Tesla Solar (Residential) Roof: Was It Worth It?

[Midpack]

Not trying to persuade anyone for or against solar - just thought it was great information for other solar nerds already interested.

I am NOT an expert on residential solar, but I have been interested and reading about it for decades. This is the most informative and complete review of a residential solar installation I can remember if you’re interested in investing 30 minutes. He answers most of the questions I would have.

Caveats:

• This is the most expensive bleeding edge system I have ever seen. Nothing like the solar panel only, no battery systems most people consider. He acknowledges same during the video, but the math includes all that.
• He drives an EV and charges at home, that is the highest electricity demand item he has by far - so that influences all his results. If you had a system without EV home charging, the math would be significantly different.
• I for one believe net metering shouldn’t and won’t always be available to homeowners, that will change everything for these grid connected systems!
• I am a fan of MKBHD for all his tech videos. I suspect some here have watched MKB videos on consumer electronics before, he’s been on YT for many years.

 

[njhowie]

I've been interested in solar for a very long time, same with DW. However, we have never been able to make the numbers work, the payback/break even period was always on the order of 20 years. There were also a number of important assumptions built in to that. Some utility companies have begun implementing a monthly minimum charge, usually $25, just for having service. This changes some of the dynamics, as this is a mandatory additional $300 annual expense for those who say their solar completely covers their electricity bill. This is similar to some states that have begun charging EV folks an additional annual fee because they don't pay taxes at the gas pump as everyone else does, which subsidizes maintenance of roadways, which they are still utilizing.

Going forward, who knows how the landscape will change? Every time we visited/revisited the topic, we decided it's not worth the investment for having to make a permanent change to our home/property and an unknown future from the financial side.

 

[Zathras]

There are a lot of aspects that vary from state to state, utility to utility and house to house.
For us, we ran the numbers prior to buying solar. We took the total cost of the panels, labor, expected replacement costs over a 30 year expected lifetime.
Prior to any rebates, the cost per kWh worked out to be $0.115/kWh, or 11.5 cents per kWh.

We basically approached it as locking in our price of electricity for 30 years as retail cost of electricity was 11 cents/kWh. Now, 7 years later, retail electricity is 14cents/kWh.

With rebates, bonuses, but not net metering, our cost for electricity dropped to about 6cents/kWh. So if net metering disappeared the day after our panels started producing, we would still come out ahead.

With net metering our price dropped to 1.5cents/kWh. This was if the retail price of electricity never went up.
We also didn’t calculate the panels keeping our house cooler in the summer.

That said, solar has a long term benefit. If you don’t know if you are staying in the same place for years, it may not make financial sense.

 

[Fermion]

There are a lot of aspects that vary from state to state, utility to utility and house to house.
For us, we ran the numbers prior to buying solar. We took the total cost of the panels, labor, expected replacement costs over a 30 year expected lifetime.
Prior to any rebates, the cost per kWh worked out to be $0.115/kWh, or 11.5 cents per kWh.

We basically approached it as locking in our price of electricity for 30 years as retail cost of electricity was 11 cents/kWh. Now, 7 years later, retail electricity is 14cents/kWh.

With rebates, bonuses, but not net metering, our cost for electricity dropped to about 6cents/kWh. So if net metering disappeared the day after our panels started producing, we would still come out ahead.

With net metering our price dropped to 1.5cents/kWh. This was if the retail price of electricity never went up.
We also didn’t calculate the panels keeping our house cooler in the summer.

That said, solar has a long term benefit. If you don’t know if you are staying in the same place for years, it may not make financial sense.

Curious if you also included the value of the money you spent on the system initially? If you had $30,000 invested in a solar system, with today's interest rates of about 5%, that would generate $1500 a year in easy no risk interest, which could offset the cost of electricity like a solar panel system does, but with no hassle or maintenance

 

[Zathras]

Curious if you also included the value of the money you spent on the system initially? If you had $30,000 invested in a solar system, with today's interest rates of about 5%, that would generate $1500 a year in easy no risk interest, which could offset the cost of electricity like a solar panel system does, but with no hassle or maintenance

No, I didn't.

