Solar Production

I did another experiment to investigate how cooling the solar panels improves production.
I built this contraption to cool the panels down:

It is a popup lawn sprinkler mounted on the side of the roof and attached via a garden hose.

Here is todays production til about 4:45PM

The blue and red lines are the DC watts from each of the 2 different strings. The green line is the total output in AC watts ( after a small 4-5% inverter loss ).
5 times between 10:30am and about 1pm I turned on the sprinkler for several minutes, then once again at 3:45pm.
The red string is 6 panels on one part of the roof and 4 on another part – and the sprinkler only hits 6 of them. The blue string has all 10 panels on the main roof and the sprinkler can cool them all. The red string has a minor improvement, but the blue string is much better.

This chart includes a purple line that shows what the production would have been if the panels didn’t receive any water cooling.
The blue string peaked out at a 15.7% improvement from this line when it was being cooled ( significant variation can be attributed to fine tuning the sprinkler to not miss any panels )

This chart includes an orange line that is a prediction of what the improvement would have been if both strings had each received the full cooling.
The light blue line is a prediction of what output would have been all day if the cooling could have been applied consistently and continuously all day long.


Latest Electric Bill

Just got the latest electric bill. It’s a little unfair because we were out of town for a while, but its fun nonetheless.
The total bill is $19.37

We used 388 kWh from PSE, but we sold them back 246 via net metering, the difference is only 142 kWh. Thats 4.7 kWh per day.
Last years bill for the same period was 812 kWh – about $81. ( we were out of town last year for the same amount of time – so its apples to apples ).
How much did we actually consume during the month? Well our panels produced a total of 405, we sold back 246 of those, so we used 159, that makes our total usage for the month 547.
We managed to use about 33% less electricity this year than last year due to conservation efforts.

Note that even though our bill for this month is $19.37, we are going to get a payment of $218.70 for our energy production credit. ( The incentive program for putting on the solar panels pays us for every kWh we produce. ) So this month we will actually net $199.33 from PSE – that goes towards paying for our solar panels.

If we had made our solar system 35% bigger then we would have been net zero kWh for the month, and with batteries ( assuming no efficiency loss from the batteries which isnt the case ) we could have been “off-grid”. Of course june-july is one of the best production periods with the lowest demand, so its not much of a victory.

Thank you Puget Sound Solar!

Compressed Natural Gas

I think the ICE and gasoline are dead and I believe EVs are the only way forward.
A friend asked me about CNG vehicles, so lets look at them.

Why would we use CNG? Because it is domestically produced, burns cleaner than gasoline, and is cheaper than gasoline ( about half the price ).

What is compressed natural gas? Natural gas is a fuel that is different from gasoline. Gasoline is a liquid at room temperature, if left uncovered it will evaporate and make flammable ( possibly explosive ) vapor, but you can store it in an airtight plastic can with a lid. Natural gas is actually a gas at room temperature, it is liquid at -250F ( -158C ). In order to store any significant quantity it needs to be kept under pressure. A CNG car stores it at 3600 psi ( 250 atmospheres ), the tank has to be big heavy and strong, because 3600 psi of anything contains a powerful explosive force if ruptured ( without even igniting the resulting vapor ).
As natural gas pipelines get further away from the source ( a station that compresses it ), it divides into smaller pipes at lower pressure.
Natural gas pipeline in your neighborhood is 60 psi, and it enters your house at about 2 psi.
Propane in a tank for your barbeque stores propane at about 128 psi, and the tank is rated to handle about 300psi. The filling stations refuse to fill one older than 10 or 20 years ( depending on your local whims ) because they dont want to rupture an old worn out tank.
The CNG tank on a car will need to be replaced after about 15 years, when it nears that 15 years it will need to be carefully inspected. An old CNG tank is dangerous. CNG Tank Expiry

