Pages

Saturday, May 2, 2026

Virtual Roadtool Museum

 

Welcome to my Roadtool Museum. 

The collection includes almost all the tools I have found, primarily while out training to race bicycles. 

As of 2026-05-02 the roadtool collection comprises 1236 tools recovered and cataloged since 2013‑12‑31. Some of the tools date to earlier times, before I began inventorying them. 

In addition over the same period I have found $171.84 and fifteen mobile phones, a few of which I have been able to return. 

The inventory of the museum is in a spreadsheet that categorizes metadata such as the find dates, the size, the find location, and a condition assessment. But they don't have accession numbers or any other proper museum-quality information. 

The tools in the collection naturally fall into four grades: 

  1. tools that live in my workroom tool cabinet, 
  2. a second set of tools that I keep in a waterproof tool box in my car trunk,
  3. useful tools that do not have a place in the cabinet, and which live in a tool bag, and
  4. duplicated, poor quality,  damaged, or otherwise unusable tools, which I have retained for their value to the collection. 

The most common size, by far, is 10 mm.

In addition, but not shown, I have a collection of individual drivers (Phillips, Torx, and nut drivers), which I not longer count toward the total number found. 

Class 1: Usable roadtools from the tool cabinet

Sockets

The collection of usable sockets that made it into the tool cabinet includes nearly complete sets of 

  • 1/4 in. drive metric short and deep sockets,
  • 1/4 in. drive US Customary in short sockets,
  • 3/8 in. drive metric short and deep sockets,
  • 3/8 in. drive metric impact sockets,
  • 3/8 in. drive US Customary short sockets,
  • five individual 1/4 in. drive socket handles, and
  • a wide range of 3/8 in. and 1/4 in. drive extensions.
The 1/2 in. drive metric impact set is only partially complete.

Unusual elements of the socket collection include 

  • 1/2 in. drive ITL electrical-rated socket handle (orange in the image below),
  • three large shallow sockets for removing various automotive oil filters, 
  • two torque-limiting impact extensions for automotive lug nuts (blue and white in image 2), and
  • a 3/4 in. drive 1-5/8 in. socket. 
Some of the set are Snap-on, but most are the more common Pittsburgh (Harbor Freight), Husky (Home Depot), and Craftsman. 

Metric sockets of sufficient quality to store in the tool cabinet

US Customary sockets from the tool chest.

Wrenches

More an more of the wrench set are the ratcheting closed-end models. 

Wrenches from the tool chest. US Customary on the left; metric on the right.

Pliers

Slip joint pliers are a very common find, and are generally in good condition. The pliers on the right, as well as the pry bar, are Snap-On. 

Pliers from the toolchest

Knives

Most of the knives are the typical Walmart grade, but the collection includes several reasonably expensive models, whose loss certainly caused previous owner distress:

  • CRKT skinning knife (black handle), and 
  • Kershaw (silver with belt clip in the center), which I use every day.



Scissors, Punches, Screwdrivers

How do people lose so many scissors? Screwdrivers seem to have difficulty surviving the roadside--only a few were worth including in the tool chest. 



Drill bits

I carried the Irwin spade-bit set for miles before makingit back to the car. Drill bits usually are too damaged to be of any use. 



Electrical

I rebuilt the Klein wire stripper, and have used it for many home electrical projects. 




Class 2: Car tool box

As the collection grew larger, it became possible to start a second collection that lives in a waterproof box in the trunk of my car. Generally the car set includes function, but more scarred elements of the collection. Periodically I revew the spare set to see if any are in condition to warrant replacing elements of the car tool box. 

Sockets

It includes mostly complete sets of 1/4 in. and 3/8 in. drive metric and US customary sockets. 



Wrenches

The 3/8 in. drive socket handle is the only one I have ever found in usable condition. The tool roll came from the trash can at work. 



Screwdrivers, etc.




