Micro-inverters vs. Central Inverters: Is There a Clear Winner?

great podcast Stephen, very interesting since this is a heated debate in the solar industry. I think the old timers / traditionalists are fearing the change that micro-inverter technology is bringing to the market. Mirco Inverters, once embedded into modules in lieu of a junction box may even make it too easy and safer to install solar, which many installers and central inverter companies fear that will eat into their businesses.

The reality is that micro-inverters are becoming more "bankable" every day. I'm not sure if Leo Casey read about SunEdison using Enphase for one of their projects. It seems like solar PPA companies are starting to test the waters, which is great market validation for Enphase.

Obviously reliability is a major concern with micro-inverters and no matter how many MTBF tests are conducted there is no substitute for the real world at the end of the day. This is were a central inverters have an advantage, they've been operating in the real world much longer. But how much weight does that factor bring to the table? Remember, Xantrex had a huge inverter recall last year on their residential central inverter lines. The bottom line is that all products will have issues in the field, the most important aspect is how the company handles the problem after the failure has occurred.

Personally, I have sold many enphase micro-inverters to our customers, and they have been operating in the field for over a year and have not experienced many issues yet. When issues do come up, Enphase is really good at following up and making sure the problem is resolved quickly.

Recently I installed a project called, "Battle of the Solar Panel Brands", a module competition that compares the power produced of several solar panel manufacturers. I could not have pulled it off without Enphase because of the module level monitoring and detailed analytics their product provides.

My opinion is that moving forward micro inverters will only get better and more economical

Anyone that has been involved in the solar industry a long time will probably not acknowledge that micro-inverters will eventually put the DC vendors out of business, excluding off-grid applications. The main stumbling block for the entire industry has been the complex mathematics and guesswork necessary to calculate potential system mismatches for everything from wiring loss to different orientations of multiple strings in a DC system.

Very soon, AC solar modules will make design pure AC solar power arrays possible. This will completely eliminated the need for expensive, unpredictable, balance of system DC components. The means no junction boxes, no combiner boxes, no conduit, no dangerous 600-volt external wiring, no central power inverters, etc.

Designing a pure AC system can be done by any handyman, electrical contractor, or building contractor who has professional experience in installing AC power in residential homes.

This completely removes solar module mismatches, string sizing exercises, calculating wiring loss, inverter efficiencies or the limit to the number of panels required to boot up a central DC power inverter.

System can be built with one or 100 modules. All it takes is to install a single 2-pole 30 amp breaker in the homes junction box.

Consumers can go down to Home Depot or Lowes, buy their solar panels, racking systems and wiring kits for a fraction of the price charged by DC solar power installation companies. They can hire a roofing contractor for $50 an hour to install racking and solar panels, then hire an electrician at $50 an hour to install a circuit breaker in the main panel.

The system is flexible, resistent to shading, can be oriented in any direction and the system can used mismatched panels with difference outputs.

The biggest problem with DC power arrays is that they have no idea of individual power production. AC power arrays resemble computer networks and are very easy to track.

@rhoskins I agree with your general statement that AC Modules will make the process of designing and installing solar electric systems much easier compared to a system with a central inverter. I think you are misinformed of the installation process of an AC module/micro-inverters based on your response. You will still need junction boxes and conduit and even though the AC wire does not carry as high voltage compared to DC wire, it's still potentially dangerous.

Even with AC modules/micro-inverters, Installations have to be done according to NEC (national electric code). The circuit breaker goes inside the home's electrical service panel not into a "junction box". You can't just simply stick a 2p 30a circuit breaker into every household electrical service panel, the 120% rule must be considered each time you're installing a system. In the NEC, the 120% rule allows the busbar to be fed from a solar electric system as long as the total maximum current does not exceed 120% of the busbar rating.

You can "make money" (create debt) or you can make sense, but you can't do both until money makes sense:

yahoogroups.com/group/GlobalRelations

JPChance.wordpress.com

RH - You are assuming that typical homeowners can (or want to) design the correct system, create permit packages, file for rebates, find and hire multiple subcontractors (basically act as a General) just because your changing DC to AC.

I respectfully disagree.

A solar project, even with micro-inverters is still as complicated as a complete bath remodel with wiring and plumbing and finish work upgrades, all while ensuring code compliance, and with the addition of a pile of administrative paperwork for rebates. Most homeowners do not attempt to tackle that themselves.

Typical homeowners can also go down to Lowes and buy concrete and rebar, but I don't see a lot of homeowners re-doing their own driveways either.

