Efficient Lighting

Hello Jim,

I recently converted much here to CFL after I found some that worked well and were cheap. Costco had a sweet deal (maybe still does) where utility subsidies brought the cost of 23W and 14W versions down to around a Dollar a piece. If you have a good reflector behind them they are really bright, especially the 23W versions which are claimed to produce the output of a 100W incandescent bulb (looks like they really do now). But mind that they produce EMI. Mostly conducted which you can muffle but also radiated which can't easily be filtered unless you place a mesh across the face. That mesh will be a dirt and spider web collector and can become a hazard.

AFAIK you can eke out a wee bit more efficiency if you build your own inverters and then use larger fluorescent tubes. Plenty of circuits on the web.

When it gets cold fluorescents need a long time to reach their target output. Meaning more than just a few minutes. When it gets really hot I had some of the industrial long tubes cycle off.

--
Regards, Joerg

http://www.analogconsultants.com

Put in some windows? Maybe skylights? :-)

Good Luck! Rich

For battery operation I'd sure be looking at compact fluorescent bulbs. El cheapo ones have crappy phosphors but I got some 15 watt for $1 ea at the local grocery store that are really high output.

Its obviously efficient to use as little light as possible, in small areas. That said, the compact florescents are pretty good but their light output is not going to equal a 4 foot fluorescent

I just installed lights in my garage and used 10 4 foot bulbs, 3300 Lumens each at 40 watt each. I wasn't worring about efficiency, just light output. I must examine some lamps installed above my head. The are very much brighter than the other fluorescents around here. I wonder what the Lumens rating is? Well I just found out. Its less than I expected, but interesting. The GE Ecolux is a 32 watt and outputs 3000 Lumens. Its still looks superbright to me SPX41.

greg

It's all in the phosphors.

You might want to ask this question over on sci.engr.lighting

If they are still speaking to us, that is. ;-)

You might also want to give a few more details about what sort of lighting levels (or what kind of tasks will be performed) in this structure. Efficiency might be a factor of lighting levels, CRI, etc.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
Bloody typical, they\'ve gone back to metric without telling us.

Since we don't have the details that Paul has mentioned, I'm going to go out on a limb and generalize here.

For general lighting it's hard to beat the efficacy of high quality 4-foot T8 (1 inch diameter) fluorescent lamps on quality electronic ballasts and mounted in high performance fixtures. Compact fluorescent lamps are less efficient than linear fluorescent lamps. Incandescent lamps are much less efficient. Metal halide lamps come close, but their have terrible lumen depreciation so their mean efficacy over life is worse than fluorescent lamps.

My generalization falls apart if your application does not need uniform light levels. If you can use task lighting in critical areas and lower light levels in other areas then there may be better solutions.

All linear fluorescent lamps are not equal and all ballasts and fixtures are not equal. The lap data provided by the "big three" can generally be assumed to be accurate so you can compare different T8 lamps so discover their minor differences.

Some people prefer T5 linear lamps, but they are no more efficient than T8 lamps of the same length when both are operated on electronic ballasts.

If you are operating from DC power I would suggest purchasing ballasts designed to operate from your DC power source instead of using AC ballasts and a DC-to-AC converter, which will only waste power.

I'm sure you will get other opinions.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail: 
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only. 
It may not be used in any publication or posted on any Web 
site without written permission.

It would be necessary to know the available power to give you a good answer. Fluorescent is the way to go, but you could limited to single 9 watt lamp to several multi tube 4' fixtures depending on available power and required runtime. John

Thanks for all the answers, and more thanks for all the questions. The "metal building" in question is an aircraft hangar that doubles as my airport electronics lab. I use the lab to validate and true up designs done for both my manufacturing company

and my monthly magazine column designs

The lighting will be used for maintenance on the aircraft and general hangar lighting for the machine shop, the parts bins, and the like. Each electronics bench has its own individual two-tube fluorescent fixture that gives me very good lighting on a microscopic scale.

As to "how much power is available", the answer is that I've got a 3 kW gasoline generator that kicks in when the voltage drops below a critical level on the 12 volt batteries being charged by the solar panels. The answer is that I'm limited to something below 3 kW over the long haul, but could give you 5 kW on a short term basis.