Class 3: Tool bag

The elements of the tool bag I keep in the basement include hammers and vise grips, as well as functional tools, but which are often scarred or otherwise cosmetically damaged.

Hammers, screwdrivers, automotive trim tools, and vise grips

Of note are the three different automotive trim removers on the left. The cylindrical object on the right next to the vise-grip clamp is a grounding-rod driver for a hammer drill that I carried for almost 60 miles. The engineer's hammer (blue head) also rode in my jersey pocket for several hours. 



Class 4: Duplicated, poor quality,  damaged, or otherwise unusable

The finite number of sizes and the salting of the roads in winter leads to a large collection of tools that are too damaged to use. 

Sockets

Many of the sockets are either damaged or corroded, particularly the impact sockets on the right. In addition, many are "Brand-X" sockets of low initial quality. 



Wrenches

I no longer count the flat, stamped wrenches (on the left) that come with Ikea kits. Most of the wrenches are also Brand-X, though the adjustable wrenches are all servicable. 


Pliers

The collection of slip-joint pliers represents only a part of the total number found, as I have given away several as well. 

I have no idea what the Frankenstein, home-made lineman's pliers with the sockets welded to the handles could possibly be used for. 



Automotive tools

Apparently many people lose the key for their locking lug nuts. The black-handled tools are tire-plug insertion tools. And the fuchsia tool is for some part of a Cadillac. 



Scissors

Scissors often don't survive the ejection, and defintely rarely survive being run over.



Screwdrivers

So many are in poor condition or were poor quality to begin with. 



Razor knifes and toenail clippers

Apparently many people multitask while stopped at traffic lights by trimming their fingernails or toenails. Possibly they lose the clipper while trying to shake out the nails. I have several other undamaged ones in regular use, after ultrasonic cleaning and disinfection. 

Razor knives rarely survive.




Hammers

And chain-saw tools. 



Thursday, February 5, 2026

2025 Year-end analysis of racing in the Midatlantic

Almost every year since 2001, I have tabulated and published an analysis of demographics and race attendance in the mid-Atlantic. The first edition was six pages long, and had seven tables and two graphs.  The 2025 edition spans 76 pages, 20 tables, and 45 graphs. 

Download the complete version, with many more high-resolution graphics,  here: https://drive.google.com/file/d/1HrT9VbiZpCvaRzCs2GSJNiBMWMy4bHlc/view?usp=drive_link

This post, a kind of executive summary, highlights some findings and plots that interested me. 

Methodology

Every year I have managed to acquire several snapshots of the USACycling rider database, though this task has become much more difficult in recent years.  I also assemble all the results for races in the MidAtlantic (DC, DE, MD, and VA.) from the USAC website (also no longer possible since October 2025) or road-results/cross-results.com.  A giant R script processes the data, creates the figures and tables. A  LaTeX file of  the final report ingests them all on every run. 

Observations

I decided not to add too many of my own speculations on the reasons for the trends. But, with hte exception of the trends for Juniors, the current situation does not look healthy. 

Club membership trends

• The number of Licensed racers in "New" clubs (formed after 2008 but Pre-Covid) has generally decreased.

• The number of licensed racers in clubs that formed Post-Covid has generally increased.

• The number of licensed racers in large clubs is generally decreasing.



The size definitions based on membership, M, of licensed racers identified in the database are 
Large                      M >40
Medium        15 <= M <=40
Small              5 <= M <15

USACycling membership 

The number of licensed racers in USACycling continued its steady decline that began in 2011



MABRA licensed racers 

The number of licensed racers in MABRA (the area governed by the MidAtlantic Bicycle Racing Association, roughly DC, DE, MD, and Northern Virginia,  has continued to hold constant at 2000 post-Covid-19, but it has not reached the peak from 2019.


USACycling demographics

The distribution of the ages of USACycling licensed racers has three peaks: 16,  40, and~55.  The peak at  55~years has been increasing about 11 months per year since at least~2005, when I first had access to the data. 