You can buy a CNG burning Honda Civic GX for about $25000 ( It qualified for a $4000 federal tax credit that may have ended at the end of 2010 – reducing the price to $21000 )
It has a 12.6 second 0-60 and gets about 180-240 miles range on its 8 GGE ( gasoline gallon equivalent ) of CNG ( compressed natural gas ) at 3600 PSI. The CNG tank is huge and eats your trunk, so you only get 6 cubic feet of cargo space. An 8 GGE CNG cylinder would be about 17.5 inches in diameter and 42 inches long – probably with a few inches of protective structure around it, and it cant be squashed into unused spaces like a gas tank. That makes it use about 4 to 8 times the volume of a gas tank in your car.
The nearest CNG station to me ( about 20 miles away ) is $1.99 per GGE.
Here is a link to a CNG filling station map and prices: CNG Map
There are 830 CNG filling stations in the U.S., but a large portion of them are not available for public use.
In fact there is no CNG station within 400 miles to the east, west or south of me, so my maximum range in those directions is 1/2 of my cars range. ( I am in Seattle )
You may not live near enough to one of them to want to go every 6 days ( 12000 miles per year = 32 miles per day ) – if I lived 20 miles further away that nearest CNG station would not be useful to me.
You can buy a home CNG filling unit that will cost you about $6000 installed. It can refill your car in about 16 hours. But it uses 800watts to do that, so in 16 hours it would consume 12.8kWh of electricity.
You can buy a $10000 unit that refills twice as fast, but I do not have any data on its electricity consumption.

So for $31000 you can have a car that will take you no more than 120 miles from your home ( probably only 100 ). If you get stuck there is no way for someone to bring you a can of some CNG.
If one of the public CNG stations is on your path, you can go further.
If you refill exclusively at home, you will spend $932 per year on fuel to drive 12000 miles ( $857 on the CNG at $2/GGE and $75 on the electricity ( at 11 cents per kWh ) to run the compressor! ).

Now lets compare it to a Nissan Leaf.
A Nissan Leaf is $32000 ( minus a $7500 federal tax credit, reducing its cost to $24500 )
It does the 0-60, in about 10 seconds and it goes about 80-100 miles on a charge ( with a 14.5 cubic foot trunk )
With a $1000 240V charge plug you can recharge the Leaf in about 8 hours ( They are going to upgrade the next years model to charge in half the time – if the Leaf had the same onboard charger as a Tesla, it could recharge in less than 2 hours from a 240V charger ). With a standard 120V plug on a 20amp circuit you can recharge it in about 16 hours. There are many dozens of the 240V public chargers within 100 miles of me if I wanted to use one.
With a Level 3 fast charger, you can recharge the Leaf in about 20 minutes, but there are very few of those right now, although they promise to add them along the I5 corridor over the next year.
The standard plug 120V can be found in every garage in America, meaning anywhere you go you can recharge your Leaf overnight and get home. So a round trip that includes an overnight stop is the same convenience as the 2x range CNG car.
The Leaf is rated at 34kWh per 100 miles, so $448 per year in electricity to go 12000 miles at 11 cents per kWh.

Neither car is useful for long cross country trips right now. The Leaf can do it, using existing infrastructure ( RV parks and other plugs of opportunity ), but it would be inconvenient. The CNG car can not do it at all.

After 5 years ( without inflation ) you will spend $4663 fueling your CNG car, or $2240 fueling your EV.
Including the home CNG filling station, they cost about the same amount of money.
In the very near future, the batteries will improve and the EVs will become cheaper and longer range, but the best you can hope for for CNG cars is more public filling stations – they arent going to figure out how to compress the CNG more.

Building a CNG station is more complex than a gasoline station, and thus likely to cost even more. You need a big compressor to compress the gas to 3600 PSI. In order to refill the car in a short time, it needs to have a big 3600+ PSI tank so it can have pre-compressed gas ready to push into the car. Gas stations are in the ballpark of a million or two million dollars to build.

A Level 3 charging station is a coke machine sized box that costs in the ballpark of $20k-$50k. They arent really very complicated, so if we buy them in quantity that price will plummet. ( However – an individual level 3 charging station can serve fewer cars per day than a CNG station with multiple pumps can. )

It is a lot cheaper ( several orders of magnitude ) to put in scattered level 3 charging stations and serve a vast geographic area ( like a few thousand and cover the entire country ) adding more capacity as the number of EVs grow than it is to try to cover the entire country with CNG stations, so I think the future for EVs in that regard is also much brighter.
Lastly note that if you took the natural gas and turned it into electricity at an efficient powerplant, the same CNG that would drive the Honda Civic GX 180-240 miles would drive the Leaf about 430 miles.

