Regarding the roof. The Home Owners Association for a townhome I own recently reroofed our entire complex over the past two years. Those roofs were gravel over torchdow or hot mopped. The roofer saved money by not tearing off the old roof. It only had one layer over 2x10 roof rafters and 5/8" plywood sheeting. The roofer installed 1" high density foam which added an R value to the energy calcs. He then used a TPO membrane and guaranteed the roof for 25 years.

In my case I already had solar and I had to remove the panels and racking at my expense. The roofer charged extra to flash each stanchion/standoff (about $25-50 per stanchion). Originally I had trouble getting bids on a flat gravel roof and after it was installed I had two leaks which the solar company fixed but after the second visit they said the leak was not theirs but fixed it temporarily any way. The point is you don't want that kind of situation where there is finger pointing. My suggestion is to coordinate these projects so the solar installer fastens the stanchions and makes penetrations for electrical. (My HOA would not allow conduit down the side of the building, which would have saved money but been an eyesore.)
Next the roofer comes on the roof and puts on the roofing and warrants the installation against leaks. Then the solar installer comes back and installed the racking and panels. I would suggest a minimum tilt so the panels self clean in the rain.

I won't go into the advantages of additional tilting and other adjustments to compensate for your unique weather. Others may have more experience than I on this subject. Various scenarios could include mixed orientations, over paneling (high DC to AC ratio) or highly efficient panels to optimize inverter production because of your unique weather.

If you are considering going from a gas to electric water heater I would seriously consider a heat pump water heater.

Thanks for your response, Ampster. Sounds like we should get the roof and the solar racking done at the same time. Not sure whether to look for a solar installer who knows good roofers, or for a roofer who knows good solar installers...

Lots to inform yourself about before lots of conversations take place here or elsewhere. The more you inform yourself, the more info you'll absorb here and the more sense it will make to you.

I'd suggest you start with a read of "Solar Power Your Home for Dummies". ~ $25 at bookstores or Amazon. There are also slightly dated but still useful versions on the net for free in PDF format.

Then, for the sake of not rehashing a lot of stuff that discusses some of the things that can make PV install on low sloped roofs different from install on roofs of greater pitch, check the recent thread: "Advice for residential flat roof installation in New England". Briefly: do not use ballast, leak potential is greater for low sloped roofs, sloping panels are almost a mandate, but sloping will limit array size for fixed roof size.

Then, get your actual usage (not $ - but kWh usage by billing period) and learn how you are billed for power, billing plans available to you and how Net Energy Metering works for your POCO (POwer COmpany).

Then, download and get familiar with something called "PVWatts". It's a residential solar modeling program that's pretty straightforward and simple to understand after you read all the help info screens a couple of times. 20 min. of reading intro stuff and a few runs and you'll be on your way.

Get any HOA you have on your side or at least be very informed about any CC & R's and/or bylaws. Do not assume you can do anything you want with PV. CA Solar Rights laws allow a lot but HOA's are not helpless. Get porky with them and they can ruin your day and project. Suck it up and ask for their help. You'll get more flies with honey than vinegar.

Same goes for local building dept. I suspect you will need to get them informed through building permits, reviews and approvals and all that goes with that hassle. Be prepared for design calcs/drawings including wind and seismic calcs.

My guess is you may have some difficulty finding a reputable installer who will install equipment they are not selling to you. That includes panels. Just sayin'.

As for gas vs. electric for appliances, for tasks that involve generating a lot of heat (primarily space heating and water heating) gas is the usual best choice. it's cheaper per unit of heat delivered, often/usually by a factor of 2 or 3. Using electric resistance to heat water is like killing flies with a howitzer. Match the fuel and its thermodynamic availability to the task at hand and save electricity for tasks that can't be done with fossil fuels. At $0.25/kWh, electric resistance generated heat costs $7.33/100,000 BTU ($7.33/therm). Divide your current per therm gas cost (from your bill) by a somewhat conservative 0.7 for energy cost/therm after efficiency losses.

Once you get PV, I suspect you'll be on TOU billing so expect your per kWh ave cost of electricity to be higher. As for heat pump water heaters, depending on the relative costs of electricity vs. gas, they can be better than electric resistance for heating water, and close to gas, maybe even a bit cheaper for the fuel, but the added cost of the equipment and the still nagging questions about equipment reliability would make me stay with gas, especially if your gas water heater is functional.

FWIW, I've found roofers usually don't know much about PV and PV installers are not the best roofers around. I'd most assuredly get the roof inspected/serviced as necessary, but I'd act as my own project coordinator between the two trades. I did that for my install. A bit more work but IMO, well worth it.

