-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Micro-inverters a better choice with solar-coupled Powerwalls?
- Thread starter abasile
- Start date
Haha, Southern California has many different climates and environments, of course, all in relatively close proximity. Keeps life interesting!
So, between my incremental efforts at snow removal and the much warmer weather, all of the snow and ice are now off our panels and we're back to normal until the next storms. The only negative is that thick ice sliding off of a couple of panels broke the dome on a tubular skylight directly beneath them. Next time there is a very heavy snow load, I'll have to be a little more aggressive about clearing those particular panels.
I do appreciate that the micro-inverters, at least the ones on the most accessible panels, enabled us to start "feeding" the Powerwalls a few days sooner than if we have been relying on a string inverter with optimizers. This was enough to slowly charge the Powerwalls at least a few percent per day, as they had dropped down from 100% to as low as 39% just keeping themselves "alive".
However, this is the other side of what sucks (to me) about micro-inverters, the clipping.
28 360W panels, 28 MPPTs, each clipping to its SolarBridge 320W inverter. That's 88% of STC, and ~92% of PTC. These are 1 year-old panels fresh after recent heavy rains. Shown with 8,682W peak .
28 360W panels, 28 MPPTs, each clipping to its SolarBridge 320W inverter. That's 88% of STC, and ~92% of PTC. These are 1 year-old panels fresh after recent heavy rains. Shown with 8,682W peak .
Attachments
nwdiver
Well-Known Member
That's about right. Most string inverters are going to have a similar level of clipping if it's a standard install. I usually oversize my systems by ~20%. The project I'm finishing up was limited to ~15kW due to the service size but the customer wanted more production so the more economic choice in that scenario was East and West facing arrays oversized by ~50%. There's not as much loss as you would expect. The annual energy loss with an oversize ratio of 1.2 is <2%. Even at 50% oversize the annual loss is ~5% compared to a 'right-sized' array.
Here's a chart I made examining the economics. The far right is the percent the last watt adds to annual production. Up to ~1.15 there's 0 clipping. At 1.2 there's 2% of the 5% added. So very little overall. Even at an oversize ratio of 1.5 added DC capacity is ~60% productive. ~1.3 is about the max I usually like to go but my last project was a little different since upgrading the service would have added ~$7k to the cost.
Last edited:
Exactly! These SP micro-inverters are undersized by 0.12 whereas you can appropriately oversize the inverter when it's separate.
I'm sure the situation is now a bit better with SP switching to Enphase, but the lesson here is to watch that panel-to-inverter ratio mismatch.
A more fair 1:1 comparison would be total production under same conditions even with the micro-inverter clipping for micro vs separate.
nwdiver
Well-Known Member
Sure... you could... but why would you want to? Oversizing is more cost effective. Look at the chart. The lost production up to 1.12 is <1% annually.
Yeah, I see the exact same thing with our 360W panels and SunPower-branded micro-inverters. I noticed this kind of clipping right after our install and complained to our installer, who of course told me that this is "normal". I'll add that the installer did end up having to honor the first-year production guarantee that they gave us.
Up at our elevation, in dry weather, we receive particularly intense sunlight. As a result, we normally see a lot of inverter clipping. I really wish I had been aware of this issue prior to signing our solar purchase contract! If possible (maybe not with SunPower), I certainly would have demanded oversized micro-inverters. I don't think it would be economical to upgrade now, unless inverters fail prematurely and need to be replaced anyway.
Oh, by the way, the "green technology" talk I was scheduled to give tonight has been postponed due to weather. I'll consider mentioning the issue of inverter clipping when I do give the talk.
Up at our elevation, in dry weather, we receive particularly intense sunlight. As a result, we normally see a lot of inverter clipping. I really wish I had been aware of this issue prior to signing our solar purchase contract! If possible (maybe not with SunPower), I certainly would have demanded oversized micro-inverters. I don't think it would be economical to upgrade now, unless inverters fail prematurely and need to be replaced anyway.
Oh, by the way, the "green technology" talk I was scheduled to give tonight has been postponed due to weather. I'll consider mentioning the issue of inverter clipping when I do give the talk.
