Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

jagec
jagec Solar Expert Posts: 157 ✭✭
Hello, everyone. Great forum you guys have here.

I've been looking into solar recently, and have done a fair bit of spec'ing, planning, and (most importantly) figuring out whether it's actually worth it in my case. Pbartko's excellent writeup on his DIY grid-tie installation helped me a lot. I hope to spawn some discussion and get some feedback on whether my system, and math, make sense.

First off, the math. The federal government, of course, has the vaunted 30% tax credit, which can actually be split over a few years. That's important to me, as I also have some painful losses in the stock market that I'm deducting, and so I don't pay enough taxes to take full advantage of the credit in a single year.
The state of Maryland has heavily trimmed back the payout on its solar grant program (shakes fist), but they are still covering $.50/DC watt for PV. Another important consideration (in fact, the only reason that solar is even close to economically viable in Maryland) is the SREC market. Right now a single SREC in Maryland (1000 kWh) is selling for slightly above $300. And I'm paying about $0.13 /kWh for power currently.

The system that I have worked out is a single string of 12 * 230W Canadian Solar panels coupled to a Sunny Boy 3000W grid-tie inverter. I will be mounting them at a 34-degree angle on my roof using the Unirac system. My roof has a slight (5 degree) incline to the east, but more importantly it doesn't get shade at any point during the day. I'll do all of the grunt work, install the panels, run cable, etc, and then hire an electrician to actually connect everything together.

I'm planning on buying panels, inverter, AC disconnect, AC and DC lightning arrestors, and a 20A fuse+holder from http://sunelec.com/, and the Unirac hardware from http://www.civicsolar.com/. Anyone used either of these guys? It's pretty hard to tell the difference between the major and minor solar players online.

My costs are projected to be:
$5,496 panels
$1,844 inverter, lightning arrestors, etc.
$1,081 mounting hardware
$8,421 PLUS shipping fees, permitting fees, electrician fees to hook system up, conduit, wires, unforeseen extras...

I budgeted an extra $2,000 for these contingencies, for a grand total of $10,421 before all incentives.

At 2.76 kW, Maryland should give me a grant of $1,380, and the federal tax credit (figured AFTER the grant) will bring the total cost to $6,329.

Based on PV-watts, this system should produce 3,450 kWh a year, which equals about $1,070 in SRECs and $448 in energy bill savings, for a yearly "income" of $1,518. This will change over time, as the price of SRECs will decrease, and the cost of energy will increase.

With these numbers, the payback period is 4.17 years, and the ROI, which some people* think is the "better number", is 24%. Not too bad.

Here are a couple quick-and-dirty mockups of what it will look like installed. (Edited post, mockups have 1 additional rack of 3 panels than my current plan)

Anything obviously wrong?
solaroverview.png

Detail of array mounting. The idea is to use 4 "racks" of 3 panels each. In theory those legs are adjustable, but I have a STRONG suspicion that not only are they not intended for regular seasonal adjustments, but that it would be way too much of a PITA for me to loosen 4 legs while supporting 120 lbs of panels, adjust the angle, and tighten everything back up, four times over, four times a year. So they won't be moved.
mountdetail.png

Obviously the L-brackets aren't shown, but they would bolt directly into the framing. Since my roof is a solid membrane, I'm open to suggestions on the best way of waterproofing the installation. I don't want to be one of those people who just slathers sealant all around, but, I don't have a shingle roof, so it's not as easy for me to flash small sections. What would make the most sense, EPDM flashing tape? Roofing tar? Or just suck it up and use the highest-quality roofing sealant I can find?

Oh, and my *edited* wiring diagram looks like this:
wiringdiagram12panels.gif
Thanks again to pbartko for posting his well-diagrammed system.

I've also got a couple of stupid questions. First, the combiner box: Is there a reason that everyone puts it on the roof? Is it just because there's plenty of space, or am I missing an incredibly-obvious reason that you don't want your leads to go too far before hitting the lightning arrestor?
Next, of course I'll be grounding all of my panels and rails, but given the angle of my rails, should I be running copper all the way to the top of each rail and making the grounding connection there, or can I just ground them all at the base? I know that aluminum's only around half as conductive as copper, but the rails will still be the most-conductive path to ground by far if lightning strikes at the top. And it would save on a lot of copper to ground at the base.

