Solar photovoltaic on the house as an investment

[NW-Bound]

not yet. Tesla wants you to believe that its new battery is the answer. But I think an affordable solution is likely in the next decade.

Speaking of batteries, the just announced 7-kWh residential battery by Tesla stores enough energy to run my AC for 1 hour, at a cost of $3000. How many charge/discharge cycles does it last?

Even without solar, I can use it to store off-peak energy at 7.41c and use it during on-peak which would cost me 22.26c, for a saving of 14.85c/kWh. This 7-kWh battery would save me $1.04 per day per battery. And I can use about 7 of these batteries (costing $21K).

If the above battery lasts 10,000 cycles (30c per day), it makes sense to use it to save $1/day in electricity. If it lasts only 1,000 cycles ($3/day), then it does not. Or if perhaps the price can drop down to $300, the economics would also work.

Maybe in the next 10 years...

The other side of the coin is that rooftop solar reduces the need for utilities to build new power plants, thus all customers money. Our power bill has fuel costs separated from base rate charges that all customers pay. I think that is likely the fairest way to pay for overhead.

I do not remember if I showed in this or another thread that here in the SW, the highest electric usage is for AC in the summer. And the demand runs flat out to 9PM, way after sunset because the temperature does not drop until late at night, after all panels have stop producing. So the peak demand is the same as it was before any solar panel.

As I explained in an earlier post, my local utility is now allowed to have a different contract for residential solar customers to recapture the cost via a somewhat complicated formula.

PS. In Gila Bend, there is a thermal solar plant called Solana. It uses parabolic trough mirrors to melt a salt, which is then used to run turbines. The salt can store enough heat to produce power for a couple of hours past sunset. This meets the flat demand to 9 PM.

 

[meierlde]

Thanks for the reminder about roof replacement etc. I'll have to get DD to explore those costs. Her roof is ~10 years old so 10 -15 years left. My quick googling suggests remove and reinstall costs are in the $1000 range for a small array like she wants.

One possibility would be to go with a metal roof in the process, they have a 50 year life and hail resistance to boot. (double the price of a regular roof, but also less likley to blow off if the standing seam variety as one sheet of roofing runs from eave to peak)

 

[meierlde]

Maybe in the next 10 years...



I do not remember if I showed in this or another thread that here in the SW, the highest electric usage is for AC in the summer. And the demand runs flat out to 9PM, way after sunset because the temperature does not drop until late at night, after all panels have stop producing. So the peak demand is the same as it was before any solar panel.

As I explained in an earlier post, my local utility is now allowed to have a different contract for residential solar customers to recapture the cost via a somewhat complicated formula.

PS. In Gila Bend, there is a thermal solar plant called Solana. It uses parabolic trough mirrors to melt a salt, which is then used to run turbines. The salt can store enough heat to produce power for a couple of hours past sunset. This meets the flat demand to 9 PM.

Actually the Ercot sites load curves (the Texas Grid shows peak usage around 5 pm the last couple of days with around a 10% drop by 8 pm. (This is likely the dinner cooking and dishwasher load).
Now of course there is a solution to this which would take some space put in large water tanks and chill the water in the morning and use it to cool from 6 pm a 10,000 gallon tank is about a 12 foot cube for example and could store 80,000 btu per degree. If you take it from say 72 to 35 you get about 3 million btu stored or a 1000 gallon tank 300,000 btu which would provide about 3-5 hours of ac type cooling. (add antifreeze and you could get more heat storage). The question is of course the economics of the situation. as the technology itself is well proven chilled water for example is used in most larger buildings the main hvac makes chilled water which is piped thru the building to the air handlers.

 

[NW-Bound]

Yes, the dinner preparation as people arrive home from work adds to the peak demand. I was making a general observation that the power demand does not drop off at the same time as the sun disappears below the horizon.

About chilled water tanks, large industrial or commercial campuses have been using them forever. They also use evaporative precoolers to enhance the effectiveness of their ACs.

 

[ERD50]

...

