San Jose / Bay Area - help me decide

[realmcoy]

can you PM me your quotes and installers please?

Yep, yet another help me decide thread. Opinions welcome as I keep going back and forth.

My current annual usage is 9500 kWh and seems to be trending up. If I divide my yearly usage by total cost, it works out to be $0.24/kWh. I've been struggling to figure out how it might change. Other than appliances, most energy upgrades have been done. I have a toddler so it will go up as he gets older. Since I'll have solar, I'll surely run the A/C more since I won't have to worry about the bill (well, not like before). I'll be replacing the A/C sooner than later so the new one will be more efficient. We'll remodel the kitchen next year so the electric wall oven will be gone in place of a gas range. Will get a new fridge as well. So maybe it is a wash...

I did a lot of searching for local installers and all should be above average (Yelp, Angies List, Solarreviews, Diamond Certified (always cynical on that one even more than the others), etc...). I'm fairly confident whomever I choose will do a good job and also figure I'm probably paying a bit of a premium for them. The panels will be install on the front of the house and I do have some shade issues. I'd prefer the look of black panels as well and sort of want to keep the number down since, as the panel count goes up, so does the sprawl. There are some plumbing stacks and an "eyebrow" (or something like it, it is nothing functional) that take up space so panels must go around it. The more panels, the more chance of shade issues.

I'll do it by company initials, size W(dc), cost (before tax credits), $/W, equipment. There are some I'm not considering but will throw in at the end just for people to see as I found it useful seeing as many quotes as possible.

Company: CS (quoted based on a 10% bump in usage during summer months)
5500w. $23,274.00 ($4.23/w). 20 SolarWorld SW275 mono black + SolarEdge SE5000
OR
5490w. $23,715.00 ($4.32/w). 18 LG LG305 + SolarEdge SE5000

Company: CP
4675w. $19,685.00 ($4.21/w). 17 Suniva OPT275 + Enphase M215

Company: HS
4560w. $18,000.00 ($3.96/w). 16 SolarWorld SW285 mono + Enphase M215 (not black panels from what I see so probably not considering)
OR
4480w. $18,450.00 ($4.12/w). 16 SolarWorld SW280 mono black + Enphase M215

Company: ST
4578w. $20,789.00 ($4.54/w). 14 SunPower E20/327 + SolarEdge (probably a no since panels are not totally black)
OR
4690w. $22,918.00 ($4.89/w). 14 SunPower X21/335 + SolarEdge

Company: TSC
5600w. $24,020.00 ($4.29/w). 16 Supreme SNPM-GxB-350 + SolarEdge SE6000 (probably not since bificial panels and I won't get the benefit).

The not considering but in case people find the quotes useful.
Company: ST
4578w. $20,664.00 ($4.51/w). 14 SunPower E20/327 + SMA TL (shade issues and would prefer per-panel monitoring).
5559w. $24,770.00 ($4.46/w). 17 SunPower E20/327 +SolarEdge (same as above plus too expensive overall).
5696w. $27,342.00 ($4.80/w). 17 SunPower X21/335 + SolarEdge (too expensive overall).

Company: CP
6050w. $23,985.00 ($3.96/w). 22 Suniva OPT276 + Enphase M215 (this was done for a possible EV in the future).

Company: PV
4000w. $24,220.00 ($6.06/w). 16 SunPower SPR-X20-250-BLK-A-AC (inverter built into panel). (yeah, cost is insane).

Company: TSC
5232w. $26,660.00 ($5.10/w). SunPower E20/327 + SolarEdge SE6000 (too expensive overall).

It is interesting the size companies quoted based on my annual usage. Most did the about 80% energy offset, so a 4500w system. I suppose I could pick some variation of all of their quotes as well. I like the idea of a slightly bigger system for some future-proofing but really wonder if it will be necessary.

TIA for your insight.

Derek,
Thanks a lot for your wonderful post. Do you mind sharing the names of your installers and your quotes with me please? I am in the south bay.

