The Tale of a PV Installation: Part 1

Back in July,I received a phone call from a couple interested in PV net metering. They were at a local HVAC business called Total Tech in Frankford, WV. As it turned out, they were finishing business concerning a geothermal heatpump system just installed at their new house. Wayne at Total Tech referred us as solar installers. I said sure, come on up, its only 9 miles up the road.
When they arrived we introduced ourselves and I began to query about their needs, desires and home site. This info gives me a better handle on what to present as options for system design.
The couple's names are Glenn Freeman and Susan Mitchell and they are in the process of moving to their new home on Muddy Creek Mountain in Greenbrier County, WV. They are retirees from New Jersey and was just a matter of a couple weeks before they say goodbye to Jersey and hello to West Virginia. It must have been a brave decision to make such a commitment. I think they will adjust well to the local mixed culture and what the mountains have to offer.
Anyhow, we sat in front of my computer and scanned Google Earth for their home site. The aerial photo was dated but we did find the clearing in the woods where the house sits now. I was a bit concerned about the ring of trees shading the array, so I suggested an on-site assessment.
Once I arrived, I was quite impressed with the style, quality, but appropriate size of the dwelling. Previously they had a solar consultant recommend siting, orientation, and roof pitch for the architect and builder. With the geothermal heatpump delivering 4kw of heat for every kw put in, they reduced their demand so that the PV array would provide a substantial contribution percentage-wise. We came to the conclusion that a 3600 watt array was a good starting point while having enough capacity built in so they could easily add 1800 or another 3600 watts over time. After taking some measurements for solar aperture, looking at wire run options, and roofing material, I told them I would be a couple days before presenting the results.
Back at the office, I plugged in the data into a program I assembled to provide performance, cost, tax incentives, and potential SREC value. SRECs stand for Solar Renewable Energy Credits and can be auctioned off to utilities in surrounding states. These utilities will pay up to $0.33 per kilowatt hour you produce instead of installing RE mandated content in their own grid.
Once satisfied with the general design and results, I emailed the docs to them. They soon called and said they were interested. Once I received an engineering deposit, I proceeded to create a detailed hard quote, wiring diagram, and agreement. I went to their place and discussed the details and answered questions. Upon signing the agreement, I requested a deposit and gave them the document to apply for net metering with Allegheny Power.
As soon as I got home the deal went into high gear. I researched availability of the parts through my various distributors to get the best prices and shipping costs. Once decided, I pulled the trigger and made the orders. It didn't take long for the smaller parts to arrive via UPS. The modules had to come by truck, so that took longer.
Once I had the mounting system in hand, which is a clamping system for standing seam roofs, I began to doubt whether they would work. So I drove back to the site and tried one on the seam. It didn't fit...crap. Instead of parallel, vertical seams, these had a truncated triangle profile. No amount of crushing or modification would make for a strong connection. Plan B.
I immediately ordered some UniRac SolarMount rails and clamps. These had to come by truck, so I put the hot rush on it. Meanwhile, I got a RMA to return the other mount system (at a 15% restock charge). The rails got here in a few days, no time lost. This was all my fault for not inspecting the seams closer. I figured standing seam was standing seam. These are the kind of gotchas that happen when you fail to focus on every detail even if you think you know what's going on.
By the time I had the rails and microinverters, it was blazing record hot for weeks around here. Add to that the brown tin roof and the installation windows were mornings only.
In a moment of inspiration, I thought why not populate the rails with the microinverters and roof anchors in my shop? That would prevent a lot of roof time precisely positioning the components on 130 degree tin. It took half a day on the workbench to mount the parts.
So we scheduled a rail install day and got them all up in a few hours. And they where pre-wired and pre-grounded boot!
In the meantime Joe, our licensed electrician, began to run wires in the house and mounted a breaker/disconnect box on an outside wall. He took the wires though the attic crawl space ready to punch through the roof once the modules are mounted.
My wife, Tenley and I delivered the modules on a Saturday since our truck had to go to a garage for a week. Susan and Glenn helped us lift them over the porch rail and made a neat stack. We strapped them together and covered them in case some rogue wind came through before the install.
I kept watching the NOAA weather forecasts online looking for a day with some heat relief. No, just relentless heat as far as you could see. I decided to break the module install into two mornings. Once we got there, there was dew on the roof (can't win for losin') After progressively toweling down the tin, we mounted the two grab ropes to assist our traction.
The first few modules established the flow required to be efficient. Like; where someone was at a given moment, what small parts in the belt bag, tools handy, knowing where the grab rope was, and getting the position of the first module straight and square. Our boots-on-the-roof guys were Joe the electrician, and Michael our electronics tech and fabricator. I was on the ground trying to think ahead about any glitches and providing parts, modules, and tools.
This is the end of part one, part two will be posted before too long.