The other side of the coin is that rooftop solar reduces the need for utilities to build new power plants, thus all customers money. Our power bill has fuel costs separated from base rate charges that all customers pay. I think that is likely the fairest way to pay for overhead.

I really don't think solar will reduce the need for power plants for a very long time, certainly not enough for customers to see any savings.

As NWB pointed out, the peak usage can last long into dusk. At this time, or any overcast day, power is needed to avoid brown-outs. So those power plants are needed. Until we get storage, but I've pointed out before, days of storage for days of overcast periods is a HUGE amount of storage.

The US presently gets 0.4% (yes, ZERO dot four) of its energy from solar PV. We are talking HUGE increases in installed levels before we start making a dent in power plant requirements, even when the sun shines.

What is U.S. electricity generation by energy source? - FAQ - U.S. Energy Information Administration (EIA)

-ERD50

 

[meierlde]

Yes, the dinner preparation as people arrive home from work adds to the peak demand. I was making a general observation that the power demand does not drop off at the same time as the sun disappears below the horizon.

About chilled water tanks, large industrial or commercial campuses have been using them forever. They also use evaporative precoolers to enhance the effectiveness of their ACs.

Looking at the ercot site when the sun goes down the wind begins to blow more in Tx The highest wind power levels tend to come in the early morning, the lowest about mid day. So wind may be a good supplement to solar from a utility point of view. IMHO much of west Tx and Eastern NM really is also good solar country, and there is little scenery to spoil. Now utility solar tends to make the ground underneath much less useful but then again in those areas for pasture it is acres per cow.

 

[NW-Bound]

Looking at the ercot site when the sun goes down the wind begins to blow more in Tx The highest wind power levels tend to come in the early morning, the lowest about mid day. So wind may be a good supplement to solar from a utility point of view...

I have read that wind power and solar complement each other, but have not dug down to see the data.

But is it guaranteed that wind will always pick up at sunset? Even if there's a chance of 1 in 1000 that it does not, well, we will have a brownout about 1 every 3 years.

In this 115-120 heat without A/C, weaker people will drop like flies. I may be among those.

 

[ERD50]

I have read that wind power and solar complement each other, but have not dug down to see the data.

But is it guaranteed that wind will always pick up at sunset? Even if there's a chance of 1 in 1000 that it does not, well, we will have a brownout about 1 every 3 years. ...

Exactly the point that so many miss. And of course it is probably more like an occurrence several times a year. So we need just as much peaker plant capacity as we have solar/wind.

People talk about solar getting to be cheaper than coal - is that the case after you pay for the peaker plants to carry you through? I doubt it.

Coal (as ugly as it is) only needs peakers for peaks. Up times are very high, so other coal plants and existing peakers can cover the occasional maintenance shut down. They don't shut all the coal plants down at the same time for maintenance, peakers can cover the one shut down while another 20 on that grid keep going, but Mother Nature is not so compliant.

-ERD50

 

[Texas Proud]

I have read that wind power and solar complement each other, but have not dug down to see the data.

But is it guaranteed that wind will always pick up at sunset? Even if there's a chance of 1 in 1000 that it does not, well, we will have a brownout about 1 every 3 years.

In this 115-120 heat without A/C, weaker people will drop like flies. I may be among those.

Heck, if you go to London people are dropping when it only hits 90....

 

[NW-Bound]

In heat waves in years past, people in NYC also dropped at temperatures approaching 100F.

But here, we can stand up to 115F. It's dry heat, you see.

 

[NW-Bound]

Actually the Ercot sites load curves (the Texas Grid shows peak usage around 5 pm the last couple of days with around a 10% drop by 8 pm. (This is likely the dinner cooking and dishwasher load).
Now of course there is a solution to this which would take some space put in large water tanks and chill the water in the morning and use it to cool from 6 pm a 10,000 gallon tank is about a 12 foot cube for example and could store 80,000 btu per degree. If you take it from say 72 to 35 you get about 3 million btu stored or a 1000 gallon tank 300,000 btu which would provide about 3-5 hours of ac type cooling. (add antifreeze and you could get more heat storage). The question is of course the economics of the situation. as the technology itself is well proven chilled water for example is used in most larger buildings the main hvac makes chilled water which is piped thru the building to the air handlers.

