Proper LCD monitor turn off time?

[Cat]

Back to those days when we were using CRT monitors, I think I heard
that turning a CRT monitor on consumed much power. What about LCD
monitors? Does turning it on consumes much power also? More precisely,
I would like to know the 'x' in the following :
"If you are not going to use your LCD monitor more than x minutes,
then it is better to turn it off." (in the view of power saving)

Thank you.

 

[Andrew Barss]

: Back to those days when we were using CRT monitors, I think I heard
: that turning a CRT monitor on consumed much power. What about LCD
: monitors? Does turning it on consumes much power also? More precisely,
: I would like to know the 'x' in the following :
: "If you are not going to use your LCD monitor more than x minutes,
: then it is better to turn it off." (in the view of power saving)


I don't know, but the TV show Mythbusters recently did a show which
tested this issue on a variety of bulb types -- regular incandescent,
compact fluorescent, regular fluorescent, and LED. They found that
with most of the bulb types, the extra juice used on startup
amounted to what the bulb in use would use in well under a second.
For the regular fluorescent, it was in the neighborhood of three
seconds. So, from purely a power-use standpoint, turn it off.

What I have no idea on is the effects of leaving it on/turning it off
for the longevity ofg the LCD backlight, and also the brightness over
time for that light (I've heard somewhere that backlights can dim over
time, but I don't know if this is true). Then there's issues about
the power switch and internal electronics of the monitor, and what
repeated on/off cycles do to them. I hope someone more knowledgable
chips in here --


-- Andy Barss

 

[rjn]

Back to those days when we were using CRT monitors,
I think I heard that turning a CRT monitor on consumed
much power.

The concern wasn't power consumption at turn-on, but
inrush current shortening the filament life - but I'm not
sure that's what you were trying to get at.

CRT aging factors included but were not limited to:
1. thermal (general heat in the case)
2. filament lifespan
3. cathode poisoning (general dimming) and
4. phosphor burn-in
5. decay of high-voltage components

Screen-savers were pretty much focused on #4, sometimes
aggravating #3 (if sync wasn't blanked, putting the CRT
into standby) or aggravating #2 if it did shut down.

What about LCD monitors?

Completely different, and may change further.

My impression is that LCD aging factors include primarily:
1. CCFL backlight (BLU) life
2. general thermals
3. CCFL color temperature decay

Burn-in is not a factor, except on LCD rear projection.

Thermals are aggravated when the BLU is on, and the
screen is dark, trapping all that light as heat. StarField
screensaver is not a friend of your LCD.

Going to standby turns off the BLU, cutting heat, but cycling
it a lot shortens its life.

"If you are not going to use your LCD monitor more than x minutes,
then it is better to turn it off." (in the view of power saving)

My view on LCD:

If you tend to run a bright screen (dark text on light background),
leave the screen up and use no screensaver. Go to standby
(BLU off) after 45 minutes or more of idle. Turn the LCD off entirely
overnight (I use a mechanical switch).

If you tend to run a dark screen (e.g. light CAD drawing lines
on black), then go to a bright screen-saver after 3-5 minutes
(to let more light out). Go to standby after 45 minutes or
more of idle. Turn the LCD off entirely overnight.

If LED replaces CCFL in monitor BLUs, the game changes
again, as LEDs are more robust at turn-on. Go to standby
after 3-5 min, and screen-savers are history (except when
they also lock the machine for security reasons).

If you are doing work where color management matters,
you probably need to leave the LCD on and bright during
the entire work period, and budget for more frequent replacements.

This being usenet, however, contrary opinions will arrive
any moment now

 

[Cat]

The concern wasn't power consumption at turn-on, but
inrush current shortening the filament life - but I'm not
sure that's what you were trying to get at.

CRT aging factors included but were not limited to:
1. thermal (general heat in the case)
2. filament lifespan
3. cathode poisoning (general dimming) and
4. phosphor burn-in
5. decay of high-voltage components

Screen-savers were pretty much focused on #4, sometimes
aggravating #3 (if sync wasn't blanked, putting the CRT
into standby) or aggravating #2 if it did shut down.


Completely different, and may change further.

My impression is that LCD aging factors include primarily:
1. CCFL backlight (BLU) life
2. general thermals
3. CCFL color temperature decay

Burn-in is not a factor, except on LCD rear projection.

Thermals are aggravated when the BLU is on, and the
screen is dark, trapping all that light as heat. StarField
screensaver is not a friend of your LCD.

Going to standby turns off the BLU, cutting heat, but cycling
it a lot shortens its life.


My view on LCD:

If you tend to run a bright screen (dark text on light background),
leave the screen up and use no screensaver. Go to standby
(BLU off) after 45 minutes or more of idle. Turn the LCD off entirely
overnight (I use a mechanical switch).

If you tend to run a dark screen (e.g. light CAD drawing lines
on black), then go to a bright screen-saver after 3-5 minutes
(to let more light out). Go to standby after 45 minutes or
more of idle. Turn the LCD off entirely overnight.

If LED replaces CCFL in monitor BLUs, the game changes
again, as LEDs are more robust at turn-on. Go to standby
after 3-5 min, and screen-savers are history (except when
they also lock the machine for security reasons).

If you are doing work where color management matters,
you probably need to leave the LCD on and bright during
the entire work period, and budget for more frequent replacements.

This being usenet, however, contrary opinions will arrive
any moment now

Thanks for your explanation.
I was thinking that bright screen was worse than dark screen for the
LCD monitor so I used to set my desktop background to
solid black. (actually, this is my old habit from CRT days)
And I was turning it off if I leave the computer only about 3~4
minutes (to
go to the bathroom)
Wow I was completely wrong. So, basically, bright screen generates
less
heat on the monitor, thus better for the monitor? Unfortunately most
of
Windows' built-in screensavers are dark. I guess I have to find a
bright
one.

