New 4.8KW installation in San Diego just finished

[J.P.M.]

FWIW, I've yet to learn any confirmed failure with SolarEdge optimizers due to defect.

Me too, and maybe (we all hope) robust design and good Q.C. will bear fruit. But, its early.

[J.P.M.]

I wonder what the fail rate is for these things are. With the warranty, you aren't taking much of a risk.

Failure rates are probably pretty/very low at this time, partly because they are somewhat new, and while becoming more common, not yet ubiquitous.

Read the warranty carefully, particularly with respect to who pays labor. With multiple failure points, the PROBABILITY of a failure increases. How many failures in the middle of an array that requires removal/reinstall of multiple panels are acceptable ? What does that removal/reinstall do to the integrity of the wiring/fixing/etc. of the affected panels and thus the rest of the array ? IMO, too many uncertainties and things to go wrong for the possibility and amount of potential increased output, but opinion vary.

[thejq]

I wonder what the fail rate is for these things are. With the warranty, you aren't taking much of a risk.

FWIW, I've yet to learn any confirmed failure with SolarEdge optimizers due to defect.

[thethingx]

String in a hot NY second. Simpler, easier to service, lower probability for problems making for, in my engineering judgment, a better, more reliable system, easier to maintain and spot problems. The extra monitoring capabilities and the potential for a slightly greater output are not worth the drawbacks/shortcoming to me.

Also, for those who like to see output and claim problems are easier to spot: True. But, I'd bet the novelty soon wears off and, given all the other variables that affect system output, a 5-10% drop in output from 1 bad micro will probably get noticed less than if/when a string inverter craps the bed and the monthly bill takes a big jump.

I wonder what the fail rate is for these things are. With the warranty, you aren't taking much of a risk.

[J.P.M.]

If you had to make a choice today, what would you go with?

String in a hot NY second. Simpler, easier to service, lower probability for problems making for, in my engineering judgment, a better, more reliable system, easier to maintain and spot problems. The extra monitoring capabilities and the potential for a slightly greater output are not worth the drawbacks/shortcoming to me.

Also, for those who like to see output and claim problems are easier to spot: True. But, I'd bet the novelty soon wears off and, given all the other variables that affect system output, a 5-10% drop in output from 1 bad micro will probably get noticed less than if/when a string inverter craps the bed and the monthly bill takes a big jump.

[thethingx]

I'd suggest the greater efficiency on a daily basis for micro equipped systems is to be expected as many, if not most arrays, micros and string either one, for the most part are subject to some form of shading over the course of a day. That's were the characteristics of micros/optimisers have an advantage. I'm pretty clear and shade free and still get ~ 1% - 4%/day output penalty in the late afternoon depending on the season. If micros/optimisers for a 327 had been available when I bought the system, there probably would be a slight advantage, reducing that loss by probably about half or so +/- some, but not eliminating it entirely.

Warranty labor questions aside, the disadvantages, which may have more long term implications, involve the greater # of failure points and system configuration to get at a problem in the middle of what are usually tightly packed arrays.

While not having looked at the data you cite, I'd suspect I'd see the advantage of increased output for micros to exist as you say, but that it may be relatively slight, with the importance of such additional output modified further by the tendency of folks to oversize arrays. To the extent an array is oversized, any extra output from micros is mostly moot, and we're back to the increased probability of problems due to the increased # of failure points as one consideration of many, maybe larger for some users than others.

The idea that micro equipped systems may, on average, be newer as a whole than string systems and therefore have less panel time dependent degradation looses may also be a contributory factor, but quite honestly, based on the relatively short amount of time most all systems have been operational, and also from what I measure around here, I'd suspect that's not as big factor as published panel degradation #'s would indicate.

Still, pay your money, take your choice. Just walk in with open eyes and mind, and know what the consequences of the choices.

If you had to make a choice today, what would you go with?

[J.P.M.]

That's good info, thanks for the reply

I'd suggest the greater efficiency on a daily basis for micro equipped systems is to be expected as many, if not most arrays, micros and string either one, for the most part are subject to some form of shading over the course of a day. That's were the characteristics of micros/optimisers have an advantage. I'm pretty clear and shade free and still get ~ 1% - 4%/day output penalty in the late afternoon depending on the season. If micros/optimisers for a 327 had been available when I bought the system, there probably would be a slight advantage, reducing that loss by probably about half or so +/- some, but not eliminating it entirely.

Warranty labor questions aside, the disadvantages, which may have more long term implications, involve the greater # of failure points and system configuration to get at a problem in the middle of what are usually tightly packed arrays.

While not having looked at the data you cite, I'd suspect I'd see the advantage of increased output for micros to exist as you say, but that it may be relatively slight, with the importance of such additional output modified further by the tendency of folks to oversize arrays. To the extent an array is oversized, any extra output from micros is mostly moot, and we're back to the increased probability of problems due to the increased # of failure points as one consideration of many, maybe larger for some users than others.

