Intensity vs power

[iresq]

I know this comes up time to time. I did this a couple of years ago but have lost the data. Needing a break from programming, I redid the test here are the results:

I used 4 strings of 100 count mini's. One of the strings is half dead (now were is my light zapper?), so we are talking about 350 minis. Power measured by kill-o-watt.

100% - .98 amps
90% - .96 amps
80% - .94 amps
70% - .91 amps
60% - .88 amps
50% - .83 amps
40% - .77 amps
30% - .69 amps
20% - .59 amps
10% - .48 amps

So with these lights, you need an almost 90% reduction in intensity to half your power draw.

YMMV

[Steven]

As someone pointed out in another thread, this may not be the whole truth. The Kill-O-Watt may not measure current correctly if the current is not a sine wave, which it isn't when a channel is dimmed.

Looking at your chart, it seems possible that the Kill-O-Watt may be measuring peak current, rather then RMS current, which is a more accurate way to measure power use. The peak current will not change (much) until you get lower than 50%, although the actual power use will drop faster. It may also be measuring average current, which is also not useful to calculate actual power use.

However, with incandescent lights, the power used at 50% will still be more than 50%. This is because when the lamp burns cooler, it's resistance goes down, which increases the current.

The best way to determine power use is to actually look at your electric meter (with everything else turned off). This may not be practical.

[JonB256]

I must agree that the Kill-a-watt is being fooled. I have looked at the output waveforms with my USB O-Scope while dropping the power. The "area under the curve" for the output seems mathematically accurate as power level drops.

Then again, that is for incandescents. LEDs do respond differently.

http://www.granburychristmaslights.com/OScope.html

[iresq]

Gee, thanks guys! I knew I should have just kept sequencing, LOL.

As this issue pops up several times a year it would be nice to get the definitive answer. It always felt like the results were not accurate. As my knowledge of electricity usually results in a few bee stings, I will leave this for the experts.

[-klb-]

Jeff Millard wrote:

If I use a true RMS voltmeter along with a calibrated AC ammeter connected in line with the source to the controller, I should be able to get values that will tell what the load actually is at different percentages.

...

The most accurate is to place the meter in series with the load.

This should be a really good combination, that we are unlikely to beat the results from. A RMS amp meter in series, and RMS voltage across the string of lights should tell us exactly what we want to know, within the limits of accuracy of the meter.

I would expect that for incandescent lamps we should be able to tell that the resistance in the bulb is sort of close to constant, at least through the upper range of intensity. I would expect to see that it drops some towards the lower end of the intensity range.

I look forward to seeing what the results are.

For the purposes of maximizing the load that can be applied to the controller and circuits, the RMS current will be what counts. For energy savings, it should be the product of the RMS volts and RMS amps.



I haven't run this test myself, as the best RMS amp readings I could get would be with a clamp meter, and I'm not sure how accurately the clamp handles the phase angle step.

[Cray Augsburg]

Cool new avatar, Jeff!

Cray

[ainsworth]

I will be very interested in the measured result. I'm not an electrician so I can only go by what i have read.

When using the "set intensity" control; is the percentage based on power to the light or the lumens that result? I have always assumed it was power. However that would lead me to the power usage would be a linear scale; which it seems is not true.

Correct me if I'm wrong; a 50% reduction in power results in a 70% reduction of brightness. This is a result of a cooler filament is less efficient at producing light.

This in no way effects the usage of my LOR, just a very interesting discussion!

[cenote]

Jeff Millard wrote:

I was in the field so I powered from the service outlet of a pole mounted recloser breaker. The sensing pots are isolated to the control only. No other load, so there wasn't anything on the transformer but this test set. The breaker and control were disconnected when I was doing this, so I had my own power source right off a 13KV primary for this test!

I would love to of seen the people's faces as they drove by and saw "some guy" in a van with a set of jumper cables from a 13kv line down to a laptop.....Hooked to Christmas Lights

P.s. Love the Avatar

[iresq]

Those numbers are incredible. I am sure there is a good explanation that I won't understand.

[ainsworth]

Could you even see any light at 2%?