CFL brightness and longevity claims

This thread is for discussing my article CFL Fluorescent Light Bulbs are a RIP OFF! - specifically the parts that talk about CFL brightness and longevity.

I must say I agree 100%!  Seriously, about 10 years ago I spent over $100 online to buy replacement CFL's for all my lights, and I never saw my electric bills go down.  I even replaced the 15 traditional candelabra bulbs in the chandelier with monster-sized ugly CFL ones, and no difference.  A few years later we installed a dimmer switch and replaced all those bulbs with regular incandescents.  Dimmer switches are worth it.

We still use CFL's in most of the other fixtures, except 2 in an antique fixture in the original bathroom (they looked horrible and provided hideous light through the frosted shades).  This summer we installed 2 outdoor frosted lantern-style fixtures at the back door and used CFL bulbs.  At first they were way too bright - almost illuminating the whole yard.  However, now, just a few months later they are quite dull - not even half as bright.  I don't know if it is because they are older or because it is cold out, but the difference is amazing.  It sucks because now (winter) is when I need the light out there - to see into the woodshed.

I want to conduct an experiment, but I need a part. 

I need something that is like a timer switch that turns light on at 6 pm and off at 9pm.  Only I need it to turn lights on for 30 seconds and then off for two minutes and repeat forever. 

This is something that is beyond by abilities to cobble together.

I want to hook it up to a panel of a dozen lights.  Half will be incandescent and half will be fluorescent.  I will then run this bank of lights and take a picture every day.  I expect that all of the lights will have burned out in six weeks of this, but I want to document how long they each made it. 

So - I need this contraption.  A switch.  I hope that somebody reads this and thinks "oh, that's easy!" - and then I want to know how much you would want for such a thing.

In computer server racks they have power strips that have ethernet and/or rs232 ports for controlling the outlets. They are expensive, but maybe you could scrounge one somehow. I'm sure you could handling scripting it to toggle the power. But you'd need some kind of computer running, but maybe you need that anyways for the time-lapse photos.

http://www.amazon.com/s/ref=nb_sb_noss?url=search-alias%3Delectronics&field-keywords=managed+power+strip&x=0&y=0

I guess I'm kinda hoping that somebody will take a 24 hour timer and somehow rig it to represent 24 minutes - or something like that.

I suppose I could do this with something that connects to my computer.  But $200 is way too much.

This should do what you seek Paul, notice the selectable time relays.

Found at Grainger, I can get you the part numbers if you like. If I recall each relay is rated at 10 amps which is more than you need. I whipped this toy up as a demonstrator for the students, each bulb on for 5 seconds rotating in one direction for 2 minutes and then rotating the other direction for two minutes.

I could draw you a crappy wiring diagram not crappy electricly I am just no artist and you could then have granger drop ship you 2 relays and bases, lol like I have said before, I am kinda lazy so I might have a student do the diagram and it will be very pretty!

Let me know if you want one or want more info on ordering the parts. If they give you trouble about shipping to a non-business let me know and I can fix that too.

I don't need anything that fancy.  And I suspect it is way beyond my budget. 

That is five switches.  I need only one.  I can wire up a board with a dozen lights on it and one pair of wires.

It's the timer that I need.  30 seconds on, two minutes off.  How might I get just the timer?

This looks like what I need:

http://www.grainger.com/Grainger/INTERMATIC-Cycle-Timer-2E355



repeat every ten minutes.  5 second resolution.  Triggers for on and off within the ten minute cycle.

Still, $159 is outside of my budget.

I only mentioned those strips because I suspect they end up in the trash often.

You can get an Arduino* for $30 and a relay for less than $5.

You can get some 555 timer chips for a few dollars, but then you need all the resistors and capacitors and such, and a case and soldering stuff... Do you know anyone into electronics? A 30 on/ 120 off 555 timer circuit is an Electronics 101 Lab Exercise.

*Very easy to program microcontroller that you could use for other projects later on.

paul wheaton wrote:
I don't need anything that fancy.  And I suspect it is way beyond my budget. 

