How's That Roof Working Out?

Our house needed a new roof when we bought it.  There wasn't anything particularly wrong with the roof, but it was 30 years old and worn out.  When we replaced the roof we went with light-colored shingles, added ridge venting, used radiant-barrier sheathing, and upgraded the attic insulation to a minimum of R38.

Well, now it's been a few years and I have some data (thanks PG&E for making usage history available for download in CSV format) so it's time to see how well all that stuff is working out for us.

Before we start, some caveats about the data.  I only have 1 year of data from before the roof replacement, so it's not very stable data.  For the 3 years after, I've averaged each month to obtain more stable information.  The graphs start in December because the new roof was put on during October.  So December is the first full bill after the roof was completed.

The gas data is very consistent.  We have a gas dryer, gas water heater, and gas furnace.  Our laundry and hot-water usage is probably fairly consistent throughout the years (with a small increase in both when Corinne joined the family), so the change in therms seen from before to after should be pretty focused on heating.  We've heated the house to approximately the same temperature throughout each winter so it should be a very stable comparison.

The electrical usage, however, is much less stable.  Usage patterns have changed as Heather has grown, we've switched TVs, upgraded computers, more hard drives spinning, more often running dishwasher, etc.  So the comparisons for usage from before and after are much more muddied.

Let's go to the data.

Gas

The gas usage shows an obvious reduction in gas usage throughout the winter months with equivalent usage during the summer.  This seems like pretty solid evidence that the upgraded insulation is making a difference.

The total therms used in 2013 was 503.  The average therms per year from 2014-2016 was 412.7.  An 18% reduction.  At an average of $1.29 per therm, this results in annual savings of $117.

Electricity

Sadly, the electrical usage doesn't show an obvious dip during the summer months.  It actually shows a dip for the winter months which I presume must be related to the cost of running the blower on the furnace which didn't need to run as much as evidenced by the reduced gas usage.

Regarding the summer months, the 2013 data is not very good to begin with.  This was our first year in the house and we were adjusting our usage.  The spike in July 2013 would have been our first hot month in the house (July bill, for June usage) and we ran the air conditioner liberally.  When we saw the power bill, we adjusted the air conditioning to reduce costs as seen in significant drop in usage in August.

The uptick in January is most likely due to Christmas lights.

Although there was not a clear drop in usage throughout the summer as I was hoping, there was an overall reduction. The total kilowatt-hours used in 2013 was 4599.  The average usage per year in 2014-2016 was 4344.3.  A reduction of 6%.  At an average of $0.17 per kWh this results in annual savings of $42.  However, due to the nature of the 2013 data the validity of this claim is suspect.

Due to the many confounding variables on the electrical usage (mentioned in the opening paragraphs), I don't think this data can say that the high-reflectivity shingles, ridge venting, and radiant barrier were ineffective upgrades, but clearly they weren't obvious wins either, about which I'm a bit disappointed.

If we assume the data is valid as presented then the net change is $159 per year in savings.  If we were to assume the entire benefit seen is due to the combined effect of the insulation and roof upgrades (and not changes in usage patterns), then the break even point of the upgrades would be ~38 years.  Which is longer than the expected life of the roof (30 years).  However, the insulation and radiant barrier are one-time expenses.  Unless the roof fails catastrophically neither one should need to be replaced when the next roof is installed.

The net effect is that my data doesn't show the upgrades to have been a definitive win compared to a standard roof, however, I believe the electricity usage data is too inconsistent from year to year to be reliable to make any strong claim.  If I had a few more year's worth of data from before the roof replacement I'd be able to make stronger claims about the effect.

Heard from the Girls

Heather, while camping: "Let's have a conversation.  What's your favorite food, drink, and, ummmmm, dinosaur? No wait, just dinosaur."

Heather, while showering after camping:
"Ummm, Mom?  Dad?  Anybody out there except Corinne?!"
[Corinne's interest is piqued and heads for the bathroom.]
"I need a washcloth!"
[Corinne wanders in, grabs a washcloth off the sink and tosses it into the tub.]
"Thanks Corinne."
[Corinne wanders back to the TV Show she was watching.]

Corinne is confident in her identity if not her language skills:
"What's your name?" -- "I Kin!" [She's missing 'r's]
"Are you a monkey?" -- "No! I Kin!"
"Are you tired?" -- "No! I Kin!"
"Are you my little girl?" -- [Throws hands up in the air] "No! I big gul"
"You're not a little baby?" -- "No! I big!"
"You're my big girl?" -- "Uh huh"

These two are goofballs.

Heather on Wheels

We gave Heather a pedal bike on her birthday last year, but she still needed to improve her balance before she'd be able to pedal.  So I took the pedals off so she could focus on balancing (she had outgrown her balance bike).  Last weekend she was ready to level up and add the pedals back.  She really improved her balancing ability and lifting her feet to the pedals went off without a hitch.  She is greatly pleased with this new skill and enjoys riding laps around the neighborhood at every opportunity now.