Now, since the consensus seems to be that a fluorescent fixture with 4' bulbs is the most efficient, how do you select the fixture. I understand that the bulbs themselves have a lumen/watt rating on them, but how do you determine the efficiency of the ballast/fixture itself?

Jim

...and get the windows from Gates...

Just pick up a two bulb fixture with a large reflector and an electronic ballast and then pick up a pair of 25 watt bulbs in place of typical 32 watt bulbs. If you are going with 2 two bulb fixtures better to get a 4 bulb ballast and wire the two fixtures together with that one ballast rather than use two ballasts. You might even consicer single bulb fixtures if you can find them with a large reflector and use a two or four bulb ballist and up to four fixtures connected together. There are even 17 watt bulbs out there if you can find them.

Richard.

NOT recommended. You just can't install any wattage fluorescent tube in any fixture even if it fits. The voltage and current characteristics of the ballast is designed for a particular bulb. John

When speaking of available power. How much is available for lighting? Lets say you supply 3 kw continuous. What is the maximum load your equipment puts on the supply. Subtract that from the total and this is what is left for lighting. Wouldn't hurt to give it a 20% cushion. You probably should have an electronic engineer evaluate your situation. You might have to deal with power factor correction to get the best out of your system.

Electronic ballasts will give an edge in efficiency. If the building is close to a square in shape, I can see four 8' tube fixtures (two bulbs per fixture) in two rows across the ceiling if it is sufficiently high. It would use under 1 kw.

John

Agree. If the lamp type is listed on the ballast label, then that lamp is O.K. to use. If the lamp is not listed, check the ballast manufacturer's web site and the ballast listing. The ballast manufacturer wants to make that ballast as widely used as possible, so if the lamp is not listed for use on a particular ballast, there is a reason.

Terry McGowan

8-foot fluorescent lamps are slightly more efficient than 4-foot lamps of the same type. I suggested 4-foot lamps because they are almost as efficient as 8-foot lamps, are much more convenient, are available in a greater variety of types and colors and may be less expensive.

To find the most efficient ballast you will need information that is available on the ballast data sheet and may also be printed on the ballast labels. To compare the efficiency of various ballasts for the same type and number of lamps, compute the Ballast Efficiency Factor (BEF). The BEF is the Ballast Factor (BF) divided by the ballast input power. You can only compare the BEF for ballasts designed to operate the SAME TYPE AND NUMBER OF LAMPS. Also, if any ballast under consideration operates more than one type of lamp, remember to use the BF and input power for the same type and number of lamps when you compare ballasts.

If you decided to use a DC-input ballast, as I suggested to eliminate the additional losses of an DC-to-AC converter, then your ballast choices will be limited.

I normally don't work with fixtures, but you can also find fixture efficiency data on fixture data sheets. I would not choose 'any" fixture, as has been suggested.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail: 
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only. 
It may not be used in any publication or posted on any Web 
site without written permission.

25-watt lamps will reduce power consumption, but will also reduce light levels. For a new installation designed for a specific light level it is not clear that 25-watt lamps provide either the most efficient or most economical solution.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail: 
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only. 
It may not be used in any publication or posted on any Web 
site without written permission.

In , Homer J Simpson wrote in part:

My experience is that in compact fluorescents, lousy phosphors are only or almost only used in ones available at dollar stores, or of brands usually found at dollar stores.

- Don Klipstein ( snipped-for-privacy@misty.com)

That's about 100 lumens per watt. The best white light LEDs can do about half that, which is worth considering for small lights eg bench spotlights etc.

--
Dirk

http://www.onetribe.me.uk/wordpress - The UK\'s only occult talk show
Presented by Dirk Bruere and Marc Power on ResonanceFM

We should keep in mind just how much difference there is between some of these options. In many cases we could be talking about fractions of a percent, that could be easily overwhelmed by other factors.

Assuming that $$ are not unlimited you may soon find a region of vanishing returns.

It is not well understood by the general public how the temp of the bulb wall effects output and therefore efficiency. In a metal walled structure I would look *very* closely at that.

For maximum efficiency you must start with how the light is distributed over the area of interest. Then (for fluorescent) the lamp and ballast as a pair! Then move to the fixture construction.

Furthermore you are using a 12VDC source and the options in ballast are quite limited.

Richard Reid, LC Lum> Thanks for all the answers, and more thanks for all the questions. The