Any other comments, concerns, or questions, please feel free to bring 'em up!

*who coincidentally work in PV sales

//EDIT: Oops, I already caught my first problem. I'd originally spec'd this system with Evergreen panels, but then moved to Canadian Solar for mounting reasons (higher wattage=less mounting hardware, with bulk price breaks it made sense). Well, those have a much higher Voc. With my intended 15 panels, I'm exceeding my inverter's max voltage.

I'm now scaling back the system to 12 panels. Updated calculations accordingly.

Comments

  • Jburgess
    Jburgess Solar Expert Posts: 130 ✭✭✭
    Re: Plan and financial analysis of 2.76kW grid-tie in Maryland--feedback appreciated

    Some initial thoughts:

    As long as the wire from the array to the disconnect is metal conduit where it is inside the structure, the PV combiner is not needed.

    The ground wire to the array only needs to #10. #6 would be needed in using micro inverters.

    The AC disconnect does not need to be fused if the breaker is sized to the inverter.

    If your sketch says 60A breaker, then that’s to big for your 200 amp service. 40A max.

    As for civic solar, My only experience with them is I placed an order on the phone, then they called me back and cancelled it and refunded my card, saying the decided the dollar amount was too small. (about $200). They wanted a $500 minimum.

    Your local codes may be different.
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    Thanks, Jburgess.

    I ran things through Sunny Design again, and the maximum AC that the system should produce (@240V) is only 12.7A. I'm downsizing the breaker to 20A, eliminating the combiner box, and downsizing the grounding wire per your recommendations.
    I also revised the layout slightly. Turns out that it's cheaper for me to use the "low-profile" Unirac system, since I'll (barely) be able to fit all 12 panels on the roof in a row, as two sections of 240" each. Any more panels would absolutely require the high-profile mounting.

    Updated sketch:
    12overview.png

    I also have a detail of the mounting setup, with 6" standoffs and Unirac "low-profile" legs, a view of the back from the roof of the next-door building, and a closer look at the shading from the AC unit.. Those lines are drawn between the sun and the corner of the AC unit, on the winter solstice. Starting at the right we have 9AM, 10, 11, 12, 2PM, and 4PM. There is a little bit of shading at the edges of the day, but it's fairly mild, and isn't a problem at all most of the year.

    Updated wiring diagram. I guess I'll need to get two boxes for the two lightning arrestors...when it comes to AC and DC wiring, one must remember not to cross the streams.
    canadian12panelwiringdi.jpg

    I plan to march on down to the permitting office next week to make sure that my plans meet with their approval, and I'll probably be able to order everything soon after that. I'm hoping I can get everything on the roof by the end of February, assuming that we don't have any weather problems or other delays. Of course, I wouldn't be too surprised if I can't make that deadline...
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    Couple of questions...
    • Snow? Do you need to clear panels in winter?
    • Roof? Panels will hopefully last 25-40 years or so--Don't want to pull the system at 10 years for a new roof.
    • Access? Can you get on the roof to wash the panels? Do you have water faucet plumbed in?
    • Planning? City going to OK the system or will you need to mitigate some of the visual impact?
    • Stack at rear roof of building? Shading? Shading can really reduce the output of your array--Anything you can do to reduce shading (at least 9am-3pm all year long) would be helpful.
    • Do you have a possibility of lightning strikes? If so, I would not bring the panel frame ground wire inside the building--I would run it down the outside the exterior of the building to a Ground Rod/Grounding system at the edge of the building. Much of this is controlled by NEC and your local building inspector... What is legal vs what is correct is not always clear.
    A couple threads about Lightning:

    Off Grid Grounding Technique?
    Another Question, this time about Lightning

    Note, the above are discussions, not a do A, B, and C--and you will be "safe". There probably is no such thing with lightning. Several different techniques are discussed--and a few of those posters even have experience with lightning. :cool:

