Really enjoyed this, the Wright's Law framing makes the solar story click in a way that energy transition language never quite does. The microprocessor parallel is particularly good because it reminds you that the interesting question isn't 'how cheap does electricity get' but 'what becomes possible that wasn't before' which is a big difference.
Also can't help noticing Ashley's comment about being stuck on oil in rural Yorkshire in a house built in 1650. That gap between the supercycle thesis and the reality of actually getting off fossil fuels in cold, old, rural housing stock is where a lot of the real story lives. The curve is undeniable at a global level but the last mile is going to be a thing...
This is just great! I've always been a solar optimist - worked 50 years in radiation measurement & different aspects of solar R&D, mulled over many of the topics you touched - really appreciate your expressing all this so well. Thanks.
Terrific analysis. Thank you. I’m passing this onto one of the Solar Energy startups I find most compelling, NEXT ENERGY TECHNOLOGIES. Their prosperity technology of using SOLUBLE SMALL MOLECULES to create transparent solar cells was first developed at UC, Santa Barbara winning a Nobel Prize. Check out the company’s commercialization progress:
Excellent article. Another issue for solar is land use. Fully embracing Rooftop and covering parking lots are best ways to optimize on land use by using “wasted space” already, especially parking lots. There is a need for commercial model innovation.
Thanks! I saw that but didn't see the parking lot opportunity. Most people (business owners and elected officials) don't realize how much land is parking lots. A good reference for further opportunity. https://parkingreform.org/resources/parking-lot-map/
Good analysis, good points that are brought together. It is a complex topic.
And I add a non negligible complexity point here. Sorry for this.
There is this question of the interconnexion cost that has quadrupled since the 2000's from which 75% are due to cover network upgrade costs: this is not only an administrative hurdle to overcome as you seem to reduce it to, today it doubles the cost of each project.
It is not negligible at all: the cost of building a grid does not follow Wrights law.
This means that the interconnexion costs will stay the same and will take more and more of the total costs: It is doubling the cost of a project today, it will quadruple it quite soon.
But not all solar generated energy needs a grid. It would be good to understand the proportion of solar generated kW for the grid (China invests a lot into building a grid) and solar generated KW that are locally used (no grid need like in Pakistan).
Interconnection costs are real... but first off, many are done by us to us. (Regulation.) Second, they are independent of the solar learning curve. BTM solar or decentralised or growth anywhere in the world will bring down the panel costs. If the US, UK, or Germany doesn't want to fix their grid economics (which might include network design, regulation or market design), that is their sovereign choice.
As you say, a lot won't need grid (or won't need grids the way we think of it). It's a good challenge. Would I satisfy you if I said "more than today but less than everything?" 😄
Sure interconnection costs are independent of the pure solar learning curve. The qualifier "pure" is important here: the solar learning curve cannot be seen in isolation: it is dependent of the solar success/volume and if the interconnection costs are (what I see) a non negligible part of the total cost of ownership (TCO), it has an impact on solar production volumes.
If 75% of the interconnection costs are due to grid upgrade, this is not done by us to us, it is real costs, which impacts the solar volume, hence Wrights law. TCO higher, volume lower. Because the solar-TCO is defined at 50% by interconnection costs, you need to take these costs into account for your Wright law.
That said, not all solar panels will need grid connexion. So the real calculation depends on the proportion of grid connected solar panels.
But this throws an important question: can solar energy be the technology that allows for decentralisation of energy production? Historically energy generation needed to be centralised, because it is too expensive to build generators where the demand is. So is the need for centralised energy generation still a must? Can we rethink the whole? This is perhaps an opportunity for edge-solar energy ... like there is one for edge-AI. Think about combining both
Re: interconnection & real estate costs — a solution requiring neither!
NEXT ENERGY TECHNOLOGIES has developed carbon based semiconducting molecules that uniquely harvest light energy in the Infrared (IR) spectrum while allowing visible light to pass through and Ultraviolet (UV) light is blocked. No other solar semiconductors can do this! The 1st commercial application — windows. Learn more at https://www.nextenergytech.com/technology Thanks!
Can’t get the tool to work on my iPhone - endless log in doom cycle! Like the analysis, mirrors my musings earlier in the week. I’m in rural Yorkshire and currently stuck on oil, in a house built in 1650!
You might see a single pop-up when you arrive, but you can close it and get straight to the model - no login required. It’s also worth flipping your phone to landscape mode to get the best view of the charts. Let me know if that works for you!
Hey EV, is there a channel for user feedback and musings? Have I just missed it?
I would appreciate feedback to the things I'm thinking and am sure there are hundreds if not thousands of readers who would like this type of feature. Of course you could use your AI tools to winnow and consolidate ideas into themes?
For example, two thoughts of mine this week, ChatGPT (Pro) still getting basic facts wrong (Dublin in UK not EU) and it's increasing use of clickbait answers...
Really enjoyed this, the Wright's Law framing makes the solar story click in a way that energy transition language never quite does. The microprocessor parallel is particularly good because it reminds you that the interesting question isn't 'how cheap does electricity get' but 'what becomes possible that wasn't before' which is a big difference.
Also can't help noticing Ashley's comment about being stuck on oil in rural Yorkshire in a house built in 1650. That gap between the supercycle thesis and the reality of actually getting off fossil fuels in cold, old, rural housing stock is where a lot of the real story lives. The curve is undeniable at a global level but the last mile is going to be a thing...