From the same fuel source, an EV will drive further than a CNG car and thus be cheaper to operate. EVs can use electricity from whatever source we choose: wind, solar, hydro, coal, natural gas, nuclear, tides, geothermal… Cars with roughly equivalent utility are already roughly the same cost, but EVs are rapidly improving in capabilities and price. The infrastucture to support a huge amount of convenient EVs is much cheaper to build.

CNG as a fuel for cars is a dead end.

Solar Panels and Temperature

I finally have been getting some good sunny days, unfortunately the solar panels haven’t produced as much as I thought they would because I didn’t factor in temperature.
The panels are rated at 190 watts at 77F and they drop off 0.32% for every degree F over 77F.

The temperature today was in the low 70s, but my roof shingles were 140F and the solar panels were at 100F ( according to my infra red thermometer )
I hosed down every panel in one of the strings until the nearest one was down to about 77F. I did this twice ( I was really thorough the first time, and did a little less the 2nd time. )
Here is the chart:

See those two spikes in the blue line? That is the increased production that came from cooling the panels down. It was about a 10% increase in output. The specifications say that it should be about a 7.5% improvement, but I was getting 10%
Notice that it takes about 40 minutes for the panels to heat back up and the output to return to the curve it was on.
( The “red” string produces slightly better – I don’t know why. It also is shaded for slightly longer in the morning, which is why its production starts later )

I think that if I could cool the solar panels down to ambient air temperature I would get about 10% more total output for the day.

Electric Car Costs

Electric cars have dramatically lower fuel costs than gasoline ICE cars.
The lower fuel costs allow you to spend more up front the electric car and pay less in the long run.
How much less?

Here is a chart of the two largest costs of owning a car, this example is a Ford Fiesta that cost $16000 and gets 33mpg and drives 12000 miles per year.
Gasoline is assumed to cost 4$ a gallon and increase at the same rate it has for the last 10 years – 8.5% per year.

Here is a similar chart for a Nissan Leaf that cost $26000, using electricity at 11cents per kWh ( national average for 2011 ) that gets 3 miles per kWh and drives 12000 miles per year.
Electricity is assumed to increase at 4% per year, and a battery replacement is assumed to be needed at 100,000 miles and assumed to cost $6000.

Lets look at a selection of cars to compete with the Nissan Leaf:

Now lets look at the cost over time for all those cars.
Fuel economy is assumed to be: 33mpg for the Fiesta, 26mpg for the Mazda6, 22mpg for the Hyundai SanteFe, 28mpg for the Mazda3, 45mpg for the Prius.
Same fuel assumptions are made $4/gallon increasing at 8.5% per year, and 11cents per kWh increasing at 4% per year.
We’re going to look at only 2 costs, depreciation and fuel. We assume that the ICE car immediately depreciates 15% on day 1, and then 15% per year after that.
Assume the electric car does the same, except we depreciate the battery even faster. Assume that the battery needs to be replaced after 100,000 miles ( this is a guess ) and the value of that battery depreciates to zero at that point ( and then you pay to buy a new one, and it starts depreciating ) The cost of the battery replacement is a guess.
There is reason to believe that the ICE cars will actually depreciate faster – depending on how bad their fuel economy is and thus how much it will cost to fuel them down the road – but we’ll ignore that for now.

The Leaf takes only 4 years to defeat its lowest cost competitor – the Ford Fiesta.
After 10 years, the total cost is dramatically lower than all the cars compared.
This comparison requires that you remember that a Nissan Leaf has significantly less utility than all those other cars for long trips due to its limited range.

Now lets look at the 160 mile range Model S.
We’ll compare the Model S to cars that cost dramatically less. ( The Model S will have better performance, occupant room, cargo room, and will only lose on range )

The fuel economy for the BMW 325 and Lexus ES350 is assumed to be 23mpg.
Again we assume that we will have to replace the battery at 100,000 miles and we make a guess at what that battery replacement will cost ( 8 years from now in 2020 )
How do the costs stack up:

The Model S easily defeats luxury cars that cost a lot less, and even defeats the Hyundai Sante Fe that costs less than half as much after about 9 years.

