Considering solar panels, looking for sizing/financing advice

[Yanquetino]

Ah… solar aficionados. I'm intrigued that some of you are actually paid for your excess kWh. Our local utility doesn't pay squat for ours: every year they SWIPE any kWh still in our "savings accounts" with the March bill (see below). Even worse, they keep trying to convince our Public Services Commission that they need to make solar homeowners pay an extra "fine" every month, because (they falsely claim) we aren't paying our "fair share" of their infrastructure. Pure bovine excrement, as I demonstrated to our PSC with this presentation.

 

[wjgjr]

Do you have a fixed rate for excess? As far as I can tell in Maryland, excess is bought via solar credits that have a floating price, and are priced per MWh: Flett Exchange | Maryland SREC Pricing

SREC credits are actually something different from excess generation. If you have solar, you earn 1 SREC credit for each MWh you generate. This can then be sold to utilities who use them to claim credit towards their mandates for generating part of their power from clean/solar energy. You earn the credit regardless of whether you use that power or send it to the grid. Note, however, that depending on the agreement with the installer, the installer may own and sell the credit. We did have an issue where one version of the contract incorrectly assigned SREC credits to Tesla.

As far as excess generation, with PEPCO, they have a great net metering policy - every kWh you send back to the grid earns you credit at the full retail rate (generation + transmission + distribution) and my understanding is they do pay you once a year if you have excess generation for the year. Since we are still in the interconnection process, I can't comment on all the details, but I believe the excess amount, if any, is converted to a dollar amount every month (at the rates for that month) and shows as a bill credit. On the prior comment about the net metering cap in MD, my impression is that once that is hit, PEPCO would not need to allow new installations to have access to this gnet metering policy, but any install currently under the policy would still be able to take full advantage of it for the lifetime of the system.

By the way, while PEPCO does not have residential time of use pricing now for solar (though there is a legacy rate that is closed to new customers) they do have a pilot rate - R-TOU-P - that looks like it would be pretty interesting to pair with solar and powerwall. Unfortunately, the pilot is full, but it may mean something will arrive in a few years. It is odd that Maryland does not have TOU pricing since they have a significant (30%) state rebate for installing powerwalls as an energy storage system but then don't incentivize customers to use it in a way that helps the grid.

 

[Zaxxon]

If that is the case for OP's net metering--paid out at full retail, and potentially able to paired with TOU--over-building is a no-brainer. You earn out at peak rates while charging your car at the cheap rate. And you've got peace of mind that you're covered for all usage even if you start consuming more in the future or your system produces less as it ages.

 

[willow_hiller]

SREC credits are actually something different from excess generation. If you have solar, you earn 1 SREC credit for each MWh you generate. This can then be sold to utilities who use them to claim credit towards their mandates for generating part of their power from clean/solar energy. You earn the credit regardless of whether you use that power or send it to the grid. Note, however, that depending on the agreement with the installer, the installer may own and sell the credit. We did have an issue where one version of the contract incorrectly assigned SREC credits to Tesla.

As far as excess generation, with PEPCO, they have a great net metering policy - every kWh you send back to the grid earns you credit at the full retail rate (generation + transmission + distribution) and my understanding is they do pay you once a year if you have excess generation for the year. Since we are still in the interconnection process, I can't comment on all the details, but I believe the excess amount, if any, is converted to a dollar amount every month (at the rates for that month) and shows as a bill credit. On the prior comment about the net metering cap in MD, my impression is that once that is hit, PEPCO would not need to allow new installations to have access to this gnet metering policy, but any install currently under the policy would still be able to take full advantage of it for the lifetime of the system.

By the way, while PEPCO does not have residential time of use pricing now for solar (though there is a legacy rate that is closed to new customers) they do have a pilot rate - R-TOU-P - that looks like it would be pretty interesting to pair with solar and powerwall. Unfortunately, the pilot is full, but it may mean something will arrive in a few years. It is odd that Maryland does not have TOU pricing since they have a significant (30%) state rebate for installing powerwalls as an energy storage system but then don't incentivize customers to use it in a way that helps the grid.

