Light bulb comparison

[gar]

130929-0746 EDT

For my notes "Reading, Collecting, and Using DTE Smart Meter Data to Help Reduce Energy Use" I have run some new comparison measurements of power and light intensity on samples of Incandescent, CFL, and LED bulbs. See http://beta-a2.com/energy_c.html for a summary of the notes.

This experiment uses an 8" rim diameter aluminum reflector assembly, a Variac for voltage adjustment, a Kill-A-Watt EZ watt and volt meter, and a GE selenium cell light meter.

The reflector rim to light meter distance was 36", and voltage 120.7 V.

Only one sample of each bulb was used, but prior experiments with bulbs do not show large differences between comparable bulbs.

The bulbs were:

1. GE standard 120 V 75 W incandescent from at least 1970 to 2010 date of manufacture. Has only had testing use with negligible hours of on time. Average life was probably 750 hours. Cost unknown.

2. Phillips DuraMax Soft White A19 830 lumens 120 V 60 W incandescent brand new. Average life 1096 hours. Capital cost $0.50. Capital cost per 1000 hours of life $0.46 . At $0.16/kWh the operating cost per 1000 hours is $9.60 . Total cost per 1000 hours is $10.06 .

3. Home Depot Ecosmart Soft White 405 101 EDXO-19 2700 K CFL 1200 lumens 75 W equivalent 120 V 19 W 0.32 A actual rating. Average life 10,000 hours. Capital cost $1.50 . Capital cost per 1000 hours of life $0.15 . At $0.16/kWh the operating cost per 1000 hours is $3.04 . Total cost per 1000 hours is $3.19 .


4. Cree BA 10-04527OMF-12DE26-1U_00 2700 K LED 450 lumens 40 W equivalent 120 V 6 W 0.05 A actual rating. Average life 25,000 hours. Capital cost $9.97 . Capital cost per 1000 hours of life $0.40 . At $0.16/kWh the operating cost per 1000 hours is $0.96 . Total cost per 1000 hours $1.36 . Multiply these values by 1.5 for an approximate equivalent for 60 W replacement. Estimated 60 W equivalent cost per 1000 hours is $2.04 .


The results:

120.7 V all tests.

1. 75 W GE old incandescent
...... 31 FC .... 75.3 W ..... 1.00 PF

2. 60 W Philips new incandescent
...... 28 FC .... 62.0 W ..... 1.00 PF

3. 19 W CFL
...... 28 FC .... 18.0 W ..... 0.56 PF

1. 6 W Cree LED
...... 20 FC .... 05.5 W ..... 0.98 PF
multiply by 1.5
...... 30 FC .... 08.3 W

The 19 W CFL has comparable light output to the 60 W Phillips at 29% the power input.

The 6 W LED when adjusted to provide an equivalent light output to the 60 W Phillips used 13.4% of the 60 W bulb's power.

Neither the CFL or LED should be operated in a confined space. Use an incandescent inside refrigerators, in range top hoods, and inside ovens. Ceiling can light fixtures may shorten the life of CFLs and LEDs.

.

 

[Dennis Alwon]

Thanks Gar

This issue of Cfl especially when they first came out made me furious. There was no info on the package and no education to the public as to the proper functioning of these bulbs. People would install them in enclosed fixtures, recessed cans etc and the bulbs would last a few months at best. Add to that the fact that they were supposed to be disposed of in a proper fashion, because of the mercury, and not a word of that was mentioned to the public. I hope they die a quick life.

Unfortunate the LED followed the same path. No education about them and also no quality control , standards or guidelines as to how they should be built. Many burned out very quickly.

We still have a way to go but it is getting better.

I installed some LED wall warts but have not found a photocell that works satisfactorily. Some are gone in months. Of the 7 units I installed 2 years ago I have replaced all the photocells once and some of them 3 times. Not sure if the LED is the issue or just the manufacturing of the photocells.

 

[mgookin]

Thanks Gar

This issue of Cfl especially when they first came out made me furious. There was no info on the package and no education to the public as to the proper functioning of these bulbs. People would install them in enclosed fixtures, recessed cans etc and the bulbs would last a few months at best. Add to that the fact that they were supposed to be disposed of in a proper fashion, because of the mercury, and not a word of that was mentioned to the public. I hope they die a quick life.