Juniors

• The most common age demographic in USACycling is Junior racers, see plot above. 

• The number of licensed junior racers in MABRA has remained constant since 2016.




Unattached racers 

More than two thirds of the total licensed racers in USACycling are unattached to any club.  And even the fraction of established racers (Cat 3 and above) has been increasing for the last decade. 




Competitors 

In 2025, more than 40 % of all licensed racers in the MABRA area did not compete in any event. That fraction had been 25 % a decade ago, but has remained constant since 2023.


This analysis depends on the submitted race results capturing every rider. That assumption seems to be pretty good, based on a comparison of the number of pre-registratered riders and the total number of classified (finisher or not) riders. 



Friday, September 5, 2025

2025 Colorado Bike Tour

Background

I'm trying much harder these days to say yes to opportunities, so when Mark asked if I wanted to do this trip, I did say yes.  The idea was to ride from Glenwood Springs back to Fort Collins in 3.5 days, mostly along the sidepath on I-70. The trip would be all motel-camping. 

Summary and Outlook

Point-to-point touring

This trip showed me that the actual commitment to the day has to be a big portion of the return to really enjoy point-to-point touring. I didn't get that return this time. That idea that "I'm here now, and I have to be there later, and failure is not an option" seems to be a necessary part of the point-to-point experience. 

I wonder if I even got the thrill in the past. Maybe I was just more interested in going to the limit. When I was younger I did have the reputation of suckering people into a lot of death marches. 

Aging

It's pretty clear that I'm not young anymore. Given that, I have to either dial back my expectation of what can be accomplished on a ride, or I have to find ways to extract more performance from what ability remains. That leads me to three main ideas: better nutrition, better equipment, and more appropriate rides.  I think I have wrung all the improvements out of the first two, so maybe days with 13 km of 7 % grade at altitude are not in the future. 

Equipment

When I returned to Virginia, always happens on the first ride after some kind of tour, I was astounded at how nice and fast my regular road bike is. My response to aging has become making choices to extract every performance advantage from the bike, so that the rides stay enjoyable. When I have toured on my green Zeus, fifty-mile days have to be the norm.  

But I definitely needed something more than a 36-32 on Loveland Pass. Ninety minutes of 70 rpm was too much. 

Travel

Maybe seven days of one-on-one was too much for me. Mark was certainly a great host, but my anxiety level started creeping up on the last day. 

Climb statistics

  • Vail Pass: 17.8 km 648 m 421 VAM 154 bpm average 159 W average 2540 m to 3178 m
  • Loveland:  13.0 km 730 m 458 VAM 150 bpm average 156 W average 2876 m to 3602 m 
  • Va Canyon: 06.2 km 328 m 328 VAM 141 bpm average 145 W average 2407 m to 2745 m

Gear


Bike setup:

  • 2013 Crux with Hed Ardennes wheels with tubeless Maxxis Speed Terrane 33 mm file-tread tires and a 48-34 with 11-32 cassette. The tires were treacherous on the paved descents, and were not necessary for the Virginia Canyon climb. The way the knobs hooked up in high speed turns was unexpectedly unsettling. I would have been better off with 32 mm road tires.
  • Lezyne Bar Caddy bag.  https://ride.lezyne.com/products/1-sb-bcaddy-v104 Although it looks like a dry bag, it isn't--not even minimally-- despite the description of "highly water resistant." But I just used it for my rain and cold gear. It came with two male buckles on one side, but it  worked OK once I fixed it.  Its size was perfect for drop-bar setups. 
  • Ortlieb 16.5 l seat bag: https://us.ortlieb.com/products/seat-pack Its comically huge size makes it really difficult to climb out of the saddle--it's always 180 degrees out of phase. 
  • Revelate Tangle frame bag. https://revelatedesigns.com/product/tangleframebag/ Didn't hit my knees on it on this trip. 

Clothing: I opted for minimalist, because I feared hauling extra gear over a 4000-meter pass.