Thanks for the local perspective!

Your description was my understanding as well, until I read this energy sage article I posted earlier: 2020 Pepco (Potomac Electric Power Company) Net Metering Rates | EnergySage

"Pepco customers who generate their own electricity will receive bill credits for any excess power generated. Using a special net-capable meter, Pepco measures the difference between energy you use from the grid and any power generated from your solar panel system. If your panels generate more electricity than you use that month, the excess is carried over onto your next month’s bill as credits. You may generate more energy than you use in a year. If so, at the end of the year you can sell that energy in the form of Solar Renewable Energy Credits (SRECs)."

The description there makes be believe that net metering is paid in bill credits that, at the end of each year, if you have any left over you can cash out for SRECs.

But I did just find a more technical sounding website here: DSIRE that states:

"Net excess generation (NEG) is generally carried over as a kilowatt-hour credit (i.e., at the retail rate) for 12 months. Compensation for any NEG remaining in a customer's account after a 12-month period ending in April of each year is paid to the customer at the commodity energy supply rate."

So maybe Energy Sage got their description wrong?

 

[charlesj]

at $.06/kWh for excess, that would take a long time to justify over-sizing the system.
Mine is sized for ~zero dollar true up bill even though I use slightly more than I produce. PG&E only pays about $.04/kWh excess

True, if economics is a 100% in the equation. Over production can pay off that base charge from PG&E.

 

[SMAlset]

Don't know if this will be helpful to you or not. A number of our cities here in the SF bay area banded their citizens together and joined the Silicon Valley Clean Energy partnership that became our energy provider instead of PG&E (still our distributor). The partnership purchases from renewal sources, saves money as a buying group and pass that savings on to the customers and cities in one form or another. I seem to recall during our public meeting as it was rolling out to customers that if you wanted to get solar they were recommending having it installed at a particular time of year due to how PG&E would do your true-up. Since we didn't even consider solar at that point in time, I wasn't paying close attention but sure someone else can provide some light on this. If your electricity company does something similar you might want to look in to it and factor in when to do your install if you can.

 

[TheTalkingMule]

Two other notes if you go for Tesla:

1) The larger systems are generally cheaper per Watt, so even if you only get paid $.06/W after 100%, you can still work out better with XL or something you negotiate slightly smaller.

2) Not long ago you could "upgrade" to Panasonic panels built in Buffalo. Maybe you could ask for multiple proposals with different higher end panels? The Q panels are fine, but might as well see what they can offer.

 

[wjgjr]

Thanks for the local perspective!

Your description was my understanding as well, until I read this energy sage article I posted earlier: 2020 Pepco (Potomac Electric Power Company) Net Metering Rates | EnergySage

"Pepco customers who generate their own electricity will receive bill credits for any excess power generated. Using a special net-capable meter, Pepco measures the difference between energy you use from the grid and any power generated from your solar panel system. If your panels generate more electricity than you use that month, the excess is carried over onto your next month’s bill as credits. You may generate more energy than you use in a year. If so, at the end of the year you can sell that energy in the form of Solar Renewable Energy Credits (SRECs)."

I think that page is quite simply wrong. Their site has another page explaining SRECs - SRECs: Understanding Solar Renewable Energy Credits | EnergySage - that seems to be more accurate, stating "For every megawatt hour (MWh) of electricity that a solar energy system produces, a corresponding SREC is created." So, their site is, at the least, not internally consistent. (It also looks like the page I linked either borrowed heavily from What is SREC & how can I get the best SREC prices in 2020 or vice-versa.)

Something that is worth noting in MD is that SREC prices are effectively capped (see towards the bottom of SRECTrade | SREC Markets | Maryland | MD for caps by year) and decrease pretty significantly, so much of the value of the SRECs will likely be realized in the next few years, assuming legislation isn't changed again.

 

[Zaxxon]

SRECs have nothing to do with excess energy, so that is indeed bizarre. SRECs are created for all energy generated (regardless of who uses the energy and entirely separate from the usual charges/costs for that energy). Bill credits are created for every excess kWh exported to the grid.

