Need to decide on a system design 6.3 kW - San Marcos

[hockeydude]

So I have met with several vendors now and trying to make sense of everything. I have read quite a few post on this site but still don't have a feel for what would be best. I have a difficult roof because there is no one big surface to put it so it will need to be split between 3 or 4 sections. Three areas are pretty good and one would not be bad just not as good as the others. I have been told because of the layout of my roof, the install costs are going to be higher than someone with the ability to put all their panels in one array.

Currently:
Average Use: 913 kWh/month, 10,953 kWh/year
Have Led/CFL in most of house

The quotes below are for cash price:

Quote 1
I won't even post because it is so high.

Quote 2
6.3 kW
21 LG300N1C Panels
21 Enphase M250-60-240-S22 MicroIntverters
Envoy/Enlighten monitoring
Vendor estimate production: 10,690 kWh (I don't get this high number when I use PVWatts with default loss settings)
Cost: $22,635
$3.55/watt DC (before tax credit)

Quote 3 - Option 1
Vendor thinks there is only optimal prime roof space for 18 panels
5.49 kW
18 LG305NIC-B3 Panels
Solar Edge Se5000 Inverter / 18 Solar Edge P320 Power Optimizers
Solar Edge Monitoring
Vendor estimated production: 8,311 kWh
Cost: $22,532
$4.10/watt DC (before tax credit)

Quote 3 - Option 2
Vendor provided me a quote upon me asking for similar system as Quote 2 with 21 panels for comparison purposes
6.3 kW
21 LG300N1C Panels
Solar Edge SE6000 Inverter / 21 Solar Edge P320 Power Optimizers
Cost: $24,825
$3.94/watt DC (before tax credit)

Interestingly enough, this vendor said if they run the conduit on the outside along the roof the cost would go down over $2,000 and he said it is because the roof work is minimized not having to bring the four arrays into the attic. That is a lot of money just for aesthetics unless there is another reason to have them in the attic. It's a two story house so not that visible from the street anyway. Then the price would be closer to the $3.60/watt.

So of course the biggest question is how many panels I can fit that will work well. I guess I will have to have one of them do their final measurements to find out for sure. The it comes down to system design, the Solar Edge Inverter with Power Optimizers or the Enphase Microinverters. I asked about having the 305 panels with the miroinverters but he said there could be clipping but that may even be the case with the the 300 panels right?

The quotes are from vendor recommendations from others on this forum. So I am close to the $3.50 price target I was shooting for but my roof layout seems to be adding a bit more cost than those that can fit is all in one or two arrays.

Sorry for all the questions I have within this post but any help/suggestions are appreciated. Thanks.

[sensij]

How did quote 2 plan to run the conduit?

[thejq]

If you have decent orientation (S or SW), you will definitely get clipped with m250 + LG300. Since we are about 10 miles away, you can look at my outputs in a sunny day, and estimate how much clip you will experience. Note, according to pvwatt, my best production month is July which is still 2 months away. Also when it clips, the inverter is running at max rated capacity for an extended period of time. Common logic says it mustn't be good for longevity. If you're going with LG280, maybe m250 is ok, but I wouldn't do it with LG300/305, especially with a saving of just $0.05/W.

As for the conduit, if the only reason is aesthetics, as long as the wife approves (as well as HOA), I'd save that $2K. You can have the installer paint it to match your exterior color to lessen the impact. In a few months, you will probably not going to notice it.

[hockeydude]

Help understanding Clipping

If you have decent orientation (W or SW), you will definitely get clipped with m250 + LG300. Since we are about 10 miles away, you can look at my outputs in a sunny day, and estimate how much clip you will experience. Note, according to pvwatt, my best production month is July which is still 2 months away. Also when it clips, the inverter is running at max rated capacity for an extended period of time. Common logic says it mustn't be good for longevity. If you're going with LG280, maybe m250 is ok, but I wouldn't do it with LG300/305, especially with a saving of just $0.05/W.

As for the conduit, if the only reason is aesthetics, as long as the wife approves (as well as HOA), I'd save that $2K. You can have the installer paint it to match your exterior color to lessen the impact. In a few months, you will probably not going to notice it.

thejg - Just so I am clear on the clipping, I took a look at your output. On a sunny day earlier this week, it looks like your panels got up to 282 watts at the best part of the day. So with the m250, anything over 250 would just produce and show 250? Then in summer they likely will be producing even more so I would see a loss due to the inverters ability to handle it. I will have a good portion of my panels S or SW so I should have similar production with 300s as you since we live pretty close.

Thanks,

Brad

[thejq]

thejg - Just so I am clear on the clipping, I took a look at your output. On a sunny day earlier this week, it looks like your panels got up to 282 watts at the best part of the day. So with the m250, anything over 250 would just produce and show 250? Then in summer they likely will be producing even more so I would see a loss due to the inverters ability to handle it. I will have a good portion of my panels S or SW so I should have similar production with 300s as you since we live pretty close.

Thanks,

Brad

Yes, with m250 anything above 250W will be clipped. And yes, if your S and SW facing panels has around 20 degree tilt, you will see similar results as mine. Just a few days ago I saw this at peak, so it's almost getting maxed out, although just momentarily.
Capture.JPG

[gvl]

Solar Edge is 2 strings only, it may not apply in this case but depending on how they are oriented it is possible it may have difficulties optimizing 4 planes I would think.

Are ABB 300W micros any good?