I prepared by assuming that we would get rained on, and it would be very cold. 

Mark's gear:

Although I felt a little like a cycling re-enactor, with my 13-year-old CX bike with QR, Mark put me to shame. He was riding the same frame he used during our 1979 three-week trip around Lake Michigan. His only concession to modernity was the 1990s-era drivetrain. 



Day 1: Glenwood Springs to Eagle

Map: https://connect.garmin.com/modern/activity/19684930964

Distance: 60km

We took the Bustang from Fort Collins to Glenwoods springs, which took about five hours including a transfer in Denver. The regional bus system in Colorado is nothing short of amazing. Both bikes rode on the bike rack on the front of the bus, though on Ft Collins to Denver, mine rode in the luggage compartment. 

The long bus ride meant that we were not even kitting up until almost 2:30. Amusingly, despite four engineering degrees between us, neither of us could discern that the bathroom door on the bus opened inward rather than outward, so the portajohn at the bus stop in Glenwood Springs was a welcome relief. We had to make a brief detour to the Target so I could get a replacement phone charger. Ugh

The first 30 km of the ride through Glenwood Canyon were spectacular, but hot. Even though you are often right under I-70, the highway noise is minimal. We even got to see the train we had originally planned to take. 






Day 2: Eagle to Frisco

Map: https://connect.garmin.com/modern/activity/19694143433

Distance: 96km

We started pretty early, and immediately ran into the situation the plagues any kind of group ride: someone is fast, and someone is less fast. The faster person tries to go slower, but inevitably tortures the less fast person. 

We had to navigate a lot of bike path in Vail, dodging pedestrians. But then it was on to the feature of the day: Vail Pass. I went out too hard, after I got passed by a gentleman of similar age but riding an e-bike. Although the scenery on the climb was spectacular, the actual pass was pretty mundane. It was just a dirt parking lot next to the interstate. But the climb was all bike path/abandoned road. I ended up waiting for Mark for almost an hour at the top. How he made it to the top with gear and a 39-28 is beyond me. I spent most of the climb in the 34-32 barely turning it over at 70rpm. 

While waiting, a big tour van disgorged 25 young women, all dressed identically in dark shirts and black, calf-length dresses.  They were all prepped to coast back to Copper Mountain, as far as I could tell, completely without adult supervision. One of them crashed almost immediately. 

After Copper Mountain we spotted a moose and calf off to the side of the path, and then it was on to Frisco.  Mark went off to look for liquids, and I explored the Frisco Historic Park and Museum, which is a collection of buildings that have been relocated from other sites around Frisco. I love local history museums, and this one really succeeds at encapsulating the history of Frisco. As always, I wish it had more maps. 

Our grade-school friend, Betsey Foley, drove all the way over from Vail to meet us for dinner. I had not seen her since 1980. 

Looking back westward on Vail Pass. The real climbing hasn't even started yet.


Waiting for Mark at the top of Vail Pass. Some kind of girls summer camp (fundamentalist? Orthodox?) 

Bill Luecke, Mark Anderson, Betsey Foley in Frisco
Mark, Betsey, and I in Frisco

Links: Frisco Historic Park: https://www.townoffrisco.com/things-to-do/frisco-historic-park-museum/

Day 3: Frisco to Blackhawk

Map: https://connect.garmin.com/modern/activity/19703846727

Distance: 98 km

At breakfast, Mark confirmed what he had hinted at the day before. He opted to take the bus from Frisco and meet me in Idaho Springs, which left me to attack Loveland Pass alone.  He graciously took my seatbag to lighten the load. 

I got lost a half dozen times before I got out of Frisco's seemingly limitless warren of bike paths, and had to use my phone to navigate to the base of the first climb.