    And our host's consolidated FAQ page:

    www.windsun.com
    Lightning Protection for PV Systems

    From other past posts here, Windsun (admin/owner of NAWS), he said that most of lighting induced failures he saw were in the Inverters' AC output section.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated
    BB. wrote: »
    Couple of questions...
    • Snow? Do you need to clear panels in winter?
    • Roof? Panels will hopefully last 25-40 years or so--Don't want to pull the system at 10 years for a new roof.
    • Access? Can you get on the roof to wash the panels? Do you have water faucet plumbed in?
    • Planning? City going to OK the system or will you need to mitigate some of the visual impact?
    • Stack at rear roof of building? Shading? Shading can really reduce the output of your array--Anything you can do to reduce shading (at least 9am-3pm all year long) would be helpful.
    • Do you have a possibility of lightning strikes? If so, I would not bring the panel frame ground wire inside the building--I would run it down the outside the exterior of the building to a Ground Rod/Grounding system at the edge of the building. Much of this is controlled by NEC and your local building inspector... What is legal vs what is correct is not always clear.

    To address your questions:
    Snow isn't huge here, but it does happen. It's not too hard for me to get onto the roof, as I have a 2nd-floor deck at the back of the house which gets me a short ladder-climb away. That's also how I'll be doing the installation. For water, I'd probably toss the hose down the front of the house and connect it there when I needed to hose down the panels.

    The roof is about 5 years old (the house was originally built in 1910, but it was completely gutted and renovated around '05. I bought it mid-2010). And it's torch-down. As long as I'm careful to seal my standoffs properly, it should be OK for a while.

    As far as planning goes, that's something that I'm going to be discussing with the city well before I start buying panels. On the one hand, Baltimore is full of boarded-up houses which are worse eyesores than a solar array could ever be. On the other hand, if I want to get the Maryland solar grant, I have to run the plans by the Maryland Historical Trust and get their approval. And on the gripping hand, with the low-profile mounts and surrounding buildings, it actually won't be all that noticeable from the street level.

    Yes, I'm aware of the shading problem, and I actually contemplated moving the AC unit. But, with the panels mounted as far north on my roof as they can go, shading is only an issue on the bottom tenth of one panel, before 10AM, and then only when the sun is at its absolute lowest. I'm not content with it, because I want absolute perfection, but any further mitigation would cost far more than the energy gained.

    Grounding...luckily there's not a LOT of lightning here, but, like you said, I'm pretty much limited to what local code says I should do. Also, it's all concrete, and so there's no place for me to add my own grounding rod. I'll be running my wires down the side of the house right next to the lines from the utility, and if the inspector gives me his blessing, I could connect my ground line to the nice, big, fat one that they're running to me.
  • Jburgess
    Jburgess Solar Expert Posts: 130 ✭✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    The sunny boy disconnect has DC fuses in it making the fuse you show redundant.

    Here’s a sample 3 line with a sunny boy inverter, it also shows the neutral and ground wires. I find it helps to have a sample that is close.

    3line.jpg

    Your service panel looks more this:

    3line1.jpg

    You had asked about flashings. These are EDPM pipe flashings that could be put over your standoffs.

    PICT0197.jpg

    I get an engineer to review my drawings to make sure the array won't blow away in the wind or burn the house down. Piece of mind for a few hundred dollars.
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    I haven't had a lot of time to deal with the project lately, but I did manage to take a couple of renderings down to the permitting office and let them do a "smell test" over whether the plans looked OK. What I'm worried about is the fire code. My roof is fairly new, and overbuilt (2x10s, 16" OC, holding up a torch-down roof and nothing else, in an area with a design snow load of 30 psf). But I wasn't sure if they were going to hassle me about setbacks from the edge (particularly given that it's a townhouse and there's another roof RIGHT THERE), or space between arrays in an alternate "high power" design that I've been tinkering with.

    Since they didn't shoot down my 20-panel design right away, I'm going to sketch it out and submit it next week, along with some calculations. If they shoot that down, I'm going with the 12-panel design which is in the first post of the thread.

    Here are descriptions and sketches, to keep you all entertained:

    12-panel setup-as described in the original post.
    12 Canadian Solar 230W panels coupled to a SB3000 inverter:
    isometric.png
    2.76 kW DC rating, panels at a 34 degree angle, some very minor early-morning shading in the winter, but none at all any other time of year.