If I had, my solar array produces more funds than that $1500, however, that is due to savings, and incentives, which may or may not exist for everyone.

I know some people do, which is their prerogative.
I also don't do that comparison when buying a car, house, or 12 pack of cola.
As far as 'hassle', I'd suggest maintaining investments are far more of a hassle than maintaining solar panels

 

[Time2]

I haven't watched the video, but I do have a case for net metering. The government is pushing for electric vehicles at the demise of the ICE. As far as the info I have we are already a little short of electricity production. If people got paid a reasonable number for the excess they produce, I believe that would help more people to have a reasonable payback period and be a good incentive for people to purchase solar. I have just not been convinced it is a good investment of my money.
As I understand, states have different rules, and I don't know what states provide net metering incentives to make it worthwhile.

 

[steady saver]

Thank you; will read. We've been interested in solar a long time. (And just bought a Tesla!)

 

[Midpack]

I haven't watched the video, but I do have a case for net metering. The government is pushing for electric vehicles at the demise of the ICE. As far as the info I have we are already a little short of electricity production. If people got paid a reasonable number for the excess they produce, I believe that would help more people to have a reasonable payback period and be a good incentive for people to purchase solar. I have just not been convinced it is a good investment of my money.
As I understand, states have different rules, and I don't know what states provide net metering incentives to make it worthwhile.

While residential solar was/is a small part of total generation the case for net metering as an incentive was arguably feasible, that doesn’t hold up as more adopters join in.

Consider the extreme long case. Not that everyone will but for illustration let's say everyone adopts residential solar, and everyone generates enough from their solar panels to net zero $ cost. They’re actually getting credits during peak solar hours, and using from the grid at night or low solar gain periods. They net zero, but they rely on the grid to do so, often completely reliant for many hours in a day. Who pays for the considerable cost of power plants, the grid and maintaining same if users are paying net zero? As long as homes expect full power from the grid whenever residential solar isn't generating, we're all going to have to pay the same capital costs for power plants and the grid - the only utility savings are fuel (nat gas, coal, etc.). In fact, the fixed cost of (existing) power plants plus home solar will be higher than just power plants...so fuel savings may be offset by increased capital costs per MwH.

The video confesses at the end that while he's not paying any net usage costs, he pays $5.75/mo for being connected to the grid. Where I live the base cost is $45/mo even if we use no electricity whatsoever. If we're all paying net zero, those fixed costs will get VERY high - and that's not fair for all users to pay the same fixed cost irrespective of usage.

 

[fosterscik]

Nice video - thanks for posting. He's a great presenter. I will have to explore more of his videos

 

[homestead]

Watched that video. He spend 93k after the tax credit for the system. Noticed he didn't say how much he spend on electricity before installing the system and what he could have made investing the 93k elsewhere.
May have been an ok decision for him since he can make $$ of the video now.

 

[Midpack]

Watched that video. He spend 93k after the tax credit for the system. Noticed he didn't say how much he spend on electricity before installing the system and what he could have made investing the 93k elsewhere.
May have been an ok decision for him since he can make $$ of the video now.

I’m not sure I follow those who want to apply investment returns to one side only. In the first place we don’t know how he is paying for the installation, lump sum or amortized. He could indeed calculate what he might have made investing the 93K instead, but then he’d also want to calculate what he’ll make investing the 93K he saves in utility expenses over the next 10 years too as an offset to be consistent…

 

[skyking1]

I've got about 21k in parts for the system that I'm intending to put on our new house. That will be 14.4 kilowatt.
I don't have any storage planned, at least beyond a very minimal amount of batteries and a small inverter to keep the food from spoiling.
That's really the only thing I want to have work is light switches and keep the food from spoiling in a power outage.
We too are saddled with a very high base rate of $40.50 to say hello to the power company.