Last edited:
nwdiver
Well-Known Member
There is an 'optimal' level of clipping. If you never have any clipping then you could have added more solar at little to no cost. You're also often limited by panel capacity. A 200A panel can only support 32A (~8kW) of solar unless you undersize the main breaker. But the DC side is effectively unlimited since the inverter is a current limited device. So you can have 12kW of solar so long as your inverter is only ~8kW.
As we get more solar on the grid clipping is inevitable and the benefits FAR outweigh the costs. The clipped energy is the least valuable in terms of supply-demand. You're losing a few kWh when the grid will most likely already have plenty of solar and gaining kWh in the morning and evening when it needs it the most.
And some clipping in the Spring doesn't mean there will be clipping in the summer when production is needed the most. All arrays should really be oversized by AT LEAST ~20%. Especially with panels being as cheap as they are...
Thank you for the informative post! Given ample roof space, it does make sense that you would generally want to oversize the array relative to the inverter(s).
Our personal situation is that we've already maxed out our south-facing rooftop solar potential, and we use more electricity than we generate. I went with high-efficiency panels and micro-inverters, and did some significant tree trimming, in order to squeeze as much energy out of our modest-sized rooftop as possible. If we were to add more panels, we'd have the following placement options where the shading wouldn't be prohibitive:
1. North facing rooftop. Easy to install and probably not that bad in summer, but not very helpful for at least half the year.
2. North facing rooftop, with reverse-tilted panels. Likely very good overall production, but could be more costly, ugly, and more prone to getting buried by snow.
3. South facing wall, with vertical orientation. Good for fall/winter production and to keep the Powerwalls charged when the roof is covered by snow/ice. Lousy summer production. Not too much of an issue with aesthetics because it's set back from the street.
4. East facing wall, with vertical orientation. Good for mornings year round. But a no-go for aesthetic reasons because it directly faces the street.
Because our Powerwalls are programmed to charge only from solar, (3) seems like it would be the best place to start, if we do anything at all.
Our personal situation is that we've already maxed out our south-facing rooftop solar potential, and we use more electricity than we generate. I went with high-efficiency panels and micro-inverters, and did some significant tree trimming, in order to squeeze as much energy out of our modest-sized rooftop as possible. If we were to add more panels, we'd have the following placement options where the shading wouldn't be prohibitive:
1. North facing rooftop. Easy to install and probably not that bad in summer, but not very helpful for at least half the year.
2. North facing rooftop, with reverse-tilted panels. Likely very good overall production, but could be more costly, ugly, and more prone to getting buried by snow.
3. South facing wall, with vertical orientation. Good for fall/winter production and to keep the Powerwalls charged when the roof is covered by snow/ice. Lousy summer production. Not too much of an issue with aesthetics because it's set back from the street.
4. East facing wall, with vertical orientation. Good for mornings year round. But a no-go for aesthetic reasons because it directly faces the street.
Because our Powerwalls are programmed to charge only from solar, (3) seems like it would be the best place to start, if we do anything at all.
I don't think any of the SP micro-inverters are generously-enough sized even up to their 370w panels. Spec sheets looks like they're all 320w max. The newer Enphase-inverter versions are probably a little better, but that was a unknown future timeline.
Looking at the LG LG365Q1C-A5 panels, not much better there either. They have higher PTC than SP, but the panels are larger than SP's, so end up being lower in W/m² under STC or PTC.
But as nwdiver has said, spring-problems. With summer heat and dust, clipping is no longer an issue. The nit is clipping is lost TOU-credits.
Re vertical installation: I did ask my installer if I can get vertically installed panels on my all south-facing walls -- as it's not street-facing and maybe would cool the house. There was refusal because permitting is harder with the city? Reduce value of house. More excuses.
I even asked if they can be mounted as "awnings" to shade south-windows.. nope! So finding an installer to do this may also be difficult.
Last edited:
For most homeowners, that may be true. At higher, cooler elevations across the West, however, and with panels like SunPower with low thermal coefficients, it seems that clipping would also be a factor during summer.
