Central Inverter vs Micro Inverters

[Volusiano]

Second, several here stated that the enphase warranty doesn't cover labor. I don't know if they changed their policy for the M215/M250 modules, but enphase paid my dealer a labor allowance for each inverter replacement.

I heard that Enphase initially covered labor in their warranty but they then changed their policy and dropped labor warranty afterward. So what you said is consistent with what I heard and you were probably in the initial wave when they still used to cover labor warranty.

I'm not surprised that after the original failures they realized how expensive it was to cover labor in their warranty so they dropped it. If you read the new current warranty statement from Enphase, it's pretty clear that labor is no longer included in the warranty.

It's great to get to hear from a real owner about their real experience with Enphase. I sure hope they've really improved the M215/M250 modules a bit by now over the older modules like they claimed. But if they really have improved them, why yank away the labor warranty they used to have? That sure is not a very good vote of confidence on their part on their supposedly improved products.

[Mike90250]

I just stumbled across this thread, I'm a first time poster. I currently have a 3.5 year old enphase system ..........

Sorry to turn this thread into an enphase rant, but I just wanted to clear up a few misconceptions.

Thanks for the info and the gory details.

[ginahoy]

Thank you for the info. Lots to consider. Clearly and objectively presented.

Thanks. I'm in the early stages of developing a 54-home net zero community in SE Arizona. I registered for this site to learn about latest developments in PV.

Notwithstanding the (potentially) better performance of the newest enphase modules, I'm considering the SolarEdge approach. My lots are ideally oriented and have no shading issues, but I can't imagine not being able to monitor individual panels. With a central inverter, it would take a long time to even suspect a failed panel with all but the smallest systems.

Also, I would very much like be able to track cumulative panel power output (DC), which you can't do with enphase. The enlighten portal only reports cumulative output (AC kWh) per inverter, so there's no way to distinguish long term trending (degradation) of panels vs. inverters. Presumably, the SolarEdge monitoring portal tracks DC kwh per panel. Can someone confirm?

Much of this thread is about whether Optimizer modules fail in such a way that doesn't take out the panel as well. This is a big deal, and I don't think was ever resolved. Assuming the SE portal tracks per-module and central inverter power output, it would be simple to verify this from historical data. Customers who have had Optimizer failures would surely have reported their experience. Hmm... I can't seem to find a link to a forum on the SE website. Do they even have one??!

[thejq]

Thanks. I'm in the early stages of developing a 54-home net zero community in SE Arizona. I registered for this site to learn about latest developments in PV.

Notwithstanding the (potentially) better performance of the newest enphase modules, I'm considering the SolarEdge approach. My lots are ideally oriented and have no shading issues, but I can't imagine not being able to monitor individual panels. With a central inverter, it would take a long time to even suspect a failed panel with all but the smallest systems.

Also, I would very much like be able to track cumulative panel power output (DC), which you can't do with enphase. The enlighten portal only reports cumulative output (AC kWh) per inverter, so there's no way to distinguish long term trending (degradation) of panels vs. inverters. Presumably, the SolarEdge monitoring portal tracks DC kwh per panel. Can someone confirm?

Much of this thread is about whether Optimizer modules fail in such a way that doesn't take out the panel as well. This is a big deal, and I don't think was ever resolved. Assuming the SE portal tracks per-module and central inverter power output, it would be simple to verify this from historical data. Customers who have had Optimizer failures would surely have reported their experience. Hmm... I can't seem to find a link to a forum on the SE website. Do they even have one??!

Thanks for the post. It's very helpful, esp from an Enphase owner (people tend to defend their choices). I used to be set on getting the M250 based setup. But after another week of research and seeing your post, I too come to favor the SolarEdge based solution. It has a smaller installed base than the Enphase. Is that the reason I can't find any complaints/failures? or they're really that reliable? Moreover, I plan to install the highly rated LG 300N1C panel. The M250 is little under powered, while the SolarEdge 300 is perfect match. Please do update what you find out about the SolarEdge optimizer's reliability and the per panel monitoring capability.

[Volusiano]

...I can't imagine not being able to monitor individual panels. With a central inverter, it would take a long time to even suspect a failed panel with all but the smallest systems.

First of all, solar panels are in general pretty reliable, so I don't see why there's a strong need to monitor individual panels per se.

They either work or they don't up front. And if they don't work or fail early up front, either a microinverter or central inverter approach can easily alert the failure right away. A failing panel in a string will for sure raise a red flag for the whole string performance right away.

If the argument is the ease of pinpointing the failing panel with a microinverter, sure it may be not as easy to pinpoint a failing panel in a string until you get up on the roof and do some debugging. But with an early failure, the job of debugging belongs to the installer anyway, and they will find it out one way or another because they're the pro and they know what to do.

If the concern is the overall gradual degradation of the panels over the years, not single failures, then that can be deduced by looking at the system performance as a whole just as effectively as looking at individual panels.

Again, solar panels are expected to be pretty reliable because there's been a good track record of their performance already. So to me, it'd be overkill to use microinverters and expose your system to higher risks just to be able to monitor individual panel performance if you don't have a shading issue to begin with.

[ginahoy]

A failing panel in a string will for sure raise a red flag for the whole string performance right away.

I'm not sure I understand how. If you have, say, 25 panels and one goes out, your system output would drop by 4%. In my experience, day-to-day weather fluctuations make that level of difference impossible to detect. I suppose you could do some regression analysis and if you have enough cloud-free days within a short time frame, with similar temperatures, it would reveal a dead panel. But I just can't see spending that much time on a continual basis just to make sure all my panels are working. Or maybe I'm missing something?