 

[ERD50]

Watched that video. He spend 93k after the tax credit for the system. Noticed he didn't say how much he spend on electricity before installing the system and what he could have made investing the 93k elsewhere.
May have been an ok decision for him since he can make $$ of the video now.

I’m not sure I follow those who want to apply investment returns to one side only. In the first place we don’t know how he is paying for the installation, lump sum or amortized. He could indeed calculate what he might have made investing the 93K instead, but then he’d also want to calculate what he’ll make investing the 93K he saves in utility expenses over the next 10 years too as an offset to be consistent…

Interesting video, I FFW through most of it to get to the numbers, and I don't get it.

So he says he has paid zero last year, and the 54 MW-hr he would have used would have cost ~$9600 ( @ 20 min mark). That works out to ~ $0.177/kWh, higher than average, but I found it's about right for New Jersey. But 54MW-hr? That's a LOT of electricity. He mentions the EV, but generously assume 2x the national average driving, and say he drives 24,000 miles, and assume a higher 400 Wh per mile, that's 9600 annual kWh, round up to 10 MW-hr, that leaves 44MW-hr for the rest of his home (heated by gas), which is 3.67 MW-hr per month. And that's 4.4x average home in NJ (not including the EV). If those are his numbers, those are his numbers. But how do you use that much? If he heats his home with gas, than surely something like a pool heater would be gas as well? Where is it going?

Maybe some conservation measures would have provided much better payback?

https://www.energysage.com/local-data/electricity-cost/nj/

Monthly electric bills are a product of how much electricity you use per month and your electric rate. In New Jersey, the average monthly electric bill for residential customers is $154/month, which is calculated by multiplying the average monthly consumption by the average electric rate: 832 kWh * 18 ¢/kWh.

As far as investing the savings, OK. ... I'm going to do a spreadsheet before I double count something, maybe after dinner...

And I can't help but comment on the $30,000 that average Americans paid so that this obviously rich guy can burn up that much electricity a month, and pay zero $. Better if the government just did a commercial sized solar install (more energy for the $), and fed the grid to lower everybody's bill!

-ERD50

 

[ERD50]

OK, spreadsheet done before dinner.

Takes 12 years for the solar savings (invested) to grow beyond the original $92,797 (invested). Takes 15 years if you leave that calculation out on the savings. Takes 17 years if you add back in the subsidy.

And, I suppose there are ongoing costs - are batteries degrading by 12 years? I've heard that inverters aren't lasting so long. And solar cells do degrade a bit over time. So anything at all will push that out a bit further.

I skimmed the first few pages of comments, can't believe I didn't see anything on the 54 MW-hr consumption!

-ERD50

 

[steady saver]

Thanks for posting. We listened to it on the drive to the farm.
We're both tempted but want to have some sense that we'd install it where we'd actually plan to live long term and that's still up in the air.
I'd be curious how resale value is on a home that has the Tesla power wall and Tesla roof installed...not enough data yet I'm sure but it's something to think about. You'd want to consider the additional cost and if that would be a selling point (and worthy of a higher selling price) or if it would be a negative for some people. Though honestly, I can't see enough downsides that outweigh the upsides. But that's another rabbit hole to follow.

 

[NW-Bound]

Interesting video, I FFW through most of it to get to the numbers, and I don't get it.

So he says he has paid zero last year, and the 54 MW-hr he would have used would have cost ~$9600 ( @ 20 min mark). That works out to ~ $0.177/kWh, higher than average, but I found it's about right for New Jersey. But 54MW-hr? That's a LOT of electricity. He mentions the EV, but generously assume 2x the national average driving, and say he drives 24,000 miles, and assume a higher 400 Wh per mile, that's 9600 annual kWh, round up to 10 MW-hr, that leaves 44MW-hr for the rest of his home (heated by gas), which is 3.67 MW-hr per month. And that's 4.4x average home in NJ (not including the EV). If those are his numbers, those are his numbers. But how do you use that much? If he heats his home with gas, than surely something like a pool heater would be gas as well? Where is it going?