 

[rjn]

I was thinking that bright screen was worse than dark
screen for the LCD monitor ...

Quite the opposite. If the BLU is fully illuminated,
you want to let the light out.

Some CCFLs are dimmable. Assume that yours is not,
unless the maker included some statements on that
and your OS provides obvious support for the feature.

... so I used to set my desktop background to
solid black. (actually, this is my old habit from CRT days)

This is literal poison for a CRT, unless the sync is blanked
(on a recent monitor only), putting the display into standby.
Cathode poisoning has been underappreciated by consumers
since the vacuum tube was invented.

So, basically, bright screen generates less heat on
the [LCD] monitor, thus better for the monitor?

If the BLU is on, let the light out.
Either save to bright, or turn out the light.

What causes LCDs to get retired, currently?
My impression is that it's BLU failure, either individual
tubes (probably from latent defects, and causing
darked bands), or the entire BLU power circuit.
Frequent power cycling of a CCFL BLU probably
hastens these failures. CCFLs may be rated for a
20,000 hour life, but nobody advertises how that
number degrades with increasing power cycles.

As new technologies creep into the market (OLED, SED,
LED BLU LCD), screen-saving techniques need to be
completely reviewed. Most of these will actually be
routine dimmable, enabling new idle-screen display and
power saving strategies (if the screen saver APIs for
dimming become widely implemented).

 

[Bob Myers]

Quite the opposite. If the BLU is fully illuminated,
you want to let the light out.

Bob, I don't have a lot of data to back this up - unfortunately,
this is one factor I don't think has been well-researched re
monitor reliability - but I suspect that whether the LCD is
letting the backlight light through or not is actually a fairly
insignficant factor re the thermal stresses on the backlight.
The majority of the light produced by the backlight doesn't
make it through the panel no matter what, due to losses
which occur whether or not the LCD is switched to the
"bright" state. You might see some slight decrease in
temperature in the "bright" vs. "dark" states, but turning
the BL completely off would make far more of a difference.

As noted, though, you have to trade off the power savings
and reliability improvement (through simple aging) from
turning the BL off with the stresses that you'll get from
the on/off switching - meaning that you don't want to be
switching the BL on and off really frequently. Switching
off if the monitor will be unused for, say, 30 minutes or
longer is probably about the right point to make that
tradeoff. For less idle time than that, I really wouldn't
worry much about the choice of "screen-saver."

What causes LCDs to get retired, currently?
My impression is that it's BLU failure, either individual
tubes (probably from latent defects, and causing
darked bands), or the entire BLU power circuit.

Exactly right - although it's not really from latent defects.
CCFLs, like all fluorescent tubes, are phosphor-based
light sources, and also rely on a form of cathode and
a small amount of free mercury within the tube for
proper operation. With time, all of these basically
"wear out". The tube loses brightness (very often, the
expected life speced for these tubes is actually the mean
time to half-original-brightness, not the time to a hard
failure) or stops lighting altogether, depending on what
goes first.

The BLU power circuit (inverter) also is a major
contributor, but generally these are longer-lived than
the bulbs themselves once you get past "infant mortality"
sorts of inverter failures (which, of course, ARE latent
defects).

As new technologies creep into the market (OLED, SED,
LED BLU LCD), screen-saving techniques need to be
completely reviewed. Most of these will actually be
routine dimmable, enabling new idle-screen display and
power saving strategies (if the screen saver APIs for
dimming become widely implemented).

I would expect to see LED backlighting getting more and
more share in the monitor market over the next few years,
although a good deal of this will be phosphor-based
"white" LEDs rather than individual RGB emitters. I'm
not holding my breath on OLEDs for the near future, and
the SED doesn't look like a viable monitor technology
"ever." (Of course, "ever" in this business often just means
we can't see it happening within the next 5 years...;-))

Bob M.

 

[rjn]

I don't have a lot of data to back this up - unfortunately,
this is one factor I don't think has been well-researched re
monitor reliability - but I suspect that whether the LCD is
letting the backlight light through or not is actually a fairly
insignficant factor re the thermal stresses on the backlight.
The majority of the light produced by the backlight doesn't
make it through the panel no matter what, due to losses
which occur whether or not the LCD is switched to the
"bright" state. You might see some slight decrease in
temperature in the "bright" vs. "dark" states, but turning
the BL completely off would make far more of a difference.

I'm not terribly surprised.

Let's run the numbers I didn't have time to research
last time (and wasn't easily able to find this time).

Let's assume 100% efficiency is, according to
some sources, 683 lumens/watt.

CCFL efficiency is reportedly 20-30 lumens/watt,
so if we use 25, right out of the box, only 3.7%
of the power becomes light. I don't know if this
includes the inverter or just the lamp.

The bulbs radiate in all directions. Absent the
reflectors and diffusers of the BLU, perhaps only
20% of the light would even hit the LCD at useful
angles. Let's assume the BLU improves this to 50%.
So now 1.8% of the power is light hitting the panel.

Then there's the efficiency of the panel at full
bright. I'm guessing that also is around 50%.
So less than 1% of the initial power actually gets
out of the panel as light. A black screen saver may
cause this 1% to get trapped as heat in the panel
(and maybe less than that, as LCD black still lets
some light leak out).

Switching
off if the monitor will be unused for, say, 30 minutes or
longer is probably about the right point to make that
tradeoff. For less idle time than that, I really wouldn't
worry much about the choice of "screen-saver."

Below 30 min, the role of the screen saver becomes
purely security with LCD.