Rather than derailing another thread I thought it was worth starting a thread on common fallacy with respect the life of solar panels. This gets lot of attention in the media and every few years there are solar panel crisis articles designed to pull in eyeballs on the web. The reality is that many solar panels that date back to the beginnings of solar's early era are still putting out power despite being over the 20 year cut off. Folk tend to forget that standard affordable solar panels were not readily available until 20 years ago . As an example my 2002 era array was built with "2nd generation affordable panels which cost $6.60 per watt (note this is the cost for the panels only, the balance of system components added a lot more). First Quality panels in bulk are now in the $0.50 per watt range today and second quality are down in the $0.30 per watt range. They are now effectively commodities.
There was a market for solar panels prior to the year 2000 but the panels were in excess of $10 a watt. They were used for high value specialty applications like remote telecom and scientific (Note I am excluding satellite PV panels that are a cost is no object applications). They also were adopted by off gridders and had some popularity with off grid pot growers in Northern California. Arco Solar built a couple of early large solar plants. Due to the cost of the panels they used reflectors to boost the amount of sun on each panel. Unfortunately the coating used to laminate the panels were not designed for this use and the panels aged rapidly. As subsidies went away, the solar farms were dismantled and the panels were graded for damage and resold to mostly off gridders. These were usually relabeled as Carrizo Solar quadlams and acquired the nickname "crispy critters" as they were visibly "cooked" with a distinct burnt look to them. Despite the damage, crispy critters are still out there and they still generate usable amounts of power 40 years after. I have some later Siemens panels built in around 1990 that I still use to run my Solar Hot Water and also have some rigged up for solar water system for my garden. Yes they have lost some rating compared to nameplate but still cranking along. They also have some deterioration to the lamination coating but still work. There are several studies on the web of older PV arrays and universally the silicon based panels are still putting power out generally with less degradation that predicted and guaranteed when sold new. My original array is 18 years old and still puts out quite a bit of power and has almost no visual aging.
I am referring to the current standard which are Silicon based panels. They are thin slices of silicon laminated to a tempered glass panel that have a very small amount of other elements added to them to make them semiconductors. These chips have an electrically conductive grid attached to the chips that are connected to an internal electrical grid that eventually comes out to a junction box. The overall assembly is then sealed from the weather with coating and mounted in an aluminum frame. Overall the components are far less toxic than what is found in household electronics. On the other hand there have been other chemistry's for solar panels. The big competitors were CIGS made from a compound called copper gallium indium diselenide and Cadmium telluride (CdTe). Both were cheaper to make although both were less efficient than Silicon. Luckily the US mostly dodged the bullet as these got deployed in Europe especially Germany as they were very difficult to dispose of and had higher long term degradation rates. Since there was not a lot of history in the US on large solar deployments much of the German experience with the other chemistries was imported to the US and applied to Silicon panels.
So where did the 20 year life come from?. Its a combination of the German experience with at least three chemistries including two problematical ones and just as importantly accountants. Accountants by nature have to be conservative. In order to calculate the long term cost of the solar panels, accountants need to do a financial model over the life of the project. They don't pick those numbers out of thin air, they base them on information they have at the time and other factors and eventually the number they pick becomes a Generally Acceptable Accounting Principles (GAAP)that get adopted across the industry. The GAAP values may not be "right" but they are standardized and documented so project economics can be compared. Additionally factored in is Mean Life Between Failure of power electronics. Behind every large array is bank of inverters and they typically have a 10 year life. Most accounting models assume that there will be an initial set of inverters and one change out 10 years into the project. They are using a 20 year life so they dont have to factor in a second inverter change out into the economics so as far as they are concerned the array is beyond its economic and accounting life at 20 years. This does not mean the array is no longer functional it just means that the original investor has picked an end date and as far as their balance sheet is concerned they have an end date. Included in the economics is what it will cost to dispose of the system. Many developers do not actually buy the land, they lease it for 20 years and then let it revert to the prior owner. Generally the original developer is long gone and if the clean up is costly the project is some sort of LLC where they just walk away and let someone else clean it up.
There is also the factor of obsolescence. My original panels are still putting out power but the inverter required is no longer available as the voltage at the time is no longer supported. I hacked up a fix but a big array owner almost guaranteed will need to reengineer and possibly rewire the array to get it to meet future available equipment. What that may look like is an unknown so its easier to just write it off at 20 years.
So in 20 years what happens to the array? Most likely some entrepreneur buys it for cheap, upgrades the electronics and keeps selling power. IMHO the biggest disposal demand to date is damaged panels due to fires, wind events and defective manufacturing. These are usually damaged beyond use and whoever owns them has to dispose of them. Where they go is dependent on what laws are on the books. If mandatory recycling laws are passed some entity will deploy already developed technology to recapture the materials. There will be costs to do this and if the revenue stream does not equal the cost to break them down then whomever owns them get to pay for it. The European approach now is that the original manufacturer needs to recycle them or put in place a system to recycle them. This sounds like a good idea until someone surveys how few companies have survived in the solar business.