Loveland Pass, which is the original road that the current I-70 tunnel bypassed in 1973,  climbs 730 m in 13 km to 3660 m at just under 7 %. The traffic was almost non-existent; much of it was tanker trucks that are banned from the tunnel. To be honest, it was not fun. Instead it was just a 1:45 slog, where I mostly thought "Let's get this over with." The switchbacks that start after the Arapahoe Basin ski area helped break up the monotony. 

I'm smiling in the picture at the saddle of the pass, but I didn't a sense of accomplishment like I once did. Instead it was just more relief that the climb hadn't killed me. 


At the highest point on Loveland Pass

The tires did not inspire confidence on the descent, which was further spoiled by a pickup truck that refused to exceed 35 mph--just fast enough for me to have to ride the brakes, but too fast to get around. 

View from the descent overlooking the I-70 tunnel

By the time I started descending, Mark had texted me his location in Idaho Springs. I felt like I had to move it along, so the next ninety minutes were another time trial, mostly on the I-70 frontage road. The first and best part was on a dedicated paved bike trail along US6, which I assume is the right-of-way of an abandoned railroad. 

The frontage road scenery wasn't great, but Silver Plume, Georgetown, and Idaho Springs looked like neat places to do more exploring, including the Georgetown Loop Railroad.


Amusingly, in Silver Plume I ran into two Kelley Benefits alumni on their way to the top of the pass. Oh to be in my early 20s again. I'm sure they were concerned when this rando old dude was yelling "Hey KBS!" at them. 

In Idaho Springs I rejoined Mark, and we set off up Virginia Canyon Road, an excellent 6-km, 330 m dirt climb. It passes quite a few active mines. 


Looking back toward Idaho Springs.


My dad said that he and friends had driven down this road at dusk from Blackhawk in the 1980s. I thought it looked pretty tame, but perhaps its wider than it was forty years ago. 

I found an unopened Gu at the base of the climb, which I consumed the next day!

Our lodgings were at the Lady Luck Casino in Blackhawk. 

Links:

Day 4: Blackhawk to Fort Collins

Distance: 144 km
Day 3 route

Mark took us on a gravel side excursion that turned out to just go directly up the fall line: hard, but it was early in the day. The first 30 miles were on the Peak-to-Peak highway. At one point, while looking at my Garmin, I apparently drifted over the white line on the shoulder and scared some dude in an SUV, who pulled over to let me have it. Surprisingly, I think Mark was more worried than I was that he was going to shoot me. I just apologized. 

After Nederland, we blundered onto the "Ned Gravel" course. Because we were riding against the direction of the race, I really had to pay attention to figure out which line to avoid. Fortunately, we were off the downhill section of the course after only a few miles, though we followed the route for about 10 miles in total. The section of Gold Hill road that we descended looked exceptionally brutal for the racers going the opposite direction. 

On the Ned Gravel course.


We lunched at the Gold Hill Store, a famous Boulder area ride destination. I had a coffee and a delicious grilled cheese and bacon sandwich. 

Rejuvenating grilled cheese at the Gold Hill Store.

The descent to Lefthand Canyon Rd on Lickskillet Rd took my brakes to the limit. But then we were on the pavement. Weirdly, a triathlete and a road guy both took some risks to get around me, but then sat up and forced me to pass them back. How they couldn't outdescend a guy with 35mm gravel tires is beyond me.

In Hygiene, Mark opted to call Mitch for a car service back to Fort Collins. Once again he took my seatbag and handlebar bag, and I rode the last 35 miles alone. I wasn't so much on fumes as just unable to go past 135bpm, though I felt like I could have done that for hours.  As usual, I got lost on the bike path in Loveland, and then hit a million cross streets on the bike path in Fort Collins. 

Day 5: Hermit Park Limber Pine Trail

We had planned a day of mountain biking, but I accidently added an extra day on the say, so we ended up with two days. Day 1 was on the Limber Pine trail at Hermit Park, a former Hewlett-Packard company recreation site. 

