Panels on both sides of roof?

[crashintoty]

My plan was to put 10 to 12 panels (305 watts each) on the south facing slope of my detached garage roof (opposite slope in pictures), but my wife asked why not put panels on the other side of the roof too for 20 to 24 panels (near slope in pictures)? I never considered that because that slope faces north and I assumed it's always shaded. But looking at the north slope in the morning I see the sun shines on it (first picture, 10:30-ish AM) and looking at it in the afternoon I see the sun still shines on it (second picture, 3:30-ish PM). I imagine the near/north slope wouldn't harvest as much sun as the opposite/south slope, but would it harvest even a respectable amount to make it worth installing on that slope?

Seattle, WA


Morning (10:30-ish AM)

IMG_20200309_103208.jpg


Afternoon (3:30-ish PM)

IMG_20200309_153006.jpg

[Ampster]

The best answer is to model those two scenerios using PV Watts. You will get some data that you can run against your cost of power to see if it is worth it to you.
Seattle is a higher latitude and in the winter that northern face will most likely not see much usable sunshine. What sunshine it will see will be at such an oblique angle that the Watts per square foot of panels will be very low.
if I remember from earlier threads, your cost of power is pretty low, but the uncertainty of the shade makes that an even riskier investment than the alternatives.

Have you looked at the payback of professionally installed panels on your two story main roof? If it were me, I would run the math to see what the difference is between a low cost DIY project with some risk compared to relative predictability of a higher cost professionally installed project on your home.

if you are looking for a DIY project, post a plot plan and maybe there are some alternatives that could make this a productive science project.

[bob-n]

Ampster gives great advice. The PVWatts website is very helpful in modeling this.

Although conventional wisdom is that North is no sun, your roof is low angle, so when the sun is high, there is light on both sides of the roof. It may work out well for you. PVWatts has an input for roof angle. It also accounts for a rough estimate of seasonal cloud cover. If you can think of an issue, they can model it. It isn't perfect, but it is very good.

Another thought: You could do the South roof first, see how it works out, and then add the North roof later if you want more. You don't need to do everything at once. If you think ahead a bit, it is easy to design a system that can be expanded.

Are you willing to take down that East side tree? Unfortunately, solar panels don't like shade, so that tree, especially when leafed out, will not be your friend in the morning. Even the East power pole will give you a tad bit of shade, which will subtract a little power.

As for DIY, three or more healthy, able adults can take on that job (four or five would be even more comfortable, but not necessary). It is comparatively easy, because you have a fairly flat, low roof. Some systems use rails on mounts, and put the panels on rails. To install a system like that, two people can comfortably install the rails on day #1, but when it comes to putting up the panels on day #2, you'll appreciate having more than two people, You won't need expensive tools or anything special, just a good cordless drill and socket set. Expect to save 50% of the cost by DIY, and get to break even in half the time.

It just depends on whether you have lots of money or lots of friends.

[crashintoty]

The best answer is to model those two scenerios using PV Watts. You will get some data that you can run against your cost of power to see if it is worth it to you.
Seattle is a higher latitude and in the winter that northern face will most likely not see much usable sunshine. What sunshine it will see will be at such an oblique angle that the Watts per square foot of panels will be very low.
if I remember from earlier threads, your cost of power is pretty low, but the uncertainty of the shade makes that an even riskier investment than the alternatives.

Have you looked at the payback of professionally installed panels on your two story main roof? If it were me, I would run the math to see what the difference is between a low cost DIY project with some risk compared to relative predictability of a higher cost professionally installed project on your home.

if you are looking for a DIY project, post a plot plan and maybe there are some alternatives that could make this a productive science project.

Oh, shade is a certainty - the south facing slope starts shading at 3:00-ish (well, 4:00-ish now with daylight savings time) and this north facing roof is just now starting to shade (5:30-ish, roughly 90 minutes later than the other slope and looks it won't be anywhere near as shaded).

As for a professional installation on the main house roof (2 story, un-shaded building), I imagine that's going to cost $25,000 to $30,000 for 10 panels ($8.20 to $9.84 per watt). I'm guessing I can't get more than 10 panels on this main roof because of my two skylights. I estimate my DIY cost for 12 panels on my garage roof to be about 90% cheaper at $2,880 ($0.79 per watt). If I add 12 panels on the other slope (and if my source still has this many left) then it'll add roughly $1,680, so $4,560 total ($0.63 per watt).

