Archive for the ‘Going Solar’ Category

Going Solar: Costs and Rewards

February 8, 2014

More gratitude for ground mount system 2/3/2014

So far we are generating slightly more electricity than we are using. In four months as of 2/8/14 here is what we have:

2929 kWh total generated, 732 kWh/month, 24.2/day
2856 kWh total used, 714/month, 23.6/day
2054 kWh meter F (Forward) reading (from grid)
2127 kWh meter R (Reverse) reading (to grid)
73 kWh excess generation

Not  bad for the coldest, shortest days of the year, and particularly cold this year.

Our total investment in this system is $9,365 after tax rebates. This amounts to $1.36/DC watt or $1.57/AC watt.

I think the way installation costs are usually compared is dollars/nominal DC watts, which for our system is 6,900 DC watts. So that would be $1.94/watt before tax rebates.

This compares very favorably with average residential installation costs. According to Lawrence Berkeley National Laboratory Tracking the Sun VI report, average installation costs for the US in 2012 were $5.2/watt ($2.6/W in Germany, $3.1/W in Australia, $3.1/W
in Italy, and $4.8/W in France). The lowest state was Texas at $3.9/watt. *1

This is our reward for all the research, study, risk taking and work that went into our Going Solar project.

Here is a breakdown of the costs for our solar system.

Total   System
6,900 Nominal DC watts
$13,378 Total Cost
$1.94 Cost per DC watt
$4,013 IRS Tax rebate @ 30%
$9,365 Net cost after rebate
$1.36 Net cost per DC watt
5,970 Max AC inverter watts
$1.57 Net cost per AC watt
87% Max efficiency (AC/DC watts)
5 Average hours sun per day
29.850 Potential kWh/day
10,895 Potential kWh/year
$1,089 Value at $0.10/watt
8.6 Years to pay off at $0.10/watt
7.5 Years payoff at 4% inflation
2,929 Total kWh in 4 months
1.06% Percent of estimated (YTD)
$8,598 Solar Panels & Inverters*
$1.25 Cost per nominal watt
$1.44 Cost per max inverter watt
$1,351 Electrical Parts
$0.20 Cost per nominal watt
$0.23 Cost per max inverter watt
$3,321 Ground mount system
$0.48 Cost per nominal watt
$0.56 Cost per max inverter watt
* Includes $480 Envoy for data

*1 A limited set of data for early 2013 showed installed prices have continued to fall, with the median installed price of projects funded through the California Solar Initiative declining by an additional $0.5/W to $0.8/W (10-15%) depending on system size, relative to
systems installed throughout all of 2012.

Going Solar: More fun with the Envoy

February 8, 2014


What is causing that little dip in the power curve?

Right away I started seeing a dip in the power power curve every day about 12:30 PM.

Enphase Energy - Enlighten  Milczarek 6.9 KW Grid-tie - Google Chrome 11212013 14455 PM

Each day, some of the inverters at the west end of the array would disconnect. It happened in full sun or dark grey skies. The number of inverters and the exact time of day varied just enough to make it even more confusing. Sometimes one or two inverters from the east circuit dropped out too.

Lets look at that example from October 20. Here is a diagram showing the average Watts generated by each inverter for the five minute period from 12:30 PM to 12:35 PM.


The 12 inverters at the west end disconnected during that period, giving lower average Watts than the other inverters.

Micro inverters are designed to disconnect when they sense the power from the grid is outside of certain specifications. When this happens, the Envoy logs the event and we can find a history of these events on the Enlighten Website. Here is our October 20 event:


So it’s a “AC Frequency Out of Range” event involving 12 inverters starting at 12:31 PM and lasting for 10 minutes and 40 seconds. We can drill further into the detail of the issue…

Enphase Energy - Enlighten  Milczarek 6.9 KW Grid-tie - Events - Google Chrome 11292013 121320 PM_thumb[1]

…and learn more about this kind of issue:

“The microinverter reports that the frequency coming from the utility is either too low or too high as specified by applicable regional standards.
AC frequency is the frequency at which voltage varies on the utility grid. Frequency Out of Range events are usually transient and self-correcting by the utility.
When the microinverter detects an out of frequency condition, it must remain offline until the utility has been within acceptable limits continuously for a short period of time (seconds to minutes, varies by region). If during that time the utility again exceeds or falls short of acceptable limits, the five-minute timer must restart and the microinverter may not begin producing power for an additional short period following the last out-of-bounds condition.
If the condition persists: Contact your installer or refer to the Troubleshooting Guide at”

A Carroll Electric engineer said he thought it unlikely the frequency of the power coming from the grid really was out of range. Wouldn’t it affect all the inverters, and why everyday about noon? Weird.