That is five switches.  I need only one.  I can wire up a board with a dozen lights on it and one pair of wires.

It's the timer that I need.  30 seconds on, two minutes off.  How might I get just the timer?

The photo was just to show you the part, all you need is two relays and two bases. 6a855 is the Grainger part number for the relay, you will have to look up the base when you scan the catalog.

I was offering to draw you a wiring diagram to use the relay to do what you seek.

I used a quad tube 96 watt compact flourescent aquarium light for a while, because my 2X100 Watt fixture was too dim for the room with the incandescents in it. It was too bright, and I had to cover it, but it worked really well for years, and before that I had it above my aquarium, on for 16 hours a day, 365 days a year, for three years, so I got at least 20,000 hours out of it, and it was still damn bright, but it broke when I was moving. It was a coralife retrofit, and it cost something like $75 in 2005 for the ballast, another 35 for the lamp.

Professor Rich wrote:
The photo was just to show you the part, all you need is two relays and two bases. 6a855 is the Grainger part number for the relay, you will have to look up the base when you scan the catalog.

I was offering to draw you a wiring diagram to use the relay to do what you seek.

A relay and a matching base work out to just shy of a hundred bucks.  So it would be about $200 for two.  The other contraption is cheaper and still too expensive.

Thanks for the offer though!

paul wheaton wrote:
A relay and a matching base work out to just shy of a hundred bucks.  So it would be about $200 for two.  The other contraption is cheaper and still too expensive.

Thanks for the offer though!

Didn't yet make an offer, but I have one now!

In the name of education I will make you a deal. Since I have over a hundred of them I will ship you 3 (its education and you might goof one up) and I will look around for an analog manual for one of them as well, you have to figure out the wiring yourself or seek local assistance, then when you are done with them, you ship them back, that way they are not stolen property.

I am really pretty sure you are going to have a hard time finding another method, adjustable time relays are not standard equipment in these days of PC boards and while I know a few analog potentiometer boards that would work, they are not cheap these days due to obscurity.

Let me know and if you want em, shoot me an address via pm to ship em to. FYI if you look at that earlier post, I hit ya from the college server in case you wanted to confirm but ask it remain confidential.

Ned

Paul
Here is a source for a five minute timer. The on time can be adjusted from 1 second to 5 minutes as well as the off time. Assembled it is $28.95. The output relay is 120 volts ac and 3 amps contact rating. This can be ran into another relay for larger loads.
Here is the source for the board.
http://tinyurl.com/6hgl6sa

Ken

Wow .... all I had to do was ask and look at this awesome bounty!

I am a bit out of my element here.  I have done some home wiring and the like, so I think I can do either. 

The important thing with the contraption that Ken talks about is that I will own it and can then use it for future wicked experiments I haven't thought of yet.

Yet it is painfully difficult for me to turn down "free" - although I guess I will need to pay for postage to send it back. 

Ken - I'm guessing that the way to modify that is to set the on time and the off time?  So while it has a capacity of five minutes, I will be setting it for 30 seconds on and two minutes off, so it will cycle every 2:30 (not every five) right?

Rich, what do you think of the contraption in Ken's link?  Does it seem easier for a guy like me to figure out? 

Ken, what do you think of Rich's relays?  Have you used that sort of thing before?

That looks like exactly what you need. You'll have to fiddle with the two knobs and observe until you get 30 seconds and 2 minutes.

A 60W bulb will draw about half an amp. I don't know how to figure it out for a CFL. But you might get close to the 3A limit if you're using 12 lamps...

travis laduke wrote:
A 60W bulb will draw about half an amp. I don't know how to figure it out for a CFL. But you might get close to the 3A limit if you're using 12 lamps...

Excellent point. 

Maybe I should scale this back a bit.  Maybe six lamps.  Two incandescent and four fluorescent. 

Of course, the next problem is that the fluorescent lights need a big power spike to get started.

Anybody out there ever measured how much of a spike?

