I'm reading up on individual PV panels that convert DC to AC without going through a main inverter. How does that work? How do they overcome phasing issues? If it's down, how do they know when to NOT feed power back onto the grid? .
Hill, do a quick search on this site for Enphase micro inverters. Dress has just completed an installation. Also look here: Enphase Data Access - Solar Electric Power Discussion Forum by Northern Arizona Wind & Sun (there are many thread there about Enphase, with lots of info both pro and con.)
Yep, just put 18 180w ET Solar panels and 18 Enphase micro-inverters on my roof last month. Been operation now for about 3 weeks without any issues. Enlighten | Encinitas, CA Residential System I would definitely recommend them. But they are currently a bit more expensive than a central inverter. The inverters themselves work just like any other DC-AC inverter - they sync up to the grid and feed whatever power the solar panels are making into the circuit. Now you have one per panel instead of one or more large inverters depending on your install. They use a number of measures to power down when the grid goes unstable. It's all standardized.
Congrats on your system! The internet view is slick. I understand you have to pay extra for that? I've seen SOME web access system that are free, but it doesn't seem to be the norm. I watched a few you tube videos on micro inverters to see how they work. Seems that each micro inverter has to be part of the disconnect scheme when/if the grid goes down, as opposed to 1 or 2 big inverters. Looking at cost (dis) advantages ... it seems more labor intensive to hang/wire 20-40 inverters rather than 1 or 2 big ones? And if I have this right ... the cost (per Kwh of inverters) is a bit more? Maybe eventually they'll be the same/less than big ones? Any studies yet on the effects of heat? Our 2 sunpower inverters hang in the garage. When they're converting at over 7Kw's ... the cooling fans really hum along from the current passing through 'em. I'd think with the heat characteristics of DC/AC conversion, along with roof temp's ... there'd be an extra issue to tackle for micro inverters. One installer commented on a you tube video that "if" a micro inverter crapped out in the center of your array, you'd have a larger issue with labor / getting to it. That assumes you'd have more the 2 rows, which is not the case for many installs. The micro inverter warranty ... similar to big ones? I suppose they have to compete with big ones, or there'd be no incentive to go this rout. Micro inverters ~ pretty interesting. My knee jerk reaction is they're different .. but not necessarily going to put the alternate out of business. .
Thanks! Unfortunately, yes. It's about $2/year/inverter. Should be half that, IMO, but they've basically got a monopoly on the micro-inverter market right now. There are a few others I've heard about - I'm hoping they help drive prices down. Each inverter automatically shuts off if the grid goes down. The inverters/panels are arranged into strings of up to 15. Each string feeds into it's own 15A 2-pole circuit breaker. Maybe - didn't seem like much extra work to bolt them down and wire them up to me when I did it. Yes, they currently cost about $1/watt (~$200/each). A central inverter costs about $0.50-$0.75/watt. I would expect them to command at least some price premium as there are more components compared to a central inverter. Only an issue if they get over 65*C. They have an internal temperature sensor (the data gets logged to the Enlighten website where you can monitor each inverter's temperature) which seems to read about 15*F over ambient at it's peak. The inverters are in the shade an inch or two under the panels. Panels are mounted nearly flat about 6" off the roof surface. Many people think it may be an issue, but it doesn't seem to be. There are no fans - they cool completely by convection. With a 200W max output, they only have to dissipate 40W. They sit in the shade under the panels, so the panels take most of the heat. Convection naturally circulates air under the panels. If my panels were mounted at a steeper angle, they'd be even cooler. Enphase has tested them in Palm Springs in the middle of summer (110*F) and found that the inverters stayed plenty cool. How many installs have more than 2 rows? Even then, it only takes 4 bolts to remove a panel. Takes 5 minutes to remove one. Better. The Enphase units have a 15 year warranty. I don't know of any central inverters with more than a 10 year warranty which seems to be the standard.
I have been considering a PV installation. Micro inverters seem like a great concept, allowing independent operation of different panels of potentially different ratings. The complexity of the system is increased a bit, but the MTBF rating for the modules (enphase) suggests a reliable system for the most part. I like the idea that I could start with just one panel and one inverter and plug it in and forget it. I don't think that the construction of the system would be more complex than having a central inverter. In some ways it is easier because the system is directly delivering usable AC power. I had some questions for Enphase so I called them today. Is there a fee for monitoring the system? Yes (as above) they want $2 per inverter per year. Solar enthusiasts are frequently a bit frugal and I know it rubs me the wrong way to continue to pay a fee to get full functionality from my system. Who knows what the fee will be next year. In addition, they 'cripple' the system if you don't use the internet features. You cannot get data out of the system unless it is internet activated. That sounds too much like something Microsoft would do! So, in effect, they have lost me right there. Maybe someone will come along soon with an 'open source' inverter or maybe we'll have a 'jailbreak' for the Enphase units someday. Another note - I asked about how to use the Enphase system for a totally off-grid application. The rep said simply "You cannot". Of course, this is not true, but he was getting a bit sassy at this point (apparently salesmanship was not part of his training). I suspect that if you simply ran a pure sine wave inverter to allow synchronization for the modules, then they would sync to the inverter. I guess you would have to take the 240 V and drive a battery charger for the times when you had enough power to do so. I also asked if it was possible to drive just a 120V line with the full power from the inverters for a 120V only application. Again he said 'not possible'. I would consider this a 'wrong answer'. Perhaps it is true that there is not a config for 120V only. ( the 240 volt lines are both hot and there is a neutral / ground ) But one could drive a 2:1 step down transformer for 120V only in order to get full 120V power, albeit with added expense and loss of some efficiency. So, with a software hack or jailbreak, I think the Enphase would do it for me, but I'll wait for a while.