Mark let me use his twenty-year-old Titus Racer-X full-suspension MTB, which was pretty high-tech and high-zoot when it was new. It included a swanky Chris King headset! And it was set up like my Stumpjumper with narrow bars.

More re-enactor mode on the Racer-X. 

The guide I consulted after the ride described the route as "fast and flowing," but that description obviously did not take me into account. It was close to the limit of my anemic MTB abilities. It didn't help that I was still figuring out the bike. I got better on the return leg, but I still bashed my shins up figuring out that I could ride over stuff. 

We went for pie afterwards, and I got the strawberry-rhubarb, because Sandra wasn't around. It reminded me of my childhood, where rhubarb was on the menu every day in the summer. 

Links

Limber Pine/Hermit Park https://www.visitestespark.com/blog/post/guide-to-exploring-hermit-park/

Day 6: Switzerland Trail

The second mountain bike day was on the Switzerland Trail, an abandoned railroad bed built to serve the mining industry and failed about one hundred years ago. That meant that the gradients were not steep, and I didn't have to contend with any rock gardens. It was technically unchallenging, but probably not much fun on a gravel bike. 

Plus the scenery was amazing.

On the Switzerland loop off the Peak-to-Peak highway. 

Switzerland trail

Links


Monday, July 7, 2025

Utility Use Deep Dive

1. Introduction

About five years ago, I noticed that our water bill seemed to be increasing beyond what I thought that the rate of inflation was. This observation led me to do a deep dive into our utility consumption.
This deep dive was possible because, as a hoarder, I had saved almost every utility bill we've ever received at 7316 Allan. 
The deep dive comprises three sections: Water, Gas, and Electricity. Within each section, plots show the total cost in both absolute and inflation-adjusted dollars. Each section attempts to determine the major sources of the usage and cost, and provides recommendations for reducing usage. 

2. Water

2.1 Usage

2.1.1 General usage notes

Values of usage for appliances come from the manufacturer specification. For toilet use I tracked the number of flushes over a three-week period and effectively doubled them, because we have two toilets. I estimated the length of the shower, but measured the flow rate of our low-flow shower head. 

Activity Use Frequency
gal/day per day
Bill showers 5 0.57
Sandra showers17 0.57
Bill toilet 4.5 3.53
Sandra toilet 2.6 2.02
Dishwasher 3.4 0.57
Clothes washer11 1.00
Total 43.6

Figure 1 below shows the daily water consumption since 1992. Colors and symbols identify the season. The annotations indicate major life changes and installation and use dates of water-consuming appliances. 

Over the past thirty years, Sandra spent 2 years living full-time in Albany, and has stayed in Baltimore first four or five days a week, but since 2023, three days per week. 


Figure 1. Water usage since 1992

2.1.2 Findings

  • Our water use has generally decreased from about 80 gal/day to the present 40 gal/day from the 1990s to present.
  • None of the high-usage quarters (Usage>100 gal/day) were in the winter quarter. 
  • Note the general drop in usage 2005—2007 when Sandra lived in Albany.
  • The largest drop in usage correlates with the replacement of the old toilet in 2017. 
  • Despite Luecke post-retirement, usage remains low. 

2.1.3 Conclusions and Inferences

  • The new washer (11 gal/wash vs. 35 gal/wash) was canceled out by Luecke retirement.
  • Heavy usage in the 1990s could be due extensive lawn watering.

2.1.4 Recommendations

  • Very little extra efficiency on water use is possible. 

2.2 Cost

2.2.1 Background

  • Water bill is issued quarterly, includes charges for both supply and sewer. 
  • Water is billed in units of 1000 gallons. 
  • In 2014, the Virginia supreme court mandated that Falls Church City sell its water system, which supplied much of Northern Virginia inside the beltway, to Fairfax Water Authority. 
  • Our sewage is handled by the AlexRenew plan on Cameron Run in Alexandria. 