Paul,
You would turn the 2 pots on the board to control the on and off times. Rich's relays are nice but overkill for you.
I have been in electronics for over 35 years and still repair commercial satellite equipment.
The relay that is in that board could be replaced by what is called solid state relay or SSR's.
Here is one such device.
http://www.opto22.com/site/pr_details.aspx?cid=3&item=120D10

This operates on 3 to 32 volts Dc to control the relay and the output is 120 Volts AC at 10 amps. You can get larger amps rating for what ever you want. You could use 2 of these relays on 1 board because these relays pull only about 10 mills of current each when activated.

Ken

paul wheaton wrote:

Rich, what do you think of the contraption in Ken's link?  Does it seem easier for a guy like me to figure out? 

Ken, what do you think of Rich's relays?  Have you used that sort of thing before?

I have, timer boards like that are not uncommon and it should do the job for the experiment you are wanting to run. Sometimes those pots are a bit sensitive to temps meaning they can operate at slightly different time cycles in differing temperature ranges, they are not used where such things are critical. A common use is equipment shut down. Think of a conveyor that has boxes traveling on it, you have a trip switch with a long arm on a light spring, once the last box passes the timer begins to time out as power has been applied, if no more boxes come by and kill the power to the timer, after the adjustable time limit, the relay shuts down the conveyor.

As to Ken's later comments about the relays I have offered, he is spot on, they are indeed overkill. They will work and you will never approach the capacity they have, well you could do your test with perhaps 20 light bulbs and get close to the limit if it suited your fancy lol.


One other comment "All I had to do was ask" well call me captain obvious here, but perhaps some folks need to hear this from the horses mouth.

Go to your local vocational training centers!!! As a teacher albeit on a slightly higher level than that, the hardest thing to do is to keep student interest. Nothing builds the students interest and pride more than a "real world project" and since the schools are rarely set up to collect any funds, your going to find they will do most stuff for free.

I can not count the number of welds a student must do on a 6" line, various angles etc but a straight line over and over, seriously boring stuff. We weld stuff for others all the time, not commercially but helping Mr. Fixit who does not have a welder but brought his bird feeder hanger in and we whipped him a new one out.

We perform clean and checks on window air conditioners a local charity gives to the poor.

You gotta talk to the teacher, not the admin folks or you will get the liability speech but a lot of items you folks might seek that are common place like this timer  set up are great projects for electrical students and I would have a student assemble such a thing, test it out and then cut the plug off the end, unplug the relays and put it in a box for you to take home without ever batting an eye. Several of my students would perk up at the real world project, others might even complain they did not get to do it etc.

I am just saying, lots of fixit guys become vocational teachers and students love real stuff compared to assignments and the learning process is enhanced. You might get a school near by to do the whole thing for free or even join in a comparable experiment themselves, I would if you were local to me, in a heart beat.

Hi Rich,
In Paul's testing he don't need a lab test setup. But I don't think I would do what he is attempting to do either. The reason is this. A 60 watt bulb when first turned on will pull a little over 1 amp then will run at around .5 to .6 amps.
A CFL bulb, 15 watt, is just a little under that. Turning on and off an Incandescence bulb is normally the failure rate of the bulb. I would just turn then on and see which last the longest and also measure the lux from each. I will guess that the CFL bulb will lose.
I am glad that there is someone like yourself teaching students electronics.  The art of troubleshooting and repairing equipment has been lost. Now its just replace modules and throw the old module away.

Ken

Oh I agree with you 100%. If I were to be running a test on something like this, I would just toss a quick ladder logic program into a plc so I could monitor the inrush current on each even evaluate performance of different brand same design bulbs etc. The old box mounted pots you had to be sure were marked "I" for Incandescent bulbs instead of other usage that did not have the inrush.

I happen to disagree with Paul on this one, I think the CFL performance will be much better than the Incandescent bulbs as that has indeed been my personal experience. I also think neither one will come remotely close to LED and I find LED more pleasing than CFL light even with diffusers.