Welcome to the neighborhood.

what they said, plus

if you are re-doing the roof, have it engineered to meet your expected loads, while it's exposed. Going to have to do it anyway to get permits
Ask the roofer & engineer about adding some 4x4's to provide an elevated spot to mount the racking. it may help, it may block drainage.

Ask the electric company (PG&E) how much solar you can install and still be a homeowner and not a commercial generator (10kw is the average limit)

To be clear you only need the standoffs installed before the roofing. But you need the racking layout designed and engineered before you install the standoffs. Mike 90704 makes a good point about knowing what the limit is on size for PG&E. Sometimes it is based on prior usage. A good installer can help you with workarounds. I contracted for a 5.7kW system which was way over what the prior years consumption would have suggested. I told my installer to mention that I was converting from a gas water heater to electric and had two EVs. That may have helped me get a larger system.

Regarding heat pump water heaters, the rumors about reliability were true only when GE was selling them. I have one of those as did a friend. It was difficult getting warranty repairs because GE sold that business. Mine is still going strong after a simple fix. It is six years old. I also have two Rheem heat pump WHs and they seem better built than the GE. What I like about them is I can set them from an app on my phone to take advantage of TOU rates. They heat the water at night during super off peak rates.

@J.P.M also makes a good suggestion about PV Watts. It does use weather data but I would run a few scenerios to see if it takes into account your micro climate. My sister used to live in the banana belt in Sausalito and homes 1/4 mile away had entirely different weather as far as clear sunny days. That is where a good local installer can be worth finding. I have not lived in the Bay area long enough to know any good referrals.

If you are in the City/County of San Franciso you may find a resource with Clean Power SF. It is a Community Choice Aggregator that provides the actual power while leaving the distribution and billing with PG&E. It is a non profit agency run by the City/County that provides other incentives. Recently there has been discussions about them taking over the distribution assets of PG&E because of the PG&E bankruptcy. I wouldn't hold my breath on that happening anytime soon.

Thanks Ampster, J.P.M. and Mike for these replies. Here are some learnings:
- We will likely replace the roof as part of the project. I'm talking with our accountant about whether new roofing that is directly underneath a solar panel can be considered eligible for the 30% solar federal tax incentive.
- We will fix tilt at 15 or 20 degrees so the rain cleans the panels, and we don't have to go up on the roof often.
- I will start looking into heat pump water heaters. This is apparently a good match for hydronic / radiant flooring. That could quadruple our energy needs and put this whole project into a different tier -- but we have the roof size and the panels on hand to support that. If anyone has estimates on how much power a heat pump uses to heat radiant floors, that would be very helpful.
- There is a concern that with these older panels, we may have a challenge complying with new zoning regulations around emergency system shutoff. We may need to install per-panel inverters to provide this shutoff in compliance.
- As far as I can tell, San Francisco has no limit on the amount of per that can be generated. There may be a 1MW limit which is way beyond what a single family home would be concerned with.
- PV Watts thinks we can go very big.
- Another roof planning app thinks we can easily fit all 48 panels on our roof, physically.
- We have no HOA to worry about. In general San Francisco is very solar friendly.
- We just switched to something called "Clean Power SF" which happens to have a very low $ electrical rate for off peak usage.

I'm reading "Solar Power for Dummies" -- thanks for that reference J.P.M.

You might want to ask him about the electrical panel upgrade.
When you look at the heat pump EPA tag note the kWh consumption and kWh savings. Ignore the dollar saving because that is based on average USA rates. Do the math based on California rates. Much faster payback in California.
I don't know all the questions to ask in order to answer your questions about hydronic systems.More data is needed. I have had two that were gas fired and i loved them. The fan coil type use the same duct work as forced air units but you dont have that smell of burnt rusty metal that I remember with an old FAU. In the floor radiant is the best in my opinion. Anectdotally I have a friend in a beach town in Southern California who used a 50 gallon heat pump and baseboard radiators.
I think you are talking about Rapid Shutdown requirements which are a electrical requirement of the building code. It is not related to the age or size of your panels but more about the system of panels and inverter. Currently in my jurisdiction Enphase Micro Inverters and Solaredge string inverter with power optimizers will satisfy the code. City/County of San Francisco may be different. The question to ask is whether they follow NEC 2014 or NEC 2017. As I mentioned earlier I don't know how specific PV Watts is in terms of micro climates. Since you already have the panels you want to optimize your system generation using all those panels. You may want a higher DC to AC ratio to make sure you have the best mix of panels to inverter. Since you have free panels it is a question of how much you want to spend on excess inverter capacity for the few sunny days that you may have. It is probably not as important since most grid tie inverters are fairly efficient from 50% of capacity to 140% of capacity. That issue is how your might differentiate installers. The ones most knowledgeable on your microclimate who are willing to coordinate with the roofer are the ones you will want on your main list of bidders. This is not your typical cookie cutter type of installation.
It is not the city limit on power that Mike was referring to but a PG&E limit and it may relate you the transformers in your area. When the installer is talking to them about upgrading the main service panels is a good time to test that assumption.