To learn even more than you ever wanted to know, we can drill further to a list of the affected inverters…

Enphase Energy - Enlighten  Milczarek 6.9 KW Grid-tie - Events - Google Chrome 11292013 122415 PM_thumb[1]

…and further into a specific inverter…

Enphase Energy - Enlighten  Microinverter 121129943792 - Google Chrome 11292013 123324 PM_thumb[1]

… then run the slider through the five-minute intervals to see the average power for each period. This inverter tested the waters after five minutes, found the world safe at 12:36 and came back online. The inverter to the left waited a minute longer to drop out, found the frequency still out of range five minutes later so staid down an additional five minutes before coming online at 12:42 when it was safe. Most (10) of the inverters affected were offline for 10 minutes; two were offline for only five minutes. See, more than you ever wanted to know.

One thing I learned is that on October 20, we lost more energy to these inverters going offline than we did to the clipping when they maxed out. And, these power dips happened every day, not just on the best sunny days when the inverters maxed out. So here was a potentially bigger source of energy loss we would not have discovered without the Envoy.

Grasping for theories I wondered about the greater distance of the west 15 inverter circuit from the main electrical panel in the field east for our home. Was the gage of the electrical cable big enough? Were electrons bunching up causing disturbances in the frequency? Now I would need an electrical engineering degree! And really, why would that happen around noon everyday anyway?

And get this: after the time change November 4th, power drop continued as before around 12:30 PM. They new about daylight savings time! So clearly it is clock time related.

An Enphase engineer I spoke to said the frequency registered by these inverters was only slightly above the specified limit and that he could adjust that limit up a little to keep them from dropping out. Wow, and he could do it from there. In the end, that did not completely solve the problem, although it does seem there has been less power loss.

See how much fun you can have with an Envoy? I don’t know if we’ll ever make up in energy savings the $480 it cost us, but what if the problems were greater and we just didn’t know? Or what if something develops in the future the Envoy will help us find and resolve. Plus we can watch for all kinds of patterns over the seasons in different weather and the positions of panels in the array.

For example, I discovered that there is a greater difference in energy production between the top and bottom rows of the array, than from the west and east circuits where we would expect the loss of the daily power drop to show. It looks like the sun hits the top row first when it rises and last when it sets giving them a little more energy to harvest each day.

Connecting the Envoy

The Envoy instructions say you need an always on broadband Internet connection through a broadband router with a spare Ethernet port. Well, that’s not me. I’ve got a cell phone with a Wi-Fi hotspot.

Below you see the Envoy with a power cord and Ethernet cable coming out of the top. The Envoy has no wireless capability. The instructions have you plug it directly into a


The Envoy receives its data from each inverter through the electrical cable connecting the solar array to your household electrical circuit. To get the strongest, cleanest signal, the Envoy must be plugged directly into a power outlet as close as possible to the connection to the solar array. Our array connects to a subpanel in the shed that feeds power to well and water equipment, so that is where we plugged in the Envoy.


Edimax 2

Going Solar: Benefits

November 29, 2013


Power and energy: How much?

Electrical power is measured in Watts, which we observe to see how powerful a light or heater might be.

Electrical energy is measured in Watt hours or kilowatt hours (kWh), the amount of power used or generated over time. A 100 Watt light left on for 10 hours will use 1000 Watt hours or 1 kWh of electrical energy.

So how much power and energy is being generated by our array?

We can begin with the “net-metering” meter newly installed by Carroll Electric. This meter is designed to measure electrical power flowing both forward (F) from the electrical grid to our home, and reverse (R),  back out into the electrical grid when our array is generating more electrical power than we are using at the moment.


Here we see that 458 kWh of electrical energy have been drawn from the Carroll Electric grid since the meter was installed October 11, 2013, when our array started generating electricity.