    I was thinking about ways to scale up the system, but those Canadian Solar panels have a ridiculously high Voc which made them difficult to design around. Finding a number that would provide enough power to feed the grid at 70 C, but not overvolt the inverter at Voc, was difficult with my roof size. Finally I went with Evergreen B panels. The payback period isn't quite as good, but it's still roughly 4 years after incentives.

    20-panel setup:
    20 Evergreen 205W panels coupled to a SB4000 inverter.
    4.1 kW DC rating
    Panels at 15 degree angle, which is sub-optimal but the only way to eliminate shading in the winter months. The plus side is that it would make the panels somewhat less visible from the street level.
    PV-watts says 4819 kWh a year.
    20paneliso.png

    However, depending on local codes, this solution *might* not leave enough space on the roof for fire crew access. The front desk at the permitting office didn't mention anything when they looked at it, but I didn't really get the impression that they knew what they were talking about, so I'll have to wait until someone a bit higher-up gets a look.

    This solution does solve the shading problem, as well as providing more power. The closer to the west edge of the roof I can get, the better with that AC unit sitting on the east side, but I get full sun from 9AM until sunset, 365 days a year, with this setup. But I'm not sure how far up I can nudge those panels before they will no longer be "in code".
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    Life has been keeping me busy, and so I haven't had a chance to deal with the solar system for a while.

    But today I went in to the permitting office with my latest plans, and had them look everything over. To be honest I was probably a lot more worried about the minutiae of measurements and clearances than they were. It became evident to me that you can't count on code enforcement to save you from yourself.

    I also updated my solar grant application. I can't actually put panels on the roof until I get approved by the Maryland Historical Trust, but since my building isn't historical or near anything historical, I really doubt that is going to be a problem. So, I'm going to start placing my equipment orders while I wait.

    The final design is 2 strings of 11 Canadian Solar 230W "B" panels, connected to a SB4000 inverter. Ideally it would be 2 strings of 12, but I simply don't have the roof space without my AC unit causing serious shading issues. When I first bought my house, I thought that it was great that the AC unit was on the roof and out of the way, but now I wish that they'd put it in the backyard instead!

    I'm at the point where inches count when it comes to winter-solstice shading, and I probably won't know the EXACT angle that I'll use until I'm actually mounting the panels, but it will be somewhere between 15-20 degrees. I would prefer 34 degrees (the year-round optimum for my location...I'm realistic enough to know that I won't be adjusting 1000 lbs of panels twice a year), but that isn't possible without some truly giant standoffs on the back row, or a high-profile mount that costs more than the power gained.

    I console myself by saying that this less-efficient angle is what enabled me to drop down to the SB4000 without throwing away extra power during high-production days.

    My panels and inverter (minus wiring and conduit) are going to cost $11,900 shipped, and the mounting system is another ~$1700. Add $1400 for wire, conduit, permitting fees, and hiring an electrician to pull the permits and check my wiring, and I'm in for $15k. Honestly, that's shockingly low, and I keep waiting for the other shoe to drop.

    After incentives, this drops to $7,970 (I was doing the math wrong: the federal 30% incentive comes first, and then the state grant). Not. Too. Shabby. Mind you, I'm omitting taxes paid on the state grant, and taxes paid on the sale of SRECs, but neither of those numbers is very high.

    With a 20-degree installation, PV-watts says I'll make 6126 kWh a year, for an income of $2,695 per year assuming that SRECs sell for $310. I'm giving up only $69/yr compared to the optimal tilt of 34 degrees. I give up another $58 at 15 degrees.

    That's about a 3-year payback period. Thank you, taxpayers. Of course, the price of SRECs will affect this heavily, but even under a worst-case scenario (the SREC program is eliminated, or I can't get certified after the installation) it's still a 10-year payback period.
  • SolarLurker
    SolarLurker Solar Expert Posts: 122 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    Have you Given thought to using a more efficient panel, such as the sanyo or sunpower? The cost is a little more however you could get a bigger system.

    You could sell your srecs upfront to recoup your money faster.
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated
    Have you Given thought to using a more efficient panel, such as the sanyo or sunpower? The cost is a little more however you could get a bigger system.

    You could sell your srecs upfront to recoup your money faster.

    My understanding was that no one in Maryland was buying SRECs upfront, but I could be wrong. I also have a feeling that, since I'm going to be installing the system myself, any utility planning on buying from me would be questioning whether I was going to follow through.