Most southern-state solar installers are probably not experienced at vertical installations, which are likely more common way up north. The biggest installation issue seems to be the potential need for scaffolding when mounting panels on taller structures. On the other hand, on walls that are easy to access, it seems to me that vertical panels might even be simpler to install than rooftop panels, as there's no need to worry about roofing material or flashings.
Right, though if our SunPower panels live up to their advertised very low degradation rate, then it seems that clipping will continue to be a factor for years.
Jigglypuff
Member
We have 255W panels and 250W converters. Supposedly we'd be more efficient with smaller converters, but who knows. They might last longer if underrun.
nwdiver
Well-Known Member
This is what my production curve looks like. The clipping doesn't bother me at all. If you don't have clipping then your panels are undersized. It's a balance.
I guess it depends on the cost of adding more panels, if that's an option, versus the cost of a more powerful inverter and appropriately-rated service panel. Eyeballing your chart, it looks like you're losing no more than 5% of your daily production to clipping, and it's not quite Spring. Suppose that translates to losing about 3% of your potential annual generation. Given a $20k solar system, would it be worth spending 3% more ($600) on inverter hardware to significantly reduce clipping? I'm guessing it would be better to spend that 3% on upgrading the panels or adding one more panel. But if you're like me and you've already maxed out your good roof space with the best possible panels, then speccing out a better inverter might still be a good long term investment.
Yes indeed, but recall that I started this thread with a discussion about keeping Powerwalls charged up. If the homeowner happens to have Powerwalls, then that midday energy can be stored on site for later use during Peak rate hours.
Interestingly, SunPower corporate just called us out of the blue to set up an appointment to upgrade our micro-inverters (to their "3.1" version) and upgrade our monitoring system. Since they're going to swap out the micro-inverters anyway (wow!), I asked if it might be feasible for us to upgrade the rated AC wattage on the inverters and pay the difference. They might not want to do this because it'd change the AC rating on the overall system, but I figured I might as well ask.
A little closer to the original topic, it seems to me that the issue of clipping would be a selling point in favor of string inverters with optimizers. Correct me if I'm wrong, but with a right-sized string inverter, it seems that clipping won't occur unless all or nearly all of the panels are in full sun. With micro-inverters, there is clipping on individual panels even if half the array is in shade and the other half of the panels are in full sun.
nwdiver
Well-Known Member
Yes indeed, but recall that I started this thread with a discussion about keeping Powerwalls charged up. If the homeowner happens to have Powerwalls, then that midday energy can be stored on site for later use during Peak rate hours.
That's actually where clipping is most 'beneficial'. Instead of adding storage to offset demand during peak evening hours it's actually cheaper to add panels... even if their output is clipped for a few hours mid-day. Spending ~$2k to add a kW of solar can displace more evening generation than spending $2k to add a few kWh of storage.
Minnesota study finds it cheaper to curtail solar than to add storage
"additional capacity coupled with energy curtailment is considerably less expensive than, and a viable alternative to, long-term or seasonal storage in a high renewables future."
Storage is still relatively expensive, so I don't disagree.
For individual homeowners, the current economics for on-site storage (including Powerwalls) are not very good. We justified the purchase of our Powerwalls based on the desire to have a whole-home backup system not dependent on an external fuel source, with TOU load shifting and self consumption being secondary benefits. (Obviously, having our solar panels buried by snow for weeks throws a wrinkle into that!)
That being said, as long as we have the Powerwalls, we might as well use them to capture lower-value solar energy for use after the sun sets. Another option for some would be to charge their EVs during the middle of the day.
For individual homeowners, the current economics for on-site storage (including Powerwalls) are not very good. We justified the purchase of our Powerwalls based on the desire to have a whole-home backup system not dependent on an external fuel source, with TOU load shifting and self consumption being secondary benefits. (Obviously, having our solar panels buried by snow for weeks throws a wrinkle into that!)
That being said, as long as we have the Powerwalls, we might as well use them to capture lower-value solar energy for use after the sun sets. Another option for some would be to charge their EVs during the middle of the day.
