IBM's paper regarding constant power on of PC's and PC displays.

Downloaded from their Technical Advisor bulletin board 06/19/1992


Constant power-on operation of displays consumes product life at a much accelerated rate, as does the operation in harsh environments. Additionally, the time to failure or the time to the end of the useful operational life of the display (when the screen becomes too dim to be easily read, or screen burns appear on the screen), or the MTTF (mean time to failure) of the display can be shortened significantly by improper adjustment or unnecessary operation.

As an example, in an environment where the display is actually used from 8:00 AM to 5:00 PM Mon-Fri but is left powered on 24 hours a day 7 days per week, the MTTF might be shortened by a factor of approximately 300%. (Assuming an MTTF of 10 years under normal 8 x 5 operation, if left powered on 7 x 24 continuously the MTTF might decrease to 3-4 years.) If the display is also subjected to harsh environments such as high ambient temperatures, the MTTF decreases further. Theoretically, this being the case, the number of failures that would normally occur over a 3-4 year period in an account might be incurred in the first year of operation, or the failures that would have occurred over a ten year period might be experienced in the first 2 or three years.

It is important to understand that there is no suggestion of multiple power on/off cycles between jobs or operators during a normal day. This is not necessary and from a business standpoint can be disruptive. Displays are designed to operate in a normal office environment. The recommendation is simple. What should be avoided is a constant ON situation where the units are never turned off at the end of a work day. Constant ON situations should be limited to unique, critical applications.

Variability of brightness and control adjustment is designed to offer users restoration of normal viewing brightness and contrast over time as the display ages. However, some users, during installation, set the brightness and contrast all the way up to their fullest extent. This is well beyond where it needs to be for normal viewing.

A word on screen saver programs -- These programs are designed to protect against phosphor burn from long interim periods where the same image is displayed on a screen. For this purpose they are excellent. However, there is a misconception that these programs will eliminate the effects of a constant power on. This is not true, as the logic elements remain energized and the cathode element continues to burn (emit). It is only the electron stream which is biased off by the program.

System Units

Field experiences appears to indicate that system units may consume life at a lesser rate in constant ON operation. This is due in part to the existence of a built in fan providing a positive air flow and the absence of factors such as brightness/contrast control drivers. However, hard files and internal fans are electro-mechanical devices that will wear. Heat and humidity are also factors.

A subtle but important item to be considered is that the self-diagnostic programs (power on self tests, POST's) are NEVER run during a constant power ON. This can cause a fault not to be evident until such time as that function is required for use. (EX: high memory error) An important note: If a large number of system units have been in constant ON operation for a considerable period of time, it can be reasonably anticipated that some failures will become evident within this population if they are powered of and back on. A "shaking out" phenomena as a result of power on self test can occur. Unfortunately, this may reinforce the customer belief in constant ON operation.

Other consideration: (Heat Factor)

A common but not readily recognized problem associated with constant power ON operation (especially in southern climates) is that air-conditioned systems are shut down in the office areas and in many cases the entire building during off hours and on weekends. Ambient room temperatures rise to equalize with the outer weather environments and can exceed the outer environment due to the uncirculated heat build up resulting from electronic equipment being powered on.

Under these types of conditions, temperatures can easily exceed the equipment design operating environment specification for prolonged and repeated periods. The reliability over time of electronic circuitry is affected exponentially as the temperature is increased.

Summary: (My words, not IBM's)


Although the above wording in this document is somewhat cautious, it basically says that if you leave displays powered on 7 x 24 when not required you will probably experience some sort of failure within year 2 or year 3 of ownership. If used in a harsh high temperature environment the failure may occur in the first year. A failure from an OEM manufacturing point of view, means something will happen to make the monitor unusable. It DOES NOT mean that the monitor will reach its end of useful life and not be repairable. If the failure is an electronic component or circuit board the cost of repair could be in the $80 to $100 range plus shipping based on current monitor depot repair prices. If the problem occurs with the flyback transformer or the tube itself, the cost of repair and shipping could exceed the cost of a new monitor.

My discussions with monitor manufacturers (while investigating this issue as it applies to all monitor products in general) concluded that powering a monitor on and off multiple times a day has no adverse affect on the circuitry or power supply but could cause premature failure of the mechanical power switch itself. A green monitor, when utilizing the DPMS on off standard may go to the "sleep or suspend" mode multiple times a day if the system is used sparingly. This continual off/on cycle is not considered detrimental to the life of the monitor so the power off/on cycling is not an issue compared to the cost of leaving a monitor powered up constantly..

One manufacturer's response was that they felt their displays could tolerate 7000 to 10000 power on cycles over the life of the product with no adverse effect. This is equivalent to 2-3 power cycles a day, 365 days a year for 10 years. I doubt that any of us use our home systems 365 days a year. (We all need some vacation time).

System Units:

System units (the PC itself) do not appear to be as adversely affected by constant power on due to internal cooling fans and the fact that there are no moving parts other than the fan(s) and the hard disk. A PC can be left powered on constantly with less likelihood of premature failure. If it is necessary to leave the PC powered on (receiving Fax's, server duties, unattended tape backups) then power off the monitor when not in use and leave the PC powered up. Not having the display illuminated has NO EFFECT on the operation of the program. The video card will send data to the tube as if it was powered on but nothing will be displayed. (Very much like videotaping a program on your VCR with the TV off.)

My recommendation:

Leave your system and monitor powered up for the day or for the evening if you plan on coming back to the computer to "surf" or do some work. The power saved by continually powering it off is negligible. Turn it off when you go to bed or if your not going to use it for the rest of the day/week.

DPMS Monitors

What if my monitor is a "green" monitor? (DPMS compliant)

If your monitor is DPMS compliant AND you have the video power saving modes enabled in your CMOS or video driver then it should power off according to the timers set in the CMOS or driver S/W.

However if you have not enabled the DPMS power saving modes the monitor will not power itself off/on and the above rules will apply.

Click here for a brief explanation of the DPMS monitor standard.

Dick Perron