Download video here: VID_20170513_140642507 (28 MB)

A Trip to the Coast

We took an extended-weekend trip out to the northern California coast last weekend.  We left Wednesday afternoon (April 5) and headed out to the Red Lion Hotel in Eureka.  We got in a little late so it was straight to bed the first night.

On Thursday we headed out to the Elk River Trailhead & Old Town of Falk Walking Tour to take the ~2 mile round-trip path out to what I was led to believe was a ghost town.  But, at least as far as we could see from the trail, there really isn't anything to see left of the old logging town.  You can see some of the remnants of the old caretaker's house from the 1970's, but that's about it.

Though the goal of the hike was a bit of a letdown, I really liked the quiet of walking through the redwood forest with mostly nobody else around.

After our hike we headed to the Loleta Cheese Factory (not to be confused with the Cheesecake Factory restaurant).  They had already completed their cheese production for the week so we didn't get to see any being made, but we had lunch at their grilled-cheese bar and then completed their garden scavenger hunt.

We bought some cheese to bring home and then headed back to the hotel.  For dinner we walked across the street to the China Buffet.  Corinne was happy to have a plate of green beans and honeydew.

On Friday we got up and headed out to the Samoa Cookhouse for breakfast.  They claim to be the last operating cookhouse (from the logging days).  They have a daily-changing, set, all-you-can-eat menu served at communal tables.  We had pancakes, eggs, sausage, orange juice, and biscuits with gravy.

We needed to fill up for our drive up north.  Our first stop was at Trees of Mystery and stereotypical tourist trap, but with the unique draw of gondola rides through the redwood forest.  We were thwarted in this plan due to a power outage and the gondolas weren't running.  So we kept on driving up to Crescent City to see the Battery Point Lighthouse.

The road up to Crescent City from Eureka (US 101) seems to have suffered some damage from California's rain this year.  Several sections were down to one lane due to landslides, with construction going on to rebuild the lost roadways.

Here the stormy weather made for a much more interesting visit.  You can see a wave crashing over the far side of the cliff upon which the lighthouse sits.  It's an operating lighthouse with a 24/7 staff.  Its cliff is an island during all but low-tide and when low-tide aligns with midday they offer tours.

We got there a little early for the tour, so we headed to the Apple Peddler for lunch where we narrowly avoided getting caught out in an impressive hailstorm.  After lunch we headed back to the lighthouse for a tour.

The stormy sea made for some moody pictures I really like:

I would have taken more long exposure pictures of the waves, but I didn't bring my tripod, so I took a few by setting my glove down on some seaweed on a rock and putting the camera on that.  This rather limited my subject options.

On the way back from Crescent City we tried Trees of Mystery again, but the power was still out.  However, I snapped this picture from the parking lot.

Dinner was at Marcelli's Pizzeria.

Saturday morning we got up and had breakfast/lunch at Kristina's Restaurant before heading out to Trinidad State Beach.  Our GPS unit doesn't understand where that is and attempted to kill us.  It wanted me to drive down some steep, dark, dirt/mud, little one-lane road. I vetoed that option and Jess used her phone to find our way to actual beach access.

I call this "Heather versus the ocean"

After the beach we ate at the Lighthouse Grill in Trinidad and then headed back in to Eureka to get dessert at Vampire Penguin (shaved snow sundaes).

On Sunday we packed up and headed on the road back towards home.  We drove through Avenue of the Giants where I snapped this picture:

Then we stopped off at Chandelier Tree to see if I could squeeze the van through a tree.  We managed to squeak through but not much room to spare.

Heather and Corinne desperately needed to get out of the van and run around for a bit, so we stayed at the tree for a little while.  Heather explored the meadow and Corinne played with the gravel.

Then it was back in the van until we managed to find a Taco Bell for a quick dinner before pushing on home.

Facebook's Image Recognition/Tagging System

I've noticed recently that Facebook has been applying an automatic image recognition/tagging process to any pictures that are uploaded to their system.  The general idea is to get a computer to "understand" what's in a picture.

In Facebook's case, images are given a textual description automatically which is used to provide meta-data that can be accessed by screen-readers (used by the visually impaired when using computers).  This isn't about facial recognition (though Facebook is doing that as well).

For example, I came across this picture from a friend of mine:

Which Facebook's artificial intelligence system automatically tagged with this information:

Image may contain: sky, twilight, outdoor and nature

Automatic image tagging/recognition has been a popular field of research, but this is the first time I've seen it actually used in a production system for something useful.

It's pretty good too.  Here's one of interest to the Dickerson family at large:

Which Facebook automatically describes as:

Image may contain: 8 people, people smiling, people standing and indoor

I just found that interesting.  That's all.