Maybe some conservation measures would have provided much better payback?

https://www.energysage.com/local-data/electricity-cost/nj/

I watched the video, and was also surprised that he used that much energy. Perhaps his 54,380 kWh/year was due to his EV. I dunno. Perhaps his appliances are grossly inefficient, and his ACs need freon.

But there are some other things that caught my attention, and that's because I built my own solar system+battery. It's regarding the solar generation rather than consumption, because the latter varies with the home owner's habit and his home appliances.

At 9:49, he observed that while his panels were rated for 29.3 kW of output, his phone app showed as high as 38.6 kW output at some point. He made the comment that the number may or may not be accurate.

Wow, that's 31.7% higher actual output than the rating. How is that possible? Let me explain something that laymen may not know.

In order to compare apple to apple, a solar panel output is measured at the standard conditions of 1000 W/square meter solar radiation, and at the panel temperature of 25C (77F). The less bright the sunlight, of course the lower output. And the hotter the panel above 77F, the lower its output. And of course the sunray has to hit the panel squarely at 90-deg incidence.

And then, you have losses in the MPPT electronic circuit. The MPPT circuit is a must, because without it the loss due to the panel characteristics being mismatched to the load would be huge. And then, you also have losses in the inverter that converts the DC power to the AC voltage.

My own experience is that in the summer, I could get only about 75% of the DC power from my panels, even at the time the sunray hits them squarely. And much of that was due to the panel temperature which I have clocked at 170F (77C), the wiring loss, and the loss of the MPPT charge controller. Going from the DC battery voltage to the AC power costs some more, but this is hard to boil down to a number because it varies with the load.

And that's the highest power output at noon. What is the energy collected throughout the day?

At 12:30, the YouTuber said his system was producing as much as 300 kWh/day.

How is that possible? To produce 300 kWh/day with a 29.3 kW solar array, you would need 10.24 hours of bright sunlight, and the sun has to park above the panels for all that time. Is the day longer where he is, or does the sun stop moving for 10 hours when it was above his house? It did not make sense.

Here's a plot of the sunlight intensity at my place, on 6/22/2023. I did not use data on 6/21, the summer solstice, because there was high cloudiness that day. The maximum solar radiation was 1023 W/sq.m., right at the specification condition. How about that?

I had a bit more than 10 hours of sunlight, but as physics dictate, the sun did not stay overhead the whole day. The sun angle did change, and the average intensity was of course a lot less than 1023 W/sq.m.

And he does not have a home where the entire roof faces south for an optimum orientation. The north/east/west facing sections will generate less than the south face.

In conclusion, I don't understand the data this guy shows. It's not his fault, of course.

 

[Montecfo]

OK, spreadsheet done before dinner.



Takes 12 years for the solar savings (invested) to grow beyond the original $92,797 (invested). Takes 15 years if you leave that calculation out on the savings. Takes 17 years if you add back in the subsidy.



And, I suppose there are ongoing costs - are batteries degrading by 12 years? I've heard that inverters aren't lasting so long. And solar cells do degrade a bit over time. So anything at all will push that out a bit further.



I skimmed the first few pages of comments, can't believe I didn't see anything on the 54 MW-hr consumption!



-ERD50

Yeah probably replacing components and reduced efficiency pushes it out further.

 

[JoeWras]

Thanks for posting. We listened to it on the drive to the farm.

We're both tempted but want to have some sense that we'd install it where we'd actually plan to live long term and that's still up in the air.

I'd be curious how resale value is on a home that has the Tesla power wall and Tesla roof installed...not enough data yet I'm sure but it's something to think about. You'd want to consider the additional cost and if that would be a selling point (and worthy of a higher selling price) or if it would be a negative for some people. Though honestly, I can't see enough downsides that outweigh the upsides. But that's another rabbit hole to follow.