When the meter flips to the “R” mode, we see that 814 kWh of electrical energy have gone back to the grid, the amount of energy generated by our array beyond what we used from the array. The difference, R – F = 356 kWh, is the energy “credit” we have in the grid, energy we can draw on nights and cloudy days when our solar array isn’t generating anything.

So the meter can’t tell us how much energy our array has generated, only the excess above our own use.

This is the purpose of the Envoy Communication Gateway from Enphase, the manufacturer of the micro inverters under each solar panel. The Envoy records the power and energy generated by the 30 inverters and sends the data over the Internet to their Enlighten website where we can see the results.

imageimageEnphase Envoy Communication Gateway

The Envoy is basically a $500 meter, more than many want to pay to learn how their solar array performs. I certainly had second thoughts.

But all the information you get is really quite wonderful: a record of every watt and watt hour generated from every solar module and the entire array every five minutes from day one, plus a record of all events that cause a module to go offline.

And all this data is available over the Internet in many helpful ways: tables, charts and a view of the array itself, for any time period from a five minute interval to the cumulative life-time total for any and all modules.



Here on the Enlighten website we see a diagram of our 30 solar panels on Thursday noon at 12:10 PM on a cloudy raining day. The 30 inverters under them are pushing electrons out at 543 Watts (the number in the lower left of the above diagram), giving us a total of 1.14 kWh so far today (upper right). We have 150 kWh for the last 7 days, 532 kWh month to date (November 21) and a lifetime total of 1.09 MWh, more precisely 1,088,275 Watt hours so far since October 11, 2013.

Imagine that, more than a million Watt hours already…from the sun. As of this blog post on December 1, we have 1.29 MWh, and 292 kWh of credit in the electrical grid. Wow. 

Will we be able to keep this energy credit after January 1?

There is a new addition to the net metering law this year that utilities must roll over up to four months of average annual kWh usage of credit remaining at the end of the year. Whether this will help us or not depends on some interpretations. The Carroll Electric engineer I talked to on the phone didn’t know when this provision goes into effect, and he thought annual usage might begin the date of our net metering contract, not the beginning of the calendar year. If so, our “annual usage” since October 11 will be very small indeed, since we have been generating more than we use.

Since I think Carroll Electric may take away some or all of this energy credit come January 1, we have been using some on a space heater to save propane. Space heaters really soak up the kWh fast. Our 1500 Watt heater draws around 1.4 kW according to our Kill A Watt electricity use monitor, so running for 10 hours on a cold day it can draw down 14 kWh from our credit all by itself. 

Lets return to the performance of our solar system.

You can see in the diagram above that each inverter is pumping out only 18 Watts, and its noon for Pete’s sake, not much on this dark wet day. In contrast, here is noon on a bright sunny day.


Each inverter is pumping out 199 Watts for a total of 5.97 kW for the 30 inverter array. At the bottom of the diagram are the power curves for seven days. I’ve moved the slider to 12:15 PM on October 20, our best day, which gave us 42.435 kWh of energy.

Our lowest day was 2.06 kWh on October 30, which I just looked up in a report created by the Enlighten website showing the kWh for each day since the beginning.

So what else can we learn from the Envoy and Enlighten website?

First, did I make a good decision to save $1,600 by buying the less powerful 190 Watt inverters? And second, what is this strange dip in power generated each day about noon?

Energy lost to save $1,600

So why only 199 Watts per inverter when the panels are rated at 230 Watts? It’s because of the lower 190 Watt micro inverters I purchased instead of the 215 Watt inverters that saved us $1,600. Was the savings worth the energy loss?

Lets look at the energy lost on that best day, October 20. Here is the power curve with the area under the curve representing the entire 42.4 kWh. Ignore the little dip in the curve for the moment; that is our second question.


As you can see, the curve flattens off between 12:15 PM and 2:00 PM for 1 hour and 45 minutes when the inverters max out.

The energy lost is what lies above our curve and under a hypothetical curve that would result from using the 215 Watt inverters. With the higher rated inverters I assume the curve would continue above the flattened area to about half way to the next line, or 6.5 kW.