The idea that micro equipped systems may, on average, be newer as a whole than string systems and therefore have less panel time dependent degradation looses may also be a contributory factor, but quite honestly, based on the relatively short amount of time most all systems have been operational, and also from what I measure around here, I'd suspect that's not as big factor as published panel degradation #'s would indicate.

Still, pay your money, take your choice. Just walk in with open eyes and mind, and know what the consequences of the choices.

[thethingx]

A lot of times, it's a personal preference. For me, I really liked to have per-panel monitoring. But I don't like the idea of having the most venerable part of your system (the inverter) up on the roof constantly under extreme heat/cold/humidity/wind etc.. Also the horrendous reputation of m190 doesn't make me feel good about Enphase either, even though the newer generations of m215 and m250 seems to have vastly improved. SMA (or other reputable string inverters) is a solid choice and probably will give you the best overall value if you don't have shading issues. Another observation which might give some performance edge to panels that're individually tracked (eg. Enphase and SolarEdge) is that there're always some variation between panels, eg temperature difference, dirt/leaf/bird poop covering, degree of degradation etc. For example, I always find my panels on the outside performance better slightly (cooler temperature?), see below.
[ATTACH=CONFIG]6395[/ATTACH]

If you go to San Diego team's pvoutput.org page (http://pvoutput.org/listteam.jsp?tid=859), pick a sunny day and rank based on efficiency, Enphase and SolarEdge are consistently on the top.

That's good info, thanks for the reply

[J.P.M.]

LG has been mass producing panels for less than 5 years (citation, and a better citation). Their experience in the industry is less than frequent whipping-boy Enphase. Although they are a big name, and are clearly pushing the envelope with reasonably priced high performing panels, I would use some critical thinking in assessing how committed they, or any other company, is to this industry. Their warranty support, along with all others that I've looked at, if you read the fine print, is pretty much worthless after the first couple years. Whether any of that is worth a premium of $0.20-0.30 / W is far from certain.

As mentioned before on other posts, product warranties, like performance warranties are probably more useful to sellers as marketing tools than to users as protection.

[thejq]

What was your guys reasoning for going with SolarEdge inverters?

A lot of times, it's a personal preference. For me, I really liked to have per-panel monitoring. But I don't like the idea of having the most venerable part of your system (the inverter) up on the roof constantly under extreme heat/cold/humidity/wind etc.. Also the horrendous reputation of m190 doesn't make me feel good about Enphase either, even though the newer generations of m215 and m250 seems to have vastly improved. SMA (or other reputable string inverters) is a solid choice and probably will give you the best overall value if you don't have shading issues. Another observation which might give some performance edge to panels that're individually tracked (eg. Enphase and SolarEdge) is that there're always some variation between panels, eg temperature difference, dirt/leaf/bird poop covering, degree of degradation etc. For example, I always find my panels on the outside performance better slightly (cooler temperature?), see below.
Capture.JPG

If you go to San Diego team's pvoutput.org page (http://pvoutput.org/listteam.jsp?tid=859), pick a sunny day and rank based on efficiency, Enphase and SolarEdge are consistently on the top.

[sensij]

Yes! I think there is a lot more to be learned in discussion than there is from any one opinion.

[thejq]

LG has been mass producing panels for less than 5 years (citation, and a better citation). Their experience in the industry is less than frequent whipping-boy Enphase. Although they are a big name, and are clearly pushing the envelope with reasonably priced high performing panels, I would use some critical thinking in assessing how committed they, or any other company, is to this industry. Their warranty support, along with all others that I've looked at, if you read the fine print, is pretty much worthless after the first couple years. Whether any of that is worth a premium of $0.20-0.30 / W is far from certain.

All valid points. It's exactly these opposing but well thought of opinions that make this forum useful.

[thethingx]

What was your guys reasoning for going with SolarEdge inverters?

[thethingx]

Done. FYI, from talking to a couple of friends who just signed with the same installer, looks like the price/W hasn't changed much. So $3.5/W is still a good starting point for negotiation in the 4-6KW system range. If you're really charming, you might do even better. One got a free extended SolarEdge warranty. When I asked my installer (in dismay), I was told that she was better looking than me . The trick is to know what you want in terms of usage, system size and equipment, so they know you're serious and have the research.

Rofl, maybe I'll have my wife do the negotiation.

[sensij]

LG has been mass producing panels for less than 5 years (citation, and a better citation). Their experience in the industry is less than frequent whipping-boy Enphase. Although they are a big name, and are clearly pushing the envelope with reasonably priced high performing panels, I would use some critical thinking in assessing how committed they, or any other company, is to this industry. Their warranty support, along with all others that I've looked at, if you read the fine print, is pretty much worthless after the first couple years. Whether any of that is worth a premium of $0.20-0.30 / W is far from certain.