Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Micro-inverters a better choice with solar-coupled Powerwalls?

This site may earn commission on affiliate links.

abasile

TSLA shareholder
Supporting Member
Here in the US for ITC (Investment Tax Credit) compliance, Tesla has programmed all solar-coupled Powerwalls to charge only from solar and not from the grid, the only exception being relatively infrequent Storm Watch emergencies. It is therefore in the interest of solar/Powerwall users to have the most robust solar production possible. Even if micro-inverters increase overall system cost somewhat, I am thinking that micro-inverters should be preferred over string inverters when installing solar systems that will be coupled with Powerwalls.

This month, we've had scarcely any solar production because of heavy snow accumulation on the roof, combined with freeze/thaw cycles and ice formation. As a result, our two Powerwalls have not been able to recharge, except for one Storm Watch event when we were thankfully able to refill them from the grid. At the same time, the Powerwalls lose roughly 3-5% charge per day keeping themselves on standby in the cold, as they are installed outdoors on a porch.

Thankfully, we have SunPower panels with micro-inverters. I was just able to clear off one relatively accessible panel today, and the rest of the snow and ice may just have to melt on its own, which could take weeks longer. However, even with just one panel clear, we should be able to generate enough power to at least keep the Powerwalls from discharging further.

At another family home, however, we have basic, commodity panels with a SolarEdge inverter and power optimizers. The roof on that house is much more accessible and has a gentler pitch, so I was able to go on the roof and clear six out of 14 panels. However, even after a couple of days of sun, we saw no solar production, and the inverter remained "asleep". Today, I went back up and cleared two more panels, or 8/14. Finally, the inverter started up. My solar installer confirmed that if the majority of the panels are not getting significant light, the inverter will not start even if a minority of the panels are in full sun. That would obviously not be ideal if we were trying to keep Powerwalls from going down to 0%.

Further, there is no issue with a single inverter being a single point of failure when you have a micro-inverter on each panel. If that SolarEdge inverter had gone bad during the last several weeks, we would have had trouble getting the installer back up to replace it due to the road conditions and snow chain requirements.

In a couple of weeks, I am going to be giving a talk on our adoption of "green technology to a local environmental group up here in the SoCal mountains. I am inclined to strongly recommend the use of micro-inverters for those considering installing solar coupled with battery storage. Unfortunately, this would rule out using Tesla to install the solar, as I don't believe they do micro-inverters. Does anyone see any holes in my reasoning?

Also, I will add that it could be advantageous to have at least a couple of panels that are tilted at a steep angle, or even vertical on the south side of the house. This could permit at least enough winter solar production to keep Powerwalls charged. We don't get heavy snow accumulation on the roof all of the time, but when we do, the likelihood of power outages is pretty significant.
 
As far as I know, only Sunpower has microinverter and its one of the reason i went with them, simply put, each panel operates on its own vs the rest are daisy chained and rely on each other which causes issues as such if one is out, the whole system can go out.
 
As far as I know, only Sunpower has microinverter and its one of the reason i went with them, simply put, each panel operates on its own vs the rest are daisy chained and rely on each other which causes issues as such if one is out, the whole system can go out.
Enphase is the #1 micro-inverter supplier in the business. There are a few others, including Sunpower, but they just recently switched from their in-house micros to Enphase.
 
As far as I know, only Sunpower has microinverter and its one of the reason i went with them, simply put, each panel operates on its own vs the rest are daisy chained and rely on each other which causes issues as such if one is out, the whole system can go out.
No, LG also has Enphase micro-inverter (AC) panels.

I had my some of my roof-mounted SP panels installed with manually-adjustable tilt. However unless the horrible backyard neighbor ever gives me clearance (nonsense glare reasons), one can't tilt it beyond the maximum height of the peak of the 24° roof -- per city permitting. I'm sure this violates my CA Solar Bill of Rights somehow, but I don't have the energy to chase it down.

Tilt on the roof is a catch-22 problem where you can't raise the panels closer to the southern/lower-edge because it'll shade your panels higher on the roof. However, tilting the upper-panels bumps against city-permitting with maximum-height of objects on your roof.
 
I would say that the few panels mounted on a steep angle should have micro-inverters while the main array can use a string inverter.
That seems like a reasonable option for cases where the main array is at a uniform orientation and angle with no shading. There's still the potential issue of the string inverter being a single point of failure, but that may be a secondary concern.

In our case, the panels in the main array don't shed snow/ice at a uniform rate. Next week, I might find that a few of the panels are clear while the majority of the array is still buried. With other fixtures on our roof and partial shading from tall pine trees, micro-inverters do offer us the most reliable approach for producing at least some power from our main array. I wish we had used micro-inverters for the other house for the same reason.