My best estimate is that we lost no more than about 625 Watt hours or .625 kWh that day, 1.5% of the 42.435 kWh total. You can see the energy lost would have to be a very small part of the total area under the power curve.

Of 48 days the array has been operational, I count 8 days that have some clipping, fewer with as much clipping as this day. This translates to about 60 days of clipping in a year, or about 38 kWh per year, and 760 kWh in 20 years. At a rate of about $.10/kWh, the economic loss would be $76, much less than the $1,600 savings for purchasing the less powerful inverters. Not a bad tradeoff.

The price of electricity will undoubtedly increase, and it is possible there will be greater energy loss to clipping in the lighter half of the year, but it seems unlikely the financial loss will come near the savings gain. I expect to revisit this question once we have a years worth of data.

I don’t think this kind of analysis could be done without the data recorded by the Envoy, data that can help us answer the next question…

What is causing that little dip in the power curve?

Find the answer in the next exciting installment! This is long enough already. I know you can’t wait. Chuckle chuckle.

Going Solar: The Array Goes Up

October 30, 2013

All the Gear


While the ground dries out from the big rain, Jerry and I take a hard look at all the gear shipped to us by, reading manuals and lining it all up in order to plan the installation.


The oval device top left is the Envoy Communications Gateway that sends data from all the microinverters to the Enlighten website where we can monitor the performance of the solar array over time.

A maximum of 15 Enphase 190 watt inverters can be installed on a single circuit, so we divide the array into two circuits of 15 inverters each. A small junction box begins each circuit, They connect to  the two 15 amp circuit breakers that will go into the combiner box (with the big red lever in the photo) and to the cable that goes to our home electrical system.  We have a 240 volt subpanel for our well and water equipment in the nearby shed, which in turn connects through a cable to the Carroll Electric meter in the field east of our home. The gear includes a box of grounding wire lug nuts, a lightning arrestor and a lockable disconnect switch required by Carroll Electric to be mounted on the outside of the shed.

Where’s the Electrician?

About this time I realize I should get an electrician involved in the project. Because I had completely wired a house a few decades ago, I thought I could get by with a consultation, and then at the end an inspection. Wrong. Mike and Faith tell me they are happy with master electrician Lyle Pinkley who installed their system recently, so I make an appointment. Lyle and his assistant review everything and give me a list of things to purchase at Maverick Supply, plus another list of things to return to He saves me $360 even after restocking fees of 20%. Well, minus the much smaller cost of the replacements from Maverick. I’ll need to go over all that later when we get to financial details. So let that be a lesson to you: get your electrician involved at the beginning. I know this is obvious to most of you.

Installation Day

Finally, on Monday afternoon October 7, Jerry joins me in beginning the installation. We start with the inverters which must be installed first since they will be covered by the solar panels. The Enphase 190 watt inverters have long connecting cables that allow all the inverters in a circuit to be connected in series together. The more powerful Enphase 215 watt inverters require a trunk cable to which the inverters are connected, an additional expensive I saved by using the older cheaper and less powerful inverters. In a later post we’ll look at the tradeoff details of this choice that saved me $1,600.

Here Jerry is bolting on the first inverter. You can see the lug nut at the bottom of the mounting plate where the copper grounding wire is to be attached. A single strand of 6 gage bare copper wire must connect every inverter, solar panel and mounting rail to an 8’ copper clad ground rod driven into the ground.


We are so excited to see what it all looks like, so to test our approach, we mount the first column of panels. It’s a total thrill to see them there at last. Wow! Alece is thrilled too.


That evening while looking at the installation manual, I realize the inverters’ longer connecting pigtails are pointing up, rather than down toward the circuit’s junction box as required. So the next morning I have to unbolt and remove the three solar panels, disconnect the grounding cable, remove the inverters and reattach them to the opposite mounting rail, reconnect the grounding wire and install the three panels again. Oh well.

Another tricky step is getting all the mounting rails spaced evenly, and the bottom end-clamps set to insure the whole array will be square. There is head scratching, wheel spinning and some frustration as we work this out. I’ve come away with ideas about how to do this better which I’ll share in a latter post.

We install the rest of the inverters, running the ground wire across each mounting plate, and on up and down the rails leaving enough wire to attach to each panel and rail. Then we tuck the wire under the grounding nut on each inverter and tighten. We have to drill holes in the solar panel frames and mounting rails for the ground wire lug nuts in our shipment.


