer) has a battery saving
feature that shuts off the hard disk after a set amount of inactivity
time.  I was surprised that it seemed to be designed to constantly go on
and off.

If the advice about leaving it on only applies to some hardware, how can I
tell which is which?  Does my habit of leaving the thing on a lot explain
why my vintage 1987 Zenith 286 died in 1992?  Is my TV doomed because I
can't stand to have it on when I'm not watching it?
20 responses total.
scg
response 1 of 20: Mark Unseen   Apr 8 04:11 UTC 1994

        I have no idea what the technical ramifications are, but my
computer hasn't been off for more than a few hours since early January. 
The times when it has been off for brief periods of time have either been
when I was doing hardware work or during thunder storms, and the time in
late December and early January when it was off was because I was out of
town for a week.  Part of it is because I've heard it's better for it to
be left on, but it's mostly because I don't want to wait the 30 or 40
seconds it takes it to boot.
srw
response 2 of 20: Mark Unseen   Apr 8 06:04 UTC 1994

This is an age-old dilemma which has been much digested in trade magazines.
Largely there has been no resolution. There are disadvantages to turning
hardware on and off, but recently the consensus is that they are minimal.
Therefore you are encouraged not to worry about shutting/restarting
if that's what you want to do. It's essential for laptops to spin the disk 
down to conserve battery time. This is a much more important consideration.

If you are going to leave your desktop computer on a lot, you should
turn the monitor off when not in use. Power cycling the monitor is not
reputed to be at all harmful, and monitors can by responsible for the lion's
share of the power consumed by a machine.

I personally cycle my home desktop down if I'm not going to use it for more 
than a few hours. If I leave it up, I shut the monitor usually.
At work, my Mac is a server, too, so it's always up. I always power
down its two monitors at the end of the day. I do the same with my VaxStation,
it takes ages to boot up, and the monitor is a power hog. In fact, I 
don't bother to turn the VAX monitor on until I need it.
danr
response 3 of 20: Mark Unseen   Apr 8 11:50 UTC 1994

Turning things off and on does stress the components.  Ten or fifteen
years ago this was a serious concern.  Minicomputers  had quite a lot
of parts and those parts weren't as high-quality as electronic parts
are today.

Today, however, there are fewer parts because the level of integration
in today's ICs is so much higher.  The parts are much better, too.
A lot of research has gone into discovering how ICs fail and how
to design and build them so they don't fail.  The result is parts
that stand up to the stresses of power-cycling better than ever.

My educated opinion is that as long as you're not turning the power
on and off several times an hour you're probably not risking
early failure.
bhallu
response 4 of 20: Mark Unseen   Sep 26 19:01 UTC 2000

well,look at it this way. if you keep the PC powered on , line surges etc
MIGHT screw it up. anyway , it has been proved almost concludsively that
powering  your Pc up/down doesn't make any difference as long as you keep a
safe time of say 15-30 minutes between each shutdown.
Hope this helps.
gull
response 5 of 20: Mark Unseen   Sep 26 22:41 UTC 2000

I've wondered about this.  Not with respect to my PC, which I suspect
doesn't care, but for my stereo equipment.  My amp is a Fisher 400, which is
from the vacuum tube era.  Each time I switch it on, there's a sizable
current pulse that goes through the cold tube filaments.  (This is part of
the reason light bulbs tend to burn out when first switched on.) This
stresses both the filaments and the power supply.  On the other hand,
leaving it on uses up some of the life of the tubes.  My compromise is to
leave it on if I'll be turning it on again within an hour or so.  Otherwise
I shut it off.
scott
response 6 of 20: Mark Unseen   Sep 26 23:19 UTC 2000

If you wanted to be truly anal about prolonging the filaments, you could put
them on a seperate supply and and feed them from a variac.

But I don't think that startup hit on the filaments really affects tubes
badly.  Most high-powered tube amps include a "standby" switch to leave the
plate voltage off until the tube is warmed up, since apparently hitting the
plate cold can cause some wear.

(but if you've got one of those Fishers with the 7591 output tubes, I'd baby
those things as much as possible since replacements are very hard to come by)
rcurl
response 7 of 20: Mark Unseen   Sep 26 23:27 UTC 2000

In fact, the linear, all triode, push-pull audio amplifier I designed and
built in college had its filaments on permanently.

gull
response 8 of 20: Mark Unseen   Sep 27 01:11 UTC 2000

Mine uses 7868's, which are almost as rare and expensive as the 7591's. It
has four of them in a push-pull arrangement.  They're running about $50 each
new-in-box, currently; the tubes in mine are all *orginal* Fisher stock. 
The FM front end also uses some NuVistors, which are even rarer.