Given the amount of snow that we sometimes receive, I don't think that increasing the tilt of some of the panels on our roof is a great solution. With three feet of snow sitting on the roof, panels will be obscured no matter what. We could potentially install a few vertical panels on south facing walls. In terms of overall solar production, the ROI on such panels would be poor due to shade from the eaves during the summer, but their autumn/winter production would be appreciated.
 
Clearly, if there is shading, micro-inverters are the right solution. I have a tall tree, probably 200 yards from my property that casts a shadow across my panels for about 3 months of the year. That is the main reason that I have micro-inverters. I can actually watch the shadow move across my array in the Enphase Enlighten monitoring portal.
 
I have a recent SolarEdge/Panasonic install using power optimizers on each panel and three strings of 12 I believe. We have also been going through snow, sun, thaw cycles and I still make power on the exposed panels while neighboring panels are covered. Could be another option to consider.
 
Last edited:
Solar Edge systems have a minimum possible string size of 6-8. If you can't get some production on that many panels in a string, then your inverter wont effectively fire up.
Microinverters are a great choice for just this reason, as well as the multiple points of failure being an advantage.
Tesla doesn't need to install every system, smaller installers make up an important market as well.
 
  • Like
Reactions: abasile
Very interesting thread, thank you for sharing. As I read your first hand account, it became apparent how much geography influences inverter selection.

Here in Houston the few folks I know with enphase wish they had solaredge due to constant inverter failures in the summer. -- I presume due to the rooftop heat causes the inverter to fail more quickly. (Which would not be an issue in your area). After 6 months the installers don't even bother to assist on the RMA process and you are stuck paying someone to swap out the inverter on the roof. Since we do not have the snow issue, convenience of the inverter location and maintenance are higher priority...

Just shows how complicated a buying experience solar can be and there is truly no one size fits all option.

Great to hear you sharing your experience with your local community
 
Thankfully, we have SunPower panels with micro-inverters. I was just able to clear off one relatively accessible panel today, and the rest of the snow and ice may just have to melt on its own, which could take weeks longer. However, even with just one panel clear, we should be able to generate enough power to at least keep the Powerwalls from discharging further.

LOL... ok... I'm not a fan of micro inverters. It wouldn't be a stretch to say I hate them (only because they're ~3x the cost). But you have found one thing they do better.... Neither String inverters nor optimizers can work on 1 panel...

Clearly, if there is shading, micro-inverters are the right solution.

*sigh*.... frustrating that this myth won't die. The only time shading is really detrimental to overall system performance is if you have 2 or more strings in parallel on 1 MPPT. If each string has it's own MPPT shading isn't a big problem. If you have optimizers shading isn't a big problem.

Here's 1 panel being shaded.... rest of the array... not affected. Bypass diodes, they work and they come installed in EVERY UL rated panel. No need to spend $0.60/w on micro inverters. A generation ~3+ $0.18/w string inverter with a MPPT for each string handles shade just fine.

Screen Shot 2019-02-25 at 10.07.20 PM.png
 
Last edited:
LOL... ok... I'm not a fan of micro inverters. It wouldn't be a stretch to say I hate them (only because they're ~3x the cost). But you have found one thing they do better.... Neither String inverters nor optimizers can work on 1 panel...
Yes, one panel on our system with micro-inverters.

However, the system at our family’s other house has a SolarEdge inverter with power optimizers. There are 15 panels (I misspoke when I said 14 panels above), all of which are on one string if I am understanding correctly. Even after clearing snow from six of the panels (in full sun), the inverter would not start up. It did after I cleared snow from two more panels.

The bottom line, I think, is that in areas of the US that may receive significant snow accumulations, micro-inverters are most likely a better choice for systems that are coupled with Powerwalls. Alternatively, one could spend that additional $0.42 per watt (the above-quoted incremental cost of going with micro-inverters) on installing a backup generator as a backstop to the Powerwalls and solar. That would not be my personal preference.
 
Very interesting thread, thank you for sharing. As I read your first hand account, it became apparent how much geography influences inverter selection.

Here in Houston the few folks I know with enphase wish they had solaredge due to constant inverter failures in the summer. -- I presume due to the rooftop heat causes the inverter to fail more quickly. (Which would not be an issue in your area). After 6 months the installers don't even bother to assist on the RMA process and you are stuck paying someone to swap out the inverter on the roof. Since we do not have the snow issue, convenience of the inverter location and maintenance are higher priority...

Just shows how complicated a buying experience solar can be and there is truly no one size fits all option.

