Subject: Lights on and off Date: Sat, 17 Apr 1993 01:13:06 GMT The U.S. Government has cho

From: phil@rahul.net (Phil Gustafson)
Subject: Lights on and off
Date: Sat, 17 Apr 1993 01:13:06 GMT

The U.S. Government has chosen not to give Terry Chan access to uk.misc.

This site gives everybody access to anything, so here's Terry's article:


Phil,

I don't have access to the newsgroup "uk.misc",
can you post the following for me in it and a.f.u.?

Thanks.

Terry

---------------------------------------------------------------------------
Newsgroups: uk.misc,alt.folklore.urban  
Subject: Re: If you're not using it, turn it off!  
Summary:  
Expires:  
References: <1993Apr14.104648.3305@lsl.co.uk>  <13723@spark.ed.ac.uk>  
Sender:  
Reply-To: TWChan@lbl.gov (Terry Chan)  
Followup-To:  
Distribution:   
Organization: Department of Redundancy Department  

  +>In article <1993Apr14.104648.3305@lsl.co.uk> snail@lsl.co.uk wrote:
  +>>If you think you're going to come back in to the room within 
  +>>the next two hours, leave the strip light 
  +>>ON if it is already switched on. This will consume less 
  +>>electrictity and prolong the life of the tube compared with turning 
  +>it off and back on again.

jaspert@cogsci.ed.ac.uk (Jasper Taylor) writes:
  +I think we have the clear signs of an urban legend here. In 1980, a
  +friend of mine renting a room in Birmingham was forced by his landlady
  +to wash up in darkness because the kitchen had a fluorescent light and
  +she was under the impression that "switching it on takes as much power
  +as running it for half an hour".
  +
  +This is demonstrably false. Let's see, [does some math]
  +This would destroy any domestic lighting circuit.
  +
  +I do suspect that those new high-efficiency bulbs with tubes inside may
  +have their lives shortened by switching off and on. Any takers?

Sorry to quote the previous articles, but I like the way they
covered this story.  If I read the above correctly, we have the
following stories:

1.  snail@lsl.co.uk suggests that if the "strip light" (by which I 
    take it to mean fluorescent lamp) is already on, leave it on
    if you expect to be back within 2 hrs to maximize both energy
    savings and lamp life.

2.  jaspert@cogsci.ed.ac.uk (Jasper Taylor) gives a 1980 example
    of a friend who was forced to endure darkness by his Birmingham
    landlady because the "startup" power would be just as much it
    would take to run the lamp for 1/2 hr.  He also performs a 
    calculation to demonstrate that this is silly.

Jasper is correct.  One qualification is that my familiarity is with 
US fluorescent lamps.  The fluorescent lamp and ballast industry is 
a worldwide market dominated by a few firms, but I don't know what 
the specific equipment saturations are in Europe.  That aside, I will 
make the following comments.  

Based on what we know here at the laboratory, this story is false.  
Interestingly enough, I work in a group that analyzes energy efficient 
technologies for various products including lighting.  

Early fluorescent lamps (pre-60s) had such short lives that rumors 
started that they shouldn't be turned off.  It is true that rapid 
switching makes the cathodes deteriorate more rapidly.  But once lamp 
lives went over ~10,000 hours, the rumor became myth.  Any analysis 
shows that for typical usages it's always better to switch off the 
fluorescents if you're away for more than a couple of minutes (a 
common rule of thumb is 7 minutes). You'll affect lamp life somewhat 
but that's more than compensated for by the increase in lamp life. 
There are several articles on this which I could dig up somewhere if 
someone really needs it.  

Note that story #1 is not correct for two reasons.  The first is
that running lamps continuously will use more power than turning
them off, not vice-versa.  Second, running lamps continuously has 
a deleterious effect on the lifetime of the lamps by burning out 
the phosphors.  Turning lamps and and off will reduce their lifetime 
because of the repeated power to the ballast will burn out the 
filament.

Btw, modern fluorescent lamps have lifetimes on the order of 
20,000 hours for 4-foot fluorescents and 12,000 for 8-foot).  

The landlady in Jasper's example #2 is clearly mistaken.  While
fluorescent lamps that use instant start ballasts (more common) 
require more startup power than rapid start ballasts, the negative 
effect is somewhat reduced life for lamps that use instant start 
ballasts rather than higher overall energy consumption.

Jasper is also correct in that the the lifetimes of the "new high-
efficiency bulbs with tubes" (by which I believe he means what is 
marked as "compact fluorescent" lamps in the US) will be considerably
shorted if they are subject to a great deal of on/off switching.
Because they are relatively new, there has not been a great deal of
field testing with CFLs to determine what rule of thumb to apply
for them (i.e., how long should they be left on?), but experiments
with models available now all show greatly shorted lives if subject 
to this type of stress.

For a good survey of the existing technologies (and future ones),
see Barbara A. Atkinson, et al., _Analysis of Federal Policy Options 
for Improving U.S. Lighting Energy Efficiency:  Commercial and 
Residential Buildings_, Lawrence Berkeley Laboratory (LBL-31469),
December 1992.  Barbara of the Energy Conservation Policy group and 
Francis Rubinstein of the Lighting Systems Research group here both
supplied me with some information on the above.

Terry "Lights out!" Chan



Phil "Do you guys use Lucas computers?" "At one time I owned six SU
carburettors, on three cars" Gustafson
-- 
|   phil@rahul.net                         Phil Gustafson 408-286-1749      |
|  How many boards would the Mongols hoard if the Mongol hordes got bored?  |