I spoke with PG&E. They have a limit of 30kW above which another type of review is required. They did not describe any trigger existing at 10kW. They allowed as to how they might need to upgrade a transformer, but that is on their side and would be transparent to the customer.

That's all good news, on the flip side, I've learned more about the rates. As I now understand thing (still a but murky), for a hypothetical month in which our system generates a 500kWh excess of power:

- Roughly 200kWh of this would be considered "generation". That is now handled, in San Francisco, by Clean Power SF (our CCA), which pays $0.0893 cents per kWh --> $17.86 credit
- Roughly 300kWh of this would be considered "distribution/transmission/delivery", which is handled by PG&E. They pay $0.035 cents per kWh --> $10.50 credit

Both of the above numbers would need to be adjusted for Time-Of-Use. PG&E mandates that we switch at least their portion to a TOU rate plan. Those rate plans go up to $0.407 cents per kWh. As I understand it (and here it gets VERY murky) for a given peak hour in which we generate excess power, we would get an energy credit at the higher $0.407/kWh TOU rate, but any cash credit would only be at the lower $0.035/kWh rate. If anyone out there has an actual PG&E bill that shows how they handle a surplus month when a CCA is in the picture, I'd be very curious to see it!

At the end of the day, I think we will recommend our installer to install our panels at a 5 degree south facing tilt. This will yield 9.89% less energy than a 30% tilt, according to PVWatts. However it will also catch less wind in a storm (6x less direct cross section -- I'm disregarding uplift). Plus, with the TOU rates, going a bit flatter could help us capture a bit more of those late afternoon credits.

I spoke with two hydronic heat installers. Neither was familiar with using a heat pump water heater, both said that gas would be cheaper. However I'm still very interested in pursuing the heat pump option. With the amount of power from this system, plus the TOU credits (see "They heat the water at night during super off peak rates." comment above), it seems actually possible that we could heat our whole house with a ~80000BTU heat pump and eliminate our gas bill. That would drastically alter the payback period. Again, anyone with experience here, I'd love to hear your advice!

Nat. gas will probably be the cheaper alternative. Consider also that as less electricity is wasted on a task more appropriately done w/ CH4, further up front savings can be realized by virtue of the array size reductions made possible by the electrical load reduction.

I'd be careful about the 5 deg. tilt. You'll get more panels in the same footprint, making the per yr. production about the same per m^2 of footprint, but you may wind up cleaning them more often and need more panels for the same annual production.

CCA billing works differently than you describe above. You are still credited or charged for the total kWhrs you use or generate per each TOU period. Instead of one rate from PG&E, the rate is unbundled and you are charged and credited a separate rate for distribution and a rate for generation. The sum of those rates has been generally less than the unbundled PG&E rate because the CCAs in California keep their generation charges low enough to retain customers. I can go into further detail later but that is the basics. PG&E does all the billing but as I mentioned above the total kWhrs are billed at separate rates for generation and distribution' both of which are at unbundled TOU rates. Every CCA handles true up slightly different when it come to paying for excess power. There are other advantages of CCAs such as subsidized EV charging stations and programs where you are paid a monthly fee for letting them turn of your charger when the grid is stressed. I have not gone through an entire year with my CCA so I can't even give you a real world example abou true up. I did purchase a subsidized EV charger and receive a $5 a month credit for signing up for their Grid Savvy program. That may be a good plan since your panels are free and a lower tilt will give you more total panel area since you don't have to space them as far apart. You mentioned you have 48 panels which is 11kW. I think the PG&E limit is based on inverter capacity. Since the panels are free the only incremental expense would be the mounting hardware and installation for overpnaling beyond the PG&E 10k limit. There could be a strong argument for overpanelling but that discussion is more appropriate when we know the square footage of your roof. In any event, my opinion is you want to optimize your free panels and the only constraint should be the 10kW inverter size and the roof area. Unfortunately most tradesmen don't take the customers goals into consideration. I converted from a gas water heater to a HPWH and from a gas dryer to electric in my current home and enjoy the thought that I am using free energy. Many on this forum would argue that that is inefficient. California may be unique in that it has an aggressive Net Zero Energy goal and I believe in supporting that goal to the extent that I am able.