    As it stands, this system should produce about as much power as I use every year. I could sell more SRECs with a bigger system, true, but the installed cost per watt would go way up and I'd be giving the power company free kWhs. Both those factors really screw up the payback period.

    The reason why I chose the panels that I did was that they were insanely cheap for poly-Si.

    It's not all gravy, mind you: On very hot days, there's a chance that my 11-panel strings won't have enough voltage to feed the grid (Sunny Design thinks that I'll get a yield loss at a panel temperature of 80C, which is, admittedly, high) If I were to choose Sanyo HIT 220W panels instead (for example), I'd be able to fit 12 or even 13 panels per string on the roof and completely eliminate this problem.
  • rollandelliott
    rollandelliott Solar Expert Posts: 834 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    http://www.civicsolar.com/product/unirac-302017?qty=1

    $460 for 64 feet of railing, that is $7.18 a FOOT!

    You can get T track aluminum on Ebay for $2.75 a foot 1x1 in size.

    If you want the really heavy duty T slot 1.5x1.5" here is some discounted stuff for $3.25 a foot.

    check out there youtube channel and web site. If you are a DIY'er than this is like adult legos. www.8020.net

    Never realized a mounting system could cost that much.

    You can get a spiral drilling tap from mcmastercarr and aluminum tapping fluid as well from there. It is easy to cut witha $40 aluminum circular saw blade off ebay.
  • pbartko
    pbartko Solar Expert Posts: 37 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated
    As it stands, this system should produce about as much power as I use every year. I could sell more SRECs with a bigger system, true, but the installed cost per watt would go way up and I'd be giving the power company free kWhs. Both those factors really screw up the payback period.

    jagec,

    Maryland changed their laws late last year. You can now produce unlimited electricity and sell all of it back to BGE. The utilities now must accept all of you electricity and pay you for it. No longer are you limited by being credited only the amount you consume. You can get a check in the mail from your utility!

    Pete B
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated
    http://www.civicsolar.com/product/unirac-302017?qty=1

    $460 for 64 feet of railing, that is $7.18 a FOOT!

    You can get T track aluminum on Ebay for $2.75 a foot 1x1 in size.

    If you want the really heavy duty T slot 1.5x1.5" here is some discounted stuff for $3.25 a foot.

    check out there youtube channel and web site. If you are a DIY'er than this is like adult legos. www.8020.net

    Never realized a mounting system could cost that much.

    You can get a spiral drilling tap from mcmastercarr and aluminum tapping fluid as well from there. It is easy to cut witha $40 aluminum circular saw blade off ebay.

    I have since changed my mounting system to Ironridge, which was quite a bit cheaper. The rails themselves were $3.36 a foot, and there were plenty of engineering calculations for snow load and wind load, Professional Engineer-stamped certification letters, etc. on their website. If I had more time on my hands, I might well have chosen to do my own mounting with a more "generic" extruded aluminum product, but I'd have to do a lot of engineering work to get it past the permitting office. If I was building an off-grid system I would certainly have fabbed up my own.

    Once all of my equipment shows up and the installation commences, I'm probably going to start a new thread. My current scheme has drifted pretty far away from the initial designs!

    pbartko, that's the best news I've heard all day.
  • rollandelliott
    rollandelliott Solar Expert Posts: 834 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated

    80/20 can provide you with all sorts of stress and load numbers as well.

    I would recommend using lock tite to make sure the bolts stay put.
  • jagec
    jagec Solar Expert Posts: 157 ✭✭
    Re: Plan and financial analysis of 3.45kW grid-tie in Maryland--feedback appreciated
    80/20 can provide you with all sorts of stress and load numbers as well.

    I would recommend using lock tite to make sure the bolts stay put.

    They can give me numbers on the beams themselves and the recommended connection method between them, yes, but I'd still have to do a LOT of calculations to get point loads and uplift for a given design, angle, wind speed, snow load, and panel size and weight. Furthermore, my AHJ might want to have my own calculations confirmed by a Professional Engineer prior to approval. This costs a lot of time, effort, and frustration, compared to using an already designed and PE-certified system, with nice little tables to show the permitting office that it will hold up under local weather conditions.

    Yes, I plan on using Loctite.