2.2.2 Rate and Total Cost

Figure 2 shows the absolute and inflation-adjusted cost of water. Annotations denote the periods in the change in ownership of the water company (see links below). Initially, I had thought that the increase in water rate was due to the Virginia supreme court-mandated sale of Falls Church Water to  Fairfax Water Authority. 
Figure 2. Absolute and inflation-adjusted water rate 

Figure 3 shows the actual billed cost of water. 


Figure 3. Absolute and Inflation-adjusted quarterly water cost. 

2.2.3 Findings

  • Sewage disposal is main driver of cost after 2008.
  • Sewage disposal costs have outstripped the rate of inflation by 4x since 2008.
  • Inflation-adjusted water supply costs have not changed in thirty years. 

2.2.4 Conclusions and Inferences

  • The payback time on a 2 gal.flush toilet over a 4 gal/flush toilet, assuming that the toilet cost is $200 and you install it yourself is only 470 days: $150/((7.1 flushes/day * 2 gal/ flush) * $0.03/gal). Paying a plumber $300 for the installation drives the payback time to about 3.2 years, which is still a rapid return on investment 
  • Despite our roughly 2/3 reduction in water use, our water bill has doubled since 1995. 
  • It seems reasonable that sewage costs have increased, presumably due to reduced tolerance for dumping raw sewage into the Potomac River. 
  • I had originally thought that the court-ordered sale of the water company to FC drove the price   increase, but the sewage price increases, which had always been controlled a different authority, started in six years before the sale. 

2.2.5 Links

This website http://www.virginiaplaces.org/watersheds/fairfax-fallschurch.html  summarizes the history of the Fairfax Water Authority/Falls Church Water Company battle. 

Sewage from our neighborhood goes to the AlexRenew treatment plan in Alexandria. Maps here: https://www.fairfaxcounty.gov/publicworks/wastewater/wastewater-treatment

3. Gas

3.1 Introduction

We use three gas-fired appliances:  a stove, a water heater, and a 80% efficient furnace, which was installed in 1995. 

3.1 Climate change in Northern Virginia

3.1.1 Summary plots

For a day, the heating degree days, HDD is defined as
HDD = 65 ℉- (Tmax+Tmin)/2 
If the average temperature is above 65 ℉, HDD=0.
Typically, HDD is a measure of the heating requirements in a given location. 

Figure 4: Heating degree days per year since 1999. 


3.1.2 Findings

  • The number of heating degree days, HDD per year  has decreased about 20% since 2000. 

3.2 Gas Usage

Figure 5: Gas use in therms in months when it is likely that the furnace is off.


3.2.1 Findings for gas use

  • Figure 5 shows that installation of a new water heater only partially reduced cut the gas use during months with no heat from about 6 therms/month to 4 therms/month. 
  • Gas, presumably for water heating, rose during the COVID-19 work-from-home era.
  • Gas use has remained at about 4 therms/month even though Sandra is only working and staying 3 days /2 nights per week in Baltimore, and William is retired. 
  • I have no explanation for high usage 2008-2014, although that era was near the end of life of our original water heater, which came with the house in 1992. 

3.3 Gas Cost

3.3.1 Summary Plots

Figure 6 shows the yearly total gas cost (heating+hot water+cooking) in both absolute and inflation-adjusted dollars
Figure 6: Yearly gas cost


3.3.2  Findings

  • In absolute terms, the cost for gas has remained effectively unchanged for the last 25 years.
  • In inflation-adjusted terms, the cost for gas has decreased almost 50%, partially due to the reduction in heating demand.

3.3.3 Discussion and Inferences

  • Of the $750 total per year charge for gas, $70 is cooking and hot water (4 therms * $1.5/therm) and  $680 is for heat.
  • In 2025 dollars, the total charge for heat over a 20-year expected life of a HVAC system is $14K. That makes it hard to justify significant extra costs for energy efficiency on a new HVAC system.

3.4 Gas Rate

3.4.1 Summary Plots

Figure 7: Gas cost rate since 2000. 