The bottom line is this:
You are always going to have "capable" homeowners that can do an install themselves. That is the case right now. Also, 90% of electricians currently can understand how to install DC string inverters and roofers can already install mounting systems. The installation is not rocket science. The hard part of PV systems is in the design, permitting, project coordination and administrative work, all of which is why people hire a specialist.

Micro-inverters are a great idea for the simple reason that they have the potential to perform better, identify issues faster and create manufacturer accountability. PV Design is slightly more simple as it pertains to shading and orientation.

Finally, Current Kills.... not voltage. DC systems are not necessarily more dangerous because you have to close a wire loop to get a shock. That closed loop usually has to cross your heart to kill you (limb to limb) with electricity, and the burn danger (of death) is low in 600 volt systems. On the other hand, AC is just as dangerous since Earth can close the circuit with just on hand or other limb on the "hot".

As a project manager, our company has witnessed some of the best inverter companies in the world struggle with large non-grid tied applications such as remote water pumping and stand alone power system. The system engineers struggle to find UL listed equipment that can handle up to 100kW in an off- grid configuration. The idea of micro inverters is intriguing, but I have learned not to over-simplify, at least conceptually. Grid tied and stand alone systems are complex, performing numerous functions, such as battery charging, power source switching, and generator starts and stops. All the while, also trying to protect itself and other system components from overloading, responding to system faults and variable environmental conditions.

I don't envy the Inverter industry the responsibility for designing products that meet our extremely various demands. There is a lot of room for new products and new solutions. I see a gap in the middle ground between extremely simple (residential, just do every rooftop and we'll be fine) and utility scale ( just let the big boys handle this) thinking. The truth is that the industry is learning and growing and creating new products. It will be a very long time before distributed renewables replace grid power. In the mean time I am very excited for the innovations we have and the ones to come. keep up the good work!

Thanks for the excellent format and content. I cannot help but mention that there is at least a third path in the architecture discussion, which involves not inverting at all, but maintaining a grid-connected DC system.

DC power systems advocates use a lot of the same vocabulary these technologists and the host used to couch the discussion of micro-versus-centralized AC inverter architecture. "Simpler, smarter, more integrated and more reliable" was the podcast's lead-in about power electronics, when all of those words favor DC systems, not AC ones. More bemusing was the description of "appliance-friendly AC", which I would argue is not true: all electronic appliances run on DC; and AC's only friend is, in fact, the grid.

More than just possible to optimize Distributed Generation like solar PV's output and electronic devices' inputs around standard DC voltages, it's happening today, and for good reasons. Inverter losses imposed on the panels plus rectifier losses at the device are large vs. the smaller loss in regulation of a DC system. The benefit: AC-based DG-fed systems lose between 10 and 40% when compared to optimized DC systems. 5 years ago, this was often dismissed as "maybe a better idea", but "impractical to implement". That started changing about 2 years ago.

That was when several large mainstream building materials and electronic goods manufacturers (and one small one) started planning the non-profit EMerge Alliance, the first national effort to establish DC power standards. It's grown to over 50 companies today, including utilities, leading lighting and controls companies, architects, and other stakeholders in buildings' infrastructure.

This path to market for solar PV is different than what Stephen and his guests explored, but if the goal is to determine "Clear Winners" in the race to optimize solar PV integration, let's not overlook the simpler more efficient method - even if it takes of a revolution in our thinking to realize it.

I do enjoy reading all the comments on the different solutions that we can come up with. Ole redskin is real sharp as the rest of the comments validity are all true as far as my own experiences.
Keep up the good comments, it sure does keep my mind on track!
Frank Howes
A-1 Energy Solutions

As a real solar installer of over 300 systems, the bottom line is that the micro inverters are mostly marketing hype. In fact, they are really more difficult to install, there's more much more wiring to do (grounding is also very tough) and in all reality - they *are* more complex. I see nothing "simpler" about them. They are typically not code compliant unless you use conduit to cover every AC wire (you see - solar DC is the only exempt wiring allowed in the NEC language)...and so forth.

Enphase and others conveniently omit the complete installation story and parse out only the positive spin. A little disingenuous.

HP said to their developers before they gave the design for a "point and click" input device (the "mouse") to Apple, "A mouse? looks like a rat, Ha Ha..."
IBM called PCs an annoyance, claiming that nothing could shake the stranglehold on IT of the mainframe and that PC Networks would never work compared with departmental mid-range systems.....posted their 1st ever quarterly loss one year later....
And GM used to say that japanese cars were put together with "spit and chewing gum". They called them, "Tinkertoys"
Hmmmm... like the iPhone to cellphones, and the MacBook to Windows, AC panels are coming........get used to it!
We need to stop throwing sand at each other and embrace ANY technological breakthrough that makes solar better, more efficient, safer, and easier to adopt. Its not AC vs DC solar, its solar vs. natural gas, clean coal, and others. We need to stick together and put the right solution in the right application!