On Wednesday, like an old-fashion barn-raising, David, Jerry, Carl, Mike and Faith come to look at the ground mount system and how its all coming along. Together we work out the alignment strategy and begin installing in earnest. David and I hustle placing the mounting hardware to keep ahead of Jerry as he carries one panel after another to the array.

Lyle and his assistant arrive and begin the electrical installation. Wow, its all coming together.


In each column, three panels are placed over two parallel mounting rails. Two end-clamps, one for each rail, anchor the bottom of the first panel, then four mid-clamps anchor the middle panel to the bottom and top panels, and finally two more end-clamps anchor the top panel at the top. Here, Jerry is tightening the mid-clamp between the second and third panels. He uses a socket nut as a spacer to keep the columns of panels spaced evenly one inch apart.


We finish laying out the first two rows and David tightens the mid-clamps.


Now, how to get those panels up to that third row? Hmmm.


Jerry just grabs a panel and reaches it up across the front while David (on a ladder) and I help set it in place and finger tighten the clamps. Just do it! Wish I had a photo of that process. It is like, woosh, and before you know it there it all is: 30 solar panels ready to harvest the sun. Amazing. Beautiful.


Below, the first inverter of the second circuit ties into a junction box with a cable you can see going to the left off to the combiner box at the beginning of the first circuit at the east end of the array. Our shipment included a connection cable for each circuit that connects to the long pigtail of the first inverter you see here. The other end has prestripped wires you see in the right side of the junction box. I’m sure Lyle could have simply cut the connector off the last pigtail and stripped the wires himself, saving me the cost of the two connecting cables.


Below is the $25 combiner box  Lyle got to replace the $191 box (with the red lever) sent by, where the two 15 amp breakers are to be installed to protect each circuit. The orange electrical cable goes down through the conduit, along the trench and up into the disconnect switch on the wall of the shed.


Below, Lyle installs the lockable disconnect switch required by Carroll Electric. The cable connects with the subpanel inside the shed and from there to the Carroll Electric grid.


Connecting Up

We connected the inverters in series to each other as we installed them, but there is a serious warning to never connect solar panels while the sun is on them generating power, which can short across the connections damaging them. I decided not to connect the panels that evening since I hadn’t  heard all week from Carroll Electric and had read they didn’t want us to connect to the grid until they had tested and installed the grid-tie meter.

Thursday morning, Jerry called to find out what kind of power we were generating. So disappointed he was that I hadn’t connected the panels that I decided to do that after sunset.

So next morning we are packing for our two week trip out of town when we get a call from Carroll Electric explaining the relevant person has been out of the office all week. When he realized I was ready to connect with application in hand, and had been trying to reach them all week, he said he would try to get someone over right away. They really came through.

On Friday afternoon, October 11, just before we left town, a Carroll Electric technician arrived to inspect and test the system. This guy was great: full of encouragement and interesting information. He was looking for the UL code on the inverters that said they meet the specifications for grid-tie systems that shut down when the grid is down. He took photos of the array and disconnect switch, tested the system and installed the grid-tie meter. He explained his testing is to be sure our solar array shuts down when Carroll Electric grid power is down, so that line workers are not harmed while making repairs. He also said there are other more effective safety measures that protect line workers from power generated by any grid-tie sources, and that making sure our system shuts down when the power is out is more to protect our own equipment which could be damaged by their other safety measures. Another point he made: a grid-tie system like ours helps them, because they have to buy more expensive power from suppliers when more capacity is needed.


The new meter records the net power generated by the array minus the power we use. When we use more power than we generate, for example at night, the meter records the kilowatt hours we use. When we generate more power than we use, the meter subtracts the extra kilowatt hours we produce: i.e. net metering. At the end of the billing year, the account is reset and we lose any surplus in kilowatt hours we might have produced. So at this time, there is no incentive to build a larger system than you use.

One additional interesting note: The technician said that connecting a generating system to the grid without the net metering meter in place results in any power you generate counting against you as additional kilowatt hours you are billed. This is done to thwart abusers who reverse their electric meters to avoid paying for some of their power.