3.4.2 Findings

  • Gas actually costs about 20% less in 2025 than in 2010 in inflation-adjusted dollars. 

3.5 Notes 

  • I cannot locate any gas bills before 1999. 
  • Question. What is the base load of the water heater if you never  draw any water out of it?
  • The rate plots include the monthly "system charge" $12.40. Electricity and Water build this charge into the rate. 

4. Electricity

4.1 Usage

Figure 8 shows electricity usage per month since 1992. Yellow bands denote summer months (June 21-Sept 21). Annotations describe major electric appliance installation and use periods. 

Notes

  • Until 2003 we used the dryer that came with the house in 1992. To hold the door shut we used a horseshoe magnet. 
  • After replacing the refrigerator during the 1999 kitchen rebuild we moved the original pre-1992 refrigerator to the basement. Neither WL or SH can remember when we got rid of it, though.

Figure 8: Electricity used since 1992.

4.1.1 Findings

  • General trend: 2X current energy use during 1997 - 2004. 
  • The current energy use is  (300 < E < 400)   kWh/mo
  • Most high-use months occur in the summer, presumably correlated with A/C use. 

4.1.2 Analysis

4.1.2.1 Lighting 

Various internet sources indicate that in 2002,  85 % of bulbs in use were incandescent.  I estimate the bulb distribution and duty cycle for two eras.

- 1992 usage: 390 kWh/mo from 27 incandescent + 4 T12 fluorescent bulbs.

- 2025 usage: 150 kWh/mo from 33 LED,  3CFL, 4 T12,  8 T8, and  6 48in LED 

Energy use estimated that the lights are on 1/4 of the time. 

The longer we have lived here, the more lamps we have installed. 

4.1.2.2 Other Electrical use

I estimate the following usage from manufacturer specification sheets

  • 50 kWh/mo Refrigerator 50kWh/mo = 0.07 kWh/hour
  • 25 kWh/mo Dryer 
  • 36 kWh/mo Cable modem/Wifi/DVR
  • A/C 3kW running = 35kWh/day running 12 h/day, 5 days month = 170kWh/mo

4.1.2.2.1 Notes

  • Wifi/cable modem consumes 50W 24 h/day (measured)
  • Dryer use estimated from 13 months of hourly electric usage. The dryer signal is a 1 hour spike at 3.2kW, which occurred 88 times. WEL air-dries most of his clothing already. 
  • Nighttime energy usage  is 0.1 kWh per half hour = 0.2 kwH/h =  200 W   4.8kWh/day = 144 kWh/month.

4.1.2.2 Usage Patterns

Figure 9 shows the energy used during selected months where no A/C use would have occurred. 
  •  The minimum power use is in May and November, when lighting use would be low.
Figure 9: Energy use during months where no A/C was used. 

Figure 10 shows a snapshot of the hourly energy consumption for one week in 20204.
  • The nighttime base load is 0.2 kwH/h = 200W, which presumably includes the refrigerator consumption of 0.07kwH/h per hour.



Figure 10: Energy use during the week of 2024-04-28. 
 

4.1.3 Inferences and Conclusions

  • Finishing the replacement of T8 and T12 fluorescent bulbs with LED bulbs would reduce the power need about 200 W and the use about 40 kWh/mo, which is 10% of the total. 
  • Very few other approaches to reducing electricity consumption appear to exist, beyond being more disciplined about shutting off lights. 

4.2 Cost

Figure 11 shows the absolute and inflation-adjusted cost of electricity since 1992

Figure 11. Electricity cost per month.

4.3.1 Findings

  • Because of energy efficiency, electric bill in absolute dollars hasn't change in thirty years. 

4.3 Rate

Figure 12 shows the cost rate of electricity since 1992. 
  •  The inflation-adjusted rate has decreased (!) since 1992 from $0.10/kwH to $0.075/kWh.


Figure 12: Electricity Rate since 1992. 


-