Now lets jump up to the super expensive 300 mile range Model S which provides a range that is almost no compromise over a gasoline car.
We compare it to cars that should provide similar luxury and performance.

And now the running costs:

The luxury cars that cost around $10000 less than the Model S are quickly dispatched, and even the Lexus that is priced $24000 less is bettered after about 10 years.

The end of the ICE Age

Rapidly increasing fuel costs are making gas burning Internal Combustion Engine ( ICE ) cars less and less affordable to drive. Can rapidly improving the fuel efficiency of cars keep them viable?
Let’s look at used car depreciation – using an average depreciation curve from 50 years of accumulated wisdom:

This chart shows an average depreciation curve for a $27,000 car ( $27,000 was the average cost of a new car this year ). The left axis is the price to buy the car used after each year it gets older. The right axis is the monthly payment to buy that used car with a 5 year car loan at 4%. The “traditional” depreciation has the car retain about half its value after 5 years and about 25% of its value after 10 years. This traditional depreciation curve shows the discounted value of the old car, based on the collective experience of the reliability and desirability of an old car.
An old car is a little more worn and creaky than a new car. It doesn’t have the newest features, but most importantly it has a much higher probability of failure, leaving you stranded or saddling you with an expensive repair.

Now lets look at a new car 20 years ago and include the cost of its fuel:

The absolute dollar amounts aren’t important, just the curve. 100% on the scale is the total outlay at the most expensive point – brand new. The red is the cost of gasoline to operate the car, assuming that the car is typical of its time ( 22mpg, drives 12000 miles a year and gas starts at 1.15 a gallon and increases at 2% per year. )
The curve is very similar for most of the years from 1960 to 1990 except for short periods of high volatility.
The green line is the total amount you pay to own and operate that car, I call this the “historical total monthly expense”. It represents a reasonable constraint on what you are willing to pay for a an old car – note that in the first year the cost to fuel the car starts out at about 10% of the total and after 12 years the cost of fuel has risen to about 40% of the total because it has ticked slightly up while the value of the car has dropped.

Now let’s look at a car today:

This chart shows the original payment number in blue ( calculated from the historical depreciation curve ).
The red line is the cost of gasoline per month, assuming a 30mpg gallon car ( todays average new car ) with $3.60 per gallon gasoline, that is projected to increase at 6% per year more than inflation.
Gasoline has really been increasing at 8.5% per year for the last 10 years, but the government claims that inflation is only 2.5%, so gasoline is increasing at 6% per year more than inflation.
The green line is the total cost to own and operate the car.
The purple line is the “historical total monthly expense” of payment + operating cost.
This is today’s scenario, where the total cost to fuel the car starts out at about 18% and after 12 years would be 75% if traditional depreciation values held.
Notice that the green line and the purple line are very very different.

Now let’s look at that same car but try to constrain what you are willing to pay for it to the “historical monthly total expense”:

This chart shows what you should actually be willing to pay for such a car, if it conforms to the traditional value proposition for a used car. If you constrain yourself to not exceed the historical total monthly expense curve, then what you are willing to pay becomes = historical total monthly expense – monthly cost of gas. The operating cost has risen so much, that at 10 years, the car is essentially worthless.
Conforming to this curve is not any kind of economic law, its more of a quantification of the feeling “I am not going to pay anywhere near as much to drive an old car than a new car”

This chart shows the same scenario for a car in 2021, if that new car now made 35 mpg – but gasoline continues to rise at the same rate.
When you initially purchase the car, the cost of gasoline is already 32% of the cost to operate the car per month.
Its used value plummets quickly and becomes negative shortly after owning it for 6 years.

The standard value proposition of the used car is quickly destroyed. This is because the cost to fuel it is very significant the very first year, but very shortly it dominates the total cost to own the used car. The fact that the car manufacturers are going to have to dramatically improve the fuel efficiency of the cars very rapidly to make them affordable at all, will make older cars obsolete so fast they will have little value. This will happen so fast that even a 6 year old car will be worth very little.

