4 ms·

> Currently panels drop to 80% efficacy after 10 years and are doorstops in 20 Oh come on. This is an absurd claim with no evidence. Don't be ridiculous.

by davedx 3y ago

defrost 3y ago
In the case of GW scale solar farms that's essentially true. Polycrystalline panels last 20 years, sometimes longer (the sales pitch is "up to 25 years") but the 20% fall off in deliverable power after 10 years is real. Monocrystalline last longer, they also gradually fall off and they cost more. Now it's a game of Backblaze storage stats only with solar panels - how does an at scale solar farm balance purchase costs and maintainance costs, where's the optimal sweet spot for buying new and rotating out on a schedule | reacting to failure. I'm not opposed to solar, I'm in Australia in one of the states with large amounts of residential solar and heavy mining industries that dwarf anything in the US currently going electric - we're costing out massive farms to run solar power to asia via HVDC cable. The supply side of all that is sourcing the copper, the lithium, the aluminium framing, the long list of other raw material inputs, the processing to get from concentrate ores to usable elements, etc.
- davedx 3y ago
  > the 20% fall off in deliverable power after 10 years is real If it's real, why don't you provide evidence?
  - aifooh7Keew6xoo 3y ago
    It's a wildly exaggerated figure. Maxeon cells are rated for 0.25% degradation per year which works out to about 8% in 30 years. An inferior cell might be rated for 0.75% which works out to 20% in 30 years rather than 10 years. I believe a lot of the panels rated for 0.75% are seeing around half that actual rate of degradation.
    - davedx 3y ago
      Yup that's what I found with a cursory google, I kind of hoped OP would at least have the courage to post some reference even if it wasn't accurate, but no. Unfortunately this is par for the course in discussions with nuclear power cultists. It honestly is that bad.
      - defrost 3y ago
        > If it's real, why don't you provide evidence? Because I went to bed, it was late in my time zone. > nuclear power cultists. Say what now? Are you an idiot? The vast bulk of PV performance studies are dominated by relatively short term studies in temperate climates - the latitudes of California and Germany eg: https://www.nrel.gov/docs/fy12osti/51664.pdf https://www.nrel.gov/docs/fy12osti/51664.pdf references Australia and links to papers from Perth and Melbourne which are cooler climates. See my comment above re: my region of interest - equatorial tropical regions with harhser temperature cycling and probable requirements for cooling to maximise returns (still part of an open study). I'd be happy to provide more details but I'm afraid your comment above doesn't warrent engagement.
      - Reubachi 3y ago
        Calling someone a cultist because they went to bed and didn't reply in a timeframe you considered proper is not suitable discussion for HN.
- shkkmo 3y ago
  > Polycrystalline panels last 20 years, sometimes longer (the sales pitch is "up to 25 years") but the 20% fal Your numbers seem off by atleast a facotor of 2. Average degradation is .5% per year so panels are guarantees are usually that you'll have 90% capacity after 10 years and 80% after 20 years. > Now it's a game of Backblaze storage stats only with solar panels - how does an at scale solar farm balance purchase costs and maintainance costs, The failure mode for solar is unlike traditional spinning hard drives. There isn't a "data loss" downside to waiting to replace. There won't be any risk mitigation, just calculations of when the opportunity costs become high enough to warrant replacement. I suspect in areas with cheap land, it'll be easier to build additional capacity than to replace existing degraded panels.
  - defrost 3y ago
    As I mentioned in a peer comment I'm more interested in tropical | equatorial performance and the vast majority of PV performance studies are in temperate climates - it's not currently clear whether monocrsytalline deliver peak power in such climates .. quoting a study I have here: His work has shown that in this tropical climate, with high ambient temperatures and high humidity during the wet season, the a-Si array produces up to 20% more energy than ... (It's an ongoing bit of work ATM) The optimisation issues include other factors, the cost of panel cooling techniques Vs their benefits for one. Your average degradation figures are the figures from big meta studies such as the US Dept of Energy NREL lab Photovoltaic Degradation Rates https://www.nrel.gov/docs/fy12osti/51664.pdf https://www.nrel.gov/docs/fy12osti/51664.pdf which you'll notice is dominated by temperate climate studies at the latitudes of Perth | Melbourne | Germany | Arcata, California etc.
    - shkkmo 3y ago
      > As I mentioned in a peer comment I'm more interested in tropical | equatorial performance Perhaps you should qualify your statements, because this: > Currently panels drop to 80% efficacy after 10 years and are doorstops in 20 This is an unqualified factual statement and is simply not true. > Your average degradation figures are the figures from big meta studies The big meta studies do show that degradation is a bit faster in higher temp environments for some kinds of panels. There is however zero evidence that solar panels will be "door stops in 20".

> Currently panels drop to 80% efficacy after 10 years and are doorstops in 20 Oh come on. This is an absurd claim with no evidence. Don't be ridiculous.

by davedx 3y ago

> Currently panels drop to 80% efficacy after 10 years and are doorstops in 20 Oh come on. This is an absurd claim with no evidence. Don't be ridiculous.

defrost 3y ago

In the case of GW scale solar farms that's essentially true. Polycrystalline panels last 20 years, sometimes longer (the sales pitch is "up to 25 years") but the 20% fall off in deliverable power after 10 years is real. Monocrystalline last longer, they also gradually fall off and they cost more. Now it's a game of Backblaze storage stats only with solar panels - how does an at scale solar farm balance purchase costs and maintainance costs, where's the optimal sweet spot for buying new and rotating out on a schedule | reacting to failure. I'm not opposed to solar, I'm in Australia in one of the states with large amounts of residential solar and heavy mining industries that dwarf anything in the US currently going electric - we're costing out massive farms to run solar power to asia via HVDC cable. The supply side of all that is sourcing the copper, the lithium, the aluminium framing, the long list of other raw material inputs, the processing to get from concentrate ores to usable elements, etc.

davedx 3y ago
> the 20% fall off in deliverable power after 10 years is real If it's real, why don't you provide evidence?
- aifooh7Keew6xoo 3y ago
  It's a wildly exaggerated figure. Maxeon cells are rated for 0.25% degradation per year which works out to about 8% in 30 years. An inferior cell might be rated for 0.75% which works out to 20% in 30 years rather than 10 years. I believe a lot of the panels rated for 0.75% are seeing around half that actual rate of degradation.
  - davedx 3y ago
    Yup that's what I found with a cursory google, I kind of hoped OP would at least have the courage to post some reference even if it wasn't accurate, but no. Unfortunately this is par for the course in discussions with nuclear power cultists. It honestly is that bad.
    - defrost 3y ago
      > If it's real, why don't you provide evidence? Because I went to bed, it was late in my time zone. > nuclear power cultists. Say what now? Are you an idiot? The vast bulk of PV performance studies are dominated by relatively short term studies in temperate climates - the latitudes of California and Germany eg: https://www.nrel.gov/docs/fy12osti/51664.pdf https://www.nrel.gov/docs/fy12osti/51664.pdf references Australia and links to papers from Perth and Melbourne which are cooler climates. See my comment above re: my region of interest - equatorial tropical regions with harhser temperature cycling and probable requirements for cooling to maximise returns (still part of an open study). I'd be happy to provide more details but I'm afraid your comment above doesn't warrent engagement.
    - Reubachi 3y ago
      Calling someone a cultist because they went to bed and didn't reply in a timeframe you considered proper is not suitable discussion for HN.
shkkmo 3y ago
> Polycrystalline panels last 20 years, sometimes longer (the sales pitch is "up to 25 years") but the 20% fal Your numbers seem off by atleast a facotor of 2. Average degradation is .5% per year so panels are guarantees are usually that you'll have 90% capacity after 10 years and 80% after 20 years. > Now it's a game of Backblaze storage stats only with solar panels - how does an at scale solar farm balance purchase costs and maintainance costs, The failure mode for solar is unlike traditional spinning hard drives. There isn't a "data loss" downside to waiting to replace. There won't be any risk mitigation, just calculations of when the opportunity costs become high enough to warrant replacement. I suspect in areas with cheap land, it'll be easier to build additional capacity than to replace existing degraded panels.
- defrost 3y ago
  As I mentioned in a peer comment I'm more interested in tropical | equatorial performance and the vast majority of PV performance studies are in temperate climates - it's not currently clear whether monocrsytalline deliver peak power in such climates .. quoting a study I have here: His work has shown that in this tropical climate, with high ambient temperatures and high humidity during the wet season, the a-Si array produces up to 20% more energy than ... (It's an ongoing bit of work ATM) The optimisation issues include other factors, the cost of panel cooling techniques Vs their benefits for one. Your average degradation figures are the figures from big meta studies such as the US Dept of Energy NREL lab Photovoltaic Degradation Rates https://www.nrel.gov/docs/fy12osti/51664.pdf https://www.nrel.gov/docs/fy12osti/51664.pdf which you'll notice is dominated by temperate climate studies at the latitudes of Perth | Melbourne | Germany | Arcata, California etc.
  - shkkmo 3y ago
    > As I mentioned in a peer comment I'm more interested in tropical | equatorial performance Perhaps you should qualify your statements, because this: > Currently panels drop to 80% efficacy after 10 years and are doorstops in 20 This is an unqualified factual statement and is simply not true. > Your average degradation figures are the figures from big meta studies The big meta studies do show that degradation is a bit faster in higher temp environments for some kinds of panels. There is however zero evidence that solar panels will be "door stops in 20".