This is just great! I've always been a solar optimist - worked 50 years in radiation measurement & different aspects of solar R&D, mulled over many of the topics you touched - really appreciate your expressing all this so well. Thanks.
Great stuff! Go solar go!
This might be your best essay yet! Well done.
This is a solution for the world. And the world will be so much better off.
Great analysis!
Thanks for reading!
Incredibly rigorous analysis!
It is fantastic post! Are you also fans of Dua Lipa? ;-)
Decarbonization was the main reason I became deeply interested in tech, starting a few years ago. This was such a great read for me 😍
Terrific analysis. Thank you. I’m passing this onto one of the Solar Energy startups I find most compelling, NEXT ENERGY TECHNOLOGIES. Their prosperity technology of using SOLUBLE SMALL MOLECULES to create transparent solar cells was first developed at UC, Santa Barbara winning a Nobel Prize. Check out the company’s commercialization progress:
https://www.nextenergytech.com/about
Excellent article. Another issue for solar is land use. Fully embracing Rooftop and covering parking lots are best ways to optimize on land use by using “wasted space” already, especially parking lots. There is a need for commercial model innovation.
We address land use under the objections section
Thanks! I saw that but didn't see the parking lot opportunity. Most people (business owners and elected officials) don't realize how much land is parking lots. A good reference for further opportunity. https://parkingreform.org/resources/parking-lot-map/
Good analysis, good points that are brought together. It is a complex topic.
And I add a non negligible complexity point here. Sorry for this.
There is this question of the interconnexion cost that has quadrupled since the 2000's from which 75% are due to cover network upgrade costs: this is not only an administrative hurdle to overcome as you seem to reduce it to, today it doubles the cost of each project.
It is not negligible at all: the cost of building a grid does not follow Wrights law.
This means that the interconnexion costs will stay the same and will take more and more of the total costs: It is doubling the cost of a project today, it will quadruple it quite soon.
But not all solar generated energy needs a grid. It would be good to understand the proportion of solar generated kW for the grid (China invests a lot into building a grid) and solar generated KW that are locally used (no grid need like in Pakistan).
Interconnection costs are real... but first off, many are done by us to us. (Regulation.) Second, they are independent of the solar learning curve. BTM solar or decentralised or growth anywhere in the world will bring down the panel costs. If the US, UK, or Germany doesn't want to fix their grid economics (which might include network design, regulation or market design), that is their sovereign choice.
As you say, a lot won't need grid (or won't need grids the way we think of it). It's a good challenge. Would I satisfy you if I said "more than today but less than everything?" 😄
Sure interconnection costs are independent of the pure solar learning curve. The qualifier "pure" is important here: the solar learning curve cannot be seen in isolation: it is dependent of the solar success/volume and if the interconnection costs are (what I see) a non negligible part of the total cost of ownership (TCO), it has an impact on solar production volumes.
If 75% of the interconnection costs are due to grid upgrade, this is not done by us to us, it is real costs, which impacts the solar volume, hence Wrights law. TCO higher, volume lower. Because the solar-TCO is defined at 50% by interconnection costs, you need to take these costs into account for your Wright law.
That said, not all solar panels will need grid connexion. So the real calculation depends on the proportion of grid connected solar panels.
But this throws an important question: can solar energy be the technology that allows for decentralisation of energy production? Historically energy generation needed to be centralised, because it is too expensive to build generators where the demand is. So is the need for centralised energy generation still a must? Can we rethink the whole? This is perhaps an opportunity for edge-solar energy ... like there is one for edge-AI. Think about combining both
Re: interconnection & real estate costs — a solution requiring neither!
NEXT ENERGY TECHNOLOGIES has developed carbon based semiconducting molecules that uniquely harvest light energy in the Infrared (IR) spectrum while allowing visible light to pass through and Ultraviolet (UV) light is blocked. No other solar semiconductors can do this! The 1st commercial application — windows. Learn more at https://www.nextenergytech.com/technology Thanks!
Can’t get the tool to work on my iPhone - endless log in doom cycle! Like the analysis, mirrors my musings earlier in the week. I’m in rural Yorkshire and currently stuck on oil, in a house built in 1650!
Hi Ashley, thanks for reading! It sounds like you might be stuck in a Substack login loop (potentially from clicking the 'Upgrade' link in the post).
Could you try going directly to the model here: https://solar.exponentialview.co/ ?
You might see a single pop-up when you arrive, but you can close it and get straight to the model - no login required. It’s also worth flipping your phone to landscape mode to get the best view of the charts. Let me know if that works for you!
Hey EV, is there a channel for user feedback and musings? Have I just missed it?
I would appreciate feedback to the things I'm thinking and am sure there are hundreds if not thousands of readers who would like this type of feature. Of course you could use your AI tools to winnow and consolidate ideas into themes?
For example, two thoughts of mine this week, ChatGPT (Pro) still getting basic facts wrong (Dublin in UK not EU) and it's increasing use of clickbait answers...
Just a thought
Ashley
>Because while politics is powerful, the laws of physics will always win. The learning curve will keep on turning. Solar has no strait.
Isn't there a potential chokepoint being that majority of solar industry and learning is happening in China?
Would the politics still have a big role in making sure solar is available everywhere?
In Sheffield, UK and surrounding South Yorkshire, there is significant community opposition to solar farms, even from the Green Party.
For example https://www.facebook.com/share/17FtZdcNEz/