Micro Inverters (My Turn)- the boring technical stuff

Published on September 9, 2008 by David Llorens.
Categories: Solar Trends.

A few weeks ago, Dan wrote a post about micro-inverters for beginners, comparing them to the inner-workings of a Peruvian diamond mine.   This article is more technical and boring, because that’s who I am, Mr. Technical and Boring.  (Remember the Mr. Men books? OK, never mind)…

I believe micro inverters will be a major new part of the residential solar energy market (the other being photovoltaic and hot water integrated into one panel, but that’s a post for another time).  There are definitely competitors getting in the microinverter game, but for now, Enphase microinverters are the only game in town. Much of this article is based on their product.

I’m going to go over all the advantages and disadvantages of micro-inverters as I see them.   The big picture that needs to be understood is that, with microinverters, we’re talking about one inverter PER SOLAR PANEL all linked in parallel, whereas a traditional string inverter configuration links several panels together in SERIES.  This allows microinverters to maximize the energy harvest of each panel.

First, let’s go over what max power is.  Well, Power (with electricity) is voltage * current, or P=V*I.  Solar panels change their voltage output based on a number of factors, the most dominant of which is how hot they are (up to 50%! that’s why some neighborhoods can get more out of Solar in San Francisco than in some places in Arizona!).  The panel current also changes based on how much light is hitting them (irradiance).  The maximum power of a panel is just the highest number you can get by multiplying it’s voltage and current.

Take a look at this graph courtesy of the NABCEP study guide for PV Installers:

You can see that light hitting panels changes their max current, and heat changes their max voltage.  The Max Power Point moves accordingly (Imp and Vmp).

With a string inverter configuration, the inverter is getting several of these panels linked in series.  In series, their voltages of the panels add together but the current does not.  It’s the inverter’s job to try and track the maximum power point, but it can only “see” a string 0f panels, it doesn’t know which panels are doing what.

The big benefits of micro-inverters come from the fact that they can maximize this power point for EACH PANEL. So that if one panel is say dusty, faulty, aligned differently, a different type of panel, etc., it does not drag down the whole string. Additionally, conventional string inverters are limited in the configurations they can accept, and disregarding those limitations can sometimes result in zero system output at times.  micro-inverters, on the other hand, are free of most of those limitations.

Due to this individualized “panel harvest,” Enphase claims you will see anywhere from a 5%-25% gain in power output when compared to a string inverter configuration.  The low end of that range would assume a well engineered string configuration with panels on the same orientation and tilt, without shading.  You would need some heavy shading or panel orientation differences to achieve the upper end.

ADVANTAGES

Modularity – Scalability

Let’s say you want to add an electric car in two years, but you don’t want to oversize your system now because you won’t get a check from the utility for excess production (or even a thank you.) Due to the nature of microinverters, you can almost always just slap on another panel/micro-inverter.

• Ideally you’d want to get it sized right the first time to avoid having to increase the racking system and number of roof penetrations, but this is still more scalable than a string inverter.
• With string inverters you have to adhere to a regimented string design, and an upgrade would require a number of things from a different inverter to a large wing of panels to fit the string configurations. For example, a 3000 watt string inverter may be limited to five specific configurations: 2-strings of 7-panels, 2-strings of 8-panels, or 1-string of 9-panels. That means you are limited to 14, 16, or 9 panels. Not a lot of scalability there.
• Up to 16 micro-inverters can be linked in parallel per AC branch. That does not mean you can only have 16 max panels, you can have literally any number of them from one to whatever your electrical service will accommodate. It is just good because you can fit more per branch than most string inverters which may cap out at 11 or 12 per string. This isn’t a giant advantage but it’s nice for the installers wiring small systems.
• No clunky box downstairs. These inverters go on the back of each panel, on the roof, so there’s no need for a big string inverter downstairs by your meter. That eliminates concerns about string inverter placement and clearance issues, conduit run eyesores, keeping the inverter out of the sun, etc. Also no need for DC switching points.

Shading

This is the big one. If you shade enough panels in a row, you can bring the voltage of the entire string low enough so that the inverter just stops, creating zero power output. However, if you have say, 5 panels shaded on an 8-panel array, those 3 panels just keep kicking with micro inverters, each one creating maximum power.

Furthermore, if dust, grime, or bird poop are getting on panels disproportionately, they can drag down the power of the whole string, much like stepping only a small portion of a water hose. Remember: the string inverter maximizes power output for the whole string, not each panel, so if some panels are suboptimal, so is the whole string.

Differences in Panels

Enphase conducted a beta test by creating massive solar installs in a checkerboard pattern. Meaning, every other panel on the roof was connected to a micro-inverter or a conventional string inverter. Over time, panels got more and more mismatched because of dust and de-rating differences. As a result, the power output of the micro inverters increased comparatively to the string configured system.

• Production methods for panels have gotten better and will get even better, but still, panel to panel mismatch is exploited best by microinverters, maximizing each panel’s max power point
• If you live in a home where certain parts of the roof are sprayed with road dust, the impact could be particularly significant
• This would be uncommon, but if you have a bunch of different panels lying around, or you have a system installed but your buddy gave you a bunch of different type of panels that fell off a truck, you can mix and match them with micro-inverters since each are independent to their own panel. As string inverters require that all panels on a string be the same orientation, they must also be the same type.