There was a market for solar panels prior to the year 2000 but the panels were in excess of $10 a watt. They were used for high value specialty applications like remote telecom and scientific (Note I am excluding satellite PV panels that are a cost is no object applications). They also were adopted by off gridders and had some popularity with off grid pot growers in Northern California. Arco Solar built a couple of early large solar plants. Due to the cost of the panels they used reflectors to boost the amount of sun on each panel. Unfortunately the coating used to laminate the panels were not designed for this use and the panels aged rapidly. As subsidies went away, the solar farms were dismantled and the panels were graded for damage and resold to mostly off gridders. These were usually relabeled as Carrizo Solar quadlams and acquired the nickname "crispy critters" as they were visibly "cooked" with a distinct burnt look to them. Despite the damage, crispy critters are still out there and they still generate usable amounts of power 40 years after. I have some later Siemens panels built in around 1990 that I still use to run my Solar Hot Water and also have some rigged up for solar water system for my garden. Yes they have lost some rating compared to nameplate but still cranking along. They also have some deterioration to the lamination coating but still work. There are several studies on the web of older PV arrays and universally the silicon based panels are still putting power out generally with less degradation that predicted and guaranteed when sold new. My original array is 18 years old and still puts out quite a bit of power and has almost no visual aging.
I am referring to the current standard which are Silicon based panels. They are thin slices of silicon laminated to a tempered glass panel that have a very small amount of other elements added to them to make them semiconductors. These chips have an electrically conductive grid attached to the chips that are connected to an internal electrical grid that eventually comes out to a junction box. The overall assembly is then sealed from the weather with coating and mounted in an aluminum frame. Overall the components are far less toxic than what is found in household electronics. On the other hand there have been other chemistry's for solar panels. The big competitors were CIGS made from a compound called copper gallium indium diselenide and Cadmium telluride (CdTe). Both were cheaper to make although both were less efficient than Silicon. Luckily the US mostly dodged the bullet as these got deployed in Europe especially Germany as they were very difficult to dispose of and had higher long term degradation rates. Since there was not a lot of history in the US on large solar deployments much of the German experience with the other chemistries was imported to the US and applied to Silicon panels.
So where did the 20 year life come from?. Its a combination of the German experience with at least three chemistries including two problematical ones and just as importantly accountants. Accountants by nature have to be conservative. In order to calculate the long term cost of the solar panels, accountants need to do a financial model over the life of the project. They don't pick those numbers out of thin air, they base them on information they have at the time and other factors and eventually the number they pick becomes a Generally Acceptable Accounting Principles (GAAP)that get adopted across the industry. The GAAP values may not be "right" but they are standardized and documented so project economics can be compared. Additionally factored in is Mean Life Between Failure of power electronics. Behind every large array is bank of inverters and they typically have a 10 year life. Most accounting models assume that there will be an initial set of inverters and one change out 10 years into the project. They are using a 20 year life so they dont have to factor in a second inverter change out into the economics so as far as they are concerned the array is beyond its economic and accounting life at 20 years. This does not mean the array is no longer functional it just means that the original investor has picked an end date and as far as their balance sheet is concerned they have an end date. Included in the economics is what it will cost to dispose of the system. Many developers do not actually buy the land, they lease it for 20 years and then let it revert to the prior owner. Generally the original developer is long gone and if the clean up is costly the project is some sort of LLC where they just walk away and let someone else clean it up.
There is also the factor of obsolescence. My original panels are still putting out power but the inverter required is no longer available as the voltage at the time is no longer supported. I hacked up a fix but a big array owner almost guaranteed will need to reengineer and possibly rewire the array to get it to meet future available equipment. What that may look like is an unknown so its easier to just write it off at 20 years.
So in 20 years what happens to the array? Most likely some entrepreneur buys it for cheap, upgrades the electronics and keeps selling power. IMHO the biggest disposal demand to date is damaged panels due to fires, wind events and defective manufacturing. These are usually damaged beyond use and whoever owns them has to dispose of them. Where they go is dependent on what laws are on the books. If mandatory recycling laws are passed some entity will deploy already developed technology to recapture the materials. There will be costs to do this and if the revenue stream does not equal the cost to break them down then whomever owns them get to pay for it. The European approach now is that the original manufacturer needs to recycle them or put in place a system to recycle them. This sounds like a good idea until someone surveys how few companies have survived in the solar business.
Last edited:
