Smart questions
Smart people
 Find A ForumFind An Expert Join Eng-Tips Forums
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Remember Me

Are you an
Engineering professional?
Join Eng-Tips now!
• Talk With Other Members
• Be Notified Of Responses
• Keyword Search
Favorite Forums
• Automated Signatures
• Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

#### Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Jobs from Indeed

Just copy and paste the

# Halogen bulb watts/lumen efficiency paradox?(4)

 markls8 (Mechanical) (OP) 9 Jul 03 0:01
 Halogen bulbs are considered more efficient than regular incandescents because the tungsten filament burns hotter and puts out more light. But to me, it seems that a bulb that burns cooler will be more efficient because it is producing less heat and thus more of its electrical energy is going to produce light, and less is being wasted as heat. In fact, a theoretical 100% efficient light bulb would be at ambient temperature.What am I missing here?Thanks in advance - Brian.
 Skogsgurra (Electrical) 9 Jul 03 12:31
 markls8You also should consider the "Stefan-Boltzmann Law". It says that radiation from a cavity (aka "black body" and very close to any radiant surface) emits more radiation the hotter it is. So your thinking about "better the colder" is not quite correct. The S-B Law actually states that the radiation goes up with the fourth power of absolute temperature to the fourth power. So an emitter that is 2100 K emits 86 percent more radiation than an emitter at 1800 K.There is also a shift from infrared into visible wavelengths as the temperature goes up. The net effect is that a lot more radiation is produced and it is produced in the visible part of the spectrum.So "hotter is better". 5000 K seems to be an optimum with a peak in the middle of the visible spectrum. 6000 K produces twice as much light - but a lot of it in the bluish part of the spectrum.
 IRstuff (Aerospace) 9 Jul 03 15:29
 The simple answer is that if the filament were at room temperature, it would be 0% efficient, since it would emit no light.  Actually 0/0 is undefined, but that's another matter altogether.The fallacy in the reasoning is that the heating of the element is what allows it to emit light.The temperature of the blackbody dictates where the peak emission is and how much there is.  As the previous post indicates, the total output emission goes as the 4th power of the absolute temperature, but Wien's Displacement Law tells you what wavelength has the highest emission, given by 2898/kelvin temperature, so a 5222K filament puts the peak emission at about 555 nm, which is the peak of the photopic eye response. TTFN
 markls8 (Mechanical) (OP) 10 Jul 03 0:59
 Understood. Thankyou.The reason for my question is that I just built a house and installed 40 - 50 halogen pot lights, on the advice of all current literature which is urging the use of halogen because of its efficiency. The bulbs are PAR20's which run around $9 CDN a piece compared to regular bulbs at around$1. Although I had few doubts, I was wondering about the technical basis for their claims. Thanks - interesting info.Regards - Brian.
 IRstuff (Aerospace) 10 Jul 03 1:18
 The efficiency comes through in the ability to get the filament hotter, which both increases the total power as well as shifts the peak power wavelength to the most sensitive portion of our visual response.  Under halogen lamps, such an operating condition could not be sustained long enough as the filament would get thermally fatigued as lose material through evaporation into the gas of the bulb.  Halogen gases tend to keep the filament intact longer. TTFN
 (3)  peebee (Electrical) 15 Jul 03 16:12
 If you only heated the filament to about 350 degrees, well, it'd be very efficient at generating infrared, but it would put out very little visible light and would be about 100% efficient as a heating element but about 0% efficient as a light bulb.On a side note:  IRstuff mentioned that halogen gases tend to keep the filament intact longer.  The way that works is really pretty cool:  the halogen (chlorine or similar gas) picks up stray tungsten molucules floating around inside the bulb.  When part of the filament starts to burn out, it gets skinnier than the rest of the filament and burns hotter.  When a chlorine molecule carrying a tungsten molecule hits that hot spot, it drops off the tungsten, effectively patching the weak spot.  This not only makes them last longer, but this is why you can burn them hotter & brighter and why they are more efficient.In a typical traditional incandescent, the tungsten constantly slowly sputters off the filament and condenses on the glass, that's why the glass on old light bulbs tends to turn grayish rather staying the clean white color they had when they were new.
 Skogsgurra (Electrical) 15 Jul 03 17:47
 Fascinating little story, peebee. Are the molecules actually attracted to the hot spots? Or do they just happen to get there by chance? I guess the latter.
 peebee (Electrical) 15 Jul 03 18:07
 They just drift there by chance.  But once they get there, they actively deposit the tungsten on the hot spot, which is the weak spot, rather than randomly scattering the tungsten all along the filament.  They fix themselves, it's pretty amazing when you think about it.
 RobWard (Industrial) 16 Jul 03 7:42
 Are you seriously suggesting that light bulbs don't fill up with "dark"?Everyone knows that light bulbs, when they are turned on, actually suck the "dark" out of a room.When they are full (the bulb turns dark when it's full) they stop working and you have to fit an empty one!Seriously though: very interesting thread. I'd wondered about the "why the hotter lamp is more efficient" thing myself but never got around to asking anyone else. "I love deadlines. I love the whooshing noise they make as they go past." Douglas Adams
 dadfap (Petroleum) 16 Jul 03 9:26
 RobWardThat's right up there with : electricity is a myth, everyone knows that "electrical" devices actually work on smoke.  The smoke leaks out through the insulation and the "electrical" device stops working!regardsdadfap
 docholliday (Electrical) 16 Jul 03 18:18
