Energy Audit Part 2: The computer room (Desktop Computer)

So after yesterdays post concerning the fact I paid a lot on my electricity I thought I would start my energy audit right in front of me, in my office.


Audit Items:
So my office is one of those tech related ones. I have the typical gadgets:

* Computer
* Printer / Scanner
* 2 monitors
* Desktop Speakers w sub woofer
* DSL modem
* wireless mouse / keyboard with a backup wireless mouse

I also have the usual things in a spare bedroom:
* Ceiling fan
* 3 Ceiling lightd (CFL’s)
* 1 desk light (CFL)

That’s about it.

Questions:
So I have set out to answers these questions:
* How much power does my office consume
* How much does it cost to run my office
* What are the potential gains on reducing energy in my office

Equipment:
 The equipment I will be using for my energy audit investigation includes
* Computer (ironic?)
* Spreadsheets (google docs / Open Office)
* Web related research
* EM100 Energy meter (Canadian Tire)
       * There are 171 Kill-a-Watt meters available through the Ottawa Public Library (press release)

Abreviations:

V- Volts
A- Amps
mA = miliamps
W- Watts
kWh- Killowatt Hours
h- hours
* – multiply
/ – divide

Equations:

W=V x A
A = 1000 mA
kWh = W * h /1000


Desktop Computer:
So today’s investigation is concerning my computer. Because I built this system from a bare bones kit from Tiger Direct there is no model number. I would have to itemize each component. That is out of scope for this post.

After 17.75 hours of moniotring I found the following info from the meter:
Voltage (V): 123 (snapshot in time)
Current (A): 1.6 (snapshot)
Max Current (A): 1.95 (over time)
Power (W): 6 (snapshot)
Max Wattage (W): 204
Time (hrs): 17.75 (@ 0.065/kWh)
Cost ($): 0.023

So we learn a few things from these numbers. First computers range in current and thus power consumption over time. Therefore any snapshot generalized over time will be less than accurate.

I ran the meter for an hour yesterday and then let it run over night, and probably for the rest of today. The longer sample time you have the more accurate your forecasting will be. This exemplifies the benefit of having an energy meter like the EM100 or Kill-A-Watt meter.

Second, we learn that snapshots also fail us perhaps with the math.

1.6A x 123V = 196.8 W

The meter says it is only consuming 6 W of power, but the math suggests it should be closer to 196.8 W. I have to better understand this meter and how to use it. The next lesson here besides learning the gear is to double check your assumptions.

The next thing to consider is that we purchase power by the kWh, not by the instance of consumption. So we take:

196.8W * 1 h /1000 = 0.1968 kWh

0.1968 kWh * $0.065/kWh = $0.012792

$0.012792 * 24 h = $0.307008 (31 cents/day)

However, if we assume the meter is accurate with the cost estimate then lets use this.

17.75 hrs = $0.023
1 h = 0.023/17.75
1 h = 0.001295775
1 day =  0.001295775 * 24 h
1 day= $0.031

so my math using a snapshot compared to the meter over time is off by a factor of 10. otherwise, my math failed somewhere (quite possible). So it either costs my 3.1 cents or 30 cents a day to run my computer.

I will assume that someone smarter than I and better at math than I designed the EM100 and will conclude it costs me 3.1 cents a day to run my computer.

Annually = $11.35
Monthly = $0.95

As this is an insignificant percentage of my monthly bill I will assume this is not a big issue.


Monitors
I run dual monitor setup. For those of you not familiar with this set up, essentially I have two monitors where the desktop spans two screens. If you spend a lot of time with computer related projects I would recomend a dual screen layout.

For anyone spending much time in the tech industry doing application development, data processing, Computer Aided Design, Graphic Arts, or spend any amount of time at the Command Line Interface.

I took the EM100 and did a couple of things. I have both moniotrs hooked into a powerbar for surge protection. This is a different power bar than my computer and other peripherals. Some suggest this delivers cleaner power to the monitors and reduces flickering.

First I attached each monitor individually and came up with similar readings. They are Both he same HD Monitor make and model by eMachines. These readings came back:
V: 123
A: 0.25
W: 30.75
kWh:0.03075

Assuming that monitors do not fluctuate much (a little surge on power up) we can use these numbers to estimate an hourly usage:

0.03075kWh * $0.065/kWh = $0.00199 (0.2 cents)

Lets assume my monitors are on an average of 4 hours a day:

4 h * 0.00199 = $0.008
Annually = $2.92
Monthly = $0.24

Now multiply that by two for a dual monitor setup:

4 h * 0.00199 *2 = $0.016
Annually = $5.84
Monthly = $0.48

To compare with a computer running 24 hrs a day:

24 h * 0.00199 *2 = $0.09552
annually = 

Therefore 1.7 cents difference between running

Conclusions:
The purpose of this blog was to start investigating my energy usage. I started with something I use all the time and an appliance that runs all the time (computer). This was my first intro into doing an energy audit. I had to review some basic high school physics

Running a computer 24hrs a day with monitors tunred off is not a significant consumption of power. The cost of running a computer as a server at home is cheaper than using external hosting services. The fact that servers are running in a virtual environment on one physical box may also contribute to savings.

Running monitors when not in use is a significant consumption of power. It is easy enough to turn the monitors off or flip a switch on a power bar before you leave your computer.

Unfortunately I already turn off my monitors before I leave the computer and the computer needs to remain on for it runs servers that are required by others. So I do not see any opportunities in changed behavior in order to reduce consumption or costs.

The peripherals such as scanner/printer, wireless mouse and keyboard, desktop speakers and DSL modem will be investigated later. We may or may not be surprised.

The rest of the room such as desktop lights, ceiling lights, and ceiling fan may not represent much savings. The lights are turned on when required and are all Compact Fluorescent Lights (CFL’s). The ceiling fan will be hard to figure out and might need a volt meter on the switch. In any case ceiling fans are recommended to help decrease heating and cooling costs.

In conclusion we discovered that a computer and 2 LCD monitors is not all that much to operate:
Monthly: $0.48 + $0.95 = $1.43
Annually: $17.16

On to the next location in the house. I am wondering about our fridge.