Differences

This shows you the differences between two versions of the page.

--- dougs_4.6kw_diy_grid_tied_array [2012/10/28 09:03] – tomgee
+++ dougs_4.6kw_diy_grid_tied_array [2012/10/28 09:30] (current) – tomgee
@@ Line 38: / Line 38: @@
 I briefly considering using wood for the rack, the 25-year warranty on panels and inverters and projected life of possibly 50 years made me want to build for long-term. As I can weld, I decided to use inch and a half (2”OD) schedule 40 galvanized steel pipes. I could buy the five lengths of pipe I needed locally for about $55 for a 21-foot length.
-  The website- http://rredc.nrel.gov/solar/calculators/PVWATTS/version1/ makes it easy to determine the best angle for year-round production, which was 30 degrees for my location. The only tool that was not already owned by me and needed for this job was a low-cost pipe notcher, which I found on sale for $24 from- http://www.harborfreight.com/catalogsearch/result?q=pipe+notcher It uses a drill press mounted hole saw to cut the proper angle in the pipe. Using the notcher to cut each 21-foot length of pipe into three sections went well. They were then welded together so there was a leg front and back connected by the 30 degree angled support piece.
+The website- http://rredc.nrel.gov/solar/calculators/PVWATTS/version1/ makes it easy to determine the best angle for year-round production, which was 30 degrees for my location. The only tool that was not already owned by me and needed for this job was a low-cost pipe notcher, which I found on sale for $24 from- http://www.harborfreight.com/catalogsearch/result?q=pipe+notcher It uses a drill press mounted hole saw to cut the proper angle in the pipe. Using the notcher to cut each 21-foot length of pipe into three sections went well. They were then welded together so there was a leg front and back connected by the 30 degree angled support piece.
 {{:dsc03072.jpg|}}
 {{:dsc03163.jpg|}}
 {{:img8.jpg|}}
-{{:imga.jpg|}}
+{{:imga.jpg|}}\\
  My array is arranged in what is called a portrait configuration with the long sides of the panels going up and down, two rows of 10 each. This uses a shorter and less expensive special connection cable for the Enphase micro inverters, creating an array about 11 feet by 32 feet. There was no reason not to allow the array to follow the western slope of the ground.
-{{:dsc03162.jpg|}}
+{{:dsc03162.jpg|}}\\
 The three point hitch posthole auger on my tractor made it fairly easy to bore 10 holes in the ground about 2 1/2 feet deep. Welding a small horizontal section of steel onto each vertical post halfway down the hole added stability. Once the concrete is poured around this piece, it makes it impossible for the pole to slip up or down, and it increases its footprint.
@@ Line 55: / Line 55: @@
 Looking at the aluminum rails that PV panels are commonly mounted to, they seemed rather expensive and difficult to attach to the galvanized pipe. The ones I found online were about six dollars a foot plus shipping. That gets even more expensive because longer lengths need to be truck shipped. Instead, sold locally is inch and a half steel angle iron for a little over a dollar a foot. I picked up seven 20-foot lengths. They were cut and welded to make four 32-foot lengths, and then measured and drilled for the holes that the panels would bolt to before welding them to the uprights. I amazed myself when all the holes lined up!
-{{:dsc03142_1_.jpg|}}
+{{:dsc03142_1_.jpg|}}\\
-{{:dsc03164.jpg|}}
+Note the angle iron PV module support rails that are
+welded to the 1.5 inch pipe mounts.\\
+{{:dsc03164.jpg|}}\\
+Closer view of the PV module angle iron support rails.
+Note the stainless steel attachment of PV modules to the
+angle iron rails.\\
 ==== Array Side Wiring ====
@@ Line 62: / Line 70: @@
 After bolting the panels to the angle iron using stainless steel hardware, I mounted a single inverter behind each panel and simply plugged the panels into the inverters, and the inverters into the two connecting cables, which ran the length of the array. The Enphase inverters cannot have more than 17 in a string. So my system with 20 panels required two Enphase cables having special terminals at each dead-end, while the ends closest to the house went into a small breaker box- a combining load center, in solar speak. Each cable back feeds to a two pole 20-amp breaker. The power then gets combined before entering the conduit to the house.  Each panel and inverter got connected in series to each other with a number six solid copper ground wire. One end went to an eight-foot ground rod under the array and the other to the ground in the combiner box. The steel rack and rails are also tied to the ground system.
+{{:dsc03164.jpg|}}\\
+PV modules just plug into the micro inverters, and the
+micro inverters plug into the cable that daisy chains
+all the inverters together.  Note the terminator on the
+daisy chain cable as this is the last inverter in the row.\\
+{{:array_load_center.jpg|}}\\
+The load center at the PV array.
+The 240VAC daisy chain cable from each line
+of inverters gets a pair of 20 amp circuit breakers.
+The red and black wires coming off the two terminals at
+the top center is the feed to house.\\
+{{:dsc03142.jpg|}}\\
+This shows how the micro inverter daisy chain cables
+go to the PV array load center.
+Code requires the top of the conduit to be buried at least 18 inches below grade, and I had tree roots and rocks to deal with so I called in my neighbor with his trencher. The trencher cut through my phone wire to the house as well as the phone wire, power and temperature alarm wires from the house to the greenhouse, and my low-voltage path lighting wiring. I strongly suggest mapping where all your underground wires go, and calling the phone company before you dig.\\
+{{:trenching.jpg|}}\\
+Trenching for the wire from PV array to the house.\\
+==== House Side Wiring ====
+The TVA requires what is referred to as a line or supply side connection, which means that the solar feed connects to the grid at their usage meter. Many parts of the country use what is called a load side connection. This would require running the cables from the solar array into the house breaker panel. Line side was much easier for me. I had to buy a meter base for the solar supply meter, and a manual disconnect switch. The wiring from the array runs through the buried conduit into the solar supply meter; into the manual disconnect switch, and then over to the usage meter. All boxes were connected with grounds and to a ground rod.
+{{:dsc03140.jpg|}}
+{{:dsc03141.jpg|}}
+{{:dsc03166.jpg|}}
+{{:house_boxes.jpg|}}
+{{:house_cut_off.jpg|}}\\
+After mounting the boxes and pulling the wires through the conduit, then wiring up to the usage meter, my wiring job stopped temporarily as I couldn't get in the sealed meter base until TVA came to cut off the power. First I had to have the electrical inspector come out to do an inspection. All was well and I got the stickers and the go-ahead to connect.
+Two weeks later, I had an appointment to get connected - it was exciting to finally get to the point where my system could start producing solar electricity. TVA sent a technician with a meter to make sure that if the grid power goes off, my solar array will not back feed into the grid possibly endangering a lineman. This is a feature built into all grid tie inverters called anti-islanding. From what I have read, there has never been a case of an approved inverter putting power back into the grid when the grid is down. However, on smaller systems like mine, TVA requires $100,000 of liability insurance coverage, more on bigger systems.
+When the connection day finally came, it was cold, snowing, windy and completely overcast, yet the system put out enough power to successfully complete the test.
+{{:dsc03141.jpg|}}\\
+The PV array disconnect at the house.
+This completely disconnects the PV array from the utility power.\\
+{{:house_cut_off.jpg|}}\\
+Inside the PV array disconnect.\\
+With the Enphase Enlighten monitoring system, anyone can go to  My Enphase Web Page  and see current and historical production. As of this writing, it's only been connected for a month of short and often cloudy winter weather. Even so, it is averaging just a little under our usage. I feel confident the longer sunny days of summer will create more power than we are using.
+==== Costs ====
+This table breaks down the cost of the project:
+{{:2012-10-28_083503.jpg|}}