 

[willow_hiller]

Does anyone have general thoughts on solar panel facing? On a walk a few days ago I noticed a neighbor of ours had 8 panels on one side of his roof, so I figured that would be a great way to see how comfortably 24-48 panels fit on our roof.

Here's our roof with 3 "8 panel" rectangles superimposed over it:

It looks like we could comfortably fit 24 panels on just the Southwest and Southeast facing sides of the roof. 36 might be a squeeze, and 48 would definitely require moving onto the Northeast and Northwest facing sides. At roughly 38.8 degrees North latitude, would panels on Northern facing sections have noticeable differences in generation from those facing South?

 

[MorrisonHiker]

Does anyone have general thoughts on solar panel facing? On a walk a few days ago I noticed a neighbor of ours had 8 panels on one side of his roof, so I figured that would be a great way to see how comfortably 24-48 panels fit on our roof.

Here's our roof with 3 "8 panel" rectangles superimposed over it:

View attachment 541942

It looks like we could comfortably fit 24 panels on just the Southwest and Southeast facing sides of the roof. 36 might be a squeeze, and 48 would definitely require moving onto the Northeast and Northwest facing sides. At roughly 38.8 degrees North latitude, would panels on Northern facing sections have noticeable differences in generation from those facing South?

That's where PVWatts can give you a really good idea of how the azimuth and tilt affect production. You can indicate the direction (azimuth) the panels face as well as the slope (tilt) of the roof. Trying plugging in the numbers and saving the results. Then go back and try adjusting the azimuth and you can see how much expected production changes.

We added 4 kW last year on a northwest facing roof. While they don't produce as well as they would facing to the south, they still have good production for over six months of the year. The nice thing is that they help us produce more during peak so the credits we earn with the electric company are worth that much more.

 

[Zaxxon]

@MorrisonHiker gave the right answer, so I'll give you one based on my specific installs, since they are very relevant to your question. Will you see a noticeable hit? Yes. If you can fit enough panels on the S-facing sides, that's what you should do. But while you'll see lots of advice saying to avoid N-facing at all costs, ignore them. Non-ideal faces will produce less, but they'll still produce a lot.

I have a fairly interesting set of experience on this topic, as I have a relatively small roof with 4 approximately-equal faces, each 90 degrees offset from the next:

The SW and SE panels were installed in 2014. They're 15 280W panels for a system size of 4.2 kW.
The NW and NE panels were installed in 2019. They're 15 315W panels for a system size of 4.725 kW.

The newer system is on the less ideal faces, and has about a 12% higher nameplate capacity. So when comparing the two systems, we'd expect the newer system to be about 12% higher in production than the old system, if the faces received equal sun.

For the first half of this month, the newer system produced 353 kWh. The old produced 327. So the new is about 8% higher than the old, meaning the non-ideal faces 'cost' me about 4% production vs what i'd have expected if they were on more ideal faces.

If I look at April's full-month numbers, the new produced 544 kWh, while the old produced 532. About a 10% hit.
March: 438 vs 522 - 28% hit.

You can see where I'm going with this. In my case, in the winter months the disparity is higher, while in the nicer months it's pretty minimal. Pretty much what @MorrisonHiker said: for half the year the north face will do just fine, while for the other half it'll have a bigger hit. In our area (Colorado) where we get significant snow, this is exacerbated by the fact that the snow takes longer to melt off of the N-facing side than the S, so when it snows we lose even more production until it clears up.

My opinion is that if you're looking to offset as much of your usage as you can, cover whatever you need to. If you're focused on quickest payback, then stick to the ideal roof faces.

 

[willow_hiller]

Uncovered another interesting legal wrinkle in Maryland that might affect sizing decisions. From MEA - Solar

"SREC quantities for PV systems under 10 kW are calculated from PVWatts model estimates. For PV systems over 10 kW, SRECs are paid based on actual production."

I can't actually decide if being paid SRECs based on model estimates is a necessarily good or bad thing. On one hand, it would lead to a predictable amount of SRECs per year; but on the other hand, in some years you could be underpaid if it turns out to be a very sunny year.