I double-checked your numbers , and indeed a 1000-gal container of water (5' x 5' x 5') can store 342,000 BTU of cooling which can tide me over the peak hours of higher electric rate. Assuming a reasonable EER of 10 for the AC, by cooling this water using off-peak rates, I will save about $5/day.

But the above is only over the 3 hottest summer months. Over a year, it is a bit harder to compute the savings.

 

[Nords]

For those interested the protected socket idea here's some blurb from the manufacturer:
Sunny Boy 3000TL-US
SMA has recently unveiled its newest line of grid-tie inverters with a variety of new and improved features to help system owners get the most out of their systems. These new UL certified inverters continue to raise the benchmark in solar inverter technology. The Sunny Boy 3000TL-US comes standard with an impressive list of features.

Transformerless for higher efficiency and lighter weight
Dual maximum power point tracking for more design options
Reduce effects of partial system shading with OptiTrac™
Secure Power Supply (SPS) provides daytime power if grid goes down
Sunny Boy 3000TL-US comes with a Secure Power Supply that is able to send up to 1,500 watts of power to a dedicated outlet in the case of daytime grid failure.

So maybe 13.5 Amps @ 110V on the circuit max. Maybe enough for a small window AC, but would be touch and go. But that amount of power would be very useful for all the other suggestions by Samclem.

I upgraded to the 4000-watt version of that inverter about six weeks ago, after our 10-year-old Xantrex died.

The SPS is actually just five connection points (inside the inverter's black lower panel, behind the LCD display) for you to wire in a switched receptacle. You still have to provide the switch, the receptacle, the conduit, and the remodel box to install it where you want it. Our inverter is on the outside of our garage wall, so we put the SPS receptacle on the inside of the wall. Unless you're comfortable doing that type of wiring, she might want to ask the electrician to take care of it as part of the installation.

I think it's more of a gimmick than a feature. It's only going to power your loads when the sun is out, so I wouldn't be eager to run an air conditioner on days when clouds are crossing in front of the sun. I don't think that I'd try to run a fridge on it, either. But it'll run our 100-watt pump for our solar water heater (which stays hot for a couple of days), and it'll run the air pumps for our aquarium (which will hopefully keep the fish alive overnight). But it only cost me about $10 in parts and an hour to connect all the pieces.

If she's interested in monitoring the array's performance, SMA also sells a "Webconnect Speedwire" dongle that plugs into a ribbon cable fitting (also behind the LCD display) and terminates in a couple of RJ-45 jacks. She can either add a wireless network adapter or plug a hardwired Ethernet cable into her router, download some software onto her computer, and monitor the array directly. It retails for $300-$350 and might be a useful negotiation point.

But the real use of the Webconnect box is the ability to configure the inverter directly. Here's another reason you might care about that.

In the six weeks I've had this SunnyBoy, it's tripped twice on an arc fault circuit interrupt safety feature. The AFCI system is supposed to look for problems on the DC side, but the inverter is trying to detect a brief signal up in the 100+ MHz frequency range. Ideally it's a loose wirenut with the current arcing across the wires, or a loose MC plug that's not making tight contact, or a wire that's rubbing and frayed on a sharp bend, or even a bad solar cell in a panel. However the inverter has also been known to trip on high-frequency transients from electronic dog fences, high-voltage transmission towers, A/C compressors starting or stopping, and even feedback from the local utility through your service connection. One company even noticed that their inverter tripped on AFCI whenever a warehouse ceiling exhaust fan (on a thermostat) started up. Once the inverter shuts down on an AFCI you have to manually shut it off (switch of the DC connection from the panels and open the AC breakers) and restart it (switch the DC connection back on and shut the AC breakers, then "knock" on the front panel to clear the AFCI fault). But the only way to know that you've had a AFCI trip is when you check the display panel, and in my case it took a few days before I noticed.