Proof of concept, Labor of love

When we first installed our evacuated tube solar hot water system, it got me thinking about the heat we had to dump because we had to keep the water tank below 170 degrees F. The other piece of intriguing information is that these tubes could attain a temperature of 375+ degrees if the glycol solution was not flowing.
As we were enjoying the abundance of solar domestic hot water available, my mind kept thinking about other potentials for this high temperature fluid. At first, I thought about steam production to run a steam engine and in turn a generator. I have done extensive prototyping of variations of steam schemes that would make a carefree power source. I never even got close. Steam power requires constant fiddling and water freezes at 32 F., not a practical source of continuous power.
A few months went by and I remembered the Stirling hot air engine.(invented in 1816) No water, no freezing, little maintenance. Not only that, but efficiencies well above steam engines. (up to 40%) My search commenced for a model Stirling motor at modest cost.
Being a machinist and needing some mental diversion, I found a kit that requires machining to assemble from Grizzly. http://grizzly.com/products/searchresults.aspx?q=stirling&new=1
The vertical kit cost 77 bucks, the premachined kit was $164. I opted for the more challenging raw kit knowing I had to convert the metric drawings and some of the material stock to decimal inches. After 3 weeks of part time machining, I got it to a testing level. I put an alcohol burner under the hot end and waited. A little nudge on the flywheel and she started spinning. WOW!
So, I went through a procedure of timing and lubricating until I achieved a speed of 1770 RPM. Wow again. I looked at the size of the flame to the output and thought, this ain't bad.
Once I had a functional engine, I thought about how to power this thing from the sun. Previously, I thought about a huge array of tubes circulating mineral oil at high temperatures to power a more powerful engine. This engine was not in that league, so a more doable solution was needed.
I had Sunda 7 foot evacuated tube with a heatpipe delivery system. The business end can approach 400 degrees when exposed to the sun. So I made an aluminum connector (heat conductor) between the output of the tube to the input of the engine. This will conduct the gathered heat into the engine. (with the aid of a heat conductive paste) I finished up at the end of the day and frantically set up the rig on a board looking at the diminishing sun.
It was already 5:30 pm and the sun was getting weak. I waited for a few minutes and spun the flywheel. It spun eagerly but eventually stopped. Again and again I nudged it hoping for something, anything. After 5 minutes or so, I nudged it and viola, it kept running. As time went by it went faster and faster. Not as fast as the alcohol burner but an impressive seven hundred-some RPM....... Success, Proof of Concept.
The video captures this moment. The video does not capture my elation. It didn't matter that it was no barnstormer, it ran from the heat of the sun. All this work leaves me with a bit of fertilizer for my brain. What I made couldn't power even a laptop, maybe a cell phone. Got that picture in your mind? How convenient.
From here, I'll think about how to upscale this deal. It might have some application, but I'm not betting on it. It will become whatever, this is how innovations evolve.
Before you get carried away with the potential, let me list the ups and downs of this apparatus:
ADVANTAGES:
> Any substantial non-petrol heat source; solar, biomass, wood, etc
> No freezing of water
> No boiler explosions (could be a bummer)
> Relatively few parts, limited wear.
> Quiet
> Clean, external combustion process.
DISADVANTAGES:
> Low power to size ratio (large engine/heavy weight to output ratio)
> Lack of off-the-shelf parts and engines ($+)
> Low torque, unless designed to make it. (at the expense of HP)
> Continuous power/speed only, hard to throttle.
> Needs 4 cylinders to start automatically. ($+, complex)
We're not going to let this stuff stop us, are we????

Progress report, net metering

The above photograph is our meter taken on August 20, 2010. This meter was installed on June 8th of this year. When it was first installed it read "00000". Now it indicates we have 182kwh in credit with Allegheny Power. Considering for the past 40 days we have been into a lot of cloudy and partly cloudy weather, we're doing OK. This number shows the surplus power we generated, not the PV total output.
Through the Enphase Enlighten website which displays and archives our production data, I have prepared a report comparing our actual total PV output to our program we use to predict performance for potential customers who have had a site assessment done by us.
I have a field in the software where I can adjust system efficiency to help make the predictions fit the real world. Normally I use 90% system efficiency for potential customers. This predicts a slightly lower number than what they may see for real. I would rather under promise and over deliver. Our system is calibrated at a brutal 95% efficiency. This is the same efficiency as the Enphase microinverters. One must allow for line losses too, however the Mage PV modules have a higher tolerance for its power rating method which tends to cancel out the other losses.
So, how did we do so far?
From June 8th through to the end of that month, we produced 390kwh. If you adjust that for a full 30 days, you would have 585kwh. We predicted 507kwh, and overshot the prediction by 78kwh. yeah!
July is our only full month so far. We produced 545kwh, we predicted 497kwh. Again overshot by 48kwh. yeah!
So far in August, up to the 19th, we produced 288kwh. If you extrapolate out to 31 days, it would be 446kwh. Our prediction is 497kwh (same as July). Here we undershot the prediction by 51kwh. boo! This was mostly due to persistent cloudy weather patterns. If we have a week of more clear weather, we will easily surpass the prediction.
To date, we only have a small data set to work with. It would take two years of data to confirm everything is true. I feel confident that both the system and the predictions are within tolerance. These two items are quite important to a customer considering making the plunge into PV.