@wjgjr have you had an opportunity to report your production for SRECs yet? Do you know what that process is like?

EDIT: Doing some more research, and evidently the permitting and meter requirements are lower for systems smaller than 10 kW. That would also explain why when set to Maryland, the Tesla Solar prices go $10,000 $18,000, $26,500, $34,500 (it must be ~$500 more for permits and meter).

 

[wjgjr]

Uncovered another interesting legal wrinkle in Maryland that might affect sizing decisions. From MEA - Solar

"SREC quantities for PV systems under 10 kW are calculated from PVWatts model estimates. For PV systems over 10 kW, SRECs are paid based on actual production."

I can't actually decide if being paid SRECs based on model estimates is a necessarily good or bad thing. On one hand, it would lead to a predictable amount of SRECs per year; but on the other hand, in some years you could be underpaid if it turns out to be a very sunny year.

@wjgjr have you had an opportunity to report your production for SRECs yet? Do you know what that process is like?

EDIT: Doing some more research, and evidently the permitting and meter requirements are lower for systems smaller than 10 kW. That would also explain why when set to Maryland, the Tesla Solar prices go $10,000 $18,000, $26,500, $34,500 (it must be ~$500 more for permits and meter).

You are correct on the SREC policy, except that what I've seen is that under 10 kW can (not must) use PVWatts. As we are still waiting for PTO, we have not yet registered our system so I have not yet dealt with the details of how that works. (We have an 8.19 kW system, so I was aware of that option.) You are also correct that PEPCO does seem to have a higher application fee at 10 kW - anything under is free.

 

[willow_hiller]

You are correct on the SREC policy, except that what I've seen is that under 10 kW can (not must) use PVWatts. As we are still waiting for PTO, we have not yet registered our system so I have not yet dealt with the details of how that works. (We have an 8.19 kW system, so I was aware of that option.) You are also correct that PEPCO does seem to have a higher application fee at 10 kW - anything under is free.

Thanks. If you don't mind me asking, what was your rough timeline? From quote to estimated PTO, about how long will that take?

 

[roofsolartexas]

Get the most solar you can get, and forget about it. The days when its cloudy, that extra solar KWH you decide to get will be worth it. You may have too much on a sunny day, but then just enough on a cloudy day.

 

[wjgjr]

Thanks. If you don't mind me asking, what was your rough timeline? From quote to estimated PTO, about how long will that take?

Can't give you a great answer, since we started the process last year and were close to moving forward with the v2 tiles and a fall install. Then, after the v3 tiles came out and they stopped the v2 installs, we had to wait until this year for them to have the tiles and for decent weather. That said, we are now 2 months out from the install, and it is looking like another 6 weeks (about 3.5 months total) before we will have PTO, as both the county and utility had to have revisions to the plans submitted and approved.

 

[willow_hiller]

Does anyone know why Tesla's SREC purchase offer is so ridiculously low? $170 per kW up front, where Maryland SRECs are currently trading for ~$70 per MWh.

A medium system would yield $1,292 up front if you sell Tesla your SRECs. But the SRECs will trade for at least $1,416 in the first 2 years, and will probably be worth over $10,000 over the 25 year life of the system.

 

[wjgjr]

Does anyone know why Tesla's SREC purchase offer is so ridiculously low? $170 per kW up front, where Maryland SRECs are currently trading for ~$70 per MWh.

A medium system would yield $1,292 up front if you sell Tesla your SRECs. But the SRECs will trade for at least $1,416 in the first 2 years, and will probably be worth over $10,000 over the 25 year life of the system.

Unfortunately, that will not be the value in MD, based on the current legislation (and note, it could change - they recently extended the length of the SREC program but also lowered the caps.) The legislation sets an effective maximum value on an SREC each year, which for 2020-2030 are $100, 80, 60, 45, 40, 35, 30, 25, 25, 22.5, 22.35. (The reason these are the effective max is that these are the penalty rates the utilities would pay for each MW of solar they are short, so they have no reason to pay more for an SREC.) SRECTrade | SREC Markets | Maryland | MD is the source for those numbers.