Thanks

Pranav.

[na9an]

congrats

i am nervous going with chinese panels right now.
who knows where those companies will be in 10 years.

sticking with known US based vendors who atleast will honor the warranty via local installers.

[derekj]

I ended going with the CP bid (original: 4675w. $19,685.00 ($4.21/w). 17 Suniva OPT275 + Enphase M215). The sales guy I worked with was very detailed in what he initially provided me and with all of his responses. In the end, he sweetened the deal by dropping the price by $500 and doing the attic install in that cost. I was sort of leaning towards SolarEdge but given the cost savings, I just couldn't pass it up. I tried doing some long term projections for SunPower using SAM (for kWh produced over 25 years) to get a lifetime $/kWh price but it was working out to be a couple of cents higher. So that cost premium wasn't worth it and kept me under my planned budget.

Now to hopefully get things installed and running before all the prime generation time is gone. I just want a taste...

[J.P.M.]

One more consideration that may still be appropriate for the back of the envelope is that some degradation may be expected in the panels, and the amount that is "clipped" would decrease each year (assuming irradiance and temperature are unchanged). For example, if a panel starts at 275 W out of the box, it might be reasonable to think of it as a 265 W panel in year 5. This will tend to stretch out the time it takes to achieve NPV equivalence, sometimes significantly.

Noted and generally agreed.

Also noted: That would be one of the several considerations (or many as deemed important for the analysis) of any thorough life cycle cost estimate in determining the net present value of the savings.

I'd suggest most of the simple "payback" nonsense usually used is not sophisticated enough to need or even make practical use of decreased annual savings from expected panel production decrease as f(time).

Even if known, I'm not real confident the significance or use of such a # would be applied in a meaningful way.

[sensij]

In this back of envelope, incomplete and simplistic example, that # is $400/$67.50 = 5.9 yrs. if on T.O.U., or $400/$30 = 13.3 yrs. if on tiered rates.

One more consideration that may still be appropriate for the back of the envelope is that some degradation may be expected in the panels, and the amount that is "clipped" would decrease each year (assuming irradiance and temperature are unchanged). For example, if a panel starts at 275 W out of the box, it might be reasonable to think of it as a 265 W panel in year 5. This will tend to stretch out the time it takes to achieve NPV equivalence, sometimes significantly.

[J.P.M.]

Whether done by individual panel/micro or by entire system, I'd think one way to do the analysis would be to first get an estimate of the difference in annual generation from one inverter to another. Then, get the net present value (or worth) of the difference in the cost of that annual production difference. If the net present worth of the (estimated) greater production from the larger inverter(s) is deemed greater than the initial cost of the larger micro inverter(s), the larger inverter would be the choice.

If, say, 150 kWh is the diff. in annual system prod. on a system with 20 micro inverters, and one is on T.O.U., the extra production may well be mostly at peak hrs. and carry a value of, say, $0.45/kWh --->>> ~ $67.50/yr.

OR, if on tiered rates, those savings will be lower assuming < 100% offset but high enough to be entirely in tier one so that the savings are say, $0.20/kWh, or, ~ $30/yr. marginal savings.

If the cost diff. is $20/micro, the task is to determine if ($20 X 20) = $400 is more or less than the net present value of the annual savings.

Most folks brain hits the wall about 2 sentences into a discussion about net present value of anything, and they usually wind up dividing initial cost by initial annual savings for something that is loosely defined as "payback".

In this back of envelope, incomplete and simplistic example, that # is $400/$67.50 = 5.9 yrs. if on T.O.U., or $400/$30 = 13.3 yrs. if on tiered rates.

[foo1bar]

I would stick with marginal cost analysis for something like this. A $500 "upgrade" that doesn't deliver $500 of value is the same whether you are talking about a $5000 system or a $25000 system. Scaling it to the total cost is how car salespeople attempt to get us to buy all kinds of ridiculous extras ("Just $5 / month more for the designer floor mats" ... no offense to floor mat aficionados intended).

