Attic Efficiency: Check!

It was February 2009, and I had just finished the satisfying job of adding insulation.
After a quick look around the attic, I finally felt a sense of completion. I had finally covered the 3 major areas that my internet-based education had told me were the major causes of inefficient attics: lack of ventilation, lack of a barrier to radiated heat, and insufficient traditional insulation. It took several seasons of projects, a little bit of money, and the help of some neighbors, but I had finally managed to get my attic into some semblance of energy efficiency.
I am now tempted to attempt to perform a better analysis of how well this has worked for me. Since the only numbers I have used so far have been my electricity usage, I have been unable to account for some known large differences in the weather over the course of my improvements. Without this accounting, the numbers look quite good, as you can see from my previous posts on the subject.
But I have been doing some research into how one accounts for the known temperature differences from one year to the next in ascertaining the performance of one's home. For example, was my energy usage drop from 2006 to 2007 entirely because of the well-known fact that in my area, 2007 was a cloudy and cool summer? Is the fact that my 2008 usage is less than my 2007 usage even more impressive since 2008 was so much hotter?
It turns out that there are some ways to account for this. The concept of "degree-days" is used in conjunction with either heating or cooling. The base for "degree days" seems to be 65 degrees Fahrenheit. The concept is deceptively simple: for cooling, you look at how much hotter the daytime average was than the base of 65 degrees, and you add that number to your total. You continue adding (T - 65) for each of the days you are interested in, until you get a total the represents how much hotter it was outside than inside over the span of time you're studying.
There are complexities about how to best measure the degrees on a given day (the average is not all that accurate; you'd ideally want a measurement every hour or more), and inaccuracies in that heat flow is not always linear (i.e. a 20 degree difference from outside to inside might well be expected to use more than 2x the energy of a 10 degree difference) and there is a baseload entirely unrelated to heating or cooling. Nevertheless, using this method should be an improvement over using no method at all.
As it turns out, there are sources of data out there for Heating Degree Days and Cooling Degree Days for a lot of places out there. There can be holes in the data, which require filling from other sources that may not agree exactly, etc. In addition, there are many schools of thought on how best to utilize the data. When looking at energy usage, one it looking at a sum of many variables, so when looking at energy usage vs. Cooling Degree Days, one will still have other factors unrelated to cooling influence the outcome.
Nonetheless, performing some of this analysis might be instructive. Gird up your brains, faithful readers, for the upcoming analysis is not for the faint of math.

Insulation School

As luck would have it, the day in late February 2009 that I chose to blow in the additional fiberglass insulation turned out to be the hottest day in months. It was 82 degrees and sunny, yielding an attic that, while cooler than it would have been with no radiant barrier, was definite cause for sweating.
Using the insulation blower was fairly simple, but certainly required two people for any reasonable amount of insulation. My volunteer neighbor stayed on the ground in the garage, wearing an insulation mask and feeding bags of compressed insulation into the blower. The blower, which had a set of rotating paddles inside it, tore up the compressed insulation into small enough bits that they would be blown out through the long hose. I stood in the attic with the other end of the hose, blowing a fine snow of expanded fiberglass insulation a good 15 or 20 feet out in front of me.
We quickly learned a few things:

1. It takes a _long_ time to feed in a bag of insulation and get it fully blown out into the attic. We had 12 bags, and it took us roughly 8 hours to get the job done. There was some break time to cool down, clear out the machine, etc. but the job was very slow. Perhaps some machines are faster, but ours was pretty slow.
   
2. You want to plan your route as if you were a painter; that is, work from the farthest edges back towards your attic entrance, so you aren't "painted into a corner". This helps keep the hose out of the way as well as keeps you from having to traverse over, and compress (which reduces the R-value) the newly blown insulation.
3. You might also want to plan the various positions in the attic where you are going to stand. As you might expect, the insulation shoots out in a nice arc through the air slowing as it gets further away, with the insulation landing in a nice linear pile under the arc, with most of it landing on the half further away from you. Therefore, you want to fill places that are _not_ right next to you, but are further away. If you point the hose at the floor less than about 10 feet away, you'll just be blowing insulation out of the way with the force of the air, reducing the insulation coverage instead of adding to it. Instead, you want to shoot the air _over_ the area you want to fill, and let the insulation drift down into the area. If you don't understand this, don't worry - you'll learn it pretty quickly as you work.
4. Since communication is nearly impossible between the hose operator and the machine operator due to distance and machine noise, you'll want to coordinate so that you know when to move to your different blowing positions to get an even distribution. I would ask my neighbor to cut off the machine after 2 more bags, for example, so I would know that I had blown enough insulation in that particular area.
5. Operating the hose is actually peaceful, almost trance-inducing work. There is a certain beauty (at least to those schooled in the ways of Energy Efficiency) to watching the myriad fiberglass snowflakes dancing down, slowly gathering into giant drifts of heat-flow reducing goodness.
   

After one afternoon and a brief morning of work, we were done. Another hour or so to clean up all the insulation detritus that drifter around in the garage, and the rental machine was returned. Project complete in less than 1 day!