No Single Point of Failure

This is really up for debate. Some would argue that no single point of failure is better because if one inverter goes out you still have power production while you’re waiting to get fixed up. Others would argue that you have simply multiplied the possibility of failure dozens of times. Currently, I believe what I hear about their design practices and think these things, although not proven, will last considerably longer than string inverters. I hope I don’t have to eat my hat.

Burst Mode

In the early morning and early evening there may not be enough current to drive a string inverter. It will “flip on” when it gets enough light. Micro-inverters (at least Enphase models) have a “burst mode” which stores up energy until they can release it in these low light conditions. This allows for SOME power output in times when you may have none with a string inverter. I don’t have numbers on how much, but I don’t think it’s a ton or they’d be making a bigger deal out of it.

Cheaper Labor Costs

Electricians do not need to be there all day wiring, they simply perform the connection at the main panel. Installers are doing most of the connections up on the roof. The installation is also quicker with less conduit bending and what not, so some money can be saved here.

Monitoring

Another huge one. Because it’s panel by panel, monitoring system sophistication is really powerful. Not only does the software know your system as a whole is underperforming, but which panels are the culprits. The monitoring system is smart enough to know that power depression on all panels is probably a cloud, whereas if one panel is blinking on and off you’ve got an issue. You currently can’t get this type of information with string inverter monitoring. There also is some other cool whizbang stuff like getting text messages if a panel is out. Your installer gets this info too, so as soon as you know there is a problem, so does the installer. The “EMU” is smart as hell and just plugs into the wall, grabbing RF signals that the Enphase modules throw out. No need to connect it directly to the system!

How much is the monitoring price? I think they have different programs but from what I understood, they recommend salespeople offer 1 year free and $10/month for the whole system after that (with lump packages that make it slightly cheaper than $10/month).

Orientation

Let’s say you have a small roof with a west and south face. You can fit 10 panels on the south face and 5 panels on the west face. String inverters require all panels on a string to be the same orientation and tilt, so a string of 5 and a string of 10 is probably not a possible configuration, and splitting them up 8 and 7 won’t work because of this requirement. Even if you were allowed to perform these two string configurations, different string orientations need to be carefully designed and can lead to disastrous situations if not carefully looked at, something which even seasoned solar salespeople may not be able to do correctly. This allows for roof area maximization.

DISADVANTAGES

Field testing data is relatively early

By relatively I mean that they have thousands of these things operating in the field, it’s just that they’re supposed to last for decades so it’s early until we’ve reached decades. This is a big issue for most people considering micro-inverter systems. As you may know I have a big beef with this sort of thing. “We are waiting for new technology! We are waiting for new technology!” Then you bring new technology and everyone is like “It’s untested! We don’t want it!” So here are some reasons why it’s not a big deal:

• Their MTBF is calculated for 119 years. They’ve got some people on their design team that would design stuff to be put on power poles in the middle of the desert. Believe me, they understand very clearly that if these have a high fail rate, they’re screwed as a company, so they put a whole ton of effort into nailing that beforehand.
• There are lots of micro-inverters in the field and according to Enphase, none have ever failed. There are more being installed every day.
• They have no moving parts and soft switching, unlike some fan cooled string inverters. That should aid in lengthening their lifespan.
• In California, inverters are all warranted for 10 years anyway, and the lifespan of a string inverter isn’t much longer than that. I could see being concerned in say, year 5, if tons of them started failing all at once and the company goes belly up. This is a legit concern, but I think it’s worth the small gamble based on what I’ve seen, plus they just got $15M in more in funding.

Exposed to the elements

So yah, they’re up there on the roof. They are shielded by the panels, but they are outside none the less. (However, I’ve already gone over how these are designed to be rugged as hell because the company’s future depends on it)

Limited in some applications

• No off grid situations, not overseas yet (50Hz), and haven’t been tested or designed for other forms of renewable energy other than solar
• They DO have a 3-phase version for commercial applications but that is outside the scope of this article

Are they cheaper or more expensive?

The answer is that it depends. I didn’t put this as and advantage or a disadvantage because it’s complicated. Currently Enphase is $200 MSRP last I checked. Let’s call a 3000 Watt string inverter $1600 for simplicity’s sake. So once you go over 8 panels you’re looking are more money on the product side, but you’re weighing that against some other things like, well, the advantages listed above.

The extra power output is worth money, and so is the decreased labor costs, so you have to do that calculation. On a tiny system, micro-inverters are a slam dunk in my mind.

What’s Next, who’s entering this market?

There are other micro inverters that will be manufactured, as well as some string inverter technology that attempts DC based max power point tracking like National Semiconductor’s Solar Magic technology, which as far as I’ve heard, isn’t in the field yet.  There’s something called “solar bridge” but there’s virtually nothing on the internet about it and I haven’t seen it in the field yet.

So that’s probably more than you ever wanted to know. My take is that for residential, you can’t beat these things, they rock. In some large, perfectly unshaded, same orientation, residential applications you could still make a case for string inverters I think.

If you’re actually still reading, here’s a good video:

Battle of the Bay – The Clean Tech Initiatives of San Francisco Vs. San Jose »« What do your U.S. Senate candidates really think about renewables?