The AFCI detector is required by recent updates to the American electrical code, and it's stringent enough to produce more than its share of false alarms. SMA has to include it to remain code compliant, but they've also added the ability to toggle it. If you have the Webconnect system then you can reset the AFCI feature... or even shut it off. You can also have your inverter e-mail you whenever it has an AFCI trip or other out-of-spec readings.

If you have the Webconnect system you can also reset your inverter to maintain its grid connection down to as low as 57 Hz instead of the default 59.3 Hz. It turns out that our local electrical utility has a habit of dropping grid frequency lower than 59 Hz, so our Oahu PV installers have known for years to reset the low-freq trip to 57 Hz.

 

[rayvt]

days of storage for days of overcast periods is [-]a HUGE[/-] an insurmountable amount of storage.

FIFY

I was making a general observation that the power demand does not drop off at the same time as the sun disappears below the horizon.

When we lived in Chicago, there were a few summers when our AC unit ran continuously for 2 weeks. Not just "ran" --- ran 24/7. For 2 weeks straight.

Some solar would perhaps have helped during the day, but ComEd's generators would have had to carry the entire load at night. So no matter how many homes had solar, ComEd would have still needed all their plants.

I can't imagine what adding electric cars into the mix would have done. Well, actually I can -- and it isn't pretty.

 

[jimnjana]

I really don't think solar will reduce the need for power plants for a very long time, certainly not enough for customers to see any savings.

As NWB pointed out, the peak usage can last long into dusk. At this time, or any overcast day, power is needed to avoid brown-outs. So those power plants are needed. Until we get storage, but I've pointed out before, days of storage for days of overcast periods is a HUGE amount of storage.

The US presently gets 0.4% (yes, ZERO dot four) of its energy from solar PV. We are talking HUGE increases in installed levels before we start making a dent in power plant requirements, even when the sun shines.

What is U.S. electricity generation by energy source? - FAQ - U.S. Energy Information Administration (EIA)

-ERD50

That said, solar is providing nearly 20 gigiwatts to the US grid. A drop in the bucket, but still a pretty big number. That includes Utility scale solar plants. There is also a big push to get renewable projects done prior to the end of the the 30% tax credit. I think the storage issue will be solved, not sure if batteries are the solution maybe flywheel technology. Too much money to be made for a solution not to be found. I do believe Nat Gas will continue to replace coal as the primary source of electricity for years to come. I just think we are not many months away from solar and wind being economic alternative to coal or nat gas.

 

[fosterscik]

I upgraded to the 4000-watt version of that inverter about six weeks ago, after our 10-year-old Xantrex died....

...I think it's more of a gimmick than a feature. It's only going to power your loads when the sun is out, so I wouldn't be eager to run an air conditioner on days when clouds are crossing in front of the sun...

In the six weeks I've had this SunnyBoy, it's tripped twice on an arc fault circuit interrupt safety feature...

I really appreciate the comments Nords. It's great to have actual, knowledgeable, user feedback when making choices about components. Lots to think about.

 

[samclem]

. I just think we are not many months away from solar and wind being economic alternative to coal or nat gas.

As long as power plants have to be bought and paid for to take up the slack when solar and wind (other other ephemeral sources) aren't available, the price of these standby resources need to be added to the cost of the ephemeral sources. So, they'll probably remain bit players. Unless the storage problem is solved, and that won't be a matter of months, unfortunately.

 

[ERD50]

FIFY ...

I can't imagine what adding electric cars into the mix would have done. Well, actually I can -- and it isn't pretty.

Well, I don't know about 'insurmountable', you can do a lot if you throw enough time and money at it! But I'd agree that we may not have any affordable, safe, and environmentally acceptable solutions for a long, long time (so far out, that I think other solutions may make this a moot point - new nukes or other tech?).

I have imagined what will happen if you add electric cars to the mix, and I also found that it wasn't pretty. You can get more details on the old, closed 'More on the Tesla' thread - some of the EV proponents there really did not like the observations (which they failed to refute with any facts and figures, just shot at the messenger - and missed! )

I really appreciate the comments Nords. It's great to have actual, knowledgeable, user feedback when making choices about components. Lots to think about.

Yes, the fact that an inverter dies at year 10 means people do need to consider ongoing maintenance costs in their 25-30 year calculations.