Unfortunate the LED followed the same path. No education about them and also no quality control , standards or guidelines as to how they should be built. Many burned out very quickly.

We still have a way to go but it is getting better.

I installed some LED wall warts but have not found a photocell that works satisfactorily. Some are gone in months. Of the 7 units I installed 2 years ago I have replaced all the photocells once and some of them 3 times. Not sure if the LED is the issue or just the manufacturing of the photocells.

Are you saying photocells as in photocontrols are failing? And are they causing the lights to stay on all day?

 

[gar]

130929-1934 EDT

Dennis:

My first experience with fluorescent light was about 1940 with a twin tube fixture, 20 W per bulb. What an improvement this was for setting type compared with incandescent. About 1958 I personally started using 8' Slimline fixtures for lighting my work area. Now I am changing to CFLs for spot lighting, and 4' T8s for somewhat greater area lighting. The 8' will gradually disappear. I am not too inclined to change from fluorescent to LED yet. The cost benefit is not sufficient relative to capital cost.

My above total cost figures per 1000 hours look good for LEDs, but that is based on a long life expectancy, not proven yet.

.

 

[zbang]

3. 19 W CFL
...... 28 FC .... 18.0 W ..... 0.56 PF

Love that PF. (Not surprised, though.) Do you have any way to measure the harmonics?

 

[gar]

130929-2419 EDT

zbang:

No way to measure harmonic content. But it is the result of a simple rectifier with a capacitor input filter. Large current spikes near the peak of the voltage waveform.

.

 

[Ragin Cajun]

Unless I'm mistaken, over driving the LED to get the equiv light output will shorten it's lifetime.

The POOR PF of the CFL is no surprise to me. Just remember to use VA to determine circuit load, NOT watts!.

Another issue with LED's is the huge inrush. Determining how many you can put on a circuit is a ____ shot. They have a long way to go! I love seeing the shock on clients when I tell them their precious LED fixture has a power supply. They are shocked! And if one LED goes they have to replace the whole fixture. Tee, hee. Don't know about your area, but many, if not most of the new fancy LED traffic lights around here have numerous LED's out withing a couple of months, or less. Then there's the problem in colder areas where snow covers up the LED's and you don't know if you have the green or red because there's not enough heat to melt the snow. That has caused accidents. Unintended consequences! hmy:
RC

 

[gar]

130930-0845 EDT

Ragin:

Yes, as current thru an LED is increased its average life is decreased. Same for an incandescent or fluorescent.

PF for a CFL is low because these are being made at absolutely the lowest cost. LEDs have a much higher price and some have circuitry to improve PF. Both CFLs and LEDs usually require a DC supply internally.

Inrush current problems can be solved at some cost.

.

 

[gar]

130930-0954 EDT

Package labeling has changed.

Voltage does not appear on the package, just on the device.

Life is now in years at an assumed 3 hours per day. So a 1000 hour incandescent is now 1000/1096 = 0.91 years. Marketing gimmicks.

.

 

[JoeStillman]

I am finding that LED lamps and fixtures are still priced as a novelty - people will buy them becuase they want them, not because it makes $en$e.

Only engineers look at life-cycle cost. Everyone else does what makes them feel good.

 

[gar]

131001-2117 EDT

Incandescents dim very well with either a Variac or phase shift dimmer. CFLs do not, whether standard or so called dimmable. The Cree I mention above is quite good with both a Variac and a phase shift dimmer.

For me, at this time, I would use a straight tube fluorescent with a dimming ballast vs LEDs.

.

 

[lakee911]

PF for a CFL is low because these are being made at absolutely the lowest cost. LEDs have a much higher price and some have circuitry to improve PF. Both CFLs and LEDs usually require a DC supply internally.

The POOR PF of the CFL is no surprise to me. Just remember to use VA to determine circuit load, NOT watts!.

Please help me understand... how does low PF impact the typical homeowner and residential service? They're not being dinged for low power factor, and they're being charged by the KW not the KVA, so wouldn't lower PF being advantageous?