A PV array integrating AC distribution on residential roofs is labor intensive regardless of system used. Design improvements implementing interface materials that can be compliant to NFPA safety standards can reduce installation costs. As the current NEC Article 338 permits the use of SE style cable for Service-Entrance conductors, PV ganged modules localized together into a single SE distribution run from the roof down to the Service Entrance Equipment without using conduit is possible.

This may be only one answer to how system interconnections can reduce time consuming applications for more economical methods.

if companies put microinverters on panels when they were shipped, this would be a great first step. grounding the inverter to the panel and then having a easy clip on grounding method (like tyco's clip) would make grounding easier. it is only sensible to track every module. how else do you watch out for warranty issues? panels make more energy, clouds create less drag, over 25 years, that is a lot of power. plug and play makes it easy to install. solar was always easy to install, aim a panel at the sun and plug it in, thats all you need to do really. a string of 16 panels is only 10 amps or less and i can imaging people just plugging it into a wall outlet one day and your meter will spin backwards.

What an adventure! I had no idea you could really build your own solar panel. That totally rocks though. After I read this post, I went on the look out to find some reviews on different programs. This site seriously helped me out. -Global Web Info

Strongly suggest use modules which are listed for safety and practical safeguarding.

Very interesting discussion. I like the elegance of the Microinverter but I see it as having a number of hurdles it will have to jump over, some of which were brought out in the discussion:

1. Reliance on discrete less reliable components - Electrolytic capacitor are notoriously unreliable, especially in extreme environment. In spite of their low cost as a consequence of their widespread use in consumer products, for power applications such as this they will fail. Whereas, the electromagnets are inherently reliable.

I think using the PC components as a model for reliability of the micro inverter is a mistake because, in general, PC's typically have lifespans of no more than 5 years and often shorter. They're not built to last like a PV system: They're built to be serviced, and more often than not, replaced.

2. Reliance of the overall system - I'm not sure that the micro inverter folks really understand the reliability issue from a system perspective. When you take the failure rates of using discrete components (1 above) into account, if you have "X" number of inverters all working together, you've exponentially not serially increased the failure modes of the system and decreased its overall reliability.

This high failure rate (really indetermined at this point since the consumer products using these components rarely are fields for longer than 5 years) necessitates replacements of system components (micro inverters) which translates into higher routine maintenance costs. Granted, they can be deferred (presuming the failure is isolated to the micro inverter and doesn't propagate because of some inherent design defect) but eventually someone has to crawl up on a roof and fix the offending unit. That's labor cost, insurance cost, etc.

3. Need for an internet connection - It's nice that the micro inverter folks are collecting data, metrics, on long term performance of these system, but how does that benefit the consumer? He has to pay fot the internet.

1. What's the thinking of: "Evaluation of Electrolytic Capacitor Application"
http://www.enphasenergy.com/downloads/Electrolytic_Capacitor_Expert_Report.pdf

2. Maybe so - not aware of rate in which occurring. May be a differentiator between one micro-inverter manufacture and another.

After thinking over this microinverter technoogy, which I had rejected last year in my solar PV design for my new home, I have completely changed my opinion and now see this step as a huge advance in bringing PV systems into more homes. I care not what the large commercial designers do, but for homes, a microinverter system simply makes too much sense to avoid. If
neither system currently existed, and two competing teams were to present their system designs to a panel of judges, I would venture to guess that
the central inverter/DC-bound system proposers would be laughed out of the room. It would be like current English spelling competing with a phonetic version of the language : no contest. The central inverter architecture is absurdly (and pointlessly) complicated and anything but user-friendly. It's a design that only the local electrician's union could love. No designer would ever come up with anything like it that if AC panels were available, which they now are, with microinverters. This advance in the PV system technology has brought such systems into the realm of many homeowners and DIY'rs, who should feel confident that they know enough right off the bat to attempt their own installation. I know I do.

Also specifically for 3-4kW systems - simple solution. For ground-mounted arrays approximately ~8 modules is the sweet spot at a distance for instance ~200 ft on a single backfed breaker.

Micro-inverter are becoming more "bankable" everyday... but, do you think that micro-inverters could be use in a close futur both in commercial and solar plant application ? Why ?

Yanick / From Paris FRANCE

One draw back I have found after testing the Enphase inverters, is when they are not producing power each one is drawing around 14 watts of energy from the grid which is very waste full . The only way I have found to fix this is add an SSR at the grid tie point which is connected to a small seperate solar panel, only 50 ma. is needed to control the relay. Has any one else seen this problem?

@schulzet - 14 watts per inverter? that sounds excessive, how did you come up with that number? Remember, the Enphase Envoy, EMU is also drawing power...

I connected a 175 watt panel to the inverter. Inserted Tektronix
TX3 DVM reading RMS AC current in series with 240 output of inverter tied to grid (240vac) and adjusted panel in sunlight for maximum AC current, around 0.6 amps output, then covered panel. Current read around 0.06 amps, which doesn't seem a lot until you do the math. P=IxE, 240 vac X .06 amps= 14.4 watts no sun.

You may wish to follow-up with the manufacture for more information.

The specifications are about ~580mW per inverter.
http://www.enphasenergy.com/downloads/8261_Datasheet_24_32.pdf

For instance, an SMA 4000 inverter would be approximately ~0.1W
http://www.sma-america.com/en_US/products/grid-tied-inverters/sunny-boy/sunny-boy-3000-us-3800-us-4000-us.html

The specifications are about ~30mW per inverter.
http://www.enphasenergy.com/downloads/Enphase_M190_Datasheet.pdf

What do you think?

I attempted to contact some one at Enphase and went through four different so called tech people, started at the low man and went up the chain, but got no explanation of this current draw. I checked out 6 other inverters with the same problem. I guess I could have six bad ones or there spec. sheet is wrong, maybe someone else can run a similar test and see what results they get.

What about phase angle? For ac systems P=I*E*cos(phase angle)

By the way, looks as if the 'night time power consumption 46mW' and a 'Warranty 25-year limited warranty' What do you think?

http://www.soldist.com/wp-content/uploads/2011/05/Enphase_M215_Datasheet.pdf

Enphase M190 datasheet also states a night time power consumption of 30mW

PVinstaller: I think you're missing a big part of the advantages of micro viewing it from an professional point of view. You as an experienced installer know all the tricks it takes to set up and design a DC system. That knowledge is what people pay for. Micro brings non professionals or crossover tradesmen into the solar game and allows them to provide better results than if they had tried to tackle a DC system. I disagree with some of the comments claiming AC is just as dangerous as DC, it's a known fact that DC voltage with high amperage behind it kills! In AC you need to get up into high voltage that locks your muscles and keeps you stuck on the hot line. The reason AC was chosen for power distribution is safety and copper wire size. As a DIY, I'd far rather install the plug in play M210 than have to work with a LIVE high amperage DC string as an amateur. As for DC being code exempt, you know very well that good installation practices are to protect all wiring DC or AC in conduit or junction boxes for reliability. This technology will cost the solar installation industry some customers. I have extensive project management experience in technical installations and seeing this leads me to believe me and my electrical contracting friend can get into this business and sell some systems. (See the advantage?)

From a consumer standpoint, the biggest hurdle we have is wrapping our heads around the "cost per watt" system of charging for panels, inverters, and labor. It's just completely insane. If PC manufacturers charged by the megabyte starting in the early 1980's, how much would a computer cost today? Once you all stop ripping off the public, the entire solar industry is going to boom much like technology and internet did in the 90's. Currently, it makes little financial sense to go solar. The biggest benefit of the internet was not having to put a stamp on every letter we write. Make solar make sense.

Well, If we are talking about science future/fiction it won't be breakers in the panel.. The homes will be solar ready with an outlet like I have for my RV right now.. you just will go to Home Depot and buy your stuff and plug it in a ready-made plug just like you do your Christmas Lights or Yard Lights. Look what has happened with swimming pools in the last 10 years. 20,000 gallon 5 ft pool that lasts 5 years for 399.00 with filter and cover. Furnaces used to have to be assembled in the homes of customers.. now we almost pack them under 1 arm into the home. They last 20 years and are literally disposable now. You are correct about needing to meet the NEC codes and obtain permits. As long as there is a potential danger to life or property a permit will need to be in place for the propler final inspection.

I like the AC solar idea and think that it will change our power rates forever.. UP.. UP and away... Someday we won't need a power company.. just like we don't buy whale blubber anymore either. I also believe the Hydrogen technology will eventually end up being the storage meduim (the battery) for solar and wind power production... When in doubt follow natures examples for the best efficiency or technology. Maybe I should buy some electric eels... (hahha anyone know how they make power?)

Damon