I claim no expertise on this. But the link below has some information that may help.
It is not known how honest are the watts and lumens posted on the packages of CFLs. I have some GE #74726 that have an outer non-glass shell similar in shape to the conventional A-19 lamps. We are using a few of them and like them very much except that warmup to full brightness is quite slow. 20 Watts, 1100 lumens for 55 lumens per watt. Check out the claims for other brands and sizes.

Here is a link that might be useful: Lighting

OK, I just found the GE product on Amazon. But I bought mine at Ollies for just $2.99 for the pack. I feel like a thief- no, not really. Some of Ollies stuff is not a bargain. This price prompted me to try a pack and then I bought more a few days later.

Here is a link that might be useful: Amazon

Thanks, at one time or another, I think I was at that first link. It looks like the tubes are significantly more efficient in pure power consumption.

What is the deal with commercial vs. residential ballasts? The FCC allows more interference in the commercial product?

It's difficult to compare long tube to CFL since you now have to consider lumens versus lux.

Simply put, lumens is the total of the measured light output, while lux is the measured light output over one square meter.

A long tube bulb fixture will illuminate a much larger area, while a CFL may provide brighter light but within a smaller area.

I've just recently really started looking at the lasted LED replacements for long tube fluorescent - they have much more potential to illuminate the same area as traditional tubes with a much lower energy cost.

While I'd suggest waiting to replace them, I'm currently replacing three, 2-tube, 40 watt/bulb fixtures in my 120 sq-ft electronics workshop with six 40 watt equivalent LED A19 bulbs. Each is roughly 7 watts, which means 42 watts now replaces 240 watts of fluorescent lighting. Since this is a small workshop, the illumination is actually better than before.

> "I have some GE #74726 that have an outer non-glass shell similar in shape to the conventional A-19 lamps. We are using a few of them and like them very much except that warmup to full brightness is quite slow. 20 Watts, 1100 lumens for 55 lumens per watt. Check out the claims for other brands and sizes."

GE now sells these in a "Bright From the Start" variety in 20w (75w equivalent, #64161), and in a 15/60w version. They too enclose the spiral in a standard light bulb enclosure (real glass, I think). They perform their trick buy putting a tiny halogen bulb in the middle of the spiral that turns on for the first 30 seconds as the CFL tube warms up. Then the halogen bulb turns off. I've used the 20w version and besides the instant-on and the more attractive appearance of the bulb when exposed, can vouch that the light quality is the best I've ever seen from a CFL - indistinguishable from an incandescent bulb to my eyes (in the standard 2700K Energy Smart version anyway - avoid the "Reveal" CFLs with their weird 2500K color temperature and low color rendering IMO). The only real issue with them is the sudden drop in brightness when the halogen bulb turns off, but you'll get used to it, and it's barely noticeable if you have a 2- or 3-light fixture and use standard CFLs in the other sockets.

Oh, and CFLs and T5/T8 fluorescent tubes are similar in terms of efficiency.

For whatever reason, T8s turn on at full brightness - wonder why it's so difficult to make a CFL do this.

Since the previous post, I have learned that the GE lamps I have do have a glass outer shell. It just does not feel like glass, so I assumed that it was "plastic". And the particular ones I have may now be discontinued, but this is not confirmed. GE brand, but made in China.

4' tubes and CFL's seem to have different applications in my experience. I haven't done any math and ignoring efficiency but I'd suspect that "long lamps" have a lower cost of ownership because you are not throwing the ballast away every time the lamp fails.

Old style is probably not as efficient as CFL's, but the newer T series are about on a par or better The ones with electronic ballast do SUBSTANTIALLY better than CFL's. Your typical CFL runs from 45-75 lm/watt where as the T8 with an electronic ballast runs 80-100 lm/watt.

LED's by the way can theoretically triple these numbers, but right now in practice they're about the same as fluorescents (their big win is MUCH longer life).

"...but I'd suspect that "long lamps" have a lower cost of ownership because you are not throwing the ballast away every time the lamp fails."

That is what one would first think, but with CFL production exploding and approaching commodity pricing, it may no longer be the case.

I'll compare two options based on what's available at my local big-box store:

A 4-pack of 13-watt Sylvania CFLs vs. a 2-pack of GE 4' T-8 bulbs

Cost: CFLs: $10, Tubes: $8
Total Lumens: CFLs: 3300 (825 per bulb), Tubes: 5600 (2800/bulb)
Life: CFLs: 12,000 hours Tubes: 20,000 hours
Total wattage: CFLs: 56 watts, Tubes: 64 watts.
Lumens per watt: CFLs: 59, Tubes: 87

To me, non-compact fluorescents still totally humiliate CFLs. They last longer, cost less, and put out a LOT more light per watt. (I very deliberately selected name-brand bulbs to compare, not "whatever was cheapest.")