But car manufacturers are going to raise efficiency even faster than 5mpg in 10 years to battle the rising cost of gasoline right?
Lets look at the scenario from the opposite point of view:
Its 2021, and you are trying to choose what car to buy. You evaluate a new car and used cars up to 10 years old.

This chart shows the average cost per month of gasoline for a new car on the left and a 10 year old car on the right. We assume that the cars are similar price but each year older car gets 1 less mpg – with a new car getting 40 mpg. ( This represents the cars improving at 1mpg per year between now and then. There are many ways to mprove the fuel efficiency of a car and only some increase the cost or decrease the size, you can change aerodynamics, weight, power, rolling resistance, engine technology. Probably multiple variables will need to change. )

This chart is my attempt at calculating a smooth curve of the relative value of these cars.
The green line is the traditional car value ( depreciated value + operating cost line ) as before.
The blue line is the new car value line ( depreciated value + operating cost )
The red line is the actual value of each car.
Notice that a 10 year old car instead of costing you about 38% of the new car to own and operate, is about 44%.
But the value of that 10 year old car is near zero.
In fact the value of the 5 year old car is under 25%, whereas in the past it was near 50%.

Is there a conclusion? Yes. Here is mine. The traditional value proposition of a car is that you pay for it with a 5 year loan and when the loan is over, you have something about half the value from what you started with. In 2021 if you buy a new ICE car you will be looking at something that is very likely to have insignificant value within 2 years of completing your loan term.
I think that by 2021 buying an ICE car will be an unattractive proposition for a great many reasons, and this is a very significant one. Nobody will want to invest such a huge amount of money into something that becomes worthless so fast.

Here is my wild prediction: by or before 2021, the sales of new ICE passenger vehicles in the U.S. will fall well below half of the per year peak ( which I believe was 17 million in the years 1999 to 2007 ) and the ICE age will be officially over.

Day in the Life of a Pump Jack

The pump jack is a workhorse oil pump used to pull oil out of the ground.
When oil stops gushing out of the ground, you pump it out of the ground, thats what a pump jack is for. Oil hasnt gushed out of oil wells in the U.S. in probably 40-50 years.

There is an electric motor that drives the pump 24 hours a day 365 days a year.
I found this website that has specs for their biggest 15 horsepower pump jack.
A 15 horsepower 90% efficient motor will use 12.4 kW, which is 298 kWhr per day and 108,916 kWhr per year.
Deeper wells produce less per unit of pumping and shallower wells produce more.
This particular model with a 5000 foot deep well can produce 100 barrels per day.
Ahh, but 100 barrels of what?

This report shows that california onshore wells produced 195 million barrels of oil and 2.3 billion barrels of water.
I think that means that out of those 100 barrels pumped, you get 7.8 of oil and 92.2 of water.

Thus over the year you would get 2847 barrels of oil from your well, and it takes 38kWhr of electricity to pump each barrel up.
Since each barrel yields about 19.5 gallons of gasoline when refined, thats 2 kWhr of electricity to pump the oil out of the ground.

Turns out that 15 horsepower is a little pump. This website describes pump jacks up to 100 horse:
How deep are wells really? Not sure but the EIA shows that the average depth of new development wells has risen from 3861 feet in 1950 to 4938 feet in 2008.

California has 42000 oil wells, it is unclear how many of them use pump jacks.
This site claims that California used 3.7 billion kWhr of electricity in 2000 to extract its 253 million barrels of oil. Thats 14kWhr per barrel, and the average producing well was 2500 feet deep.

By 2009 onshore production was down to 208 million barrels. How much is energy per barrel up? The average depth has probably gone up, not sure how much. But the report shows 2000 new wells drilled per year, if 2000 new wells are drilled each year and 2000 are retired, that turns over more than 40% of the wells over 9 years. The average new well depth in that span has been near 5000 feet. ( If you replace a 2500 foot well with a 5000 foot well and get the same volume you need more horsepower which means more kW ) The total amount of liquid pumped is probably the same while the percentage that is oil has gone down. If the amount of electricity used is the same, then we’re up to 18kWhr per barrel.

Of course this energy is used to pump the oil up to the surface. How much energy is used to pump it to the refinery and separate the water from the oil is still unclear.