I have used the relays in the picture to replace extremely overpriced pc boards on more than one occasion.

I can't fall into the trap of teaching the students replacement criteria, they need to know not only what it does, but how it does it, and if they are truly interested I like teaching em WHY it works that way as IMHO once I can get them into "why" they no longer need me and they will be on the path to self teaching.

I have ordered a light meter and a really fancy kill-a-watt.  I plan to mount both of those (for the sake of video) and then test several light bulbs to get the lumens per watt and compare it to the packaging.  I think the first 30 seconds of use for an incandescent will be very predictable.  I think the first 30 seconds of use for a CFL will not comply with what the package says.

As for the test where the lights are kept on for 30 seconds and then off for two minutes.  I suspect that the incandescents (labeled for 1000 hours) will outlive the fluorescents (labeled for 10,000 hours).  I also think they will all peter out before 500 hours. 

Paul,
My alma mater has one of the foremost lighting research centers on the planet (or so they told me). They have a searchable database of CFLs with performance metrics located here: http://www.lrc.rpi.edu/programs/nlpip/screwbase.asp Thought this might help you get the data you wanted. There's also a place to search their publications: http://www.lrc.rpi.edu/searchpublications.asp here's an example publication on CFLs (not screw base models, though) http://www.lrc.rpi.edu/programs/NLPIP/PDF/VIEW/SRCFLResidentialDownlights.pdf
My GF took some courses at that place for her lighting concentration and raved about the place.
-TL

Trying to load this page and it keeps giving me an error.

paul wheaton wrote:
Trying to load this page and it keeps giving me an error.

Huh, for a tech school, you'd think they'd have better server maintenance...  lol
Sorry bout that. Best bet 'till it's back up would be to search the publications. There should be a pdf or two on screw type CFLs. Maybe even one comparing them to incandescents.

I don't know where this fits into the discussion between CFL and incandescent lamps, but here in the SW incandescent lamps contribute to the heat load that our air conditioning must deal with many months of the year. That was one of the reasons I originally began using CFL's many years before they became popular. 

paul wheaton wrote:
I have ordered a light meter and a really fancy kill-a-watt.  I plan to mount both of those (for the sake of video) and then test several light bulbs to get the lumens per watt and compare it to the packaging.   I think the first 30 seconds of use for an incandescent will be very predictable.  I think the first 30 seconds of use for a CFL will not comply with what the package says.

As for the test where the lights are kept on for 30 seconds and then off for two minutes.  I suspect that the incandescents (labeled for 1000 hours) will outlive the fluorescents (labeled for 10,000 hours).  I also think they will all peter out before 500 hours. 

Paul,
That is the way to test these. I have found out that on the CFL's when testing them, Let them run for about 3 minutes before shutting them off. I have noticed when using one that in that length of time they get brighter.
Si I would also keep time as another monitor point.

Ken

I ordered the timer contraption from canada.  $38 with shipping.  Although I'm not sure if it comes with a power supply or if I need to provide one. 

I used to have a box full of power supplies, but I got rid of them about a year ago.

Can't decide if this is hype or if this product seems worthwhile:

Original Article: "Hybrid Light Bulb Arrives on Earth Day"

"Some folks just don’t want to give up incandescent light bulbs, so GE developed a “customer-inspired” bulb to ease the transition.  The GE Energy Smart hybrid halogen/CFL bulb is being shipped nationally with soft white and Reveal options.  The new bulb — a blend of three light bulb technologies — has the shape of an incandescent, the efficiency of a CFL, and the immediacy of a halogen.

The halogen technology offers instant, no-delay lighting, a feature missing in some CFLs.  To save energy, though, the halogen element shuts down when the CFL element reaches full brightness.

GE started shipping the hybrid light bulb and it should be available nationally by Earth Day, April 22, 2011, according to a recent press release.  Pricing is subject to retailer discretion, but GE expects something in the range of $5.99-$9.99 per bulb, depending on the line and wattage.