Completion of the barrier(!) and adventures in insulation

So as we come to this point in the winter of 2008-2009, I have realized that my attic has a newly found problem.
Problem: Insufficient depth of insulation for current building codes
Solution: Simply add insulation!

At first, I began by buying a couple of rolls of fiberglass insulation. However, after the very time-consuming process of removing the blown-in insulation from an area, measuring the space, cutting the insulation roll, placing the cut piece, then covering that with the old blown-in insulation, I realized I might not complete both the insulation job and the radiant barrier job during the nice, attic-work-friendly cold weather. The installation of the insulation, if you will, was taking too long.

New Problem: Installation of insulation pieces taking too long
New Solution: Rent an insulation blower and blow in more insulation. However, due to the sheer volume of insulation that I planned to add, this needed to wait until I was pretty well done with everything else in the attic.

I completed the installation of the barrier at long last in January 2009. There was much rejoicing in Energy Efficiency Land after that relatively inexpensive but long-lasting project. The total cost of the amount of barrier foil that I ended up using was under $200.00, but the amount of labor was a goodly number of (uncounted) hours over the course of a couple of winters.
After rewarding myself with a few weeks of break from breathing the attic air, in late February I decided to go ahead and try to blow in the new insulation.
The process seemed simple enough:

1. Estimate the total volume of insulation I would need. I could do this by multiplying the attic floor area by the depth of insulation that I wanted to add. This would only be an estimate, but blowing in insulation is far from an exact science. The bags of insulation at the store should tell me their total volume.
   
2. Rent a blower at the hardware store where I would buy the insulation
3. Arrange the borrowing of a neighbor's pickup truck to carry the blower and the insulation
4. Arrange the time of another helpful neighbor to feed the insulation into the blower in the garage, while I waved the hose around the attic to distribute it appropriately.

But how well would this go? How long would it take? What new things would I learn?
As usual, read on to find out!

The important Science of Insulation

As most readers of this blog probably already know, traditional insulation is only a part of reducing heat flow into and out of a house's living area, and should be looked in concert with other means of blocking heat flow such as radiant barriers, and cooling mechanisms like attic ventilation.
Nevertheless, insulation is an important warrior in the heat flow reduction battle whose role should not be taken lightly. In fact, good insulation greatly reduces heat flow from the mechanisms of both conduction and convection, and perhaps even a small extent by radiation as well.
The heat flow reduction properties of insulation are quantified by its "R-value", with higher R-values being more able to reduce heat flow. There's a good layman's description (with full equations and a nifty table of materials) here on wikipedia.
The recommended total R-value for my home's attic when it was built in 1996 was R-30. In terms of my fiberglass blown-in insulation which has an R-value of about 2.5 per inch of thickness, that would mean I should have had 30 / 2.5 = 12 inches of insulation. That actually would have covered up most of the wooden joists on the floor of the attic, but in fact, it did not. The depth of insulation varied, of course, but my best guess as to the average depth was about 9 inches, yielding an actual R-value of 9 x 2.5 = R-22.
So not only was my insulation insufficient for the building codes in 1996, it was now 2009, and we as a society have realized (to some small extent) that we're underinsulating our buildings. The new code seems to be R-49 for my area, which would equate to 49 / 2.5 = about 20 inches of insulation.
So I needed at least 20 inches of insulation, but I had less than half that. What would I do about this discrepancy? And would it keep me from finishing my radiant barrier for yet another summer?
Next: Find out!

Winter of '08- '09: Mission creep part 2

After experiencing the dramatic cooling capability of even a partial radiant barrier installation, I raised the priority of finishing the barrier over the coming winter. As the weather started to cool, and the fierce Texas sun began to travel south for the winter, leading the Canadian geese and Mexican free-tail bats to warmer climes, I began to spend longer sessions in the attic in an attempt to finish the work.
As I continued to work in the attic installing radiant barrier, one thing that I noticed was that looking around the attic, I could see a lot of wood sticking up out of the insulation, wood that made up the joists for the ceilings below me. This can be a bad sign; typically, as I understand it, one should have enough insulation depth that the wood is pretty well entirely covered. This reduces heat flow from the attic air into the wood, which is important because the wood is in very direct thermal contact with your ceiling. Plus, unless your ceilings joists are _really_ thick, the fact that the wood is showing through means that your insulation is not really very thick.
Now, perhaps some of you are thinking, "what does this have to do with installing radiant barrier?" And you would be right to ask that question. In fact, I had distracted myself once again from the primary mission through another bout of mission creep, this time, to improve the insulation while I was in each particular area of the attic.
Next: read on, to determine what this little diversion into the fascinating science of insulation has to teach us, and learn what efficiencies are in store!

Energy saved by 40% of a Barrier?