And the tripping breakers and monitoring does support the idea that these may not be a great resale deal - most homeowners are going to be put off with that level of tinkering.

That said, solar is providing nearly 20 gigiwatts to the US grid. A drop in the bucket, but still a pretty big number. ...

Which is it, a 'drop in the bucket' or 'a pretty big number'? It's both (it's a dessert topping AND a floor wax!)! Which one matters? The 'drop in the bucket'!

A million teaspoons is a big number too! But it would not have bailed out the Titanic! Perspective and per cents are everything.

Here's a great read which covers this concept:

19 Every BIG helps

... I think the storage issue will be solved, not sure if batteries are the solution maybe flywheel technology. ... I just think we are not many months away from solar and wind being economic alternative to coal or nat gas.

How will the storage issue be solved (in an affordable, safe, and environmentally acceptable way)?

Months? Hmmm, maybe as long as it is small enough to not require storage and can use the current peaker capacity. But then it isn't big enough to make much difference, so it's not really that important. So now we are back to my earlier statements - if these variable renewables become big enough to displace baseline power, we need storage/backup - and that is an added cost that raises the bar considerably on being an 'economic alternative'. Is solar PLUS storage cheaper than Nat Gas?

edit: cross-posted with samclem!

-ERD50

 

[jimnjana]

As long as power plants have to be bought and paid for to take up the slack when solar and wind (other other ephemeral sources) aren't available, the price of these standby resources need to be added to the cost of the ephemeral sources. So, they'll probably remain bit players. Unless the storage problem is solved, and that won't be a matter of months, unfortunately.

I don't disagree entirely, but many of those standby generators are already bought and paid for and are in place. They just don't operate at full capacity all of the time. Likely to remain bit players for many years to come, But the growth in installed capacity is real and the economics for the first time are beginning to be such that a homeowner will not need a subsidy to consider a pv system economically feasible. The other matter besides storage is the build out of the electric distribution system to move power from renewable rich areas to where the demand is.

 

[jimnjana]

ERD50

I disagree with what constitutes "important". I think that its important to establish wind/solar as a substantial part of the countries electric generating capacity. They are not yet, and we are years away. For the first time they are both becoming economically feasible without subsidy. That is important. With nat gas prices low and supply plentiful it will remain/become the primary fuel source. Coal is the loser.

 

[ERD50]

I don't disagree entirely, but many of those standby generators are already bought and paid for and are in place. They just don't operate at full capacity all of the time. ...

Yes, the peakers are sitting idle much of the time, but it doesn't matter much to this point. When we hit the peaks for the year, they are bringing them all on-line to avoid a brown-out. Would they just sit there, when we are being threatened with a brown out? The utilities are not going to install more peak power than they need (other than a safety margin, which will need to be maintained in any case).

We can't count on solar producing at the same time as the peak. There were some earlier charts showing that peak extending past dusk. Maybe some small offset, but small.

ERD50

I disagree with what constitutes "important". I think that its important to establish wind/solar as a substantial part of the countries electric generating capacity. They are not yet, and we are years away. For the first time they are both becoming economically feasible without subsidy. That is important. With nat gas prices low and supply plentiful it will remain/become the primary fuel source. Coal is the loser.

I'm not saying that getting off coal is not 'important'. I'm saying that some of these sources simply do not offer enough power to be 'important' in contributing to getting off coal.

Again, show me that solar is economical when it becomes a big enough source to make a significant environmental difference, at which point you also need to factor in the cost of peakers/storage. Coal plants (and nukes) cannot be taken on/off line fast enough to adapt to hours/days of cloud cover, that's an important factor in all this.

BTW, you were off (low) by a factor of ~ 1000 on your estimate of solar - it's closer to 20 TWh than to 20 GWh. And it's still a drop in the bucket!

-ERD50

 

[jimnjana]

BTW, you were off (low) by a factor of ~ 1000 on your estimate of solar - it's closer to 20 TWh than to 20 GWh. And it's still a drop in the bucket!

-ERD50

Solar Industry Data | SEIA t

that is the source for the GW estimate.