[Mike90250]

I'm in the early stages of developing a 54-home net zero community in SE Arizona

1) PV panels are pretty dead simple, compared to a box of electronic gear. The chances that your "box of gear" (inverter or optimizer) will fail first, is way higher than the PV itself failing. If you have no shade issues, then you need no extra gear to fail. As was said earlier, take a fingerprint for a week of each install, and compare it monthly. If a panel fails, you will see a 20-40V drop in PV input to the inverter, easy to take action on. Also, sometimes panels start to show failures just before the failure with a dark or burnt spot on the PV. A good viewing with a pair of binoculars once a month could be informative.

2) develop a plan or method to wash bird poop and dust off the panels. With a large community install, some sort of shared system (bucket truck with wash nozzle ?) would be useful.

3) Don't forget user education. Like if there is a grid failure, the grid tie inverters shut down till the grid comes back. Maybe consider the SMA line of inverters that have a solar powered outlet on them for a fridge.

[ginahoy]

If a panel fails, you will see a 20-40V drop in PV input to the inverter

Thanks. I was only thinking about power, not voltage.

Still, panel-level monitoring has its advantages. Is there such an animal as a panel monitoring add-on without MPPT?

I totally agree that in a non-shaded single-orientation install, the disadvantages of panel-level MPPT (failure risk and associated costs) far outweigh the benefits, which as I understand it, would likely only be on the order of a few percent (top-to-bottom output variance among panels, and uneven degradation).

[bcroe]

I'm not sure I understand how. If you have, say, 25 panels and one goes out, your system output would drop by 4%. In my experience, day-to-day weather fluctuations make that level of difference impossible to detect. I suppose you could do some regression analysis and if you have enough cloud-free days within a short time frame, with similar temperatures, it would reveal a dead panel. But I just can't see spending that much time on a continual basis just to make sure all my panels are working. Or maybe I'm missing something?

Loss of a panel or inverter isn't an everyday event. I have been able to divide my system
into 2 identical halves. In good sun, the momentary performance of the halves is a close
match. If a problem is indicated, I can check DC and AC system currents with a clamp on
ammeter, tracking down the fault. This requires access to individual wires, not pairs, from
any power source.

Strings show up with lower voltage into the inverter. Once a fault is detected, it may be
necessary to take apart some connections to track down a panel. A non contact thermal
meter can also be useful. Bruce Roe

[russ]

The real advantage of panel level monitoring is that you can have a "I'll show your mine and you show me yours" party.

As has been pointed out, the task can be managed with string inverters.

If explained properly a customer will be perfectly happy with string inverters and a set up like what Bruce has.

[thejq]

The real advantage of panel level monitoring is that you can have a "I'll show your mine and you show me yours" party.

As has been pointed out, the task can be managed with string inverters.

If explained properly a customer will be perfectly happy with string inverters and a set up like what Bruce has.

It's all true, if only for technical reasons. But the cool factor of panel level monitoring and the ability to have web access to all the performance figures is just too hard to resist. For a small system like mine (4.8KW, 16 panels), it's $70 extra per panel to have the SolarEdge Optimizer and SolarEdge inverter is much cheaper than SMA. So the total costs are not that much different for me. Another major consideration is expandability. I just found out that where I live (San Diego/Carlsbad) as long as the inverter is the same, I don't need to get new permits to install a few extra panels. So for my 4.8KW, I'm going with SE6000A-US that's good for 7KW. So in a few years when everyone is driving a plugin/EV, I can add a few more panels of higher wattage and/or completely different brand without getting new permit ($500 today). I don't think you can do that even with micro-inverters, although technically it's not a problem.

[kwilcox]

So what's the answer for string/central inverter systems when counties adopt the 2014 NEC?

[silversaver]

It's all true, if only for technical reasons. But the cool factor of panel level monitoring and the ability to have web access to all the performance figures is just too hard to resist. For a small system like mine (4.8KW, 16 panels), it's $70 extra per panel to have the SolarEdge Optimizer and SolarEdge inverter is much cheaper than SMA. So the total costs are not that much different for me. Another major consideration is expandability. I just found out that where I live (San Diego/Carlsbad) as long as the inverter is the same, I don't need to get new permits to install a few extra panels. So for my 4.8KW, I'm going with SE6000A-US that's good for 7KW. So in a few years when everyone is driving a plugin/EV, I can add a few more panels of higher wattage and/or completely different brand without getting new permit ($500 today). I don't think you can do that even with micro-inverters, although technically it's not a problem.

Is your solar install yet? You are on the lowest output (80%) of 6kW inverter. I'm just wondering if that hurts the performace on initial output. Have you check into that? For example if your solar array is 4.8kW, often you choose inverter size from 3.8kW to 5.28kW. The inverter is actually 20% larger than your solar array. I'm just wondering if the installer say anything about that.

I think you can add up to 10% of original output or less than 1kW without re-apply for another permit or net metering contract. This is for Orange County, but not too sure about San Diego.

[thejq]

Is your solar install yet? You are on the lowest output (80%) of 6kW inverter. I'm just wondering if that hurts the performace on initial output. Have you check into that? For example if your solar array is 4.8kW, often you choose inverter size from 3.8kW to 5.28kW. The inverter is actually 20% larger than your solar array. I'm just wondering if the installer say anything about that.

I think you can add up to 10% of original output or less than 1kW without re-apply for another permit or net metering contract. This is for Orange County, but not too sure about San Diego.

Not yet. I'm about to sign the contract. He did mention that the invert is bit larger than needed, but didn't think it would affect the output that much, especially after I told him that I might want to add more panels later.