"I'm currently replacing three, 2-tube, 40 watt/bulb fixtures in my 120 sq-ft electronics workshop with six 40 watt equivalent LED A19 bulbs. Each is roughly 7 watts, which means 42 watts now replaces 240 watts of fluorescent lighting."

Er, what? "40-watt equivalent" is equivalent to 40 watts of *incandescent* lighting, not fluorescent. I found " EcoSmart A19 8.6-Watt (40W) LED Light Bulb (E)* can replace a standard 40-watt bulb." If that's the kind of bulb you're talking about, then each bulb is rated for 430 lumens. Six of them will thus be about 2580 lumens. That's less light than ONE of the six fluorescent tubes was providing. A typical 40W T-12 fluorescent tube is delivering 3300 lumens, so your existing shop lights were providing just under 20,000 lumens. You will need forty-six of those LED bulbs to provide the same amount of light.

The EcoSmart bulbs are rated for 8.6 watts. Forty-six of them would thus use 395 watts, versus the 240 used by your current fixtures.

A far smarter thing to do would be to just replace the ballasts in your current fixtures with ones designed for T-8 bulbs. Same amount of light, but 32 watts per bulb, or 192 watts total. Less than half that required by the LEDs.

snarke,

Thank you for your post and I completely agree with your numbers. No, I wasn't attempting to imply I was getting the number of lumens and yes I know the 40 watt equivalence is to incandescent bulbs. though I can see how my two uses of the term 40 watts could be misinterpreted.

I was simply attempting to point out that while the fluorescent tubes are the traditional choice, sometimes other type of lights can work better.

Like I said, this is in a small electronics workshop (built to look like a garden cottage on the outside). It's 16'x8' and contains test equipment such as waveform monitors, oscilloscopes and video & computer equipment - the two benches also have their own lighting - so high brightness overhead lighting was not a goal, but lower cost and cooler operation was.

I was originally going to use 6) 60 watt incandescent bulbs on a dimmer but that's still 360 watts. So, as typical, the solution was to go with 3) 2-tube fluorescent fixtures which brought it down to 240 watts & cooler operation. Still if I'm working out there for 8 hours, that's 2 kW just for lighting.

When I said the illumination was better, I was referring to the warmer color temperature and less harshness, which works better in my application - a small electronics workshop.

Now in my woodworking workshop, it's all fluorescent, low-temperature, tube fixtures, big time.

Your solution (other than perhaps cold temperature performance) would be done cheaper and at the same (or better) efficiency with just using fewer or lower wattage CFLs

Yeah, but what am I going to do then with that case of LED bulbs I bought?!! : )

I haven't had any luck with the spiral cfl bulb in any 'overhead' application that is in either an enclosed fixture or in a recessed light application. I do have over 12 in redneck candeliers (porcelain/plastic) socket with bulb exposed with no issues. I have 4 in the garage and 8 in the basement that are probably close to 5 years old with no issues. If I put one in a recessed light they will not last a year. I tried the enclosed ones that are supposed to be floods with a little better luck but not much. Almost all were GE or Sylvania with a few china's. The China bulbs did die the fastest. So far I would have been better off to just keep incandescent floods installed in money spent.

I have had two light switches fall victim to the cfls and slow switchers which I am not sure if its the fault of the cheap switch... the slow switcher or if there is some static or feed back when a number of cfls are slowly switched. It starts out as a little arc buzz/crackle and gets worse and then one day the switch won't do anything. The builder used the cheapest stuff he could when the house was built so that doesn't help. Oh I also had a garage door opener that had a cfl in it with a poor connection and the lead was ate off the bottom of the bulb and 1/8" or so of the spring tab was burned off. I don't know... I know that some things can be like a capacitor and arc and really do interesting things in bad connection. The garage door lights had been installed in the units for less than a year.

While power factor has been an issue with CFLs, often around 0.55, the inrush surge current on startup has been much more of a problem.

The AC input goes directly into a bridge or half-wave rectifier and charges a filter capacitor for the DC input to the switching power supply. That instantaneous inrush current can be around 2-5 amps, and laboratory tests have measured as much as 14 amps - if only for a few milliseconds.