To be clear, that 5% increase (or $20/panel) *IS* the marginal cost. So if you're talking 20 panels, it's a question of whether that extra $400 is going to have a reasonable ROI or not. And IMO at that price differential it likely will. If it were a $40/panel difference, I think it likely won't. But there are a lot of items involved in that analysis.

FWIW, when I did the analysis of using an enphase system vs. solaredge, solaredge came out ahead. But that was mostly ignoring labor costs.

[sensij]

BUT if you look at the total system, it's likely less than 5% additional cost - and it's plausible the performance improvement would justify it.

I would stick with marginal cost analysis for something like this. A $500 "upgrade" that doesn't deliver $500 of value is the same whether you are talking about a $5000 system or a $25000 system. Scaling it to the total cost is how car salespeople attempt to get us to buy all kinds of ridiculous extras ("Just $5 / month more for the designer floor mats" ... no offense to floor mat aficionados intended).

[foo1bar]

However, if there is a premium for them over the M215's, chances are good they won't produce enough to justify it.

So a quick price check I just did shows $145 for M250s vs. $125 for M215's.
That's a 16% higher cost.
For 280w panels, I'm sure you won't get 16% more production...

BUT if you look at the total system, it's likely less than 5% additional cost - and it's plausible the performance improvement would justify it. With ~300W panels, IMO it's very likely that it makes economic sense. (Just got an email offering $.97/w for 300W panels... Sad I didn't get that - but OTOH I would have been up on the roof in august instead of Feb. And my production is already going - 6 months earlier than if I had waited)

[sensij]

There are many ways to get things done.

Yes, there are. The advice you gave had none of that nuance, and I called it out. Your response just now is more thoughtful, and I agree with most of what you wrote.

[dat]

See this post, and others. In the real world, clipping does occur, but the value of the energy "lost" doesn't typically justify the extra expense of the M250's. If they cost the same, sure, go M250. However, if there is a premium for them over the M215's, chances are good they won't produce enough to justify it.

Next time I want to make a point, instead of using real world data and proven models, I'll just cite Amazon.com reviews.

What you say is true, but it is only your view ( and it is the most economic view). But there are other factors that you don't see or consider. For example, I believe the M215 life span will be shorten if the clipping occurs more with larger panels. If you pair it with more than 270 watts panels, will the manufacturer still warranty its product (recommended maximum 270 watts)? One important factor is our feeling. If I see it clipped a lot, I definitely will not be happy. I will pay more to make myself happy. Just like some people pay more to fly business or first class, to stay in a 5 stars hotel instead of 3 or 4 stars hotel.
Edit: The OP does not necessary to pay more for the M250. If the larger panel is used, just ask the installer to use M250 for the same price. The price should be $/w. If the installer does not want to use M250 then ask him to use smaller panels. There are many ways to get things done.

[sensij]

Why do you think it is a bad advice?

See this post, and others. In the real world, clipping does occur, but the value of the energy "lost" doesn't typically justify the extra expense of the M250's. If they cost the same, sure, go M250. However, if there is a premium for them over the M215's, chances are good they won't produce enough to justify it.

Next time I want to make a point, instead of using real world data and proven models, I'll just cite Amazon.com reviews.

[dat]

Bad advice.

Why do you think it is a bad advice? Enphase recommends maximum input power for M215 is 270watts. Even my 255watts panels clipped sometimes with M215. Here is a review from Amazon.com "Those M250's are awesome !! I upgraded from M215's and gain 25W per solar panel. I have LG 260W solar panels and my old M215's were generating 225W per solar panel. Just upgrading to M250 micro inverter on the same panel I get now 250W per panel. By replacing my old inverters to those new ones I get more energy generated on the same solar panels. "

[araghava]

You should also consider purchasing all the equipment yourself and hiring a GC to do the install. It will be significantly cheaper.

[sensij]

By the way, do not pair M215 with 260 watts or higher panels, use M250 instead.

Bad advice.