So, assuming the $70 holds for the next two years and then trades at the maximum after, each MW generated for the first time this year would earn about $445 or about $375 if generated for the first time next year (and if the roof is not up yet, good chance most will not be generated for the first time until next year.) In addition, this assumes the same generation every year, and it will, on average, be expected to decline several percent by year 10. And, the amounts in future years really should be discounted to account for inflation. On top of that, there are costs associated with selling the SREC. I cited SRECTrade above, and they take a 7% cut, minimum of $2.50, for each SREC sold.

So, with all of that factored, if a system got PTO on 1/1/21, and a 2% inflation rate, the present value of each MW Tesla is estimating for year 1 is probably worth a little over $300 at maximum, but it could be less if credits trade below the maximum (which is plausible since utilities get no benefit by buying at the maximum, so they may expect some discount.) Whether or not $170 seems reasonable is debatable (and in MD it is also really time-sensitive since most of the value is in the next couple years,) but it seems like it is not entirely absurd and may interest individuals who don't want to deal with trading themselves, are risk averse, or who need the credits paid upfront to pay for the system.

We kept our credits (and are really hoping to get PTO by 6/30) because it seemed far more likely than not to be the better choice (and not likely that we could lose significantly) but I could see where others could come to the opposite solution for them.

I will also note this is part of why solar can be so complicated and frustrating. There are a ton of programs - federal, state, local, utility - that provide a variety of incentives that can be hard to figure out as well as rules and regulations that trip up companies like Tesla, let alone consumers. Much of it is well-intentioned, but I think it makes the prospect of going solar overly complex for some.

 

[willow_hiller]

Unfortunately, that will not be the value in MD, based on the current legislation (and note, it could change - they recently extended the length of the SREC program but also lowered the caps.) The legislation sets an effective maximum value on an SREC each year, which for 2020-2030 are $100, 80, 60, 45, 40, 35, 30, 25, 25, 22.5, 22.35. (The reason these are the effective max is that these are the penalty rates the utilities would pay for each MW of solar they are short, so they have no reason to pay more for an SREC.) SRECTrade | SREC Markets | Maryland | MD is the source for those numbers.

So, assuming the $70 holds for the next two years and then trades at the maximum after, each MW generated for the first time this year would earn about $445 or about $375 if generated for the first time next year (and if the roof is not up yet, good chance most will not be generated for the first time until next year.) In addition, this assumes the same generation every year, and it will, on average, be expected to decline several percent by year 10. And, the amounts in future years really should be discounted to account for inflation. On top of that, there are costs associated with selling the SREC. I cited SRECTrade above, and they take a 7% cut, minimum of $2.50, for each SREC sold.

So, with all of that factored, if a system got PTO on 1/1/21, and a 2% inflation rate, the present value of each MW Tesla is estimating for year 1 is probably worth a little over $300 at maximum, but it could be less if credits trade below the maximum (which is plausible since utilities get no benefit by buying at the maximum, so they may expect some discount.) Whether or not $170 seems reasonable is debatable (and in MD it is also really time-sensitive since most of the value is in the next couple years,) but it seems like it is not entirely absurd and may interest individuals who don't want to deal with trading themselves, are risk averse, or who need the credits paid upfront to pay for the system.

We kept our credits (and are really hoping to get PTO by 6/30) because it seemed far more likely than not to be the better choice (and not likely that we could lose significantly) but I could see where others could come to the opposite solution for them.

I will also note this is part of why solar can be so complicated and frustrating. There are a ton of programs - federal, state, local, utility - that provide a variety of incentives that can be hard to figure out as well as rules and regulations that trip up companies like Tesla, let alone consumers. Much of it is well-intentioned, but I think it makes the prospect of going solar overly complex for some.

Super helpful. Thank you. That does make projecting the present value of SRECs much easier.

Even with a 7% transaction fee and assuming a 2% interest rate and a 0.5% annual degradation rate, I still get a projected present value for a Medium System's SRECs of about $4,500. I would need to assume an interest rate of about 20% before the present value of the system's SRECs works out to what Tesla is offering.