I'm in the process of installing a PV array on my home. Because my house does not have a perfect southern exposure, I will be forced to split the array across two sections of the roof, one that has a 30 degree angle to south and one that has 60 degree angle to the south. The 30 degree to south roof has the best exposure, but it is not big enough to hold the entire set of solar panels. This means that the two sets of panels will have different outputs at any given time of the day. According to my installer, if we use a centralized inverter, it would sync to the panel with the lowest power output, thus causing the the entire array to run at the half set of panels that happen to be generating the least energy. Instead, we are going with the Enphase individual inverters, and thus i get the max from each panel.
That's the single biggest advantage of Enphase inverters, IMHO. Other options (in your case) would be to use two inverters, one for each set of panels on different facets.
Yep. Over on the wind-sun forum there is a guy with a 8 kW array running off a single inverter with the panels facing two directions and some morning/late day shading. He only gets the output of a 5 kW array, but if was using micro-inverters he'd probably boost output at least 20%. Here's the forum thread. Pretty informative and lots of pics of the install. He also has a lot of data monitoring of his home, too. Illustration of 8 KW (DC) GT Solar PV System Install for Residence in Urban Area - Solar Electric Power Discussion Forum by Northern Arizona Wind & Sun
Drees, The thread you note about the 8kw array is pretty informative. If you go back to ~post #3 (in his design phase) he is warned that his out put is going to suffer with his arrangement. He spent the next couple of dozen posts trying to defend his own design, in spite of so critique from a few world class experts! When the system finally got up and running, it performed exactly as poorly as the experts said it would! It sort of reinforces my axiom, (on the solar forum) "there are some really smart folks there, who have forgotten more about PV than most of us will every know,,, and one ignores their experience at one's peril. One can learn from that experience such that one doesn't have to reinvent the wheel with each installation" Glad to see your system up and running! Icarus
Once the costs are on par with big inverters, I see micro's as being the most popular way to go. Yet once you pick one inverter type or the other, you're stuck. Example - our $1,000 panels ... 18% efficient 225 watts each were so pricey, we decided to buy 4 panels less than we originally wanted (for an EV down the road). Now that the cash outlook is a bit better, we still have capacity/room on 1 inverter, for 900 watts more ... and the whole system is already set up and framed for them. I can't see how it'd be possible to use both standard & micro inverters (without becoming a rube goldberg mess) ... right?
Hill, I think the as arrays get bigger, the issue gets complicated. In most cases, most systems are designed to take as much advantage of the inverters installed. Your case is a pretty good one, you built your system to have some headroom to add a few panels as your budget allow(ed). One other advantage of Enphase (as I see it) is it allows (within certain limits) a system to grow as one can afford it. One panel/inverter this month another next month and so forth. (utility/inspection issues not withstanding!) If you did that with a conventional inverter you would start with an inverter running very inefficiently until you reached the PV capacity of the inverter. Where I think the Enphase begins to loose relative to conventional inverter is in a larger fully built system. For example, a quick search reveals the retail price of Enphase is ~$1.18/watt (190 watts per unit, assuming you mate perfectly to a 190 watt panel to get near 100% design efficiency). SMA 7kw inverter run (retail) $.52/watt. In the net net, it is probably easier to size the array to 100% to 7kw than to 190 watts. So there are some advantages of Enphase single inverters, but I don't think they are going to take over the market. If nothing else, in that same 7 kw array that the SMA would power, you would have ~36 Enphase inverters, and ergo 36 potential points of failure, over the life of the inverters, as opposed to a single point of failure with the SMA. Icarus
I think that once the discussion was over, he was fairly convinced that it would perform about what it is performing now. In the end - he can't do much about it because of the type of program he's in (he doesn't own the system). Thanks - 3.24 kW DC system has averaged 17.5 kWh/day over about almost a month of operation now and that includes a decent amount of cloudy weather. PVwatts estimates 445 kWh for my system in April (14.8 kWh/day), so I'm pretty happy that it's performing about 18% better than expected. PVwatts estimates about 520 kWh/month from May-July - we'll see if the 18% holds up through then and I generate 600+ kWh or not. I'm thinking that PVwatts is underestimating the output thanks to the microinverters, but I don't have enough experience with PVwatts and conventional systems to know for sure. Enphase claims 5-25% more output with microinverters thanks to the ability to avoid panel mismatch issues, but the upper end is very likely only true when there are partial shade issues. For my case, I only have a bit of shade late afternoon (5:30pm) when system output drops below 1000W and then partial shading only occurs for about 30 minutes so I'm sure the microinverter benefits are at the lower end of that range. There's no reason you can't mix/match central/micro inverters as long as the wiring isn't done yet. Either way you have to run another set of wires down from the roof to your main service panel. Even then, depending on how the wiring was done, you either may be able to use the existing wiring, perhaps with slight modifications. Typical central inverter setup needs 3 wires run from the roof - DC positive, DC negative, ground. Typical micro inverter setup needs 4 wires run from the roof - AC hot 1, AC hot 2, AC neutral, ground. Wire sizing requirements should be very similar, though with micro inverters you can run romex once under the roof - with central inverter you have to run individual wire in conduit all the way to the inverter.