Before we look at the numbers, I should note that it may not be entirely useful to look at the 2008 energy use vs. the 2007 energy use as a measure of radiant barrier effectiveness, simply because I had only installed the barrier over (a bit less than) half the house's total area.
It may well be that installing 40% of a radiant barrier (which is my best guess for the percentage I had put in) does not give you 40% of the performance of a full radiant barrier. Why? Well, for one thing, the radiant barrier concept relies on rejecting heat before it enters the serious thermal mass of your insulation. Having a huge gap in the barrier allowed my insulation to heat up greatly during the day. Why would that be a problem? Because at night, when everything continues to radiate heat in all directions, a lot of the heat radiated by the insulation is going to bounce off the radiant barrier above it and back down into the insulation, rather than eventually working its way out of the house. Thus, a 40% radiant barrier coverage may not give you 40% of the full radiant barrier benefit.
But I have the numbers, so we might as well look at them.

Partial radiant barrier comparison:
-----------------------------------------------
Total electricity usage for 2007: 8056 kWh
Total electricity usage for 2008: 6982 kWh
Energy savings: about 13%

Unfortunately for our accuracy in using these numbers, this was not a controlled experiment, but instead was affected by at least a couple of major changes in 2008. First, the careful reader will recall that the summer of 2007 in my area was the coolest and cloudiest in quite some time, probably depressing the usage numbers for that year. 2008, on the other hand, had a much more "normal" summer with a lot of sunshine and many 100+ degree days.
Second, in late July my old relatively inefficient (11 SEER) air conditioner finally kicked the bucket. Fortunately, I replaced it with a more efficient 14 SEER unit. Unfortunately, the old unit died in a way that made it run more and more frequently over a period of days, working harder and burning more energy to cool less and less.

So considering the difference in weather, the electricity savings of my partial barrier probably saved a good bit more than 13%.
But, considering the fact that the air conditioner got upgraded about halfway through the summer, the electricity savings must be at least partly due to the more efficient air conditioner, reducing the apparent savings from the barrier. Mitigating the air conditioner factor somewhat was the fact that the slow failure of the old unit burned a lot more electricity over a period of days than would normally happen.
It is impossible me to say for certain which of these factors was larger. Suffice it to say that I was impressed enough with the barrier's performance to stiffen my own resolve to spend more "quality time" in the attic over the winter to complete the job.
I will include a look at my total electricity usage for 2005-2008, with 2008 on the left, in case you can't tell :)

2008: The summer of comparison

Our narrative now winds into the summer of 2008. I had just completed radiant barrier coverage in the attic of a little less than half the house, including a solid section over the east-facing and mostly unshaded master bedroom suite, giving me a very good opportunity to determine whether this formerly hottest area of the house would make a noticeable difference as the fierce Texas summer came on. The other barrier-covered section was west-facing in the attic over the garage.
The first thing that I noticed, as the sun began to warm the house in the summer mornings, was that the air conditioner did not start as early in the morning. In fact, it hardly ran at all before 10 am. This unusual development seemed to indicate that the east-facing "artificial shade" of the radiant barrier was working.
Next, from about 10 am to about 3 pm, the house heated up rapidly. The air conditioner would start to run more and more frequently, as expected, as the barrier located at the eastern and western edges of the house, didn't do much good for blocking the almost directly overhead sun angles. In the late afternoon and early evening I could not tell much difference as the roof and attic were fully heated up, and the air condition ran a lot, although certainly not constantly. The west-facing section may not have helped me as noticeably, first because it was over the garage, and second because I do have partial shade already on the west side of the house.
But how was the feel of the master bedroom, which for more than a decade has been the room that we would avoid in the summer until the last possible moment, due to the heat buildup every single hot day?
The difference was nothing short of dramatic. The hottest area of the house immediately became the coolest area due to the radiant barrier. In fact, for the first few weeks, without ever planning it, we found ourselves retreating to the master bedroom after dinner to read or play on the computer because it was the most comfortable place to be.
Even more amazing was the fact that as the summer wore on and the days got even hotter, we had to RAISE the thermostat setting a degree because the master bedroom was getting too cold. Since the thermostat itself was in an un-barrier-covered part of the house, it was exposed to the full heat load and ran the A/C accordingly. However, the parts of the house not exposed to the full heat load due to the apparently incredible heat-rejection properties of the barrier still received their full measure of cold air from the A/C, dropping the temperature dramatically. I recall at least a 6 or 7 degree difference from the central (uncovered) area to the master bedroom.
These minor miracles were made even more miraculous by the fact that the master bedroom area has two of its four sides as vaulted ceilings, which were inaccessible for installation of the radiant barrier. In other words, this remarkable turnaround from hottest area to coldest area happened with only about 60% radiant barrier coverage of the room. (The only way to get radiant barrier over the sloped part of the vaulted ceilings would be to replace the roof decking with decking that has radiant barrier along the bottom of it, and the only time that it makes any sense at all to do that is when you're replacing a roof (which I in fact did in 2009: foreshadowing!))
In short, the half-done barrier worked incredibly well. The comfort level was immediately noticeable, and resulted in our actually setting the thermostat a degree warmer for that summer. The uneven heating of the house was quite noticeable and caused me a good degree of regret at my late start that winter, and renewed my resolve to complete the barrier over the entire house once the summer heat was gone and the attic was once again safe to work in.