Also note that the Carroll Electric representative I spoke to said they want your grid-tie application early, as you are beginning. This is different from what I had been told in a phone call to them before I even ordered our system — that I should contact them when I was close to being done — at least as I remember the call. This was confusing to me for another reason: the application calls for such things as the date of installation, the size of the system and the name and credentials of the person who did the installation. I had read everything about net-metering on both the Carroll Electric website and the website of the commission in Little Rock. There it said the utility has 30 days to respond to the application. So I am grateful to them for coming through at the last minute with such flying colors.

And thank you Jerry for that extra push to connect up all the panels Thursday evening. The result was more than 400 kilowatt hours generated while we were away.

And so, ladies and gentlemen, here it is: our beautiful solar array collecting the free energy of the sun. As Alece says, its a whole other experience when you turn on a light!


Going Solar: Cement & Backfill

October 8, 2013

Not much to say here. I ordered 3 1/2 yards from Thompson Ready Mix in Berryville, and the driver showed up a couple of hours later. Maneuvering the cement chute around all the pipes was a challenge. We really needed a third person to center and top off the last tube poured, while one of us guided the driver.


With only four tubes left to pour, while I centered the last one, the driver backed up to reposition and snagged one of the pipes with the chute, pulling the top of the entire structure several feet to the north. It took quite a while to try to get it all back in place and we never did get it completely plumb as it had been before the snag.

Jerry Landrum, bless his heart, showed up toward the end to help with the last four.


A few days later, Bob Kelly was able to squeeze us in to back fill the trenches.


What an artist; that man can tread a needle with that big machine.



That very night we got 3 3/4 inches of rain, so we let it all dry out for a couple of days.


Next step: begin mounting the inverters and solar modules. Yeah!

Going Solar: Solar Angle and Mounting Scaffold

October 5, 2013


Best Solar Angle?

Many websites said simply point the array to solar south at an angle equal to your latitude to optimize annual energy production. According to, the latitude for our home is 36.19°. This is where the sun would be at noon on the spring and autumn equinoxes.

I liked the website Optimum Tilt of Solar Panels for simple explanations and the formula I chose for our array:
.76 * latitude° + 3.1° = .76 * 36.19 + 3.1 = 30.6°

Another delightful website exploring science and technology for teachers discusses Optimal Tilt Angle and compares various methods and formulas. The site has links to other great explanations of many related concepts such as insolation, tilt angle, sun angle and insolation, air mass, day length, and clouds and pollution. Reference is made to the PV Watts calculator that takes into account weather data to give estimates of the actual dollar value of an array of a given size at a given location.

According to the PV Watts calculator,  our 6.9 KW array should average abut 9198 KWH per year and return a value of $860 assuming local energy costs to be 9.4 cents/kwh. With all taxes and fees, our charges seem to be about 10 cents/kwh, which would be about $920/year.

The GroundTrac mounting system allows a maximum tilt angle of 30°, which fits the angle I had chosen. The maximum height of the rear pipe is 60” above the cement pier. To get the 30° angle, the front piers needed to be 73” from the rear as shown in the diagram below. If the hypotenuse is 84” as shown, the height of the upper pipe must be:

sin(30)*84” = .5*84” = 42”

The height of the front pipe will be the maximum of 60” – 42” = 18”

So we needed a scaffold that will hold the front pipe 18” above the top of the front piers, and the rear pipe another 42” above that (60” above the rear piers) at a horizontal distance of 73”.


We leveled a string 12” above the grade to represent the top of the cement footings and David placed bamboo poles 6’ apart where each footing would go. Next he cut a point on three 2×4 posts to place on each side of the trenches and in the middle. He used a heavy digging pole to make a hole and a sledge hammer to drive the posts into the hole in the ground for a stable base.


Next we clamped a 12’ 2×4 across the three posts and leveled the top 18” above the string representing the top of the footings. This cross beam will support the front 55’ row of 1-1/2” galvanized water pipe that in turn will support the lower end of the solar mounting rails. Leveling this beam across the back trench gave us a reference point for measuring the height of the back row pipe supporting the upper end of the mounting rails. I figured the height of the back row above the front row needed to be 42” to make the solar array tilt 30 degrees to the sun.