[sensij]

thejg - Just so I am clear on the clipping, I took a look at your output. On a sunny day earlier this week, it looks like your panels got up to 282 watts at the best part of the day. So with the m250, anything over 250 would just produce and show 250? Then in summer they likely will be producing even more so I would see a loss due to the inverters ability to handle it. I will have a good portion of my panels S or SW so I should have similar production with 300s as you since we live pretty close.

Thanks,

Brad

No, you've midunderstood clipping, and thejq is misrepresenting it. The power is limited by the 250 W AC output, not DC input. It is true that in thejq's data, the power above 4 kW would clip if he had used 250 W microinverters. However, that clipping will fade as we get into summer and temperatures increase. The end result is probably only 1-2% loss for ideal systems, and less than that for most. As the panels foul and degrade over time, the clipping will disappear that way too.

If the enphase system is 0.05 / 3.55 = 1.4% less expensive, that pretty much covers the possible difference in energy generation. Your decision between Enphase and SolarEdge would be better informed by focusing on the technical differences between them. One thing I've come to appreciate as a SolarEdge owner is that their energy reporting is garbage, several percent off. Enphase's reporting checks out much better against calibrated sources (like your power company's meter). SolarEdge's power reporting is better, but still doesn't hold calibration throughout the day.

[sensij]

Solar Edge is 2 strings only, it may not apply in this case but depending on how they are oriented it is possible it may have difficulties optimizing 4 planes I would think.

Are ABB 300W micros any good?

The SolarEdge optimization occurs per panel. The number of roof planes doesn't matter at all.

[gvl]

The SolarEdge optimization occurs per panel. The number of roof planes doesn't matter at all.

If you have 2 planes on one string and one is in full shade, will the other work? I read there is a limit on the number of shaded panels, more than it you get no power from the string.

[thejq]

No, you've midunderstood clipping, and thejq is misrepresenting it. The power is limited by the 250 W AC output, not DC input.

Ok, small difference. SolarEdge claims 98% DC->AC efficiency, so technically at 250/0.98=255W DC clipping will happen.

It is true that in thejq's data, the power above 4 kW would clip if he had used 250 W microinverters.

Not completely true. Since panels vary, some panels will produce more and clip earlier, in my observation they can be 10-15Ws. But at above 4KW, everyone are clipped.

However, that clipping will fade as we get into summer and temperatures increase. The end result is probably only 1-2% loss for ideal systems, and less than that for most. As the panels foul and degrade over time, the clipping will disappear that way too.

Again, we differ in opinions. My rough calculation says more than 2%, but since I haven't had my system long enough, I will reserve my judgement until the summer is here. If TOU is factored in, the percentage in $$$ terms would be even higher.

Your decision between Enphase and SolarEdge would be better informed by focusing on the technical differences between them. One thing I've come to appreciate as a SolarEdge owner is that their energy reporting is garbage, several percent off. Enphase's reporting checks out much better against calibrated sources (like your power company's meter). SolarEdge's power reporting is better, but still doesn't hold calibration throughout the day.

In almost all engineering disciplines, clipping means really BAD (eg. harmonic distortion in audio processing, non-linearity in power amplifier, loss of orthogonality in digital communications like LTE etc). In this context, it means the inverter is stretched to the limit, any more would have catastrophic consequences. Clipping is a built in protection mechanism. Normally I'm used to designs that operate with some margins from clipping just to be safe. To me extended clipping alone is enough reason to move away from m250, if not for the lost production, but for the integrity and longevity of your system.

[sensij]

Solar inverters don't clip in the way that you are thinking. They have control over the power generated by the panels. If the AC power approaches the limit, the operating voltage of the array is adjusted away from the maximum power point, making the panels less efficient generators. three is no excess power that needs to be handled as there is in those other systems.

really, before the wild speculation on inverter failure modes, you should improve your understanding of how they work.

[hockeydude]

How did quote 2 plan to run the conduit?

Heard back from vendor 1. Conduit will run through the attic. In a discussion with him about micro inverters vs string with power optimizes, he indicated that both are newer technology and he actually prefers the simplest approach with just a string inverter. He said SMA is a proven inverter that has been in business for 30 years and that Enphase and Solar Edge are not profitable companies at this time.

So now he has confused me even more on what would be best for my situation. Regardless of 1% or 2% efficiency differences, I want reliability the most since I will own it and have to deal with a problem should one arise and the companies don't exist anymore years from now.

[thejq]

Solar inverters don't clip in the way that you are thinking. They have control over the power generated by the panels. If the AC power approaches the limit, the operating voltage of the array is adjusted away from the maximum power point, making the panels less efficient generators. three is no excess power that needs to be handled as there is in those other systems.

really, before the wild speculation on inverter failure modes, you should improve your understanding of how they work.

So you agree with me that the reason for clipping is because the system is approaching the limit. Raising the voltage is a way to maintain the equilibrium so the system doesn't go over the edge (too much current). The clipping examples were analogies to illustrate clipping is almost always bad no matter what modes you're talking. It don't think it takes a genius to figure out which is a better system -- one that produces less and constantly at the hairy edge of its limit, and one that produces more and always have margin to spare. So remind me what part of my reasoning is "wild speculation"?

[gvl]

One benefit of micros if one goes out the rest will still work, even if Enphase goes belly up you should be able to find an equivalent replacement without a rush.

[thejq]

One benefit of micros if one goes out the rest will still work, even if Enphase goes belly up you should be able to find an equivalent replacement without a rush.

Yes, to me that is actually the biggest advantage of micro inverters.