PV Watts estimates 4,053 kWh per year with the south slope panels and 1,984 kWh per year for the north slope panels, but that latter number is without tilting the panels. I did a test with a cheap 3 watt 12v solar panel and got 13.6 volts with it laying flat on the roof and then got the full voltage of 14.9 when I tilt it roughly 45 degrees off the roof. So I assume if I tilt the actual panels on the north slope I may actually get a little bit more kWh than the south slope since it gets at least 85% less shade in the afternoon. So with an optimistic combined estimate of 8,000+ kWh per year of solar and an average consumption of just under 20,000 kWh per year, we're looking at a ROI of 2.5 years or less (without factoring in any net metering benefits stuff yet). I guess that's good....?


P.S. - That 20,000 kWh per year consumption will actually be decreasing. We recently bought a Emporia energy monitor and we've already discovered some major energy hogs and practices. We're in the process of eliminating those hogs and practices and have shed about 20% so far, so I guess our new consumption number is just under 16,000 kWh per year. There's more to eliminate which may get us down to 12,000 kWh per year which would be amazing (then our ROI would be 1.5 years or less).

[Ampster]

I had momentarily forgot that other thread where we discussed PV Watts ad nauseam. I revisited it and now understand why you didnt need to bump it back to the top. I had drafted a response, including an apology for my part in the hijack but never posted it.

My additional advice is to initially forget about batteries because of the cost. Grid Tie can be the science experiment with the grid as a very low cost battery. Perhaps go with micro inverters on the South garage and spend the money on a micro inverter monitoring system so you can see the impact of shadows real time. A SolarEdge system with power optimizers will also give you similar data. The hardest part of a DIY system is the building permit and application to the utility for the permission to sell back to the grid.

[Ampster]

........

As for a professional installation on the main house roof (2 story, un-shaded building), I imagine that's going to cost $25,000 to $30,000 for 10 panels ...... I estimate my DIY cost for 12 panels on my garage roof to be about 90% cheaper at $2,880 ($0.79 per watt). If I add 12 panels on the other slope (and if my source still has this many left) then it'll add roughly $1,680, so $4,560 total ($0.63 per watt).

Thanks for the update on the roof of your main house. I understand why you wouldn't want to consider that.

PV Watts estimates 4,053 kWh per year with the south slope panels and 1,984 kWh per year for the north slope panels, but that latter number is without tilting the panels. I did a test with a cheap 3 watt 12v solar panel and got 13.6 volts with it laying flat on the roof and then got the full voltage of 14.9 when I tilt it roughly 45 degrees off the roof. So I assume if I tilt the actual panels on the north slope I may actually get a little bit more kWh than the south slope since it gets at least 85% less shade in the afternoon. So with an optimistic combined estimate of 8,000+ kWh per year of solar and an average consumption of just under 20,000 kWh per year, we're looking at a ROI of 2.5 years or less (without factoring in any net metering benefits stuff yet). I guess that's good....?

Yes, that is good. 45 degrees is close to your latitude and in two weeks we will be at the equinox, which confirms your experiment.

P.S. - That 20,000 kWh per year consumption will actually be decreasing. We recently bought a Emporia energy monitor and we've already discovered some major energy hogs and practices. We're in the process of eliminating those hogs and practices and have shed about 20% so far, so I guess our new consumption number is just under 16,000 kWh per year. There's more to eliminate which may get us down to 12,000 kWh per year which would be amazing (then our ROI would be 1.5 years or less).

The best return on investment will be energy conservation.

If the north facing roof proves too challenging, I still have some DIY ideas for more south facing panel installations.

[bcroe]

Here at 42 deg Lat the sun actually rises and sets slightly to the north in summer. But the
energy to be had from those times is small. Bruce Roe

[J.P.M.]

1,984 kWh per year for the north slope panels, but that latter number is without tilting the panels.

If I correctly understand what you write, why "without tilting the panels" ?

FWIW, north facing 20 deg. slope w/PVWatts for Seatle gets ~ 800 kWh/yr. per installed STC kW but that's w/out any shade. That tree shading will probably penalize you more and in ways you may not fully understand. Shading from that power pole off to the SE won't do you any good either, or for either garage roof although it'll be less of a problem than the tree.