They do use a current limiting fusible resistor before the rectifier, but it's mainly for the safety factor. Too much resistance would simply create heat and then just reduce efficiency.

Put 8 CFLs together on a switch, and you can have a pretty good size arc that can take out switches, or burn poor wire nut connections. I personally have just had to replace a three-way table lamp switch that failed from what appeared to be just that type of inrush current.

In a garage door opener, it has vibration that can loosen the lightbulb in the socket, and the socket also has to carry the inrush current. Some looseness, oxidation and such and I think your results are what's to be expected.

I was previously unaware of the large inrush current involved. One of the things not usually recognized by the populace is that the electricity suppliers must have online the capacity to handle any load prior to that load being imposed on the system. The alternative is a crash of the "grid". So the switch to CFL does not lessen the need for conventional generating capacity as much as one might expect due to the starting inrush.

"Simply put, lumens is the total of the measured light output, while lux is the measured light output over one square meter."

Measuring per square meter is not the same as "measured light output over one square meter."

The area is not limited in the lux measurement (lumens/square meter).

I didn't even think about the inrush current and that would explain a lot since there are 4 CFL's on that circuit. I am noticing more and more switches becoming noisy. I have one in the basement with 8 cfls on it but I used a higher grade switch than the ones laying loose for less than a buck and I can't even make it buzz intentionally.

Good switches are a must if there are many cfl's on the circuit.

"I was previously unaware of the large inrush current involved."

The older magnetic ballasts had a similar issue.
An inrush current to set the magnetic field.

The newer electronic ballasts have have an inrush fr charging the input capacitors in the electronics power supply and maybe a sealer inductance depending on how the power supply is set up.

They need to create a stable DC voltage to then operate the high frequency switching step up supplies for striking voltage and operating voltage.

They end up like many electr4onic loads with a mostly capacitive appearing loads often with VERY non-sinusoidal current resulting in 'power factors' not really being very valid (they draw short bursts of current at the peak of the voltage waveform, especially after they are up and running).

The entire power system was not really designed to handle these type of loads. Loads have been mostly inductive (motors, magnetic ballasts, etc.) or a long time.

The odd loads have already created problems with 3-phase neutrals in large buildings with lots of PCs and their power supplies.
Neutrals had been allowed to be smaller, but it was found the non-sinusoidal current had a higher RMS value meaning larger conductress are required, not smaller (essentially the phase current failed to cancel in the neutral).

this is one of the things that lead to PC power supplies being required to have 'cleaner' power factors.

The same issue applies to ballasts.
The magnetic inrush was pretty well understood, now we have a different current demand.

It took a little searching but here's an accessible link to a UL white paper done in 2010 about some of the technical electrical problems with changing over to CFLs.

This is the one I was searching for, but if you Google something like "inrush current of CFLs" you'll find many studies that were done regarding the very same problems "countryboymo" observed.

Here is a link that might be useful: UL Whitepaper

That is awesome Yosemitebill.... I thought I was nuts and it was coincidental but I did swap some 'regular' bulbs on one of the noisy switch circuit and it went from very audible to more buzz.

"I did swap some 'regular' bulbs on one of the noisy switch circuit and it went from very audible to more buzz."

Using T12 tubes in a T8 fixture may result in poor performance, including damage to the tubes and ballast.

The strike voltage, heater current, and operating current are not correct for T12 tubes.

I grew up in a rural area, in near poverty for a while, and we had a few domestic animals of various kinds to help provide food for our table. Fluorescents are somewhat like a stubborn steer that we had. In my early teens, it was my job (among other things) to take care of the steer. Often that meant moving it from one place to another with a chain. The thing was stubborn and would not move with just a tug on the chain. So I had to slack the chain enough to get mostly behind the animal and kick him in the a-double-rear to get him moving. He would then try to run and I had to hold on to the chain for dear life to keep him from getting away or from pulling me off my feet.
The ballast first provides the kick and then the restraint for the fluorescent tube.

"kick him in the a-double-rear to get him moving."

You should see them move the first time you hit them with a 'stun stick' (the big 6 D-cell model).

Do not use a stun stick to move the through a dip trough if you are already wet.
My cousin split his scalp open once when he stunned a steer to get it moving the rest of the way through a dip trough and fell off the fence he was perched on.

Sorry for the typo. I meant to say NO more buzz with the incandescent bulbs. I tried some more to 'slow switch' the cfls on the higher quality switches and still can not get an audible noise.

If you have noisy switches with cfl's be prepared to replace them.

Sorry for the earlier typo.