The Hidden Downside of Microinverters No One Talks About 

This is a blanket response to all those below saying I’m wrong about the Christmas tree analogy. To those screaming that it’s wrong and dangerously misleading to use this analogy because it no longer applies because of bypass Diodes, please keep reading. The the Christmas tree light analogy works perfectly in explaining how solar was traditionally wired and installed. Therefore, is both useful and applicable. To say that it is not as like arguing that explaining how internal combustion engines work is no longer necessary. With the advent of modular level electronics such as micro inverters or optimizers used technology to overcome that very problem. Understanding that basic issue will help consumers in understanding how solar works. Bypass diodes are just a new form of a modular level electronics that helps overcome the shade problem. String inverters coupled with panels with bypass diodes may be just as good at controlling the shade as any other technology. That’s a moot point today here in the US where it’s not offered to residential consumers because of the virtual duopoly between Enphase and SolarEdge. Plus, if you just use a string inverter, you lose the advantage of modular level electronics that give you monitoring access at a panel level, which is super important to consumers. If you don’t have that, you won’t know if you have an individual panel out easily. Also to be 100% clear the most important thing for a consumer to do is use an installer that they trust to do a good job and take care of the system for a long time with a good warranty that covers labor. That’s far more important than an individual technology choice you make. Hope this helps clear things up.

Totally incorrect regarding string inverters. Most panels from the last five years have bypass diodes. Some are even half or tri-cut. So, while shading a single panel, you will only lose power on that panel, or even a portion of it if it is half cut or tri-cut. And in high heat zones, optimizers or microinverters fail every few years, if manufacturers go out of business, you must find the identical optimizer or microinverter to ensure that the system functions properly. But string inverters can be replaced with any manufacturer's product. The main current disadvantage of string inverters is that you cannot monitor panels individually, however manufacturers like as Huawei are working on solutions for that as well.

Thanks for this. A bit late for me, since my system was installed six years ago, but still enlightening. I don't know if the DC optimizer was even an option back then. But I have found that the microinverters on my system are _incredibly_ *unreliable*. In six years, five have failed so far. There are only 20 to start with, so that's a 25% failure rate in just the first six years, averaging one failure per year. And of course, they are way up on the roof, where it's hard to replace. Fortunately, the installer has been gracious about the whole thing. The inverters are under warranty, but the manufacturer doesn't cover labor. I had to pay labor for the first two repairs, but the installer has agreed to cover labor as these continue to fail, given how unreliable they have been. Who knows if I'll still be in this same house if and when the panels have degraded enough to require replacement, but seems that already, and probably even more so by then, the technology has been advancing quickly. If it weren't for the cost, I'd love to upgrade my 300W panels with 400W ones, and switch to the DC optimizer approach you suggest here.

I'm glad to find out we have another way to avoid string inverters. EEVblog on episode 1426 went over an issue he noticed because a tiny amount of shade on a sting inverter system was causing it to produce significantly less power when the shade was there. An easy way to think about the problem with a string inverter is to not mix new & old batteries. This is because the batteries with a lower output is going to encourage the other batteries to try to recharge that one battery & the same thing would happen with a string inverter system.

I’m an engineer that actually has designed solar systems from the utility scale to commercial and residential applications for over 15 years. Dc optimized systems like the ones offered by SolarEdge are garbage. Not because of the technology but because SolarEdge, the largest manufacturer optimizer technology refuses to replace the components when they fail, unless you threaten to take them to court. Microinverters such as those from enphase have the lowest failure rate of any technology and there’s plenty of data to back that up. They actually test for that. But for the average homeowner, that doesn’t have a lot of shading of their roof by trees etc, then latest generation of solar panels (with bypass diodes) coupled with a high quality string inverter will be just fine. It’s not usually worth adding 20%+ to the system cost for a 5% improvement in performance. Also, black on black panels are purely an aesthetic choice but one that makes sense for residential applications as it has wider curb appeal. Most tier 1 manufacturers are making such panels these days at affordable prices. Finally, the statement about subcontractors is hit or miss. It would be ideal to have all work performed in-house by a company with a long track record of solar specific work. But that’s not the reality everywhere. The more realistic solution is to have a dedicated solar installer that has a sales arm that drives business to them. So, technically it’s not subcontracted since all the work is being performed by the contractor but the consultation/sales part is outsourced because those are two different specialties.

I think my one year old ground mounted emphase micros are warranted for 25 years so maybe they’ll last a long time? I went with ground mounted solar because the roof needs replacing in a few years but I’m starting to think it’s the best way to go anyways.

My experience is the opposite, DC optimizers seem to fail way more often then the micro inverters. I've yet to do an enohase replacement, but have run into many soleredge optimizers that failed. With rapid shutdown rules, micros work really well. Batteries, yeah that's marginally better, but it's a fairly minor percentage.

Your video was very informative, you obviously know your solar systems. We are retired and live in the UK and we had a 10 panel system installed less than 2 years ago. It was fine at first then the inverter (Solis) started sending alarms which all stated that there was an arc fault which should be dealt with and these have got more frequent lately. The installers came back once and said they found a faulty connecter under one panel and fixed it but the problem persists so I think that more connectors must be faulty. They want to charge me a lot to come back and fix these so I am waiting for another installer to look at the system. I suspect that these connecters are faulty or have not been correctly wired so my question is - is this a common fault with roof panels or do you think we have just been unlucky? I would appreciate your expert opinion on this.

I respect most of what you're putting out here but the claim that optimizers have lower failure rates than microonverters runs counter to basically all industry data on the actual failure rates of these rooftop products. Do you have data on a specific optimizer showing long lifespans on them?

Sounds like a very good point. Thanks for helping crystalize my understanding of solar technology. I heard Tesla has just released a great integrated battery system that accepts DC from the panels and is easy to install. With utility companies not wishing to pay much for produced electricity, sounds like batteries might be the future anyway.

At least in my area N. California the discrete panel minoring is disabled to the consumer. I think the main reason is to cut down on service calls. People that don't understand the system would be calling their installer right and left when they saw something they think is a problem when it's not.

Optimizers are fine but increase cost just a tad over micro inverters. Secondly You'll have to replace the inverter aprox every 10 to 15 years on an optimizer system and the warranty is usually 10 years which means out of pocket for a new inverter. Another disadvantage is on a larger system you may need to or more converters which means more money when they fail. Those inverters are very expensive and fail due to the high heat and amount of wattage they are converting thus the reason for 10 year warranty. Micro inverters slightly more efficient , Warranty is 25 years, and according to installers I have spoken with extremely rare to have one fail. One installer that put in well over 200 systems says he had to replace two one which was bad from the get go the other failed in warranty and is a plug in connection. Now for charging a battery advantage does go to optimizers. They deal DC to DC with minimum loses though micro inverter systems even with the loses of converting is extremely low percentage.

Interesting. Thanks for the education. Are all microinverters the same? Are there no ones with added tech that voids the issues you talk about like this new one that comes with Anker Solix X1?

What’s your recommendation on dc optimizers?

Just normal string inverters work very well. Where i live circumstances (normal yield) are worse then in Ottawa. Still a string inverter works very well here. I am in the solar business for more then 10 years and the chistmas tree explanation is just wrong. I prefer not using micro-inverters and really do not like Solaredge (Tigo optimizers are ok, they dont fault)

What about bi-facial PV panels?

Deye hybrid inverter solve the problem with storage of energy in the battery because of the generatorport. My oldest micro inverter is 22 years old. If you don't buy a Chinese micro inverter it will last forever.

Agree with Niall, you lost me when you mentioned about christmas light effect. There are plenty of videos showing actual tests , showing bypass diodes do a lot of the work to bypass the panels which are shaded. Sure microinverters /optimisers may give better/optimal output but these generally only make 3-5% difference, yet cost is around 10-20% more, not to mention, as you say, you may have more maintenance over the long run replacing the microinverters/optimisers.

Can Tesla inverters have optimizers?

Videos like this one do the most damage because the channel presents well, and the presenter sounds like he knows what he's talking about. 80% of what he says is correct, and is valuable information. This makes it far more likely that the 20% of downright false information imparted will be accepted, absorbed and acted upon by viewers. As others have said, the reality today is that the Christmas light effect really no longer matters. Most panels I'm looking at have three bypass diodes per panel, meaning a 12 panel array is in fact 36 independent solar production zones. Shade one of those zones, and the 35 other zones continue to produce completely normally. You'll only see production drop for the zones that receive shade. The video also fails to specify that even back in the Dark Ages when bypass diodes were not being used in solar panels, the shading issue only really affects situations where there is shading. I know it sounds obvious, but it needs to be spelled out for some folks. MANY installations, if not most, won't suffer from any shading at all. So yeah - really, DC optimizers only really make sense to me if you want per panel monitoring. Otherwise they don't seem to be worth the considerable expense. Just get a good set of modern panels and a good string inverter, and you're good to go.

This is a blanket response to all those below saying I’m wrong about the Christmas tree analogy. To those screaming that it’s wrong and dangerously misleading to use this analogy because it no longer applies because of Bypass Diodes, please keep reading. The Christmas tree light analogy works perfectly in explaining how solar was traditionally wired and installed. Therefore, it is both useful and applicable. To say that it is not is like arguing that explaining how internal combustion engines work is no longer necessary. Modular level electronics such as micro inverters or optimizers use technology to overcome that very problem. Understanding that basic issue will help consumers in understanding how solar works. Bypass Diodes are just a new form of a modular level electronics that helps overcome the shade problem. String inverters coupled with panels with bypass diodes may be just as good at controlling the shade as any other technology. That’s a moot point today here in the US where it’s not offered to residential consumers because of the virtual duopoly between Enphase and SolarEdge. Plus, if you just use a string inverter, you lose the advantage of modular level electronics that give you monitoring access at a panel level, which is super important to consumers. If you don’t have that, you won’t know if you have an individual panel out easily. Also to be 100% clear the most important thing for a consumer to do is use an installer that they trust to do a good job and take care of the system for a long time with a good warranty that covers labor. That’s far more important than an individual technology choice you make. Hope this helps clear things up.

In the US, if you are roof mounting, either DC optimizers or micro-inverter are absolutely required under every panel for the emergency shutoff feature. Simply stringing the panels is not an option and the whole shading problem therefore disappears.

"choosing an installer you trust to do good work and service the system for many years" - that's an impossible task for an average consumer. I think, one can get the most peace of mind if he gets technical and supervises and inspects the work himself, and then, instead of paying A LOT upfront for a 25-y warranty (which he may or may not get), chooses instead to pay out of pocket for all ongoing repairs (or continue education and even do them himself - it's another option). I'm just trying to make solar economically viable. Pay 10s of thousands upfront and wait years to recover the cost IF all goes well - this doesn't sound very reassuring.

The whole point of micro inverters is to minimize wiring and eliminate routing dangerous levels of DC on the roof.

Sounds like you haven't heard of bypass diodes ... most modern panels have 3 so this all 'Christmas light syndrome' really is more myth than fact.

Why, for a start, don't they install the microinverters downstairs?

OK! This is a rather dumb slant on what is probably intrinsically poor product design from certain vendors. Yes, Micro-inverters can fail if they are not used inside the specified range of current and voltage in/out. Yes, DC optimizers can fail if they are not used inside the specified range of current and voltage in/out. The bottom line is that BOTH of these devices are actually electronic switch-mode inverters. And BOTH can and will fail if used beyond their claimed ratings. The input from solar panels to the DC optimizer or micro-inverter is the same. The OUTPUTS are not. And there are minor differences in the way such systems can be wired. The main advantage of the DC optimizer over Micro-inverter is greater simplicity. Because it is basically a tried and true DC-DC buck-boost switch-mode technology coupled with MPPT input load optimization to "sum energy" into a common bus back to the actual DC - AC inverter powering your home. The more common topology is to use DC optimizers under solar panels to directly charge a local battery storage system. This is cheaper, more efficient and more reliable overall. The DC-AC inverter to power your home then is from the battery only. Some readers may find this helpful... www.mysolarquotes.co.nz/blog/how-solar-power-works/micro-inverters-vs-string-inverters-a-comprehensive-guide/#:~:text=The%20power%20rating%20of%20a,of%20235%2D440%2B%20W.

Batteries cost a lot. How long do they last anyway? Perhaps the most profitable approach is to just use inverters to help run A/C during the sunniest time of the day. A maniacal desire I see in some people to turn every ray of light into electricity might be eclipsing common economic sense.

Why does black on black matter?

Why would I install ANY expensive, critical devices that, as you say, are guaranteed to fail. DC, micro or otherwise? I have yet to see any ROI projections that include these costs. Solar as sold today is a fraud.