Great to hear you sharing your experience with your local community

Solar Edge systems are no stranger to failure of power optimizers either, we see them regularly fail, and have over a thousand solar edge systems installed. Solar Edge is a great product and Enphase is as well but rooftops are brutal enviromennts.
 
Could you ask your installer how the strings are set up on the other home? Maybe if the lowest panels along the roof were all on the same string, then clearing those would allow for some solar generation. We have 3 inverters with 9 strings. Whenever we clear snow off the lowest panels, we immediately start seeing out panels produce.
 
The bottom line, I think, is that in areas of the US that may receive significant snow accumulations, micro-inverters are most likely a better choice for systems that are coupled with Powerwalls. Alternatively, one could spend that additional $0.42 per watt (the above-quoted incremental cost of going with micro-inverters) on installing a backup generator as a backstop to the Powerwalls and solar. That would not be my personal preference.

My preference would be to clear the snow off at least ~7 more panels. OR have a ground mount at a >30 degree angle so snow won't accumulate.
 
I wonder if it could be efficient to install a blower near the edge of your panels to run when it's snowing outside to just keep blowing the snow off of your panels/roof. Because in reality you want all of your panels clear of snow, not just one or two, even with micro inverters.

Sure it would use more energy, but maybe the net gain would be higher?
 
Good news - I've confirmed via the Tesla app that the single cleared panel at our primary residence is indeed charging the Powerwalls. I had to change the Powerwalls to Backup-only mode, though, as the mere 200+ watts of power was previously being allocated to our home consumption even though I had the backup reserve set at 97%.

Could you ask your installer how the strings are set up on the other home? Maybe if the lowest panels along the roof were all on the same string, then clearing those would allow for some solar generation. We have 3 inverters with 9 strings. Whenever we clear snow off the lowest panels, we immediately start seeing out panels produce.
I do like the idea of creating a separate string for the panels on the portion of the roof that is harder to clear of snow.

It looks like all 15 panels at the other home are on a single string. As per the spec sheet for the SolarEdge P320 optimizers, the minimum string length is eight panels. If we were to add a couple more panels, though, perhaps we could put a bunch of the panels on a second string. I don't know whether or not our SE3800A-US inverter supports multiple strings, however, as I don't see that information in the spec sheet.

My preference would be to clear the snow off at least ~7 more panels. OR have a ground mount at a >30 degree angle so snow won't accumulate.
For our other home where we have the SolarEdge inverter and optimizers, clearing snow from at least eight panels is probably the best option. So, after further consideration, I am OK with the fact that we did not pay extra for micro-inverters there.

For our primary residence, clearing snow from most of the roof is not easy and would probably need to be done by professionals with harnesses. That sounds expensive. On our lot, ground mounted panels would not produce much because of shading. So I'm happy that we have micro-inverters here. As soon as the next panel or two become mostly free of snow, they'll add to our generation, and so on.

I wonder if it could be efficient to install a blower near the edge of your panels to run when it's snowing outside to just keep blowing the snow off of your panels/roof. Because in reality you want all of your panels clear of snow, not just one or two, even with micro inverters.

Sure it would use more energy, but maybe the net gain would be higher?
That would have to be some powerful blower! This month, we had over six feet of snow fall here. What's on the roof today is 2+ feet of compacted snow and ice.

Probably the most practical solution for snow removal here would be to have heating elements under the panels and on the roof just below the panels. But I'm thinking that money would be better spent adding some vertical panels on south-facing walls.

During the winter, our primary goal is to maintain a decent backup charge in the Powerwalls. While we'd prefer not to miss out on any renewable energy generation, the economic and environmental value of the missed energy due to snow is comparatively minor. Its value is certainly not enough to justify the time I have spent trying to remove snow (though I'll do it anyway). But keeping our home powered through "bad" weather (and reckless drivers sliding on ice and slamming into power poles) is of considerable value.

Like so many things in life, the right answer seems to be "it depends". I think this is what I will present when I speak to the community group. I do appreciate the input here.
 
  • Like
Reactions: Shygar
The only reason why I was thinking of a heating element is I wouldn't want warm water running down then refreezing somewhere else down the line. Thinking a blower at the edge of each panel could just push it before it hits the surface. But yea probably would have to be powerful.
 
The only reason why I was thinking of a heating element is I wouldn't want warm water running down then refreezing somewhere else down the line.
Heating elements would be needed on the portion of the roof that's below the panels, in addition to the bottoms of the panels themselves.

In fact, some people install heating elements on their roofs just to prevent ice dams from forming. Our area doesn't generally stay cold enough for that to be needed. We had a good cold snap this month, but we're talking weeks, not months.
 
  • Helpful
Reactions: Shygar