If cost effectiveness is any consideration, vendor installed/priced PV on the north garage roof will knock that out of contention, although I'm not sure where you got the $8.20 to $9.84 /W vendor price. That seems high by a factor of about 3 or so.

DIY on the north roof will be less cost ineffective. If you do put anything on that north roof, place it as far west as possible and minimize the shading from the east tree.l

[crashintoty]

The utility pole does not shade the roof.

The tree that does shade the roof is dead and will be cut down before panels are installed.

I'm putting together my battery at $21 per kWh (so far), so I'm leaning toward a battery plus grid-tie/net meter system.

The 45 degrees I mentioned is not from 0 or level(?), but relative to the sloping roof - a 45 degree angle from the sloping roof itself, of which degree/pitch I do not know yet.

Yes, $8.20 to $9.84 for a vendor install is high - I mistakenly used a 20+ panel pro install guesstimate price with a 10 panel (305 watt each) install scope.

I'm really confused about the doubt others have demonstrated here - is it based on theory or actual experience with projects of this size in this region? I've come across a couple of northwesterners/Seattlites online whom shared their data and experience with solar in the northwest and it seems it is indeed worth it especially with 22 or more panels. I also just spoke to my neighbor, they installed 24 panels years ago, says solar shaves 70% off their bill and don't start paying a bill until December (because of the net metering credits that build up throughout the year). Coming into this project I would be happy with half that performance considering how cloudy this region is and how much forum users say we can't harvest much.

[SunEagle]

The utility pole does not shade the roof.

The tree that does shade the roof is dead and will be cut down before panels are installed.

I'm putting together my battery at $21 per kWh (so far), so I'm leaning toward a battery plus grid-tie/net meter system.

The 45 degrees I mentioned is not from 0 or level(?), but relative to the sloping roof - a 45 degree angle from the sloping roof itself, of which degree/pitch I do not know yet.

Yes, $8.20 to $9.84 for a vendor install is high - I mistakenly used a 20+ panel pro install guesstimate price with a 10 panel (305 watt each) install scope.

I'm really confused about the doubt others have demonstrated here - is it based on theory or actual experience with projects of this size in this region? I've come across a couple of northwesterners/Seattlites online whom shared their data and experience with solar in the northwest and it seems it is indeed worth it especially with 22 or more panels. I also just spoke to my neighbor, they installed 24 panels years ago, says solar shaves 70% off their bill and don't start paying a bill until December (because of the net metering credits that build up throughout the year). Coming into this project I would be happy with half that performance considering how cloudy this region is and how much forum users say we can't harvest much.

Most people are confused when it comes to solar. All I say is take a look at what you will be saving and see how many years it takes to re-coop the initial investment. For some it is less than 8 years for other more than 10 years. Each person has to determine what they feel is a good investment and is solar a financially smart direction to go.

Good luck with your install and I hope it all works out for you.

[J.P.M.]

The utility pole does not shade the roof.

The tree that does shade the roof is dead and will be cut down before panels are installed.

I'm putting together my battery at $21 per kWh (so far), so I'm leaning toward a battery plus grid-tie/net meter system.

The 45 degrees I mentioned is not from 0 or level(?), but relative to the sloping roof - a 45 degree angle from the sloping roof itself, of which degree/pitch I do not know yet.

Yes, $8.20 to $9.84 for a vendor install is high - I mistakenly used a 20+ panel pro install guesstimate price with a 10 panel (305 watt each) install scope.

I'm really confused about the doubt others have demonstrated here - is it based on theory or actual experience with projects of this size in this region? I've come across a couple of northwesterners/Seattlites online whom shared their data and experience with solar in the northwest and it seems it is indeed worth it especially with 22 or more panels. I also just spoke to my neighbor, they installed 24 panels years ago, says solar shaves 70% off their bill and don't start paying a bill until December (because of the net metering credits that build up throughout the year). Coming into this project I would be happy with half that performance considering how cloudy this region is and how much forum users say we can't harvest much.

Get a copy of "Solar Power Your Home for Dummies". You need an education in the basics. You may come to see why others here have doubts. I give up.

Good luck on your future endeavors.