This set is about 36 years old; I wonder how many amps made today will still
be working in 2036?
wolfg676
response 9 of 20: Mark Unseen   Oct 18 12:22 UTC 2000

I'm one of  the "leave it on" types. I leave most , if not all of my systems
on all the time, because, like scg, I don't like waiting for bootups. :)
I do turn some of them off durning thunderstorms, but will leave the ones that
are on a UPS turned on.  For example, there's been several storms since I last
rebooted my Linux box 103 days ago, and it's still going. I even lave my lptop
in suspend mode most of the time, sometimes for up to 2 days.I've also found
an interesting setting that's appearing in newer motherboards where it will
turn itself on after a power failure. Nice thing to have, especially for a
server machine. 
On the subject of monitors, in my case I've seen an odd thing. I have an old
CTX 17" that I got from property dispo, and aside from a minimal bit of
burn-in and a jitter problem at low refresh rates, it's a good monitor. I
leave it on all the time, except when it goes into sleep mode when the screen
is blank. My roomie purchased a 19" Komodo a few months back, and he turns
it off whenever he's not at the machine, but with it being on a computer that
everyone in the house uses, it usually gets turned on again rather soon. Then
off again, then on.... Now it's developed this problem where all of the red
in the image takes 5 or so minutes to "pop" in, and the problem is only
getting worse with every power cycle. I tried to tell him to either leave it
on, or take it in to get it fixed, but he won't listen. He also turns his
computers on and off all the time, an they're about the same in how unstable
they are. 
Guess you could take it two ways. Leave the computers on, or don't let my
roomie build a computer. :)
gull
response 10 of 20: Mark Unseen   Oct 20 01:43 UTC 2000

I tend to leave my laptop suspended when I'm not using it, often for days on
end.  It works fine that way, though lately I've noticed that the clock has
started to lose about 8 seconds a day while suspended.
ball
response 11 of 20: Mark Unseen   Feb 20 01:43 UTC 2001

For me it depends what the machine's doing.  A lot of older
machines lacked power management, but can be coaxed into
spinning down the hard disk if it's not been used for (say)
twenty minutes.

Wherever possible I enable display power management
signalling (DPMS), to blank the screen and go into a 'light
sleep' after five minutes, and go into a 'deeper sleep' or
stand-by after some longer period, say another ten minutes.
Where DPMS isn't an option, I make the screen-saver blank
the screen, and ask people to turn off their monitors before
leaving for the night.

I would like to configure more of our PCs to go into standby
if left alone for half an hour, but because of existing
inter-dependencies they have to be powered up in a specific
sequence and it's proven easier to leave the system unit on
and have the disk an monitor power down in an effort to
conserve power. The machines run the SETI@home screen-saver,
so their spare processor time is put to some use.  Most of
them will be retiring from the SETI@home project though, and
will be switched off over-night.

Incidentally, anyone know how to configure a PC with 'soft-
power' to automatically turn on at a given time?
gull
response 12 of 20: Mark Unseen   Feb 20 04:56 UTC 2001

Sometimes it's a CMOS setting.  My laptop is that way.  I imagine it's
machine-dependent, though.  On my machine it's called "alarm power on".
ball
response 13 of 20: Mark Unseen   Feb 20 18:22 UTC 2001

Thanks gull, I'll see if I can find that one.
arthurp
response 14 of 20: Mark Unseen   Dec 15 02:57 UTC 2001

Hm, I have a fisher 400 as well.  DOn't know what tubes it uses although 
I should.  I had to buy one six or seven years ago.  Nice unit.  Very 
few people I talk to understand why I keep using it instead of buying a 
new one.  :)
gull
response 15 of 20: Mark Unseen   Dec 18 01:06 UTC 2001

I've replaced mine, but only because I wanted an amp with a Dolby 
Digital and Dolby Pro Logic decoder.  The Fisher also needs new filter 
capacitors, it's developed a nasty hum.  I'll probably replace the 
filter caps and then sell it.
arthurp
response 16 of 20: Mark Unseen   Jan 12 05:48 UTC 2005

I talked to a guy at newyears who has a fisher as well.  He says that
they need to get refurbished in a couple different ways to improve the
sound, and to prevent one failure that smokes most of the unit.  I guess
I have to look into that.
gull
response 17 of 20: Mark Unseen   Jan 12 14:36 UTC 2005

The biggest failure items are the rectifier diodes.  These sets were
originally equipped with selenium rectifiers.  When they go, they
release noxious fumes and can start a fire.  My set had already had them
replaced with modern silicon rectifiers.
arthurp
response 18 of 20: Mark Unseen   Jan 23 00:45 UTC 2005

That sounds like what he said.  I have to admit I only listened to the
level of (note to self, look into this when you get some spare money). 
I also don't use the set very often anymore.  It kicks enough that I
feel bad using it in my apt.  That thumping upstairs is *not* my
neighbor dancing.
gull
response 19 of 20: Mark Unseen   Jan 23 22:56 UTC 2005

Heehee.

It certainly puts out a lot more volume than you'd expect, based on its
specifications.  It's rated in good, honest RMS watts, though, not the
"peak" watts that you see in ads these days.
arthurp
response 20 of 20: Mark Unseen   Jan 25 00:23 UTC 2005

Yup.  This baby has powered a couple really kickin' house parties.  (Not
at the current location.)
 0-20          
Response Not Possible: You are Not Logged In
 

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss