New testing conducted at France's oldest PV system have shown that its solar modules can still provide performance values in line with what the manufacturers promised.
To give you an idea, my 12-panel PV system installed in 2011 has put out 3.5 MWh per year at its peak and now produces between 3.1 and 3.3 MWh yearly, depending on the weather.
It’s hard to attribute that just to panel degradation, though. It could be differences in weather (cloudier or snowier this year) for example.
To measure degradation, you’d need to track the peak output of each panel. Enphase microinverters let you get per-panel metrics but I’m not sure which other brands do.
I’m pulling data from my panels into Home Assistant via Enphase’s local API (directly from the device), then into VictoriaMetrics (which is similar to Prometheus but with a more efficient file format). I’ve got per-panel production data at 5 minute granularity from when I installed them until now.
We’re getting solar installed very soon, with Enphase micro-inverters. This gets me all kinds of excited. I’m stoked to be getting per-panel metrics, and real keen to shove even more metrics into my Home Assistant.
The inverters all use their serial number as their name by default, but I renamed mine based on array and location to be more useful:
If your installer installs the consumption CTs (optional but the good installers usually include them for free), you’ll also have data on total power consumption for your house. It works really well with Home Assistant’s built-in energy dashboard.
Yeah I am. I like self hosting. I might add some friends to my server at some point.
I’m an Aussie living in the USA. I’ve been living in the US for 11 years but I’m still Aussie at heart. I’m still eligible for .au domains since I’m an Australian citizen.
Newer panels are generally much more efficient and produce more electricity compared to old panels.
The 80% after 25 years might be their warranty, my panels have a similar warranty on them. If they start producing less than 80% of their original output before 25 years, the manufacturer will replace them (or something like that).
Yes, I’m looking at the paperwork and I’m realizing that you’re right, 80% at 25 is the warranty guarantee, so I’m guessing they’re confident it’ll typically be much better than 80 at 25
The warranty is provided by the manufacturer of the solar panels, not by the installer. If your installer goes out of business, you can claim the warranty via another installer.
Of course, it’s possible the manufacturer will go out of business or sell their business.
yeh, thats what I mean, who knows what the state of the market will be in 25 years, unless its an insurance backed guarantee, be very suspicious of it.
For what it’s worth, a lot of the major manufacturers (QCell, REC, Canadian Solar) have been in business since the late 90s or early 2000s. SunPower (now Maxeon) has been around since 1985.
It’s a very conservative estimate, as manufacturers really don’t want to deal with warranty claims.
One part of solar system design is often to undersize the inverter a bit, so you might not see any degradation in the first few years at all. The logic there is that the electricity lost to clipping during peak times of the year doesn’t make up for the higher cost of a larger inverter.
Try to avoid SolarEdge inverters if you can. Enphase is really good and has a local API (so you can pull data into Home Assistant and other apps without going via the cloud) with a bunch of data, including per-panel production. Fronius and SMA are good too. Everyone I know with SolarEdge inverters has had issues with them though.
This is amazing, because I do have HA running and I do want to pull data. I’ll look into what their plan is to see if I can make any last minute adjustments.
Assuming data is sent in the clear over any medium which I can get a receiver for I think I’d be ok, I am planning on getting a software defined radio on my HA in the near future
For Enphase, the data is available via a HTTP API. The combiner box (their hardware that all the inverters connect to) connects to your network via wifi or Ethernet, and you can hit its IP directly.
Unfortunately it’s kinda-cloud-dependent these days. The API used to take a basic username and password, nice and simple. Now, you need to hit a cloud API to get an access token (and periodically refresh it). The API is entirely local once you have the access token though, and there’s no problem with polling the data once per second.
I’m getting some new panels installed this year, and I think they’re suggesting they’ll be at 80% after 25 years.
It looks like there is disagreement between the title and content of the article. Title says 75.9, content says 79.5
Either way, does this suggest that new panels might do better than expected over a 30 year timespan?
Looks like there’s a typo in the English title. The French one has 79.5%.
To give you an idea, my 12-panel PV system installed in 2011 has put out 3.5 MWh per year at its peak and now produces between 3.1 and 3.3 MWh yearly, depending on the weather.
It’s hard to attribute that just to panel degradation, though. It could be differences in weather (cloudier or snowier this year) for example.
To measure degradation, you’d need to track the peak output of each panel. Enphase microinverters let you get per-panel metrics but I’m not sure which other brands do.
I’m pulling data from my panels into Home Assistant via Enphase’s local API (directly from the device), then into VictoriaMetrics (which is similar to Prometheus but with a more efficient file format). I’ve got per-panel production data at 5 minute granularity from when I installed them until now.
We’re getting solar installed very soon, with Enphase micro-inverters. This gets me all kinds of excited. I’m stoked to be getting per-panel metrics, and real keen to shove even more metrics into my Home Assistant.
It’s an official integration and works really well. https://www.home-assistant.io/integrations/enphase_envoy/
The inverters all use their serial number as their name by default, but I renamed mine based on array and location to be more useful:
If your installer installs the consumption CTs (optional but the good installers usually include them for free), you’ll also have data on total power consumption for your house. It works really well with Home Assistant’s built-in energy dashboard.
Dan you sound like a rad dude ad I’d love to have a beer with you
Hah, thanks (if you’re serious)! Feel free to DM me if you’re ever in the San Francisco Bay Area.
Removed by mod
Yeah I am. I like self hosting. I might add some friends to my server at some point.
I’m an Aussie living in the USA. I’ve been living in the US for 11 years but I’m still Aussie at heart. I’m still eligible for .au domains since I’m an Australian citizen.
Removed by mod
Newer panels are generally much more efficient and produce more electricity compared to old panels.
The 80% after 25 years might be their warranty, my panels have a similar warranty on them. If they start producing less than 80% of their original output before 25 years, the manufacturer will replace them (or something like that).
Yes, I’m looking at the paperwork and I’m realizing that you’re right, 80% at 25 is the warranty guarantee, so I’m guessing they’re confident it’ll typically be much better than 80 at 25
Or that they were expecting to be out of business by then…
The warranty is provided by the manufacturer of the solar panels, not by the installer. If your installer goes out of business, you can claim the warranty via another installer.
Of course, it’s possible the manufacturer will go out of business or sell their business.
yeh, thats what I mean, who knows what the state of the market will be in 25 years, unless its an insurance backed guarantee, be very suspicious of it.
For what it’s worth, a lot of the major manufacturers (QCell, REC, Canadian Solar) have been in business since the late 90s or early 2000s. SunPower (now Maxeon) has been around since 1985.
It’s a very conservative estimate, as manufacturers really don’t want to deal with warranty claims.
One part of solar system design is often to undersize the inverter a bit, so you might not see any degradation in the first few years at all. The logic there is that the electricity lost to clipping during peak times of the year doesn’t make up for the higher cost of a larger inverter.
Try to avoid SolarEdge inverters if you can. Enphase is really good and has a local API (so you can pull data into Home Assistant and other apps without going via the cloud) with a bunch of data, including per-panel production. Fronius and SMA are good too. Everyone I know with SolarEdge inverters has had issues with them though.
Any thoughts on a set of APsystems DSL3-L inverters? That’s what I’m quoted for. I’m in Canada so no idea what the inverter market is here
Never heard of them so I’m not sure, sorry!
This is amazing, because I do have HA running and I do want to pull data. I’ll look into what their plan is to see if I can make any last minute adjustments.
Assuming data is sent in the clear over any medium which I can get a receiver for I think I’d be ok, I am planning on getting a software defined radio on my HA in the near future
For Enphase, the data is available via a HTTP API. The combiner box (their hardware that all the inverters connect to) connects to your network via wifi or Ethernet, and you can hit its IP directly.
Unfortunately it’s kinda-cloud-dependent these days. The API used to take a basic username and password, nice and simple. Now, you need to hit a cloud API to get an access token (and periodically refresh it). The API is entirely local once you have the access token though, and there’s no problem with polling the data once per second.
Probably like tire warranties: prorated for expected life, and not including installation cost