Our neighbor sold his home. It has a Tesla power wall, but a different conventional small rooftop array. His investment was in the low 20s.

It is really hard to tell what the effect was on the price in this dynamic housing market. I will guess about $15k. Not bad. The new owners had to have the arrays temporarily removed for roof work. So that was a new cost. Perhaps that was part of the calculation for the sale price.

 

[ERD50]

I watched the video, and was also surprised that he used that much energy. Perhaps his 54,380 kWh/year was due to his EV. I dunno. ...

As I pointed out earlier, even allowing for a 2x average miles, using 24,000 miles per year at a generous 400 Wh/mile, that accounts for less than 10MW-hr/year. the remaining 44MW-hr is still 4~5x the average for his area.

The fact that his only explanation of this outlier number is his EV (which as I just pointed out, doesn't cover it), it makes me very skeptical of his whole presentation.

BTW, I try to remember to change youtube comments (and Amazon reviews) from 'top comments' to "most recent". I don't trust their algorithms to choose 'top' for me. This guy is giving Americans a bad name, lots of comments like "Whoa - you Americans sure use a lot of energy! Here in Europe, we use 20% of that, with electric heat!"

One commenter came up with an explanation for the extreme use:

@user-im9zp4yp9x 200kwh a day? Dudes mining crypto ?

Hmmmmm?

....

At 9:49, he observed that while his panels were rated for 29.3 kW of output, his phone app showed as high as 38.6 kW output at some point. He made the comment that the number may or may not be accurate.

Wow, that's 31.7% higher actual output than the rating. How is that possible? ...

At 12:30, the YouTuber said his system was producing as much as 300 kWh/day.

How is that possible? To produce 300 kWh/day with a 29.3 kW solar array, you would need 10.24 hours of bright sunlight, and the sun has to park above the panels for all that time. Is the day longer where he is, or does the sun stop moving for 10 hours when it was above his house? It did not make sense. ...

And he does not have a home where the entire roof faces south for an optimum orientation. The north/east/west facing sections will generate less than the south face.

In conclusion, I don't understand the data this guy shows. It's not his fault, of course. ...

I was so focused on the unexplained consumption, I hadn't looked at the production side.

One reach-out explanation would be - since the nature of roof tiles means they will not be at the optimum angle/azimuth and may be subject to some shading, perhaps Tesla under-rates them, so customers don't complain? Not typical for Musk, he generally over-promises, but they may have considered this. For some history, James Watt did this - his 1 HP steam engine would do much more work than 1 horse, he didn't want to have to make excuses on a new technology, he wanted customers to be 'wowed".

So maybe those numbers aren't his 'fault', but for all the time effort he put into this, it should have been explained. Again, not explaining it has left me skeptical of everything.


And just look at all the gushing positive comments. The problem is, the average person has no idea how to validate what has been presented. They eat up the sizzle, not the steak. Sad that people are so easily misguided.

-ERD50

 

[ERD50]

.... He drives an EV and charges at home, that is the highest electricity demand item he has by far - so that influences all his results. If you had a system without EV home charging, the math would be significantly different. ...

How did you come to this conclusion? My generous math/estimates show the EV was maybe a 20% factor overall. It might be his single largest use, but we don't have enough info to even determine that, I don't think. Could be his A/C, or ? Hard to imagine this level of consumption (outside of crypto-mining!).

Are my assumptions way off? Is my math off? I had enough caffeine at that point, so if I made a mistake, I have no excuses!

What assumptions did you make? Did you find it odd that the only explanation for such extreme consumption was the EV, and he gave no actual numbers for that?

No, I'm not impressed with this guy, at least not based on this video - too much hand-waving for me.

-ERD50

 

[NW-Bound]

As I pointed out earlier, even allowing for a 2x average miles, using 24,000 miles per year at a generous 400 Wh/mile, that accounts for less than 10MW-hr/year. the remaining 44MW-hr is still 4~5x the average for his area.

