The Kill-A-Watt experiment

So, as part of the requirements for the Tweet-A-Watt, I also purchased a P3 Kill-A-Watt (model#: 4400).  So, before I open the thing up, I was curious what kinds of numbers it produces.  So I started a few tests.  I have completed the first test so far, and I have published the spread sheet that I am tracking the data in.  So far, I am monitoring Wattage, Amperage, and I will be doing some quality of service(QOS) tests too... just for kicks.

And so far, I think the results have been interesting.  So, most of this data should be self explanitory.  I hope.  I have completed one test, so the results should be fairly accurate for test 1, however, test 2 is still in progress, and because I haven't created a way for the Kill-A-Watt to transmit the data to my computer, the kW section has to be entered in manually.  One more piece of data that is not clearly present in this chart is the cost per kilowatt.  On a separate sheet (which I did not publish) I recored my power usage in kilowatt hours for the past six months, and compared that data to the dollar amount of my bill (data provided by puget sound entergy.)  Then I took the average from the past six months, and came to the amount of $0.105695 as the average price per kilowatt hour.  Then I used that number to get the price per watt hour, and then multiply that number by the watt hours that I use.  As an extra and not yet useful piece of information, I have the % of my hourly average.  This is also based on my Puget Sound Energy bill, except I divided the months into day (because each bill tells me how many days it covers) and then days into hours, and take the average of all six months.  I imagine that this data will become more accurate over time, as I enter more electric bills into the table.



In the above table, you should see a section for QOS, however, until I create a database to log that data in regular intervals, the data provided is useless.  It merely shows a snapshot that represents a small fraction of time.As far as QOS data is concerned, I am monitoring Volt, Frequency, Watt meter, Volt-Amps (VA), Power Factor (PF).  I still don't know what the difference between the Watt meter and kW functions are;  I assume is is probably the case that the watt meter uses a consistent amount of time, and the kW funciton meters kW hours over the total time that the unit was plugged into the wall. 

Volt-Amps are constantly changing with the load put on the system.  For most purposes they are the same thing as watts, however for determining power quality, they can be used to determine is you home handles low power loads better than high power loads.

Power Factor (PF) is usually expressed as a percent because it is a number between 0 and 1.  It is the ratio of real power to apparent power, and is a fairly good indicator of the efficiency of your power grid.  Normally a residential PF is higher, and commercial is lower.  To compensate for a low PF, the utility companies will need to install additional equipment such as power regulators and capacitors. 

So, now you know!

Anyway, as far as my power quality.. I know that I have seen the Voltage from the outlet my entertainment center is plugged into as high as 122 Volts and as low as 116 Volts.  The frequency is usually pretty close to 59.9Hz, but I did see it dip to about 57Hz.  My power factor is 88%, which makes me think my connection to the grid is fairly close to a power sub-station (like the one over by the Boeing plant), or at least is some what efficient. My Volt-Amperes seems to be fairly consistant around 120VA.

I know it really doesn't mean a lot now, however, I will be able to collect this data over time, and hopefully soon I will be able to create a picture of what normal power consumption is supposed to look like.

Anyway, more to come soon.  I'm still troubleshooting the whole XBee software issue  I guess I need to compile the driver as a kernel module, or recompile the kernel with the driver in it.  So... I'll post some instructions for anyone who is interested in the project.

So, that's all for now.

Later,
     SteveO