 There's a lot to this question of why tungsten halogen bulbs last longer - while the redeposition of tungsten on the hotter (i.e. thinner) parts of the filament is one cause, it's not the only one. In fact, with some halogen bulbs it doesn't happen at all; it depends which halogen you use - I remember when they were called 'quartz iodine' lamps, and the redeposition on the filament in lamps using iodine isn't usually at the hotspots.The other effect that I'm thinking of - which is claimed by most of the references I've read to be the only significant one, though I'd be surprised if things hadn't moved on a bit in recent years - is that the halogen bulb operates at a much higher internal pressure than a standard bulb, and this reduces the rate of evaporation of the filament. Where the halogen comes in is as follows:-1. the bulb operates at a much higher pressure, so the envelope has to be smaller to withstand the pressure without unfeasibly thick walls.2. Although the rate of evaporation is reduced, the internal area has reduced much more, so the bulb would become useless due to blackening of the wall from the tungsten deposit.3. Using some halogen in the fill, combined with a high temperature at the bulb wall (which has resulted from using a small envelope) means that the halogen will combine with the tungsten on the walls, and then the compound will dissociate again, redepositing the tungste, on the even hotter filament and releasing the halogen to repeat the cycle. Unfortunately in a lot of cases it will do this before it has got to the hottest part of the filament, resulting in a thickening of other parts of the filament.Incidentally, a lot of halogen bulbs have similar lives to standard tungsten bulbs - it's just that they'd die much faster without the halogen! The ones I use seem to have about twice or three times the life of a standard GLS bulb (the claim varies depending on which manufacturer they came from).Going back to the original question about temperatures; there's been a lot of discussion in the UK over the years about 'long life' bulbs, which commonly last twice as long as standard bulbs - including a load of speculation along the lines of 'the manufacturers know how to make bulbs that last ten times as long, but they don't do it because they want to sell more bulbs' . In fact the standard life bulbs probably give you more light per total pound/dollar/euro expended because theire greater efficiency counterbalances the cost of buying twice as many - but there are places where actually replacing the bulb is such a pain that it's worth paying more for longer life.In the same way the more expensive halogen bulbs give greater efficiency, and they also give a slightly different colour balance, which adds an aesthetic dimension to your choice if it's for home lighting, but could be a critical factor if you're a graphic designer! Mike
 peebee (Electrical) 16 Jul 03 19:56
 In response to 'the manufacturers know how to make bulbs that last ten times as long, but they don't do it because they want to sell more bulbs' Sounds like the typical corporate conspiracy theories, and this one almost sounds believeable, but compact fluorescents, HID's, and now LED sources seem to put that one to rest pretty quickly.  Not only do they all last longer, but they all have a way higher efficacy than incandescent or halogen.
 IRstuff (Aerospace) 16 Jul 03 21:18
 Actually, they, the manufacturers, have a higher margin on the long-life bulbs, since they sell them for more than 10 times the cost of the standard bulbs.As for the compact fluorescents, I've not been impressed.  I've been using them for about two years and the failure rate appears to be roughly the same, but they cost about 5 times as much, and I'm not even sure the payback period is short enough for lifetime that I'm actually getting. TTFN
 markls8 (Mechanical) (OP) 16 Jul 03 22:49
 Also -(A tip if you're concerned about maximizing bulb life) - when using a dimmer switch with a halogen bulb it is a good idea to run the light at full power either occasionally or right before turning it off, to ensure that the temperatures for the redeposition on the filament are attained.
 peebee (Electrical) 17 Jul 03 11:50
 IRstuff -- your experience with the compact fluorescents is interesting.  Several ears ago I had some similar high failure rate problems and had all but given up on them.  But more recently (2-3 years ago) I broke down and tried them again and have been very satisfied.  Even more surprisingly, the older ones were major brands, and the newer ones were off-brands.  In all instances, the lamps were burned continuously or at least switched infrequently (1 or 2 times per day).  Strange.  Maybe we both just hit some bad manufacturing lots or something.
 IRstuff (Aerospace) 17 Jul 03 20:06
 Possibly...I'm using them at home... Anyone need some slightly used incandescent bulbs? TTFN
 60hzhmmm (Electrical) 17 Jul 03 22:38
 Speaking of bulb life.  The life of an incandecent filament is proportional to the 12th power that's right one, two, of the voltage put to it.  This is how they make long life bulbs.  They just design them for slightly higher voltages.  If you use a 130V bulb in 120V service it will last 2.6 times longer than otherwise.  I have a rejected street light bulb...186W of all the odd values, in my laundry room that I screwed in 36 years ago.  We use it many times everyday as the room is more hallway than laundry and has a utility sink in it that is used alot.  It is less bright then I would expect 186W to be but still plenty bright!  Of course the efficacy, (usable lightfor money in), is less than your average lamp.  Compact flourescent lights have been a big disappointment to me.  I have a 150gal marine aquarium that needs light, lots of light!  I had six 180W metal halides over it until the CA energy market melted down.  They suddenly started costing over \$150/month so I changed to 6 pathetic 28W compact flourescent spot lights.  7 years!! the boxes said.  All but one burnt out in 11 months.  Reading the box they disclaim the whole thing by stating in tiny print 7 years based on 2hrs per day... Another words 0.5833 years not 7!   Bring on the LEDS!
 rumbaldking (Electrical) 18 Jul 03 17:30
 http://www.centennialbulb.org/ to link to a filament-based curiosity - "world's oldest continuously running bulb", 101 years, carbon filament, over 800,000 hours.  In a firehouse in Livermore, CA 1 hr est of San Francisco.  They moved it a short while back and this website got put up.  Enjoy.

Close Box

# Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

• Talk To Other Members
• Notification Of Responses To Questions
• Favorite Forums One Click Access
• Keyword Search Of All Posts, And More...

Register now while it's still free!