The Energy Smart hybrid lasts about eight times longer than a typical incandescent (e.g., 1,000 versus 8,000 hours) and has less mercury than a typical CFL (e.g., 1 versus 2-4 milligrams).  The soft white bulb is 2700 Kelvin with 800 lumens (15 watt) and 1,100 lumens (20 watt).  The Reveal is 2500 Kelvin with 740 (15 watt) and 1,050 (20 watt)."

Any thoughts from someone more informed on the lightbulb scene?

Why fool with this when you can get an incandescent for much less?

If I were to use this in the one light that I leave on for any length of time, I might save a few cents a month and get lower quality of light.  For all of the rest of the lights, I get more light per watt with incandescent.

Agreed. The CFL are disappointing. So are LED's. What does work is the high intensity fluorescence used in the marine aquarium trade. I have had two IceCap electronic ballasts for a decade and been very pleased with them. You can get all sorts of spectrums and they are very, very bright.

Cheers

-Walter
Sugar Mountain Farm
Pastured Pigs, Sheep & Kids
in the mountains of Vermont
Read about our on-farm butcher shop project:
http://SugarMtnFarm.com/butchershop

Here is one of appropedia's pages on CFLs. They try to present pros and cons, and clarify some beliefs about CFLs, although they are mostly in favor of them. I think some of the info in this forum would be good for the folks at appropedia to get to hear too.
http://www.appropedia.org/CFL

I'm throwing in this little snippet from a 5 Apr 2011 grist.org article:

Every year, Americans recycle only 2 percent of the compact fluorescent light bulbs they toss. The unrecycled portion leads to the release of four tons of mercury into the environment annually. That’s almost 10 percent of the amount of mercury released by coal fired power plants, which are this country's No. 1 source of the fish-and-pregnant-woman-contaminating pollutant.

If you go to the link for the article above, there are links to additional background on the 2 percent and pollutant parts.

I bought a light meter.

I bought a brand new 40 watt incandescent bulb and three brand new fluorescent bulbs labeled "40 watt equivalent".

I made a video of the first 30 seconds of each.

The incandescent was very predictable, putting out almost exactly the same level of light.

This is the level of light the fluorescents put out relative to the level of light that the incandescent put out.  Followed by a further 25% reduction when the light reaches its half life point.

paul wheaton wrote:
This is the level of light the fluorescents put out relative to the level of light that the incandescent put out.  Followed by a further 25% reduction when the light reaches its half life point.

Is light measured on a pretty linear scale, or is it logarithmic like decibels? 

Excellent question.

I suspect that it is a straight scale because the meter has the ability to go up or down by a factor of ten.

paul wheaton wrote:
This is the level of light the fluorescents put out relative to the level of light that the incandescent put out.  Followed by a further 25% reduction when the light reaches its half life point.

Could you explain what you mean by half life?

Jocelyn Campbell wrote:
Could you explain what you mean by half life?

In this case, I mean that the life of the bulb is half over.

Professor Rich wrote:
Didn't yet make an offer, but I have one now!

In the name of education I will make you a deal. Since I have over a hundred of them I will ship you 3 (its education and you might goof one up) and I will look around for an analog manual for one of them as well, you have to figure out the wiring yourself or seek local assistance, then when you are done with them, you ship them back, that way they are not stolen property.

I am really pretty sure you are going to have a hard time finding another method, adjustable time relays are not standard equipment in these days of PC boards and while I know a few analog potentiometer boards that would work, they are not cheap these days due to obscurity.

Let me know and if you want em, shoot me an address via pm to ship em to. FYI if you look at that earlier post, I hit ya from the college server in case you wanted to confirm but ask it remain confidential.

Ned

Rich,

Does this offer still stand?

I got the other contraption and a friend hooked it up all proper.  It worked flawlessly for two hours and then stuck in the on position.  I cannot seem to get it unstuck.  So, this seemed to be a bust. 

My friend says that if you send these pieces, he can rig it up to do what I want.

this sums it up really well: the light bulb conspiracy