[bcroe]

I did a test with a cheap 3 watt 12v solar panel and got 13.6 volts with it laying flat on the roof and then got the full voltage of 14.9 when I tilt it roughly 45 degrees off the roof. So I assume if I tilt the actual panels on the north slope I may actually get a little bit more kWh than the south slope since it gets at least 85% less shade in the afternoon.

P.S. - That 20,000 kWh per year consumption will actually be decreasing. We recently bought a Emporia energy monitor and we've already discovered some major energy hogs and practices. We're in the process of eliminating those hogs and practices and have shed about 20% so far

I should have said, the open circuit voltage of a solar panel is not a good indicator of
potential performance, because it is fairly constant despite some shade. What IS
meaning full is the current capability, which varies roughly in proportional to sun
intensity. What I have done is connect the panel leads to a 10W, 0.1 ohm power
resistor, essentially shorting the panel out. The voltage across this resistor times
10 will give you the current flow, a very good indicator.

20,000 kWh per year consumption is a lot, here years were spent eliminating waste
to lower the number. Go after the obvious ones, lots of my stuff got replaced with
Energy Star. HVAC for 2 building year around is totally solar powered, the 5 heat pumps
are twice as efficient as the old AC unit, and 4 times as efficient as resistance heat.
And do go after Vampire loads, 100 small ones here really added up. good luck,
Bruce Roe

[crashintoty]

I don't understand - this is a project for me to tinker - I'm possibly looking at a 90% savings by doing it myself or at least $0.63 per watt vs $3.30 per watt (81%) and re-cooping in as little as 2 years. I am so far under the cost threshold.

Yup, just under 20,000 kWh/year ($2,500) is a lot but every single power draw has been identified in my house and documented (the old basement fridges and my homelab are the culprits). The 24/7 baseline (things that run continuously 24/7) was 1850 Wh, but now down to 1100 Wh (with room to go down to at least 700 Wh). All of this was simply powering off unnecessary office computers and servers in my homelab. Which is ironic being I originally started this solar project with the hope of using it to pay or at least offset my homelab power costs.

[bob-n]

OK, you probably do understand the economics. Forgive us for assuming wrong. We see such a wide range of people and sometimes assume the worst.

For me, going solar was great in so many ways. I learned about many different fields: energy, electrical, structural, etc. I met and helped a bunch of great people. I got my hands dirty. I'm now making more energy than I'm using, which is cool. I'm using less fossile fuel and polluting less, which is also cool. And in the long run, it's going to save me money, which is cool.

For each of us, the vision and goals are different, and that guides us in some key decisions. Do we want the shortest time to break-even, the best return after x years, the most watt-hours possible, or something else? How long do we expect to be in the house? What is our vision for the future, such as changes to energy costs, changes to netmetering terms, and changes to our physical health as we age and may no longer be able to climb on the roof. We can't help you much in these questions.

You mentioned $0.63 per watt. DIY retail for a complete BOM seems to be in the range of $1.00 to $1.50 per watt, depending on your choices. If you are able to get to $0.63 per watt, that's amazing. Panels themselves tend to run from $0.50 up per watt, without inverters, without racks and without electrical.

I'm not recommending for or against any particular vendors, but the website www.altestore.com is a good place to check out panel prices. Today, their lowest price is $0.50 per watt for a low-end Peimar poly-cell panel. Super-premium Panasonic HIT panels are $0.95 per watt. It's up to your needs and risk tolerance whether the premium price is justified or where in between to compromise.

My general rule of thumb is to budget 50% of your money for panels, 25% for electronics, 20% for rails/structural and 5% for electrical. These are very rough guidelines. You could wind up 60% for panels or 40% for panels.

And don't forget to add $5 for all of the required warning stickers!

[bcroe]

Minimizing waste is a good start. I found an old fridge and an ancient freezer were
consuming over 2000 KWH a year, replacing them with a single energy star fridge-freezer
got that down to 366. Half a hundred old, very inefficient wall warts and control
transformers were using some 2600 KWH a year, some effort eventually reduced
that to more like 500. Having gotten down to diminishing returns efficiency, bringing
on solar to cancel everything else has kept me from buying any electricity, heat, or
air conditioning for 7 years. You results will vary. Bruce Roe