When the critical measures were correct we screwed the scaffold in place, removed the clamps and repeated every 10 feet for a total of six frames.

Next we slid on the Hollander T-fittings, threaded the lower and upper pipe segments together and lifted them into place. After making the pipes level and strait, we measured and marked where each T-fitting will center over the 20 cement pier footings. The four footings on the two ends are 24” from the end of the pipes, the others each 68” apart spaced evenly along the 55’ length of the front and rear horizontal pipes.

The next step was to attach the Groundtrac support rails every ten feet to assure the front and rear pipes were parallel and spaced the correct 84” apart. That was when I discovered the support rails had not been pre-notched as shown in the installation manual. Now I would have buy a Unibit step drill bit and drill 80 half inch holes before I could proceed. The bit set me back about $38 at Acord’s Home Center and several days because of weather, putting off pouring cement into the next week. I also discovered that did not carry the pre-slotted railings and suggested they make that clear when putting together a complete solar package.


Once the Groundtrac railings were in place, we set the vertical front and rear support pipes into the cement tube forms and attached the pipes to the T-fittings making each perfectly plumb. We laid out and leveled another string for the tops of the rear footings.


Finally we centered each tube form around the pipe just under the string and screwed it in place against two 4’ 2x4s cut from econostuds from Meeks, 20 at $1.39 each.


Now we were ready for the cement truck.

Going Solar: Breaking Ground

September 25, 2013

Breaking Ground (and cotter pins)

Here’s David Pettit helping me layout markings for the 20 footings. We spent a fun half day with compass, stakes, tape measure, strings,  string levels and a plumb nailing bright markers to the grass for for the precise location of each footing.


Our first attempt at digging the footing holes using our neighbors backhoe auger resulted in more broken cotter pins than holes. The auger just bounced on the famous rocks of Rockhouse Road.

Bob Kelly, bless his heart, came over the evening of Thursday September 19, 2013 after a long day at work. He thought it would be faster and easier to dig two long trenches than to try to dig out the individual holes every six feet. I stretched a string along the markers while he spray painted a line.


Here is Bob Kelly breaking ground with his big backhoe and 18” bucket.


We finished well after dark: two trenches 30” deep and 55’ long. In the dark I didn’t think about the footing cement forms and coming rain. It was an all night soaker that warped and possibly damaged some I might need to replace.


The next step is setting up the scaffold that will hold the pipe.

Going Solar: Voting with our Dollars

September 23, 2013

Solar Cost Benefit Dilemma

Each year, it seems, I drop in on our local hero and mad scientist Jerry Landrum for a tour of his latest eco-experimental projects demonstrating how we can live comfortably while making a very small carbon footprint. We always explore projects for bringing solar power to our home. Then I do more research and learn again how much more we’ll pay for solar vs. local grid power, and with the continuous scientific and technical advances how much less we’ll pay if we wait another year. And another year. And another… Four years now, each time feeling anxious about committing a major part of our limited resources. When considering only short-term personal financial costs, the analysis still favors waiting.

SWEPCO ABCs: Voting with our dollars

The prospect of major transmission lines running through our land has finally raised the cost benefit analysis to A Bigger Context (remember your ABCs!). When we consider the environmental and social costs of conventional power generation and distribution, it becomes clear we cannot continue to support them with our dollars. We are switching our vote to solar and the entire movement toward more people and earth friendly technologies. Our goal is net zero: to generate as much electricity as we use.

How Much Electricity?

Based on the last four years, we need a system that averages about 30 KWH/day or 920 KWH/month, a little below the national average of 940 KWH/month according to the US Energy Information Administration.

We do not live in an energy efficient home. It seemed the most sensible choice at the time: buy a relatively well made manufactured home to retire and settle here debt free. With our hot humid summers, air conditioning accounts for about half of our annual electricity use, and is required by our home warranty to keep humidity, mold and mildew in check. Our unit is at the bottom of the efficiency scale and eventually we will replace it.


Grid-tie, Off-grid battery or AC Coupling with both?