Natural gas may be less expensive than using a heat pump system. However since the OP's goal is to maximize the benefit of the free solar panels the decision matrix could change based on how one evaluates all the inputs. An important factor that he has mentioned is the ability to program the HPWH to take advantage of lower rates. I also enjoy the ability to put it in vacation mode from my phone after I have left on vacation. That benefit may be unique to my failure to remember everything before i leave for vacation.LOL.

As far as the optimum angle to prevent dust from pooling and drying and reducing out put the answer may be more specific to the panels. My only experience with horizontally mounted panels is that I wish I had given them some tilt. I can't tell you if 5% is optimum. It has to do with the height of the frame lip and the distance between that lip and the edges of the cells. The last bit of dust tends to settle in the pool of water formed between that lip. When that pool dries it leaves a thin film of dust. The less tilt the more likely that that pool of dusty water will extend onto the cell and obscure the output. This is probably more art than science..
Earlier the OP mentioned that he gets early morning fog and that may influence his decision about orientation to south or west for what ever tilt he decides on. If the fog clears in the afternoonshe may consider a more westerly tilt to take advantage of more production. Current TOU peak periods generally favor a westerly orientation but there is no guarantee that those time frames will stay the same. It is anyones guess whether the weather or the TOU periods will change significantly.

As far as adding a HPWH, one other idea might be to wait until the free panels are installed and see how the annual usage goes. If it approaches current usage and (further) conservation is not in the cards, A HPWH may not utilize any additional "free" generation.

The optimal elevation angle for an array needs consideration of several different factors that I won't regale you with here, but they'll vary for each application. On the specific subject of fouling vs. tilt, I've done some work including considerations of precip. frequency, duration and intensity. Hard to describe in this format. But since about a 1/32" or so lip is pretty common between frame and glazing, the bathtub ring will tend to be more f(other things). But, I can suggest that the bathtub ring depth or dist. from bottom of panel for common panels seems to be a weak function of tilt and not linear, but usually larger, at least in my neighborhood. Since most all panels have the sane edge lip, that small 1/8" - 1/4" of permanent discoloration.may well be a is a permanent disability. .

Obviously, a horizontal panel in the deser will be a sandbox that cakes w/every rain and a vertical panel in Maine will probably never need cleaning more than any other vertical glazing.

I did some math and your roof is 1,440 Square feet (24x60) and your solar panels are about 21 sq feet (3.5 x 6) each so 48 of them would take up about 1008 square feet completely horizontal. That does not account for any required access setback(s) for firemen per the code. Once you figure out how much usable space you have for panels after access setbacks then that may be the parameter that determines the maximum tilt you can set the panels at. I would go for the maximum tilt that you can fit on the roof because it will make your system more productive. Wind loading will be greater with higher tilt but that usually only means heavier rail and more frequent lag bolting of the rail to your rafters. . That assumes that the structure of your roof rafters can handle the wind load.

I am not an expert on maximum inverter size for a particular main service panel but that is a question you may want to get answered soon since it may affect costs. I recall some threads where 7600 Watt inverters were the limit that did not result in some increase in service panel size. PG&E wont advice you on that because it is a code issue that is behind the meter. Having a 7600 Watt inverter is not the end of the world with a 11,000 kW system. It is a 1.44 to 1 DC to AC ratio.

I hope the above helps you frame the questions for potential installers and/or electricians that may be looking at your installation. It is an interesting project that has unusual weather, possible roof size constraints, and possible main service panel constraints.

There are two heat pump issues that have been discussed. One is to replace the Water Heater and the other was a hydronic whole house heating using an additional larger HHPW that would only be used during winter..I agree before thinking about the heating issue more analysis would need to be done. The no brainer, at least from my perspective, would be to replace the water heater with a HPWH. I estimated that the OP will have significant excess production with a 11kW system even with some mostly foggy mornings. Using a conservative capacity factor an 11kW system might produce 10mWhrs of electricity or 900kWhrs per month. He mentioned that his electrcal usagecwas aboout 432 kWhrs per month.

The average 50 Gallon HPWH uses less than 100kWhrs per month based on EPA ratings. That would leave about 350 kWhrs excess generation even after installing a HPWH. The math gets even better with a TOU rate because the HPWH load can be shifted to a lower rate period increasing the financial payback.