This exchange reminded me of the paper I wrote in 1981 for English 101 in college. The theme for the paper was that one day computers would be in every household. The only thing the prof wrote on the paper was "they said the same thing about air planes".

 

[ERD50]

Some rough numbers to (hopefully) illustrate the peaker paradox:

Let's say on some hypothetical grid on a near-peak hot summer day the utility has cranked up their coal & nuke baseline plants to max (which takes hours) in anticipation, and they make up 60% of the power, and the other 40% is made up of NG peaker plants. Some of these peakers are large and kind of 'in-between' in response times, some are smaller and can respond very quickly. That looks like:

60% COAL & NUKES
40% NG Peakers

So now we come to some hypothetical future with enough solar to provide 30% of the power on this hot summer day, enough to avoid running coal at max. That looks like:

30% Solar PV
30% COAL & NUKES
40% NG Peakers

But now we get cloud cover, and solar drops to 5%, and we have not had time to bring coal back on line. We need an additional 25% more peaker power - and it isn't available!

And this might be a near daily event if that peak extends past dusk (some of that might be able to be accounted for by keeping the coal burning longer into the evening).

One alternative is that we simply keep the coal plants burning whenever some cloud cover is predicted. That means some of that solar will not be utilized - making the break-even point further out.

Hey, I love the idea of clean energy, but we have to face facts on what it can accomplish. I'll even pay more (how much I'm not sure) for cleaner energy - it doesn't need to be exactly on par. In fact, higher rates would promote more conservation, which would increase the ratio of renewables, and help even more. But it still is looking like a drop in the bucket for a long, long time.

edit/add: Also note that supplying 30% from solar on a summer day means that the average annual power from solar would be far less. No power at all for much of the 24 hour day, and far lower average power towards winter. So even as solar approached 10% average annual power, we might start hitting these peaker/storage limits? And 10% is ~ 25x where we are today! That's a lot more solar to get to even 10% averages, where maybe we actually get some noticeable environmental effects?

-ERD50

 

[ERD50]

Solar Industry Data | SEIA t

that is the source for the GW estimate.

OK, it looks like you are mixing the terms 'capacity' and production. They are very different numbers. One is watts, the other is annual watt-hours.

This exchange reminded me of the paper I wrote in 1981 for English 101 in college. The theme for the paper was that one day computers would be in every household. The only thing the prof wrote on the paper was "they said the same thing about air planes".

Solar PV is not comparable to computers. Moore's law does not apply, they are fundamentally different.

Why Moore’s Law Doesn’t Apply to Clean Energy Technologies : Greentech Media

Solar PV is ~ currently 15% efficient. A single cell has a theoretical efficiency of ~ 34% - we can only about double it at best (not likely), There is no such limit for computers. Even if we consider an infinite number of stacked cells and overcome every possible inefficiency (we can't), we can only get abut 6.67x better with solar.

And you still need that pesky storage.

-ERD50

 

[Texas Proud]

I don't disagree entirely, but many of those standby generators are already bought and paid for and are in place. They just don't operate at full capacity all of the time. Likely to remain bit players for many years to come, But the growth in installed capacity is real and the economics for the first time are beginning to be such that a homeowner will not need a subsidy to consider a pv system economically feasible. The other matter besides storage is the build out of the electric distribution system to move power from renewable rich areas to where the demand is.

I had a friend who worked at an electrical plant.... his plant used steam to produce their electricity... do not know what fuel used to make steam, but think it was nat gas...

He told me that they had to spend $20 million to refurb a turbine that had never been used!!! Yes, it just sat there... why? So they could use it in negotiations on purchasing power from others..

I would bet that there is good money spent on plant and equipment that is not used much.... it is there 'just in case'... so the sunk cost is real, but it is not zero going forward... and BTW, it was maned 24/7/365 'just in case'....

 

[Texas Proud]

Just a question to throw out...

How long does it take to get a coal plant running from cold

IIRC, my friend was talking like a day or two...

So, when you talk hours to produce I am assuming that it is from a running plant and you are talking increasing production....