Jerry Landrum, our local clean energy guru, suggested we consider a grid-tie system with AC Coupling providing energy independence. Each module has its own micro inverter that converts DC to AC and connects it to the Carroll Electric grid through a special meter which they supply. It amounts to using the utility grid as a battery, drawing power from the grid as needed (at night for example) and running the meter backwards as you generate more power than you use. The downside of this system is that the entire solar system shuts down when the utility grid is out—this is to protect line workers making repairs from harm by your system. So you have no power.

AC Coupling is a solution to this problem. The grid-tie system is attached to off-grid batteries through another set of inverters. The AC Coupling inverters basically trick the micro inverters into sensing the utility grid is up allowing them to charge the batteries and any house circuits needed for emergency power backup. The AC Coupling inverters also do the job of preventing the backflow of electricity to the utility grid when it is out, protecting workers. More on AC Coupling.

We decided against AC Coupling at this time because of the substantial added cost of the required inverters and batteries. We loose the benefit of energy independence as we are tied to the grid. It is more economical to have a backup generator for temporary outages, but this leaves us dependent on a limited supply of fuel sources such as gasoline or propane and assumes outages will be temporary.

Which System?

Jerry emailed a link to a solar panel price survey ranking modules by price per watt. He had recently purchased a pallet of 20 230 watt Canadian Solar modules at $0.72/watt. had a current pallet price of $0.75/watt for these modules, still historically low  in order to sell excess inventory as higher watt modules came on the market.

The 230 watt modules pair with the slightly older Enphase M190 micro inverters also being sold at a discount. We average about five solar hours a day in this area, so the 190 watt inverters should generate about  950 watt hours or a little less than one KWH (kilowatt hour) per module. Thirty modules should get us close to our net zero goal. We bought two pallets of 20 of the Canadian Solar modules. Michael and Faith bought ten of them from us. They choose the higher rated Enphase M215 inverters.

East West Roof? Which Ground Mount?

Our roof faces the wrong direction for a roof mounted system. I spent days researching different mounting systems. We considered creating a solar shade over our western deck, or the north patio, or even over windows on the east and/or west sides of the house. In every case awkward tilting angles and/or structural complexities and uncertainties left us shaking our heads. We do have a carport roof with space for half of the modules, but when I reviewed the wind and load specs I was reluctant to add up to a thousand pounds and void the warranty. Plus, after pricing roof mounted possibilities for the carport, I wasn’t convinced we would save enough to feel worth the risk. In the end, an engineered ground mount system felt most safe. I discovered there were many kinds of ground mount systems. recommended the ProSolar GroundTrac mounting system. A 10’ x 55’ array consists of 3 rows of 10 modules in landscape mode. The GroundTrac mounting rails are attached to two 55’ lengths of 1-1/2“ galvanized water pipe running the length of the array. These are supported by vertical pipe set in 12” diameter cement piers 42” long. To meet engineering requirements for wind and snow/ice load, piers must be placed no more than six feet apart. This requires two rows of ten piers each. The back row horizontal pipe is 42” above the front row to tilt the array at a 30 degree angle to the sun. This foundation requires 240’ of 1-1/2” galvanized water pipe and 4 couplings purchased at Maverick Supply for $815.54 including tax. Because I was not able to transport the four lengths of standard 21’ pipe, I had them cut the pipes in half, thread them and add 4 more couplings for an additional cost of $39.46 for a total of $855. The piers require 20 48” 12” diameter cement forms at $11.96 each at Lowes for a total of $252.62 including tax. I estimate the piers will require about 2 to 3 yards of concrete.

Here is a diagram of the GroundTrac system showing the top of the cement piers, the vertical and horizontal 1-1/2” galvanized water pipe, the GroundTrac mounting rails in pairs supporting the columns of solar modules. With no snow-load requirement, longer rails can be used that support four modules, and the piers can be 10’ apart. Thinking of the ice storms we get, I chose the snow-load engineering specs: 6’ pier spacing and mounting rails that support three modules.


Linh Tran of GoGreenSolar prepared our order for a complete system price of $10,603 to include the 30 solar modules and inverters (including connecting hardware and switches, and a monitoring system to track the performance of the array), and the GroundTrac mounting rails and related hardware. The order did not include required copper grounding wire and rod, electrical cable from the solar system to tie into our electrical panel, nor the materials for the pipe and cement foundation. It should include everything else needed to complete the system.

And…here it comes!