Jason

 

[GoldDigger]

Please help me understand... how does low PF impact the typical homeowner and residential service? They're not being dinged for low power factor, and they're being charged by the KW not the KVA, so wouldn't lower PF being advantageous?

Jason

1. Neutral currents in MWBC.
2. I can see not being a disadvantage, but never an advantage. Low PF does not result from reducing the watts, it results from increasing the VA.

 

[gar]

131002-1344 EDT

lakee911:

At the present time residential customers and small business are not charged for bad power factor. Residential customers are charged for kWh used. However, low power factor is a big cost to the electrical supplier, and in turn thru the rate structure a cost to the customer. Also the power company does do some power factor correction along their lines.

The power company can not easily correct for bad power factor from waveform distortion. That is the kind of bad power factor that comes from computers and CFLs. Look at today's sine wave vs 60 years ago.

Bad power factor hurts everyone, but is not directly seen in your bill. Are we worse off using CFLs with bad power factor compared to incandescents? No, because the use of CFLs has greatly reduced total load on the grid and power generation.

.

 

[lakee911]

I was always under the impression that W = PF * V * A. I had assumed (and we always know what that does) that in the case of a CFL, the low PF was simply the reason why it used less energy. But what you're saying is that the wattage is low and the PF is simply increasing the VA?

Let me ask this, if there was a magic widget that I could put on my service that would lower my power factor (opposite of power factor correction capacitors), would my bill go down?

 

[JoeStillman]

I was always under the impression that W = PF * V * A. I had assumed (and we always know what that does) that in the case of a CFL, the low PF was simply the reason why it used less energy. But what you're saying is that the wattage is low and the PF is simply increasing the VA?

Let me ask this, if there was a magic widget that I could put on my service that would lower my power factor (opposite of power factor correction capacitors), would my bill go down?

Your residential bill would go down by a miniscule amount, but only if you corrected the power factor at each load. By making the correction at the load, you reduce the current in the wiring up to that point, and thereby minimize your I²R losses. I don't say eliminate because you still have to carry the unity power factor current all the way to the load.

When power factor is corrcted at the service, the savings in I²R losses accrue to the Utility only. You're not reducing the losses anywhere in your own wiring.

 

[GoldDigger]

Your residential bill would go down by a miniscule amount, but only if you corrected the power factor at each load. By making the correction at the load, you reduce the current in the wiring up to that point, and thereby minimize your I²R losses. I don't say eliminate because you still have to carry the unity power factor current all the way to the load.

When power factor is corrcted at the service, the savings in I²R losses accrue to the Utility only. You're not reducing the losses anywhere in your own wiring.

Just to play devils advocate on a trivial matter, the I²R drop on the utility side can still hurt you by dropping the voltage to your motors and making them run less efficiently while also increasing the I²R drop on your wire because of the increased current for same HP . That miniscule difference will be billed to you in increased power at the meter.

Not a good reason to pay $1000 for central PFCC though.

 

[JoeStillman]

Just to play devils advocate on a trivial matter, the I²R drop on the utility side can still hurt you by dropping the voltage to your motors and making them run less efficiently while also increasing the I²R drop on your wire because of the increased current for same HP . That miniscule difference will be billed to you in increased power at the meter.

Not a good reason to pay $1000 for central PFCC though.

I stand corrected, but only by a hair's breadth (and I already have too few of those!)

 

[lakee911]

Interesting ... I need to think on this a bit more....

 

[GoldDigger]

I stand corrected, but only by a hair's breadth (and I already have too few of those!)

More importantly you, like me, did not read the OP's question carefully enough.
He asked whether lowering the PF seen by POCO would decrease his bill. And the answer to that is a resounding NO. It would actually increase his bill slightly for the reasons we discussed. And if he were a commercial user it would increase it a lot because POCO does not like supplying power to low PF loads.

lakee911: There is nothing good about low power factor. It has only neutral or bad consequences, depending on what part of the whole power distribution you look at and to what level of detail. The combination of low and power in sequence gives many people the wrong impression of what it means. Low Power Factor has nothing to do with Low Power.