The fact that his only explanation of this outlier number is his EV (which as I just pointed out, doesn't cover it), it makes me very skeptical of his whole presentation.

I live in Phoenix, where the record high was 122F. My annual consumption prior to my DIY solar installation was 16 MWh/year (vs his 54 MWh). And my home is all electric, while he has gas for heating in the summer.

It's mind boggling.

One reach-out explanation would be - since the nature of roof tiles means they will not be at the optimum angle/azimuth and may be subject to some shading, perhaps Tesla under-rates them, so customers don't complain? Not typical for Musk, he generally over-promises, but they may have considered this. For some history, James Watt did this - his 1 HP steam engine would do much more work than 1 horse, he didn't want to have to make excuses on a new technology, he wanted customers to be 'wowed".

I thought about this possibility. Perhaps they gave him much more than 29 kW worth of solar panels? The entire roof was covered with solar tiles, and they miscalculated its area? Or they knew he was an influencer, so did this purposely?

I guess one can search YouTube for other home installations to see what they say they get for their system. A kWh is the same whether it comes from a Tesla installation or another brand. And the sun does not know who is receiving its ray.

 

[NW-Bound]

I think this YouTuber should invite NREL (National Renewable Energy Lab) to come study his power usage and his solar system.

It would make an excellent case study, being an outlier in so many aspects.

 

[ls99]

I skimmed through the video. Way too much talking.

My house in SW PA is all electric except for heat, which is oil.
I run dehumidifier in basement most of the year, AC in summer, sauna with 3KW heater running 4 to 5 times a week 2 to 3 hrs per session, (my time inside about 25 minutes) year around.

Typical yearly electric use per power company records varies 6902 to 7503 KWH yearly. Most lighting is by incandescent bulbs, CFLs and fluorescent in garage.

 

[Ian S]

When I installed solar on my Phoenix home in 2012, rather than a zero electric bill, I was more interested in the best bang for the buck. At that time, this meant offsetting most (preferably all) summer peak hours usage using the net metering in use at the time while minimizing the end of year excess kWhs in your account that were paid off in peanuts ~$0.03 per kwh. In the seven years that I owned the home, this worked well as all my peak usage (@$0.25/kWh) was eliminated and so was much of the lower cost usage. I ran spreadsheets that showed I was saving about $1500 (out of an annual electric cost of about $2200 per year prior to solar) with a 6.9 kW Sunpower system. Anyone interested in solar really has to run the numbers for your particular home! Also, you really should first look at making your home more energy efficient. I didn't do that with my home in Phoenix because I got an unbeatable deal on the solar (<5 years payback) and didn't think so much about, for example, installing more energy-efficient windows.

As far as this video is concerned, there really seem to be some strange numbers. Here's a recent CNET article on the Tesla Solar Roof https://www.cnet.com/home/energy-an...f-is-the-sleekest-solar-option-your-best-one/ and the cost of it:

If your goal is to generate solar for the least amount of money possible, a Solar Roof isn't for you. Tesla's online estimate for a house in Dallas offered 12.46 kilowatts of solar generation and a Powerwall battery for $168,400. Tesla estimates 12 kilowatts of regular solar panels and a Powerwall at the same location will cost $26,966.

This is way more than the cost claimed in the video for a system and I can't believe it costs that much more per kW to install a system in Dallas than in New Jersey. What am I missing?

 

[njhowie]

This is way more than the cost claimed in the video for a system and I can't believe it costs that much more per kW to install a system in Dallas than in New Jersey. What am I missing?

The guy in the video cost was in November 2020. We're now almost 3 years down the road. Also, as he indicated in the video, Tesla has not been making a big push behind the solar tiles, periodically stopping sales. Maybe they raised the prices since then so they could be more profitable and make it worth their while to continue offering it to those who really want it?